Jump to content


Photo

An alternative 'championship' (1932-2005)


  • Please log in to reply
91 replies to this topic

#1 jpm2

jpm2
  • Member

  • 121 posts
  • Joined: December 04

Posted 19 October 2005 - 09:14

As some of you will know, this new thread was born elsewhere, on Louis Chiron's thread, to be precise.
And as you also know, I promised to explain my point system and how I compared and ranked Grand Prix drivers results and performances.
That's what I'm here to do.
Allow me, first, a lenghty reply to Doug Nye’s message in Chiron’s thread. Doug surely deserves an answer. Not on account of his somewhat paternalistic and arrogant style (“borderline idiotic”, “palpable nonsense”, are just some of the niceties that seems to go along with it), but because, behind that style, there are one or two ideas worth discussing.

Here’s what Doug wrote:

What an entertaining and challenging - if entirely inconsequential - thread this has been. jpm2's hobby produces results - repeat results, not the activity of arriving at them - which I certainly regard from what I have read here as being as borderline idiotic as some of the 'fanzine' inanities one finds (but seldom bothers to read) in 'Racing Comments'.

Everybody, thankfully, is entirely free to spend his or her time as they see fit, and I certainly applaud your industry jpm2, and I'm really happy to see your line reading "MY SYSTEM IS ONLY A WAY TO COMPARE RESULTS AND PERFORMANCES. It is something I devised, as you say. Some may find it interesting, others may find it boring. No big deal."

"MY SYSTEM IS ONLY A WAY TO COMPARE RESULTS AND PERFORMANCES" indeed? Herein surely lies the fatal flaw.

A valid method of comparing results and performances for sportsmen of different periods would be a minefield even in such relatively uncomplicated individual activities as sprinting, distance running, shot, javelin, swimming, boxing...

When applied to a sport in which the mere human being is utterly reliant upon the reliable performance of the machine which is carrying him to war, it is palpable nonsense unless major comparative factors are built in to include the car as a driver/car equation.

Even then mere statistical comparison will be nothing more than of passing interest, as you say "no big deal".

What counts - and what has always counted - between drivers is a human quality which is numerically unquantifiable - flesh and blood, grit, guts, talent, personality, perceived stature amongst his peers, and skill.

And these qualities - thank God - cannot in truth ever be captured and convincingly deep-frozen by mere numbers.



Doug is happy to see me underline that "MY SYSTEM IS ONLY A WAY TO COMPARE RESULTS AND PERFORMANCES”.
Why the admiration, Doug? I have always said it from day one, frequently correcting those who said differently, in order to avoid misunderstandings.
The troubling fact remains that, for some reason, several TNF members (Doug’s included) refused to read the lines and when I wrote results/performances they started talking of things like “talent”, “personality”, and so on.

The fact is that things like human qualities do matter but they are the object of a different approach, or of a different perspective. I've said it myself in TNF in April when we were discussing Damon vs. Graham. So, Doug, you are preaching to a convert.

Anyway, Doug takes a further step forward, and he says that I cannot compare drivers’ results and performances with a system of points.

And why not?
Because a “valid method of comparing results and performances… would be a minefield” particularly “when applied to a sport in which the mere human being is utterly reliant upon the reliable performance of the machine which is carrying him to war”. As Doug says “what counts - and what has always counted - between drivers is a human quality which is numerically unquantifiable - flesh and blood, grit, guts, talent, personality, perceived stature amongst his peers, and skill”.

Well, Doug, I think you are confusing things. Let me pick the idea of war that you brought into this discussion. War, like everything else (surely more than everything else) is also the terrain of “flesh and blood, grit, guts, talent, personality, perceived stature amongst his peers, and skill”.
But isn’t it possible to judge and compare the results and performances of armies (or individual soldiers) at war? Aren’t they in a way comparable by numbers? I guess that’s a part of what strategists and military historians do.
Here too we have man confronting each other, “on machines that carry them into battle”, men whose performance in combat, and its final result, will depend on the quality of their grenades or their spears”. But the performance translated into numbers still comes creeping in. A German flying ace that killed 100 enemies in WW2 is highly ranked then, say, an average French pilot that shot down only 3. A lot of things enter the equation: experience, chance, the theatre of war, and, of course, the quality of machinery…. Perhaps the Messerschmitt was a better “war horse” than the French fighter planes.

Yes, maybe... but who cared? The NUMBER of victories was highly praised, as we all know, and that’s why they usually translated in small signs painted on the fuselage of war airplanes.

And, of course, there are the statistics of war. How many killed, how many wounded, how many rewarded for their performance, and so on.

Is this number or ranking game “what counts”, to use Doug’s wording? There is no such thing, when you want to look at events in an historical perspective. It all depends of our perspective and personal inclinations. Just imagine I wanted to study the problem of slavery. I could focus on the Slave Trade and its statistics (how many went across the ocean, how many died, how much they cost, the profits of the trade); or I could concentrate in human values and qualities, the emotions and ideas that led to abolition; or I could…
One way of looking at Slavery in no way excludes all the other. There is no philosophical hierarchy that says this way is better than the other. There simply is no recipe to understand the past and to give account of it.

Only God has an overall and total view of events.
And that is why Doug is wrong when he says that “what counts - and what has always counted - between drivers is a human quality which is numerically unquantifiable - flesh and blood, grit, guts, talent, personality, perceived stature amongst his peers, and skill”. That’s a part of the truth. He wouldn’t be wrong if he had said “what counts FOR ME is…”. But, Doug is not God to impose upon us, simple mortals, the criteria of what’s important and what’s not.

And now let's go on "my system" of points, shall we?

Advertisement

#2 ian senior

ian senior
  • Member

  • 2,165 posts
  • Joined: September 02

Posted 19 October 2005 - 09:27

Originally posted by jpm2
)
And now let's go on "my system" of points, shall we?


No.

#3 David Shaw

David Shaw
  • Member

  • 1,734 posts
  • Joined: August 02

Posted 19 October 2005 - 09:35

Well I am interested to see what you come up with.

#4 Kpy

Kpy
  • Member

  • 1,259 posts
  • Joined: February 01

Posted 19 October 2005 - 09:36

Originally posted by ian senior


No.

I'll second that.

#5 jpm2

jpm2
  • Member

  • 121 posts
  • Joined: December 04

Posted 19 October 2005 - 09:40

As promissed before, here’s an explanation of my point system to compare Grand Prix drivers performances and esults. I hope it will be sufficiently clear to avoid misunderstandings.


1) The underlying concept:

Many attempts have been made in the course of time to compare performances and results of Grand Prix (to make things easier, let’s call it this way) drivers. Yet, the comparison has been generally made in a somewhat straighforward way, either by adding the points obtained during the racing seasons, or by the number of victories or pole-positions, rarely taking into account the diversity of the periods considered. This, greatly distorts the intrinsic value of statistics and of the comparison exercise.
Grand Prix racing complys with rules that vary in the course of time. The same can be said of the points systems of championships; penalties or even the disqualification, have been subject to different judgements; the races themselves, as well as their settings, have changed over the years. In 1930s-1960s, as we all know, a Grand Prix was a lot different from a nowadays one. Circuits were usually longer, races were longer too, practice times didn’t always correspond to grid positions (although that habit disapeared in the early 1950s), different drivers could share the same car, the number of major races was lower, but, and above all, there were lots of important races that did not count to the championship. Some years there was even no championship at all.
Therefore, to compare the results and performances of different epochs in a balanced and fair way, there is the need to create a point system to unify the evaluation criteria, enabling the drivers from different periods to be on a RELATIVELY similar footing. This would lower the handicap of drivers that raced in different times. To cut it short, there is the need to adopt a point system that may lessen the differences of the various epochs and that may be applied without MAJOR distortions to all of them.
Besides trying to establish a relatively uniform criterium that does not penalize greatly one era in relation to others, this system should go beyond the simple attribution of points to the best classified in each race, as it´s usually done in official championships. There is much more in a Grand Prix (or any race, for that matter) than final classifications, and, in my view, other aspects should be valued. I have therefore decided to value the performances in practice and during the race itself. My aim is that the points system will reflect in a more representative way what actually went on during the race, and that often is not reflected in the final result. I would also like to reward the performance, toning down the influence of the good or bad luck factor. A driver that was fastest in practice, that led the race from the start and that went out of gas in the last lap, ending in 10th place, will have no points in official championships; in my system he will be attributed the points referring to the pole-position (or best practice time) and to his leadership during the race.


2) The method: how it works

I only consider races with a minimum competitive level, not because those races were a part (or not) of the official championship, but because those who raced were relevant, i. e., they were a more or less a significative part of “the cream” of that year.
To determine the competitive level of a race, I start by drawing a standard list of the participants each year. From that list comes the 25 best car/driver that participated in the races that year, and I attribute a value or rate (between 6 and 1) to each of them, depending on the quality of the car and the one who drives it. As the driver of each car may change in the course of the year, this may (or may not) affect the rate attributed to the said car.
A race is considered valid for “my championship” whenever it has at least 10 car/drivers of the established ranking, and the total sum of their rates cannot be less than a certain level. According to the value of the participants, the race may be minimum, medium or high level. This means that certain races that were not part of the official championship are considered and may even have na higher level than some of the official championship. A few races of the official championship were even disregarded, as they did not meet the minimum competitive criteria. That’s what happened with the Indy 500 and with the 1960 Italian Grand Prix, in which almost none of the cars and drivers of the ranking participated.
So I give points to drivers who make the best practice times (who are not necessarily the drivers in the front rows of the starting grid. Naturally, the points awarded varies according to the level (minimum, medium, or high) of the race. I also grant points (that also varies according with the level of the race) to the drivers that the best positions during the race. The determination of those positions is not made on a lap by lap basis. As the various circuits are quite different, the perimeter of each lap being very different, I consider the positions every 50 kms (aprox.). Finally, I grant points to the best classified at the end, and on the track. Those points also differ according to the race level.
I only consider races that had a minimum of 125 kms. If for some reason, the organizers had to stop a race, and declare it finished, before that distance was run, I only consider, when awrading points, the practice times and positions held at 50 kms and, if it’s the case, at 100 kms, but not the final result.
The final classification of the championship is obtained by adding up all the points. These points, added year by year, will amount to the total obtained by each driver during his Grand Prix racing career.
But, the main goal of “my championship” is a comparative one. If one considers only the adding of points gained, modern drivers would be favoured when compared with drivers of the 1930s-60s, because these participated in fewer and sometimes lower level races.
So, in order to balance the comparison of performances and results, I consider each season to be equivalent to 100. Drivers which gained points in that year, wiil have a percentage of that number, and the overall classification of drivers from 1932 until now will be best determined by the adding of those percentages.
To conclude, the comparison of drivers performances and results can be achieve in various ways. Those are the 5 ways I propose: victories, best practice times, and the adding of points (those 3 ways greatly favouring more recent drivers); plus the ratio points/race and the adding of percentages. When considered as a whole these 5 criteria may give a fair account of Grand Prix top drivers performances and results throughout time. At least, hope it gives a fairer perspective than the official statistics.

Now, getting to the real things, are some examples of final standings:

Points/race (the top 30)

1- Juan Fangio (25,620)
2- Michael Schumacher (21,931)
3- Ayrton Senna (19,895)
4- Rudi Caracciola (19,275)
5- Alberto Ascari (18,877)
6- Jim Clark (18,717)
7- Bernd Rosemeyer (18,185)
8- Alain Prost (17,225)
9- Jackie Stewart (16,478)
10- Hermann Lang (14,517)
11- Stirling Moss (14,407)
12- Damon Hill (13,123)
13- Froilan Gonzalez (12,658)
14- Tazio Nuvolari (12,119)
15- Pablo Montoya (11,812)
16- Nigel Mansell (11,807)
17- Achille Varzi (11,720)
18- Luigi Fagioli (10,480)
19- Niki Lauda (10,461)
20- Hans Stuck (10,429)
21- Jean Wimille (10,421)
22- Mika Hakkinen (10,406)
23- Mike Hawthorn (10,179)
24- Kimi Raikkonen (10,011)
25- Nino Farina (9,990)
26- Nelson Piquet (9,967)
27- James Hunt (9,786)
28- David Coulthard (9,742)
29- Gigi Villoresi (8,821)
30- Jack Brabham (8,669)



Added percentages (the top 50)

1- Michael Schumacher (310,786)
2- Alain Prost (236,359)
3- Ayrton Senna (214,317)
4- Juan Fangio (204,063)
5- Jackie Stewart (155,225)
6- Nigel Mansell (151,009)
7- Jim Clark (148,722)
8- Nelson Piquet (143,035)
9- Jack Brabham (137,695)
10- Alberto Ascari (137,553)
11- Stirling Moss (135,522)
12- Rudi Caracciola (135,164)
13- Graham Hill (134,306)
14- Niki Lauda (127,919)
15- David Coulthard (122,693)
16- Tazio Nuvolari (114,181)
17- Nino Farina (112,500)
18- Mika Hakkinen (110,911)
19- Damon Hill (103,443)
20- Rubens Barrichello (98,935)
21- Gerhard Berger (97,154)
22- Gigi Villoresi (94,099)
23- Achille Varzi (82,872)
24- Riccardo Patrese (80,198)
25- Carlos Reutemann (77,568)
26- John Surtees (76,764)
27- Hermann Lang (72,403)
28- Denny Hulme (71,608)
29- Emerson Fittipaldi (70,034)
30- Jean Wimille (69,774)
31- Dan Gurney (67,812)
32- Jacky Ickx (67,629)
33- Clay Regazzoni (66,170)
34- Jody Scheckter (63,432)
35- Ronnie Peterson (63,430)
36- James Hunt (62,616)
37- Luigi Fagioli (61,709)
38- Pablo Montoya (61,074)
39- Alan Jones (60,432)
40- René Arnoux (60,295)
41- Bernd Rosemeyer (59,419)
42- Mike Hawthorn (57,184)
43- Louis Chiron (56,297)
44- Mario Andretti (54,874)
45- Hans Stuck (54,017)
46- Ralf Schumacher (53,381)
47- Kimi Raikkonen (52,003)
48- Jean Alesi (51,834)
49- Froilan Gonzalez (51,178)
50- Jacques Villeneuve (50,679)


The winners of « my championship » (from 1950 onwards I refer only to the cases that don’t match the official WDC)

1932- Tazio Nuvolari
1933- Tazio Nuvolari
1934- Hans Stuck
1935- Rudi Caracciola
1936- Bernd Rosemeyer
1937- Rudi Caracciola
1938- Hermann Lang
1939- Hermann Lang
------
1946- Raymond Sommer
1947- Gigi Villoresi
1948- Jean Wimille
1949- Alberto Ascari
1950- Juan Fangio
------
1964- Jim Clark
-----
1967- Jim Clark
------
1970- Jackie Stewart
------
1973- Ronnie Peterson
1974- Niki Lauda
-------
1977- James Hunt
-------
1979- Gilles Villeneuve
-------
1982- Alain Prost
-------
1984- Alain Prost
-------
1986- Nelson Piquet
1987- Nigel Mansell
-------
1989- Ayrton Senna



Just a final word to tell you that “the system” was created in the early 70s. I kept it as na hobby and perfectioned it whenever time allowed. Although it’s a creation of my youth, the basic principles still seem correct to me. But you’ll be the judge of that. I think these and other numbers, and, of course, the whole method, as a tool of analysis, arise interesting problems. The system still faces several difficulties and conundrums, that I hope to discuss when (and if) discussion arises. But, please don’t expect na immediate eply. This topic has been time consuming, and my work may keep me away for a few days.

Regards

#6 Doug Nye

Doug Nye
  • Member

  • 11,524 posts
  • Joined: February 02

Posted 19 October 2005 - 10:03

Originally posted by jpm2
The system still faces several difficulties and conundrums...


So true... :confused:

DCN

#7 Catalina Park

Catalina Park
  • Member

  • 6,768 posts
  • Joined: July 01

Posted 19 October 2005 - 10:10

I measure drivers by height. ;)
Dan Gurney is looking good.
Sorry Art. :

#8 roger_valentine

roger_valentine
  • Member

  • 208 posts
  • Joined: October 02

Posted 19 October 2005 - 12:28

You pre-determine who the "important drivers" are, and then define the "important races" as those which had a fair number of important drivers in them.

So, in 1952 Ascari missed an "important race" because he was off in some backwater somewhere competing against a bunch of non-entities like Troy Ruttman and Jim Rathmann.

I feel that Doug's words "fundamental flaw" may reappear quite frequently in responses to this thread.

#9 KJJ

KJJ
  • Member

  • 702 posts
  • Joined: February 02

Posted 19 October 2005 - 12:31

Didn't one of the mags devise something similar a few years ago, post 1950 of course.

I wonder why the system only starts in 1932, clearly this is going to affect the scores of the Nuvolaris and Chirons? Why include a score for practice times, this was surely not that important in the past as in more recent years? Why include a score for race positions at various stages in the race, should a driver be rewarded for breaking the car?

Regarding Doug's car/driver equation, should a driver who has an outstanding car in a particular season not have his score for that year reduced in some way? I would also penalise a driver who is second best in a team and limit the number of qualifying seasons that a driver can score, best five seasons only perhaps.

Interesting game though.

#10 Wolf

Wolf
  • Member

  • 7,883 posts
  • Joined: June 00

Posted 19 October 2005 - 13:48

I have actually contemplated very similar idea but I have neither resources nor persistance to follow it up, so I gave up on the idea...

Your way of 'evaluating races' seems to be more or less OK*, but I'd reconsider the notion that all seasons are alike (I understand each season is 100%)- some sort of 'weighing system' would IMHO be in order, similar to the one for individual races.

* Roger's remark on Indy being in 'backwater' stands, and I'd dearly like to see more comprehensive list, but the sad fact is that EU and US based racing indeed seem to be worlds apart, and there's too little exchange between them to make valid comparisons and- unfortunately, hence make Indy relatively insignificant when considering GP racing and drivers (and certainly much more trouble than could be reasonable)... :

One particular thing I miss in this way of scoring is incorporated in Williams' stats (I'm fairly certain I got the similar idea independent of him)- is considering the cause of retirement. He has more stats based system, and considers e.g. wins/race but more important criteria is wins per race in which car did not fail. This 'principle' rewards people who didn't wreck themselves or/and others (*looks where MS, AS and AP are on Your table*;)). This is a nice idea but would be hard to implement in Your system, except by multiplying race score with say 0.9 in case of mechanical DNF and 0.1 in case of accident, spin, driver error...

This is few ramblings from me for now, but I'm certain I'll have another look into Your system and report my opinions.

#11 ensign14

ensign14
  • Member

  • 61,756 posts
  • Joined: December 01

Posted 19 October 2005 - 14:33

I think it demonstrates the difficulty and impossibility of scoring systems. The recounting above does not mention Moss, for a start. But in any system you think of how can you equivalence a win to, say, two 4ths?

I have elsewhere advcated not having a World Championship at all. After all, there is no season long WC in, say, tennis (there is a number 1 ranking but it's less important than the Grand Slam events) or athletics. The individual events stand out more as a result.

A World Championship point is a fiction. You can't hold a point, it has no taste, you can't show one to me, it's purely arbitrary whether you get one or three for a 6th place, or even none at all. A fighting 7th from the back of the grid is worse than a flukey 6th cos you qualified just before the rainstorm. Coming last out of 8 finishers gets you something whereas 9th out of 20 is worthless. And the World Championship is decided on who has the most of these fictions.

It's probably fair enough to say that the best driver in the world* in the early 50s was JMF, late 50s/early 60s SCM, middle 60s Jim Clark, late 60s/early 70s JYS, then a mixture of Lauda/Hunt/Villeneuve, Prost in the mid-80s, Senna in the late 80s/early 90s and Schumi until about 2004. Now Alonso/Raikkonen according to taste.

So why do we need a World Championship? We all "know" who the best driver is [at least to one's own personal preference]. A paper crown does not add much to reputations, only to numbers. If Raikkonen is the best in the world, the fact that he did not win in 2003 or 2005 is neither here nor there. It's almost as if the WC is merely a device to get everyone to turn up.

Perhaps the most interesting period for "best driver" is that starting in about 1930. Looking at the reputations of Nuvolari, Caracciola, Rosemeyer and Varzi, you can barely put a fag paper between them. Maybe Chiron as well :p. And Williams had a few wins, Benoist was just past his peak...


* Eurocentric world, obviously...

#12 D-Type

D-Type
  • Member

  • 9,699 posts
  • Joined: February 03

Posted 19 October 2005 - 22:22

In earlier times, a driver's season comprised more than just the World drivers' championship so how about giving races a weighting. Without thinking too deeply, something like

Grandes Epreuves, or later on World Championship races - 5
World Sports Car Championship, lesser grands prix at least 75% of the average Grande Epreuve- 4
Races for grand prix cars of less than 75% of the average Grande Epreuve (and F2 cars 1952-53) - 3
International Formula 2 races, prewar vouturette races, Formula Junior/F3 1961-63, Formula 3000 - 2 (but 1952-53 score as grand prix cars)
European Mountain Championship - 2 (prewar many aces drove in it)
Other International races (sports car, Formula 3, Formula Junior, GT) - 1

Obviously this would have to be applied with care, but it allows recognition of the likes of Stirling Moss winning the World Sports Car Championship for Mercedes in 1955 in between grands prix and Fangio and Gonzalez turning out for BRM in 1952. It also allows for the many different races of varying importance that drivers raced in earlier eras.

#13 Wolf

Wolf
  • Member

  • 7,883 posts
  • Joined: June 00

Posted 19 October 2005 - 23:12

Duncan, I believe jpm2's system only includes GPs (including sportscar races and FL events, at least, would be nice too, but I understand that it would mean a lot of work), that qualify with his criteria (over 125km IIRC and with reasonable competitive level)...

#14 simonlewisbooks

simonlewisbooks
  • Member

  • 2,118 posts
  • Joined: January 02

Posted 20 October 2005 - 10:36

The problem of awarding points for qualifying is that it fails to take in things such as..

- Team polotics (Peterson alledgedly being sent out on half tanks while Mario was running nearly empty in 1978 for example)

- rain or oil affecting those who had yet to run

- those using practice as a set up session for the race in the full knowledge that they could ensure a better race pace that way

- Tyre companies not supplying qualifying rubber to every team (as happened in the 70s)

- Tyre wars in general

- Drivers wrecking a car after a good qualifying lap and having to take the back up car and a back up time, as happened in the 60s and early 70s when the actual chassis you used was the one that earned it's own grid slot

It also penalises someone like, for example, Watson, who more than once had a low-grid start and passed everyone to win. I felt that made his performance all the greater but the system here would have the opposite effect and penalise an epic drive. Schumacher, unopposed, from Pole in a superior car scores much higher although he was merely cruising by comparison.

For points at specified distances one can again use the Watson example(or that of this years Japanese GP) as a climb through the field would mean missing out on points for leading at specified laps, yet what the race is all about is being in the lead as you cross the line at the finish , which is a strategy in itself (Gethin at Monza in 1971 for further example).

But then whatever point system anyone devises there are probably more unfair aspects than fair ones so I'm not picking on this system alone. But applying a new system in retrospect means discounting the premise under which everyone has previously raced (ie. pure and simple - first past the flag). If only they had known - would they have driven differently?

Simon Lewis

#15 jpm2

jpm2
  • Member

  • 121 posts
  • Joined: December 04

Posted 20 October 2005 - 11:29

Interesting questions and comments I will try to answer them, following the order, of course. :)

Roger_Valentine says:
You pre-determine who the "important drivers" are, and then define the "important races" as those which had a fair number of important drivers in them.
So, in 1952 Ascari missed an "important race" because he was off in some backwater somewhere competing against a bunch of non-entities like Troy Ruttman and Jim Rathmann.


It is true, although your example of Ascari in 1952 is not. Ascari knew he was missing an “important” race not because I decided it was “important” but because it was already a WC race (I suppose you’re talking of the Swiss GP). Still, he chose to race at Indy, and the same thing happened with other drivers (Clark and Gurney, in 1965, for example).

But let’s go back to the main issue. It is true that my choice of “important car/driver” is subjective, although, I believe it to be an informed one.
How else would we determine the “intrinsic value” of an entry list (or starting grid), particularly in the 1930s-1960s when entrants lists varied so much?
The fact is that in WDC that “intrinsic value” WAS PREDETERMINED in advance, regardless of who showed up. And so, we faced things like 1960 Monza…

My system has tried to avoid that. And, in order to lessen the risk of letting my personal feelings interfere too much, the ranking is a tight one, i.e., four drivers are rated 6, other four rated 5… and the last 5 drivers are rated 1. The grand total amounts to 85. Between 85 and 70 a race will be high level; from 69 down to 54, medium level; from 53 to 38, minimum level. Below 38 it’s not eligible. Let me show you how it’s done. Let’s pick, for example, the 1964 ranking:

Lotus: Jim Clark (6)
Pete Arundell (5) or Mike Spence (4)
Gerhard Mitter (2) or Walt Hansgen (2) or Moisés Solana (2)

Brabham: Dan Gurney (6)
Jack Brabham (5)

Ferrari: John Surtees (6)
Lorenzo Bandini (5)
Ludovico Scarfiotti (3) or Pedro Rodriguez (3)

Brm: Graham Hill (6)
Richie Ginther (4)
Dick Attwood (2)

Cooper: Bruce McLaren (5)
Phil Hill (4) or John Love (2)

J. Siffert: Jo Siffert (4)

Rob Walker: Jo Bonnier (4)
Eddy Barth (1) or Jochen Rindt (1) or Geki Russo (1) or Hap Sharp (1)

Brp: Innes Ireland (3)
Trevor Taylor (2)

D.W. Racing Enterprises: Bob Anderson (3)

Reg Parnell: Chris Amon (3)
Mike Hailwood (2)

Scuderia Centro-Sud: Tony Maggs (2) or Phil Hill (3)
Giancarlo Baghetti (1)

P. Revson: Peter Revson (1)

Honda: Ronnie Bucknum (1)

M. Trintignant: Maurice Trintignant (1)


And the 1964 races I considered (and their level) were the following:

Daily Mirror Trophy (minimum)
News of the World Trophy (minimum)
200 Miles Aintree (medium)
International Trophy (medium)
Monaco GP (high)
Dutch GP (medium)
Belgian GP (high)
French GP (high)
British GP (high)
Solitude GP (medium)
German GP (high)
Austrian GP (high)
Italian GP (high)
United States GP (high)
Mexican GP (high)


Now let me say something about KJJ comments:

I wonder why the system only starts in 1932, clearly this is going to affect the scores of the Nuvolaris and Chirons? Why include a score for practice times, this was surely not that important in the past as in more recent years? Why include a score for race positions at various stages in the race, should a driver be rewarded for breaking the car?
Regarding Doug's car/driver equation, should a driver who has an outstanding car in a particular season not have his score for that year reduced in some way? I would also penalise a driver who is second best in a team and limit the number of qualifying seasons that a driver can score, best five seasons only perhaps.


My Championship starts in 1932 because races changed a lot by then. It was the start of single-seaters period, of shorter races where a driver could do the full length of the Grand Prix, and so on. It would have been too complex to consider the previous years, though it is possible to do it, of course.

And why include a score for practice times, you ask? You are absolutely right when you say that grid positions were not that important in the past. There were no points involved and drivers still overtook each other regularly on the track (not in the pits), I might add. And why include, you ask, a score for race positions at various stages?

Well, consider these three ideas:

- This is a system designed to be a balanced one, not to favor the past (nor the opposite).
- Although, in the past, the grid positions had not the critical importance they now have, the curious thing is that, when you analyze practice times of, say, the 1950s or the 1960s, you still find, as a rule, the best men in front.
- The same can be said of the actual race. When looking at a lap by lap chart, you usually find the top guns in front.

Why is that? Two reasons, I think. First, because of the psychology of racing drivers, who are generally intrinsic competitive persons, keen on taking risks, facing challenges and defying their limits. The glory of being faster or the glory of leading a race were (I’m not sure if they still are) powerful incentives. And, secondly, because my system is based on the concept of winning, i.e., it is designed to reward those who won, or gained the top places. That is why (as I will explain later) only the 3 top positions are taken in consideration. The system awards no points to those who held places below 3rd. Once again, if you look at the final results you can generally see that the winner of the race, were most often in the front line of the grid (whenever best practice times determined grid positions, of course) and were already in the leading positions in the early stages of the race.
In a certain way we can say that frontline grid positions (or best practice times, to be more precise) and leading positions during the race, were “victory in the making”, and, for that reason, I have included in my system a score for front grid and race positions at various stages.

Your idea of considering only the best 5 seasons is interesting. If I understand correctly, its purpose is to prevent the “system” from favoring a lengthy but only average career of a “not so good” driver. If it is so, I think the problem is somewhat solved in my system because, as I’ve just said, I’m focusing only on the “cream”, so to speak, and there’s little risk that a just regular driver with a long career and medium/low results and performances will “overtake” the scoring of a very good driver with a much shorter career.
I will develop this point a little further in my answer to Ensign14

And now, what about “outstanding cars”. Good idea but there are several objections to it. Who decides (and on what basis) what is an outstanding car? And what about the role of that particular driver in making that particular car so good? Shouldn’t that role be rewarded? And how?
I feel that would make the point system more and more subjective, and too complicated, when the idea is to keep it as simple as possible, in the circumstances.

The same can be applied to WOLF’s comments and suggestions. WOLF said:
Your way of 'evaluating races' seems to be more or less OK, but I'd reconsider the notion that all seasons are alike (I understand each season is 100%)- some sort of 'weighing system' would IMHO be in order, similar to the one for individual races...
One particular thing I miss in this way of scoring is incorporated in Williams' stats (I'm fairly certain I got the similar idea independent of him)- is considering the cause of retirement.



You have a point, WOLF. 1946 was clearly not as competitive as 1982, for example. The problem remains that the introduction of another subjective “weighing system” would increase subjectivity to a high level, while the main idea is to be as objective as possible. That is why, a criterion like the cause of retirement may open the door to great problems, and irresoluble ones. There are some causes of retirement that are unknown. There are also the cases (particularly before computers took over) where it’s not easy to tell if the driver is to blame. And the mechanical failures are not all equivalent. And the contrary also happens. Remember Clark winning at the Glen in 1967. Should we give him an extra point for that? It is (or at least it was) part of what was expected from a great driver, to be able to bring his sick car to the finishing line. I think that not finishing the race and not scoring points at each of its stages, is penalty enough in “my system” for those car/drivers that goofed it. On a broader perspective, it could be said that the nature of “racing environment” has its own way to deal with the matter, meaning that a driver who abandons constantly through faults of his own, will be sacked by the team. There’s another, and sometimes fatal, penalty.


Ensign14 said:
I think it demonstrates the difficulty and impossibility of scoring systems. The recounting above does not mention Moss, for a start. But in any system you think of how can you equivalence a win to, say, two 4ths?
I have elsewhere advcated not having a World Championship at all... A World Championship point is a fiction. You can't hold a point, it has no taste, you can't show one to me, it's purely arbitrary whether you get one or three for a 6th place, or even none at all...
So why do we need a World Championship? We all "know" who the best driver is (at least to one's own personal preference).


Yes, a WDC point is a fiction, an abstraction, merely a way to quantify something. And you right about the arbitrarity of point systems. Why not give points to everyone? It was done, as you know, in various championships.
I have followed the opposite route, precisely to avoid the somewhat perverse effect of “accumulation” or equivalence you underline (a win being equivalent to two 4th places, as it now happens).
For that reason, and others too, I have focused my “system” on the tree top (i.e., the 1st, 2nd, and 3rd in practice, in each 50 kms of the race, and at the finish – the podium).

My point system is the following:

Best practice times Minimum level race: 4-2-1 points
Medium level race: 5-3-2 points
High level race: 6-4-3 points

Points awarded at each 50kms of race: Mimimum level: 3-2-1
Medium level: 4-3-2
High level: 5-4-3

Final points: Minimum level: 9-5-3
Medium level: 10-6-4
High level: 11-7-5

My system enhances victory (and pole), so it's difficult for usual midfielders to gain much points.
The issue was already discussed abov. Here’s an added percentages “scoring” example of the long but only average Grand Prix career of Jo Bonnier:

1957: -----
1958: -----
1959: 4,091
1960: 6,129
1961: 5,038
1962: 0,914
1963: 0,723
1964: 1,313
1965: 0,473
1966: 0,591
1967: -----
1968: -----
1969: -----
1970: -----
1971: -----

After his peak in 1959-61, Jo continued racing but he didn’t score, or scored a few points only


D-Type said:
In earlier times, a driver's season comprised more than just the World drivers' championship so how about giving races a weighting. Without thinking too deeply, something like

Grandes Epreuves, or later on World Championship races - 5
World Sports Car Championship, lesser grands prix at least 75% of the average Grande Epreuve- 4
Races for grand prix cars of less than 75% of the average Grande Epreuve (and F2 cars 1952-53) - 3
International Formula 2 races, prewar vouturette races, Formula Junior/F3 1961-63, Formula 3000 - 2 (but 1952-53 score as grand prix cars)
European Mountain Championship - 2 (prewar many aces drove in it)
Other International races (sports car, Formula 3, Formula Junior, GT) - 1

Obviously this would have to be applied with care, but it allows... allows for the many different races of varying importance that drivers raced in earlier eras.


Well, yes and no. Your idea's quite ambitious but it would mean a different approach to the issue.
An approach that would obviously favor the drivers of old, which raced a lot in very different types of cars. Nowadays, stars are prevented by iron contracts to do likewise. They test a lot, though. And that wouldn't be part of the equation.

In my view, categories should be separated. But I applaud your idea of bringing other type of racing cars into the scene. This “system”, or, more advisably, some other, could be applied to different categories and championships that had at least some continuity throughout time. And Moss and all the others would have their positions reckoned in an overall list of results/performances of those different categories. I'm thinking of things like Sports Cars/Endurance (whatever their name), Rallyes, NASCAR, or US Championship (although it would be difficult because of the secession into two separated). Minor formulae would be a nightmare to deal with. In F3, for example, it would be quite difficult (if not impossible) to establish a ranking of drivers, in order to define the real quality, the real weight, of the entries lists.

The "real weight" of entrants is crucial, for me. That's why I do not agree with the idea that ALL WC races should be weighted 5 and ALL lesser Grand Prix, weighted 3 or 4. The example of 1960 Monza is well known but there are other examples, even in more recent times.

Regards

#16 jpm2

jpm2
  • Member

  • 121 posts
  • Joined: December 04

Posted 21 October 2005 - 09:30

Sorry for the delay, Simon. :( When I wrote my previous message your comments were not visible yet.


Originally posted by simonlewisbooks
The problem of awarding points for qualifying is that it fails to take in things such as..

- Team polotics (Peterson alledgedly being sent out on half tanks while Mario was running nearly empty in 1978 for example)

- rain or oil affecting those who had yet to run

- those using practice as a set up session for the race in the full knowledge that they could ensure a better race pace that way

- Tyre companies not supplying qualifying rubber to every team (as happened in the 70s)

- Tyre wars in general

- Drivers wrecking a car after a good qualifying lap and having to take the back up car and a back up time, as happened in the 60s and early 70s when the actual chassis you used was the one that earned it's own grid slot

It also penalises someone like, for example, Watson, who more than once had a low-grid start and passed everyone to win. I felt that made his performance all the greater but the system here would have the opposite effect and penalise an epic drive. Schumacher, unopposed, from Pole in a superior car scores much higher although he was merely cruising by comparison.

For points at specified distances one can again use the Watson example(or that of this years Japanese GP) as a climb through the field would mean missing out on points for leading at specified laps, yet what the race is all about is being in the lead as you cross the line at the finish , which is a strategy in itself (Gethin at Monza in 1971 for further example).
Simon Lewis


Yes, of course, but those “team politics”, the rain and so on that you speak of, and that we might generally define as “circumstances”, apply also to races themselves, and to their final results (particularly to their final results, as we know, with team orders defining who should win and who should be 2nd). Still, in the official championship, we DO count those points, regardless of the circumstances, don’t we? Why shouldn’t we do the same in practice or through the race ?
Best keep things in different bags. Such things as “circumstances” exist and they sometimes have great importance, but they are the object of another historical approach.

As to "Drivers wrecking a car after a good qualifying lap and having to take the back up car and a back up time, as happened in the 60s and early 70s when the actual chassis you used was the one that earned it's own grid slot", I’m afraid you misunderstood, Simon (or maybe it’s me not understanding perfectly what you mean). That particular problem does not exist in “my system”, and precisely to avoid what you've just criticized.
In “my system” practice times and grid positions ARE NOT NECESSARILY EQUIVALENT. If my memory is correct, in the 1972 British GP, for example, Stewart come up with 2nd best time in practice. Then, with another car, Stewart obtained 4th best practice time. He started the race with this second car, and in the 4th position of the grid. But in “my system” he’s still 2nd best in practice (and he is awarded the points corresponding to that place).

You say the system here would penalize Watson and other drivers epic drives from the back of the grid. I fail to see how. What penalizes Watson and others like him, is the official system, that makes them start from the back of the grid, and awards them nothing more than the final position points.
My system does the same, PLUS points for their practice time (if one of the top three), PLUS the points they may be able to grab during their “climb through the field”).
The “system” may not reward Watson and others as much as you would like it to, but it does not penalize them. Quite the opposite.

Take one of the races that saw Watson coming from the back. The 1982 Detroit GP, for example.

Best practice times:
– Prost
- de Cesaris
- Rosberg

Positions held throughout the race
50 kms – Prost, Rosberg, Pironi
100kms – Rosberg, Prost, Pironi
150kms – Rosberg, Lauda, Watson
200kms – Watson, Cheever, Pironi

Final
Watson, Cheever, Pironi

With “my system” this means Watson gets 19 points, Rosberg 17, Prost 15, Pironi 14, Cheever 11, de Cesaris 4, and Lauda 4

If we followed the official system it would be Watson with 9 points, Cheever 6, Pironi 4, Rosberg 3, Daly 2, and Lafitte 1 point.

In my view “my system” rewards not only Watson, but also the fine performances of Rosberg, or Prost, that received little or nothing at all in the official charts. Of course, it also does NOT reward Daly, Lafitte and all those who were never near the front.


Originally posted by simonlewisbooks
But then whatever point system anyone devises there are probably more unfair aspects than fair ones so I'm not picking on this system alone. But applying a new system in retrospect means discounting the premise under which everyone has previously raced (ie. pure and simple - first past the flag). If only they had known - would they have driven differently?

Simon Lewis [/B]


Your final point is a very important one. :up: :up: This is one of the difficult and interesting issues I was hoping “my system” would bring about. There are others, more practical ones (like practice times in the 1930s and 1940s that sometimes are not available) that we may discuss further on.

But let’s get back to the problem of premise, goals and rewards. If only they had known, they would have raced differently. Would they really? Perhaps not as often as one might think.

Anyway, I don’t want to ignore the problem. As a matter of fact this is one of the objections that “my system” faces. It’s a “philosophical” problem, of course, it’s something we should not ignore, but it’s not as overwhelming as one might think, at least not in my view.

The main question is the following: can we evaluate performances and results achieved by human beings, when their intention and their action was molded by criteria that are different from the ones we are using ? How much can it affect and distort the outcome of our evaluation ?

This is open to discussion, of course, but, in my oppinion, we can.

Colombus wanted to reach India, he prepared himself for that and he acted accordingly. Well… he discovered America…

Enough of historical examples. Let’s stick to nowadays life. A footballer plays as he knows how and as he was instructed to play by his coach, according to a set a rules and with a targett in view: to win, to score, to prevent the opponents from scoring in, and so on.
But, as he runs in the pitch, and as he does his best to follow the instructions of his coach, he’s being observed and evalued by someone of a rival team that might be interested in contracting him; and that observer may be using a set of evaluation criteria the footballer is unaware of.

A lot of similar examples can be drawn from our daily lives.

Obviously “my system” is not valid for the drivers, once it does not grant them any recognized title. For them, the only valid system was the official one (when it existed)

But this does not mean that “my system”, if it's a competent one, isn’t valid for the outside spectator or analyst. Otherwise, few histories could be written.
Summing it up, the “system” I propose, is not THE championship. It’s something else, a rational exercise on that championship. It is self-evident that it has no practical effect, as well as a history book has no effect on people and events from the past. Still, some of us continue to write history books with interpretations of the past, instead of just repeating constantly what we were told in the chronicles and documents of the epoch under study.

Regards

#17 HDonaldCapps

HDonaldCapps
  • Member

  • 2,482 posts
  • Joined: April 05

Posted 21 October 2005 - 11:59

Since this "alternative championship" is clearly one of any number of approaches that have surfaced over the eons with the intent of establishing some system "to compare results and performances," the issues, as seen in the "Chiron" thread, are both technical and philosophical.

On the technical side, there are some questions of how far the metrics can be stretched to accomodate varying periods with data that is not necessarily comparable. Awarding points for qualifying performances along with actual race performance is an approach that more than a few have explored over the years. One minor quibble is that not until well in to the 1930s were grids in Europe determined by as a matter of routine by practice or qualifying times. How is this accounted for in this system? Also, for a system that depends greatly on the accuracy of its data, how reliable are the metrics generated for those events where reliable lap charts are not readily available?

While it is interesting to see how you assign your driver co-efficients to determine the "worthiness" of an event for inclusion the "alternative championship" for the 1964 season, there is a bit ex post facto metrification inherent in this, since you are using co-efficients generated at the end of that season versus those that would have generated from the previous season. This means that your system could not provide contemporaneous data. That is to say, you couldn't figure out how 2006 went until 2006 was over. However, it seems that your system was intended to be a retrospective one.

While one and all can nit-pick or critique the system until the cows come home, an esential element to keep squarely in mind is that this was originally devised to provide an individual with a mechanism to do a comparison of results of performances for his own edification. He designed it to reflect his viewpoint and where he places value in the area of performance.

It is only when the author takes it beyond that point that the cows might wish to hasten back to the barn. While this scheme might work relatively well within a single season or over a fairly narrow spread of seasons, it is questionable as to whether it provides the level of validity claimed in comparing performances over a span of generally dissimilar seasons.

I think that the one aspect of this approach that triggers my generally negative response is that it borders on the notion of reductio ad absurdum. That is, it reduces everything to just numbers. Yes, an old song played many times with a wide variety of instruments. While many seem to see a need to analyze, quantify, produce metrics, and generally reduce everything into numbers, perhaps a few of us are diehards and see things through a different lens. While there is perhaps some reasonable arguments made for an ordinal worldview, that tends to slew both the past and the present in directions that really don't reflect how dependent upon context and many of those other "squishy" factors that make one raise an eyebrow at life being reduced to mere numbers.

While I might be derided by most here and elsewhere as being a hopelessly out of touch romantic with a view of things that runs is not only obsolete but non-operable then as well now, so be it. I have no end of race data on no end of series; certainly no where as much as some here, but a rather sizeable racing database nonetheless. I would suggest, however, that a database and whay I would vaguely refer to as a "base of knowledge," while certainly not mutually exclusive, are oriented in very different directions.

A long-standing schism in other aspects of history is not a stranger here: the New or quanitative, technical or analytical school versus the Traditional school. Residing somewhere between the two -- but generally closer to the Traditional school -- is the more recent "Holistic" school while the Cliometricians are probably among the core of the corps for the New school.

We have benefited to a very great degree in the last decade or two from an influx of information that was generally restricted to only a very, very few and even then quite compartmented with few having access to most of it, much less "all" of it. This began in the years immediately before the introduction of the Web and the internet. While many had bits and pieces of the puzzle, few had lots of pieces of the puzzle and virtually none outside a small cadre of writers and researchers. Needless to say, this has changed.

Just a few years ago, defined as perhaps a decade or so ago, finding a fairly comprehensive list of results and related data for the CSI World Championship for Drivers (and constructors) and the FIA F1 World Championships was a challenge; now it is a staple item of a seemingly infinite number of Web sites, with errors, operationalizations and all still present in most cases from those whom the information was taken.

In the final analysis (pun intended), Doug Nye says it best:

What counts - and what has always counted - between drivers is a human quality which is numerically unquantifiable - flesh and blood, grit, guts, talent, personality, perceived stature amongst his peers, and skill.


While I can appreciate the effort and hard work involved in generating this "alternative championship," I have to admit that I fail to not so much see the point of it, but rather the 'why does this matter to me' of it.

#18 roger_valentine

roger_valentine
  • Member

  • 208 posts
  • Joined: October 02

Posted 21 October 2005 - 13:51

Originally posted by Catalina Park

I measure drivers by height.
Dan Gurney is looking good.


A flawed system, I'm afraid. Jackie Stewart was pretty good too.



jpm2, have you read "The Hitchhikers Guide to the Galaxy"? People love to debate the great unanswerable questions of life; that doesn't mean they want somebody to find the answer and hand it to them on a plate. In sporting terms, people will always ask 'is Schumacher better than Nuvolari? Is David Beckham better than Stanley Mathews? Is Hurricane Higgins better than Joe Davis?" Everybody realises that these questions can never be answered, but the debates themselves provide enjoyment, nostalia, education. Then you come along with a quasi-statistical formula which ends all debate and proves catagorically that Michael Schumacher (or should that be Fangio - I'm not really sure from your findings) is absolutely the best racing driver ever. Leave your system running long enough and no doubt you will eventually prove that the answer to the ultimate question of Motor Racing, the Universe and Everything, is 42 (which, I guess must mean - Johnathan Palmer!)

In fact, there is so much subjectivity in your choice of criteria that even the pseudo-accuracy of Schumacher scoring 310,786 percentage-points doesn't alter the fact that your system is of no more value than, say, Murray Walker's purely subjective Top 20, where the sole criteria was, as I recall, 'I saw these people race and thought they were good'. (Indeed, as most of us have never seen many of these people race, maybe Murray's appraisal is of more value).

#19 HDonaldCapps

HDonaldCapps
  • Member

  • 2,482 posts
  • Joined: April 05

Posted 22 October 2005 - 12:26

Originally posted by roger_valentine
Leave your system running long enough and no doubt you will eventually prove that the answer to the ultimate question of Motor Racing, the Universe and Everything, is 42....


....which would mean Lee Petty, of course.

Advertisement

#20 Hans Etzrodt

Hans Etzrodt
  • Member

  • 3,188 posts
  • Joined: July 00

Posted 22 October 2005 - 14:15

Originally posted by roger_valentine
...In fact, there is so much subjectivity...

For those of you who are interested in a totally unbiased approch on this comparison subject, you can look here. BTW, journalists have been talking about this subject already 100 years ago.

#21 jpm2

jpm2
  • Member

  • 121 posts
  • Joined: December 04

Posted 22 October 2005 - 19:50

Originally posted by roger_valentine

jpm2, have you read "The Hitchhikers Guide to the Galaxy"? People love to debate the great unanswerable questions of life; that doesn't mean they want somebody to find the answer and hand it to them on a plate. In sporting terms, people will always ask 'is Schumacher better than Nuvolari? Is David Beckham better than Stanley Mathews? Is Hurricane Higgins better than Joe Davis?" Everybody realises that these questions can never be answered, but the debates themselves provide enjoyment, nostalia, education.



I agree. For me there is no point in discussing personal preferences based on "magic", or idolatry or emotions. But I understand that others feel differently. :up:


Originally posted by roger_valentine
Then you come along with a quasi-statistical formula which ends all debate and proves catagorically that Michael Schumacher (or should that be Fangio - I'm not really sure from your findings) is absolutely the best racing driver ever.



As you say, you’re not really sure from my findings who was “absolutely the best racing driver ever”. I think you can draw two conclusions from that “ambiguity”, a happy one, and a sad one. :)

The happy conclusion is that discussion can continue. The horrible “system” of jpm2 hasn’t kill discussion after all. All happy members will be able to discuss whether it is Schumacher or it is Fangio.

The sad conclusion, I regret to say, is that you missed the main point altogether (or, my English isn’t clear enough, I have to admit that possibility). And the main point is that there are several ways of “measuring” performances and results of Grand Prix drivers. I thought I had explained it clearly, and enunciated them (victories, best practice times, points, added percentages, points/race). The combination of those 5 can, in my oppinion, give you a fair and balanced (two words I’ve been repeating since the beginning of this discussion) perspective of the tree top.


Originally posted by roger_valentine
In fact, there is so much subjectivity in your choice of criteria that even the pseudo-accuracy of Schumacher scoring 310,786 percentage-points doesn't alter the fact that your system is of no more value than, say, Murray Walker's purely subjective Top 20, where the sole criteria was, as I recall, 'I saw these people race and thought they were good'. (Indeed, as most of us have never seen many of these people race, maybe Murray's appraisal is of more value).


Although I saw my first F1 race in 1959 Murray Walker is older and more traveled than me and he surely has seen much more than I did.
But you misunderstood me again. Here, I’m not interested in judging from what I’ve seen (I couldn’t be an historian if I needed to see the things I’m writing about, could I?) nor from what my preferences are or were back then. I’m classifying from results, practice times, race positions, as they were and as they are recorded in documents.
Is “my system” more subjective than the official point systems? As Wolf said in this thread, what is a point anyway? And why those points and not others? And why have they changed throughout time? Well, because some people wanted to reward the fastest laps, or because they wanted to reward 6th place, or because they wanted to enhance victory (hence 9 and, later, 10 points awarded to the winner, instead of just 8 as was done in the 1950s). Subjectivity is all around.
So, I would be sincerely grateful if you could step down from generalities and could point precisely where are the unbearable subjectivities of “my system”.

Regards

#22 jpm2

jpm2
  • Member

  • 121 posts
  • Joined: December 04

Posted 22 October 2005 - 20:11

Originally posted by HDonaldCapps
[B]Since this "alternative championship" is clearly one of any number of approaches that have surfaced over the eons with the intent of establishing some system "to compare results and performances," the issues, as seen in the "Chiron" thread, are both technical and philosophical.

On the technical side, there are some questions of how far the metrics can be stretched to accomodate varying periods with data that is not necessarily comparable. Awarding points for qualifying performances along with actual race performance is an approach that more than a few have explored over the years. One minor quibble is that not until well in to the 1930s were grids in Europe determined by as a matter of routine by practice or qualifying times. How is this accounted for in this system? Also, for a system that depends greatly on the accuracy of its data, how reliable are the metrics generated for those events where reliable lap charts are not readily available?

The accomodation of varying periods is, precisely, the biggest challenge of all. Have I succeeded? I don’t know. The idea was to replace, for personal use, the official statistics for something more balanced and fair, and that would include, pre-1950 single-seater Grand Prix.
Please do not forget that official statistics that give us total of points gained and other data from 1950 up to now have to accomodate also, because things changed a lot since 1950.

Practice times in the 1930s and 1940s are not a “minor quibble”. They’re a big big problem, in some cases an unresolved one. I relied on Sheldon and Snellman, but as we know back in the 1930s and 1940s grids were often decided on different criteria (car number, ballot), and, in those cases, time-keeping was usually a private one, it was made by the teams themselves, and it’s hard or impossible to get.
Hans Etzrodt gave me what he had (thanks again, Hans :up: ), I got two ot three starting grids in the Spanish forum (thanks fellows :up: ) but there are still gaps to fill.

Which means that some of the final standings and points from the 1930s-1940s are still provisional (and probably will be forever).

Whenever there is no lap by lap charts, the definition of the positions held during the race came from the chronics of the race itself. Please don’t forget we are talking of race stages of 50kms approximately, and that I’m focusing on the first 3 only. It would have been impossible to achieve if I wanted to draw a complete lap by lap chart.

There could be errors, but I hope the margin of error will not be great, and that it will be eroded bit by bit. I hope some members here will be willing to help with those practice times that are still missing.
Anyway, the reliability of results is still questionable for the 1930s-1940s.


Originally posted by HDonaldCapps
While it is interesting to see how you assign your driver co-efficients to determine the "worthiness" of an event for inclusion the "alternative championship" for the 1964 season, there is a bit ex post facto metrification inherent in this, since you are using co-efficients generated at the end of that season versus those that would have generated from the previous season. This means that your system could not provide contemporaneous data. That is to say, you couldn't figure out how 2006 went until 2006 was over. However, it seems that your system was intended to be a retrospective one.

Yes, absolutely retrospective. Never was it intended to be anything else.
Absolutely “ex post facto metrification” as you say, and, I add, as historians do, when applying modern formulae or concepts to the analysis of the past.

The expression “alternative championship” is not my own. I opened this thread with a different name “my championship”, “my system”, or something like that. Afterwards, that title was changed (by the editor, I suppose) into “alternative championship”. In my view it was not a good choice because it conveys the idea that “my system” is meant to substitute the actual championship as it exists. It is not, and it never was.

As to contemporaneous data the answer is, surprisingly, yes. I wouldn’t be able to figure out how 1959 went until 1959 was over. The ranking of car/driver would have to wait the final race of the season.
That isn’t really needed for nowadays races. Extra-championship races have disappeared from our horizon and the championship GPs have always the same or almost the same entries. They are all high level races. Schumacher can be replaced by Salo, or Montoya by Wurz, the “worthiness” of the event may drop slightly but it will always be within the accepted margin, it will always be high level.

Originally posted by HDonaldCapps
While one and all can nit-pick or critique the system until the cows come home, an esential element to keep squarely in mind is that this was originally devised to provide an individual with a mechanism to do a comparison of results of performances for his own edification. He designed it to reflect his viewpoint and where he places value in the area of performance.

Absolutely.

Originally posted by HDonaldCapps
It is only when the author takes it beyond that point that the cows might wish to hasten back to the barn. While this scheme might work relatively well within a single season or over a fairly narrow spread of seasons, it is questionable as to whether it provides the level of validity claimed in comparing performances over a span of generally dissimilar seasons.



Well… it’s up to the cows, :lol: but I don’t see why they should haste to the barn. The methodology was precisely what I wanted to discuss, alongside what you call “technical” issues.

Originally posted by HDonaldCapps
I think that the one aspect of this approach that triggers my generally negative response is that it borders on the notion of reductio ad absurdum. That is, it reduces everything to just numbers. Yes, an old song played many times with a wide variety of instruments. While many seem to see a need to analyze, quantify, produce metrics, and generally reduce everything into numbers, perhaps a few of us are diehards and see things through a different lens. While there is perhaps some reasonable arguments made for an ordinal worldview, that tends to slew both the past and the present in directions that really don't reflect how dependent upon context and many of those other "squishy" factors that make one raise an eyebrow at life being reduced to mere numbers.

While I might be derided by most here and elsewhere as being a hopelessly out of touch romantic with a view of things that runs is not only obsolete but non-operable then as well now, so be it. I have no end of race data on no end of series; certainly no where as much as some here, but a rather sizeable racing database nonetheless. I would suggest, however, that a database and whay I would vaguely refer to as a "base of knowledge," while certainly not mutually exclusive, are oriented in very different directions.



What is “my system” being accused of? It appears it “reduces everything to just numbers”. Well, the word “everything” is not an issue here. It never was (although a few fellow TNFrs are constantly pushing me in that direction). :rolleyes:

“My system” is just my personal substitute for official statistics. And what are the official statistics? Well… just numbers. And could or should they be anything else? No, I don’t see why.
That “anything else” I’m speaking of, would be in another league and would need a different approach. When you write a book, sometimes it’s useful to include annexes, generally documents or quantitative data. Well just imagine “my system” was the annexes of an unwritten history of drivers lives and deeds, of their fears, their representations of death, their social backgrounds and economic interests, and so on (by the way, who talked of “life being reduced to mere numbers”?)

But that was not what was being discussed here. Let me remind you that it all started with fines saying he personally rated Chiron far “above (ahead?) of Nuvolari, Varzi and Caracciola, not to mention Rosemeyer” and as being “on a level with Fangio, Clark and Prost”.

Why was Chiron so high rated by fines? Was it because he was more intelligent driver? Or because he had higher moral standings? No, we were talking performances and results. Just that. There is no ground for an accusation of “reductio ad absurdum”. The only “reducionists” I see here, I regret to say it, are precisely some TNF members (you included) that are trying to reduce my views on Grand Prix History to “my system” or to “just numbers”.


Originally posted by HDonaldCapps
A long-standing schism in other aspects of history is not a stranger here: the New or quanitative, technical or analytical school versus the Traditional school. Residing somewhere between the two -- but generally closer to the Traditional school -- is the more recent "Holistic" school while the Cliometricians are probably among the core of the corps for the New school.

It is ironic that those members who say I cannot “judge” drivers by their results and performances, regardless of the context, are precisely those who are taking the lead in “judging” me as a “cliometricist”, knowing so little about what I think or do. :)
Isn’t it the British that say you must never judge a book by its cover? And that you shouldn’t jump into conclusions?
You’ re entering “Terra Incognita” here, Don. Allow me to offer you a map to guide your way. I am an historian, my research is not in any way related with motor racing, and I am not a “cliometricist”, quite the opposite. You might have guessed that, by some of my previous posts here (the little work on the practical jokes of Mario Cabral and friends hardly matches the paradigm of “cliometrics”, don’t you agree?), and you may see for yourself if ever you read my books (one of them being published in England and in the States as well).
Motor racing is a lateral interest, and “my system” is an hobby from my youth. It will never be more than that. I am not trying to discover the moon with it. It is surely not history, nor did I ever pretend it to be. It is just an analytic instrument to quantify and classify, performances and results of Grand Prix drivers. KJJ called it an “interesting game”, it’s alright with me
Still, the “cliometricist” label comes back frequently (perhaps brought by those who are historians themselves). This is not about history with capital letters. It’s only about the way to “measure” results/performances of Grand Prix drivers. In my view, historiographic schools have no place here. But since you have brought the theme along I’m certain you won’t mind my asking: isn’t that also a method of judging historians? The “traditionalist”, the “cliometricist”, the “holistic”? Is it a fair method? What does it shows us? Or would it be better to read their books, and leave those labels aside?;)
As a matter of fact, this has little to do with historiographic schools. If anything, and as you correctly said, it may be related with what might be called two different ways of seeing the world, the “rationalist” and the “poetic” or “romantic” views, you name it. The main problem is that, in most cases, those dichotomies are over-schematic. One can imagine a “Linnaeus-type” sort of fellow, one who wants to classify, to put order in what he has before his eyes, and a Shakespeare-type” fellow, who preferes to contemplate human feelings, their emotions and dramas. I suppose Shakespeare wouldn’t be particularly interested to know that an elephant was a proboscidea or that a horse was an equidae. He might have enjoyed a horse-ride, though, or the beauty of that white mare galloping in the field. And probably Linnaeus would have too. Usually, there is a bit of Linnaeus and a bit of Shakespeare in each one of us.

Originally posted by HDonaldCapps
We have benefited to a very great degree in the last decade or two from an influx of information that was generally restricted to only a very, very few and even then quite compartmented with few having access to most of it, much less "all" of it. This began in the years immediately before the introduction of the Web and the internet. While many had bits and pieces of the puzzle, few had lots of pieces of the puzzle and virtually none outside a small cadre of writers and researchers. Needless to say, this has changed.

Just a few years ago, defined as perhaps a decade or so ago, finding a fairly comprehensive list of results and related data for the CSI World Championship for Drivers (and constructors) and the FIA F1 World Championships was a challenge; now it is a staple item of a seemingly infinite number of Web sites, with errors, operationalizations and all still present in most cases from those whom the information was taken.

I don’t see where this bit of prose fits in. :confused:
Nevertheless, this kind of discourse has appeared here three times already. Could it be that some members of this forum are troubled? Could it be I’m being seen as some kind of poacher, no one knew about, who entered the temple of knowledge reserved only for the chosen few, i.e. to that “small cadre of writers and researchers” you talk about?

Well, I really don’t know, but it sure looks like it.
So let me get this thing straight once and for all. The way I have built “my system” is no mystery. I explained it myself. I also said that “my system” benefited much from the early 1980s on with the colored leaflets and the black books of Paul Sheldon, that have allowed me to fil the gaps and to go back until 1932. Then, in the last 3-4 years, internet came into the fore. Through it I obtained some 1930s and 1940s practice times that were lacking, and learned a lot in Leif Snelman’s web site, and got lots of pictures (thanks all :up: ).

The system itself has nothing to do with internet. But what if it wasn’t so? Would it make it more valid (or invalid) for that reason?

I’m not selling anything here. I explained “my system” because other members in the forum said they would like to hear about it. I came in the open and said, this is the way I do it. It has flaws, it is incomplete, I would like to correct it, if possible, but for me it is a better, fairer and more balanced way of rewarding drivers results and performances than the official system. You can agree or disagree. You can ignore it. You can look the other way, you can say no, or you may keep silent.
But no suspicions or innuendos, please. You don’t need to kill the messenger just because you don’t like the message.

Regards

#23 jpm2

jpm2
  • Member

  • 121 posts
  • Joined: December 04

Posted 22 October 2005 - 20:15

Originally posted by Hans Etzrodt
For those of you who are interested in a totally unbiased approch on this comparison subject, you can look here. BTW, journalists have been talking about this subject already 100 years ago.


Read the thread Hans. Interesting stuff. :up:
It’s funny to see there are a lot of people touched by what I just called the “Linnaeus” spirit, and making “attempts to compare the top drivers”, as you say, on a results/performances basis (in your case on a ratio races per win basis). It’s even interesting to read the messages of some of those who are now participants on this and Chiron’s thread.;)

I already knew you had a system (you told me so in one of those emails we exchanged some time ago) but I was not aware of how it worked.

It is interesting also to compare your system with mine. Not only the ranking of drivers, but mostly, how the systems were conceived and built. There are a lot of differences beteen them, and also different ways to handle problems. The way we handle shared drives, for example, is quite different. But there are similarities too. We both introduced “safety valves” in our respective systems. Mine, has to do with the ratio points per race (when determining that ratio, I always consider a minimum of 20 races; otherwise drivers with one or two races that went particularly well, would be able to get a ratio that could be largely influenced by a given set of rare events and that would not be representative of their real value); you have introduced a similar “safety valve”, for the same reasons, I suppose, and you excluded from your list those with less than 12 races. I agree with the methodology, but I have to underline, Hans, that it is entirely SUBJECTIVE.  ;)

As a matter of fact, your system is not a "totally unbiased approch", as you say. Sorry Hans. :(

And it even suffers from several important problems, not on the "technical" side but on the "philosophical" side of your project. This is probably not the place nor the time to analyse them. But I would be happy to to point some of them to you, if you want me to, of course.

Regards

#24 HDonaldCapps

HDonaldCapps
  • Member

  • 2,482 posts
  • Joined: April 05

Posted 22 October 2005 - 21:03

Originally posted by jpm2
The combination of those 5 can, in my oppinion, give you a fair and balanced (two words I’ve been repeating since the beginning of this discussion) perspective of the tree top.


I will openly admit that I instinctively flinch whenever I hear or read "fair and balanced" since that seems to be something quite open to question here.

I will admit that I was not certain as to laugh or merely shake my head I read this:

As a matter of fact, your system is not a "totally unbiased approch", as you say. Sorry Hans.


I decided on the former since it seems to also apply in very large part to your work since your choice of metrics is equally subjective. Plus, given the nature of the lack of precision of some of your data from the 1930s and 1940s -- and perhaps even the 1950s for that matter -- there is some legitimate question as to your ability to accurately judge "performance" over such a large span of time.

Having said that, as well as reading your rebuttal to my earlier comments -- nothing on Lee Petty I noted, I am glad you are finding a wonderful way to fill up your time and stay busy, but all the statistics and quantification in the world simply cannot do anything but fail to reduce what is esentially at heart -- or was is perhaps a more accurate way of stating it -- a rather chaotic and messy convergence of men, machines, and no end of other things which creates a tangled mess of stories and storylines that generally seems to be far more interesting than seeing bloodless lists that have names of drivers with some cryptic numbers beside them and arranged in some ordinal fashion.

Once in a long, long while, I see a scheme such as yours that seems to make sense and which produces some results that are offered as food for thought. The project that produced a listing of competitive world championship seasons was one of those, not necessarily for its accuracy, but because it provided a rich source of stories to delve into and mine for articles, which produced my series on 1961 as an instance.

The bedrock of my oppostion to this scheme is that it is just numbers and the crunching thereof. Spare me. We need historians to be exploring the many unanswered or murky or uncertain questions of racing lore, not generating yet more reams of numbers in a field where the numbers types seem to be successfully imitating kudzu with generally the same level of success.

#25 Tim Murray

Tim Murray
  • Moderator

  • 24,592 posts
  • Joined: May 02

Posted 22 October 2005 - 21:06

I am one of those who believe that it is essentially impossible to come up with a system that can reliably compare drivers from different eras. However, I should like to commend jpm2 for his robust defence of 'his' system. He has responded to all attacks in a determined, logical and (most importantly) courteous fashion. Bravo!

#26 Wolf

Wolf
  • Member

  • 7,883 posts
  • Joined: June 00

Posted 22 October 2005 - 21:11

If I may I'd like to recapitulate some of the things here, just so that we're on the same wave-length. Needless to say, as Don may remember, some virtual dust was thrown around when I tried to survey (semi-serious) Top 10 GP drivers' list some time ago, so I appreciate the position jpm2 is in right now.;) And just to remind folks, I'd like to mention that Leif keeps track of modern F1 results on his superb site in pre-War system that gave winner of the race remarkable 0 points. He isn't being 'accused' of rewriting history or collapsing everything into a single-digit number- and I'm certain many a member of TNF has sneaked in there to take a peek.;) I too find it amusing (certainly in no derogatory sense), and interesting- but I'm also willing to take a look at jpm2's method and give him the benefit of the doubt. And that, I must admit, my trust in our member has been increasing- he is treating 'nay-sayers' with due respect, and that he made his system because he feels he is 'fairer' than existing stats (let's face it, Fangio must have won 10 times to earn as much points as MS did with 8 wins) and has opened himself to both benevolent and not so benevolent critique, and not because he knows the answer (of course it's 42 as has been said earlier).

And I believe that's saying enough- remarkably, despite the fact that Moss isn't in top 10 on his list... :lol:

#27 Doug Nye

Doug Nye
  • Member

  • 11,524 posts
  • Joined: February 02

Posted 22 October 2005 - 21:18

I agree. Bravo indeed - jpm (Juan Pablo Mon- who??????) - I really genuinely applaud your industry, I merely regard its end product if accepted without comment or qualification as being potentially a dangerously distorting lens through which to view historical realities.

Enjoy. Nobody's perfect.

DCN

#28 Hans Etzrodt

Hans Etzrodt
  • Member

  • 3,188 posts
  • Joined: July 00

Posted 23 October 2005 - 03:51

Originally posted by jpm2
...As a matter of fact, your system is not a "totally unbiased approch", as you say. Sorry Hans. :( ...

This is a true observation. I must have been wearing my pink sunglasses, when I typed that.  ;) Please change that to 'comparatively unbiased approach'.




Originally posted by jpm2
...And it even suffers from several important problems, not on the "technical" side but on the "philosophical" side of your project. This is probably not the place nor the time to analyse them. But I would be happy to to point some of them to you, if you want me to, of course

Because I work presently on other projects I do not want to be sidetracked by this topic here at discussion, and because I already lost interest some years back. Thank you for your generous offer to share your erudition with me but I have no desire to further discuss my flawed listing because I just lack the interest to do so. Long time ago it has provided me with the answers I wanted.

There was a time in the past when I was really interested in this driver comparison subject and when I had spent way too much time on this project before even obtaining the necessary data for each driver who ever won a Grande Épreuve. But this was a good exercise and not really a waste of my time. Then – after several attempts I used the system posted now at TNF thread –here discussed– with an aim to remove all bias and strive for fairness to all drivers. To compare the Grand Prix drivers, I very soon realized that I could only apply Grandes Épreuves and thereby eliminate all other Major Grands Prix, whose status is another controversial matter. While compiling the data I realized the need to set a limitation in order to avoid absurd results. Therefore I decided to only accept drivers into the list who had participated in 12 or more Grandes Épreuves. By doing so, MY system and yet unborn list was already flawed from the onset and I had no fix for it. I had to accept this shortcoming because I knew, at the time, that this was the best compromise I could reach after my beginning trial and errors period. I had seen before other driver comparisons in magazines of the thirties and forties and even earlier comparison attempts but also some years ago in the nineties magazines from Autosport and MotorSport. Although all were quite entertaining, each of those systems and lists contained one or the other fault and none was perfect. So, my list was going to be just another blemished and controversial exercise.

In answer to your system I can only state again that it is way too complicated, inaccurate and time consuming for me. I could never do this and you have my utmost respect for your stamina and perseverance. It is definitely a brave effort to search and establish new criteria although some of those are far removed from the established understanding of this matter by motor racing historians, which will obviously get you in hot water. In General you have my praise for your courageous labor. Lots of luck.

#29 Roger Clark

Roger Clark
  • Member

  • 7,502 posts
  • Joined: February 00

Posted 23 October 2005 - 07:40

I'd like to join those who have congratulated jpm2 on the courteous and articulate way he has explained his system and replied to his critics. It seems to me that he has a thorough and balanced understanding of what he is trying to achieve and of the limitations of the system. It is an interesting way of assessing drivers' performance and jpm2 has been careful to stress that it is no more than that. It is complementary to the essential analysis of what actually happened and why they happened in the way they did.

I quite like the idea of a contest in which the competitors don't know the rules until it is over. If it could be adopted in the official championships it might make people more interested in the races themselves rather than the gathering of points. It would be interesting to extend the idea to include categories of racing other than the prevailing Grand Prix formula, particular in years when drivers typically raced in several categories. Obviously, there would be major difficulties in gathering and analysing the data for such an extension.

I don't think that jpm2 has told us how much relative weight he gives to practice time, mid-race positions and final results. My apologies if he has.

In any case, there is a difficulty in doing this. A driver who starts at the back of the grid, but who recovers to a win or a good position may have produced a more worthy performance than one who was in the top three throughout the race, as long as all those in front of him didn't retire, of course. This was a point made by Simon Lewis and which jpm2 countered with the example of John Watson in Detroit. Yet the fact that there are circumstance under which the system could penalise a great recovery drive seems to be a flaw; how, for example, would it deal with Clark at Monza in 67, or Raikkonen in Japan this year?

Of course, the system does not take account of the equipment available to the driver. Perhaps we should follow Pomeroy and divide each driver's score by the sixth root of the ratio of his car's engine power to frontal area.

(Please don't take the last point seriously)

#30 Roger Clark

Roger Clark
  • Member

  • 7,502 posts
  • Joined: February 00

Posted 23 October 2005 - 07:47

I forgot: in 1964, the dutch Grand Prix was a medium coefficient race. Why was that?

#31 HDonaldCapps

HDonaldCapps
  • Member

  • 2,482 posts
  • Joined: April 05

Posted 23 October 2005 - 10:29

jpm2 has done a good job of both articulating and defending the scientific approach to this knotty question. he has the courage of his convictions. However, when racing is reduced to "Science," when "Art" is rendered irrelevant, I have no use for the "Science" and will, inevitably, opt for the "Art." That is at the hear of my objections, since you can mangle the numbers however you wish as long as they are mangled in a consistent manner and call it Science and have people nod their heads.

By the way, you have two lists with different people at the top of each one -- which is "THE" list? Not that it matters to me, but others might care.

Roger, the Dutch GP apparently did not accumulate enough "points" to be a "high" race.

Personally, I will stick with the messiness of the "art" of motor racing since the "science" seems to be a bit too dry and sterile for my tastes.

#32 Roger Clark

Roger Clark
  • Member

  • 7,502 posts
  • Joined: February 00

Posted 23 October 2005 - 11:05

Originally posted by HDonaldCapps

Roger, the Dutch GP apparently did not accumulate enough "points" to be a "high" race.

I guessed that much, but the only significant difference between the entry at Zanvoort and Monaco a fortnight earlier was the loss of the BRP entries and Trintignant's BRM, and the addition of de Beaufort. Monaco also had the usual non-qualifiers due to the limit on the number of starters, and Innes Ireland's non-starting due to his usual crash. de Beaufort for Trintignant seems a fairly neutral substitution. I am a great admirer of Trevor Taylor, but was his absence enough to downgrade the race?

#33 HDonaldCapps

HDonaldCapps
  • Member

  • 2,482 posts
  • Joined: April 05

Posted 23 October 2005 - 12:29

Roger,

I rest my case....

#34 Roger Clark

Roger Clark
  • Member

  • 7,502 posts
  • Joined: February 00

Posted 23 October 2005 - 15:22

Originally posted by HDonaldCapps
Roger,

I rest my case....

:lol: That's where we differ Don, because I saw it as an interesting topic of discussion. The process is more interesting than the result. Nevertheless, I hope we can agree that the topic is closer to the original spirit of TNF than is a discussion of Alain Prost's sex life or Nelson Piquet's urinary habits.

#35 HDonaldCapps

HDonaldCapps
  • Member

  • 2,482 posts
  • Joined: April 05

Posted 23 October 2005 - 18:13

Roger,

This is definitely the sort of thing I originally envisioned and I can only thank jpm2 for his graciousness and well-thought out rebuttals and responses along with the others chiming in.

I would be willing to be the Devil's Advocate simply to have a discussion of this sort once more. I have enjoyed this very much and hope it continues since it does involve a topic that deserves more discussion.

Don

#36 Wolf

Wolf
  • Member

  • 7,883 posts
  • Joined: June 00

Posted 23 October 2005 - 18:55

Speaking of rankings, I just remembered a friend of mine has made a system that was (unfortunately briefly) employed in GPL community. It was similar to chess and FIFA rankings, and the idea might be of some use for purposes like this. With GP racing one could easily use WWI or WWII as starting point, because of preceeding breaks. But back to basic idea behind it...

This system is based on 'rankings' (at start everybody is awarded same amount of points, as are newcomers later on), and one earns or loses points according to whom he has beaten (in this case finished in front of) or lost to (finished behind). For example, Moss would earn certain amount of points for finishing in a race in front of Fangio (more than Fangio would earn by finishing in front of Moss), but say Trintignant would earn more points than Moss because he'd presumably be lower ranked than Moss and his feat would hence be greater. System has further features like that older scores 'fade' with time, so that e.g. beating Graham Hill in say '63 would earn one more points than beating him in '73...

This method incorporates 'weighting' system for races similar to one jpm2 uses but is IMHO more objectively and 'up to date'.

P.S. He has also used 'weighing' factor for cars (I was involved with proposing and implementing it) and tracks*, which was relatively easy in terms of GPL and could pose a problem in 'real life'. But simpler and cruder method for cars could probably be used so that beating Fangio in Merc yields more points than beating him in P15.

* we made coefficients based on World records at the time to evaluate those things- but the idea was that one would earn more points for e.g. driving a '67 Cooper and beating Eagle at Spa than at Monaco...

#37 jpm2

jpm2
  • Member

  • 121 posts
  • Joined: December 04

Posted 23 October 2005 - 18:57

First of all I would like to thank Tim Murray and Wolf for their compliments and support. :up: Thank you too, Doug Nye for your comments. :up: By the way Doug, jpm2 has nothing to do with Juan Pablo Montoya (I’m too old for that). It is just my name: Joao Pedro Marques. As to the “dangerous distorting lens” you talk about, I hope to develop that issue further on.

Thanks also to Don Capps for his appraisal. :up: But, Don raises a few questions that need to be dealt with.

Originally posted by HDonaldCapps
I am glad you are finding a wonderful way to fill up your time and stay busy, but all the statistics and quantification in the world simply cannot do anything but fail to reduce what is esentially at heart -- or was is perhaps a more accurate way of stating it -- a rather chaotic and messy convergence of men, machines, and no end of other things which creates a tangled mess of stories and storylines that generally seems to be far more interesting than seeing bloodless lists that have names of drivers with some cryptic numbers beside them and arranged in some ordinal fashion.


Have you ever heard Berlioz’ Symphonie Fantastique, Don? The Symphony is built around a musical fragment that comes and comes again. It is called the “idée fixe”. It seems the word “reduce” has, for you, the same value the “idée fixe” has on Berlioz’ Symphony. As I’m an admirer of Berlioz, I understand how “idées fixes” can be important, once they get to our brains, and I will try to explain again. Here it goes: racing was not reduced to science. An aspect of racing was (if we could call “science” to this exercise). I’m referring only to the results/performances aspect of racing. All the rest, i.e., the real lives of the actors, their affections, their fears, their pains, their triumphs. There is open ground for what you call “art”.
“Art” and “science” are not mutually exclusive. You can have “statistics” or “systems”, whatever you like, and still tell us the story of how, and what and why (I hope this may answer part of the message you have posted today; the rest of the answer will be coming soon).

Originally posted by HDonaldCapps
Once in a long, long while, I see a scheme such as yours that seems to make sense and which produces some results that are offered as food for thought. The project that produced a listing of competitive world championship seasons was one of those, not necessarily for its accuracy, but because it provided a rich source of stories to delve into and mine for articles, which produced my series on 1961 as an instance.


I’m sorry, I never read your series on 1961. I would love to, though. :up:


Originally posted by HDonaldCapps
I will openly admit that I instinctively flinch whenever I hear or read "fair and balanced" since that seems to be something quite open to question here.

I will admit that I was not certain as to laugh or merely shake my head I read this:
quote:
--------------------------------------------------------------------------------
As a matter of fact, your system is not a "totally unbiased approch", as you say. Sorry Hans.
--------------------------------------------------------------------------------

I decided on the former since it seems to also apply in very large part to your work since your choice of metrics is equally subjective. Plus, given the nature of the lack of precision of some of your data from the 1930s and 1940s -- and perhaps even the 1950s for that matter -- there is some legitimate question as to your ability to accurately judge "performance" over such a large span of time.


Don, I hope you’ll forgive me, but one of the historian golden rules is to quote in full, or in a way that will not distort what other people said or meant.
If you had quoted what I have written to Hans, every one here would have understood that I agreed with the introduction of “safety valves” (like I said, I do it myself), and that I simply underlined the fact that “safety valves” were subjective criteria, while our friend Hans was stating his system is totally unbiased. No, it isn’t, it is subjective as mine is. It was clear and simple. No need to confuse this issue again.

As to the lack of precision of some of my data, and as to my ability to accurately judge performance, you are entering “Terra Incognita” again. But I won’t make a stand about that. At this stage, it is not as important as you want us to think it is (please read my comment on that in my forthcoming reply to Roger Clark's important questions; I hope to post it later today or tomorrow morning). I would only be too glad if that accuracy could be checked and its errors duly pointed and corrected. I’ll gladly put that accuracy to the test.

And speaking of test…

Originally posted by HDonaldCapps
... reading your rebuttal to my earlier comments -- nothing on Lee Petty I noted,


I have said nothing on Lee Petty? I think that remark is amazing, coming from someone who was not capable or willing to refute a single one of my arguments (apart from the "art vs, science", or "emotion vs. numbers" rhetoric). :rolleyes:
You’ll forgive me, Don, if I will answer ideas only, and not “braking tests” and little tricks destined to test my knowledge. That’s what the Lee Petty excursion was about (by the way, why not Richard, “the King” himself, or the more ephemeral Swede Savage?). To use your own words, “spare me”.

Originally posted by HDonaldCapps
The bedrock of my oppostion to this scheme is that it is just numbers and the crunching thereof. Spare me. We need historians to be exploring the many unanswered or murky or uncertain questions of racing lore, not generating yet more reams of numbers in a field where the numbers types seem to be successfully imitating kudzu with generally the same level of success.


I will not return to the “just numbers” thing again. As for historians “to explore the many questions of racing lore”, I’m afraid I’m not your man. I’m not a motor racing historian; just someone who’s been following it since the late 1950s/early 1960s, and which, by coincidence, is an historian by profession.

Originally posted by HDonaldCapps
By the way, you have two lists with different people at the top of each one -- which is "THE" list? Not that it matters to me, but others might care.


There is no such thing as THE list. There are different lists that, COMBINED, may give you a fair and balanced way of looking at things. Please excuse the word fair, again, but no other synonymum occurred

Regards

#38 jpm2

jpm2
  • Member

  • 121 posts
  • Joined: December 04

Posted 23 October 2005 - 19:03

Originally posted by Hans Etzrodt
Thank you for... but I have no desire to further discuss my flawed listing because I just lack the interest to do so. Long time ago it has provided me with the answers I wanted.


Ok ;)

Originally posted by Hans Etzrodt
There was a time in the past when I was really interested in this driver comparison subject and when I had spent way too much time on this project before even obtaining the necessary data for each driver who ever won a Grande Épreuve. But this was a good exercise and not really a waste of my time. Then – after several attempts I used the system posted now at TNF thread –here discussed– with an aim to remove all bias and strive for fairness to all drivers. To compare the Grand Prix drivers, I very soon realized that I could only apply Grandes Épreuves and thereby eliminate all other Major Grands Prix, whose status is another controversial matter. While compiling the data I realized the need to set a limitation in order to avoid absurd results. Therefore I decided to only accept drivers into the list who had participated in 12 or more Grandes Épreuves. By doing so, MY system and yet unborn list was already flawed from the onset and I had no fix for it. I had to accept this shortcoming because I knew, at the time, that this was the best compromise I could reach after my beginning trial and errors period. I had seen before other driver comparisons in magazines of the thirties and forties and even earlier comparison attempts but also some years ago in the nineties magazines from Autosport and MotorSport. Although all were quite entertaining, each of those systems and lists contained one or the other fault and none was perfect. So, my list was going to be just another blemished and controversial exercise.


No need to explain, Hans. I do it myself. It’s the only reasonable way, in my view. Yours was a subjective way, but not a flaw. Nothing is perfect and subjectivity is all around us. What matters is the level and the nature of that subjectivity.

Originally posted by Hans Etzrodt
In answer to your system I can only state again that it is way too complicated, inaccurate and time consuming for me. I could never do this and you have my utmost respect for your stamina and perseverance. It is definitely a brave effort to search and establish new criteria although some of those are far removed from the established understanding of this matter by motor racing historians, which will obviously get you in hot water. In General you have my praise for your courageous labor. Lots of luck.


Thanks :up:

#39 roger_valentine

roger_valentine
  • Member

  • 208 posts
  • Joined: October 02

Posted 24 October 2005 - 09:16

Originally posted by HDonaldCapps
....which would mean Lee Petty, of course.

No Don, it couldn't be him. He never took part in any important races! Looking at the 'highest number in F1' thread, I see that 42 could also be Tom Pryce, David Purley or Ian Ashley. As I live in Bognor Regis, perhaps I should plump for David Purley.

Originally posted by jpm2
The happy conclusion is that discussion can continue. The horrible “system” of jpm2 hasn’t kill discussion after all. All happy members will be able to discuss whether it is Schumacher or it is Fangio


Of course the jpm2 system doesn't end all discussion - like any artificial ranking of drivers (official or otherwise), it actively stimulates such discussion. But hey, this is TNF, where discussion is the name of the game, we don't require stimulation. (Perhaps I should rephrase that)! I just find it unfortunate that, in this thread, the discussion has remained fixed on 'the system', rather than wandering OT into topics thrown up by the application and results of that system.


(According to my completely objective system for ranking TNF threads, this one is running 17th in the 'number of posts before getting totally OT' catagory; 7th in the 'wordcount divided by number of posts' catagory; 3rd in the 'forcing Don to use long words' catagory; and, by my single-handed efforts, I have raised it to 22nd in the 'serious discussion punctuated by fatuous comments' catagory).

Now, about Alain Prost's sex life...

Advertisement

#40 jpm2

jpm2
  • Member

  • 121 posts
  • Joined: December 04

Posted 24 October 2005 - 12:01

Originally posted by Roger Clark
[B]I'd like to join those who have congratulated jpm2 on the courteous and articulate way he has explained his system and replied to his critics. It seems to me that he has a thorough and balanced understanding of what he is trying to achieve and of the limitations of the system. It is an interesting way of assessing drivers' performance and jpm2 has been careful to stress that it is no more than that. It is complementary to the essential analysis of what actually happened and why they happened in the way they did.

Thanks Roger for your comments and your good understanding of what was at stake here. :up:

Originally posted by Roger Clark
I quite like the idea of a contest in which the competitors don't know the rules until it is over. If it could be adopted in the official championships it might make people more interested in the races themselves rather than the gathering of points. It would be interesting to extend the idea to include categories of racing other than the prevailing Grand Prix formula, particular in years when drivers typically raced in several categories. Obviously, there would be major difficulties in gathering and analysing the data for such an extension.

I think I would not follow you on that “blind contest” approach. In my view the rules have to be there, to orient the competitors towards a well defined target, but, from the outside, as observers, we can utilize different analytical tools to “judge” or to “measure” their performances and results.

Originally posted by Roger Clark
I don't think that jpm2 has told us how much relative weight he gives to practice time, mid-race positions and final results. My apologies if he has.

I had, as a matter a fact, when I replied to Wolf (I think). No problem, here they are again:

Best practice times
Minimum level race: 4-2-1 points
Medium level race: 5-3-2 points
High level race: 6-4-3 points

Points awarded at each 50kms of race:
Mimimum level: 3-2-1
Medium level: 4-3-2
High level: 5-4-3

Final points:
Minimum level: 9-5-3
Medium level: 10-6-4
High level: 11-7-5


Originally posted by Roger Clark
In any case, there is a difficulty in doing this. A driver who starts at the back of the grid, but who recovers to a win or a good position may have produced a more worthy performance than one who was in the top three throughout the race, as long as all those in front of him didn't retire, of course. This was a point made by Simon Lewis and which jpm2 countered with the example of John Watson in Detroit. Yet the fact that there are circumstance under which the system could penalise a great recovery drive seems to be a flaw; how, for example, would it deal with Clark at Monza in 67, or Raikkonen in Japan this year?

There are several difficulties, not just that one, I’m afraid.
My concept of Grand Prix is a broad one, i.e., for me (and, I guess, for most of us), it is something that is not limited to the race, but extends to practice also. If for some reason, a driver is demoted to the back of the grid, I will consider it a Grand Prix incident, unfortunate maybe, but not unlike any other occurring in the race itself.
It’s true that some performances from the back of the grid are perhaps more “worthy” than those of the drivers who started from the front line, and held leadership from the beginning. It is true that, for a time, I considered the possibility of an additional criterion, a kind of “man of the race trophy” that would reward (with points, of course) the most outstanding drive in each race, regardless of whether the driver finished it or not. But, that criterion would have created more problems than it would have “cured”. In certain cases it would be very difficult to award; what about equal or very similar “outstanding” performances in the same race? And who would be the judge of that? A jury, like we have in Olympic gymnastics, for example? I’m afraid that it would introduce a high degree of subjectivity and that is what I am trying to leave outside “my system” as much as possible. In my view, the less subjectivity, the better.

As a general rule, “my system” doesn’t PENALISE a great recovery. It may not reward it as much as one would like, but I wouldn’t go as far as saying it penalizes (my term of comparison being the official criteria and point systems, of course). On the other hand, it can be said it penalizes “insufficient recoveries”, i. e., the recovery from the back up to 4th or 5th place, for example.

Now let’s see the case of Clark in the 1967 Italian GP

Points awarded in “my system” (Monza was a medium level race)
- Brabham 26
- Hill 24
- Surtees 16
- Clark 13
- Hulme 10
- McLaren 4

Official points
- Surtees 9
- Brabham 6
- Clark 4
- Rindt 3
- Spence 2
- Ickx 1

Apparently the system does not penalize Clark. He stands more or less where he was with the official system. But, the “system” rewards the great race of G. Hill (that received nothing in the official system) and also the performances of Hulme and McLaren (nothing gained in the official system also). It penalizes Rindt, Spence and Ickx, though.

Let’s go now to the 2005 Japanese GP

Points awarded in “my system” (Suzuka was a high level race)
- Fisichella 31
- Raikkonen 20
- Button 15
- R. Schumacher 11
- Coulthard 6
- Alonso 5
- M. Schumacher 5
- Webber 3

Official points
- Raikkonen 10
- Fisichella 8
- Alonso 6
- Webber 5
- Button 4
- Coulthard 3
- M. Schumacher 2
- R. Schumacher 1

In this case, the drivers who made a good recovery from the back are “penalized”, and perhaps Alonso is more “penalized” than Raikkonen. The benefit goes to Fisichella and, in a lesser degree, to R. Schumacher and Button as well.
It can be considered an exception to the general rule, and is typical of races where a driver that led most of the time, loses the race in the last lap or so. Other examples are the 1973 Sweden GP, or the 1971 Italian GP, or the 1964 Belgian GP.
But, it’s also true, that nowadays races have certain characteristics that tend to introduce some confusion in the system (namely, the 2, 3 or even 4 pit stops during a relatively short and close race, which may allow a midfielder to find himself in the top 3 sometime during the race).
Generally speaking “my system” rewards the ones that manage to stay in the top 3 throughout the race.

Originally posted by Roger Clark
I forgot: in 1964, the dutch Grand Prix was a medium coefficient race. Why was that?

My mistake, sorry Roger, I don’t know why I wrote “medium” instead of “high”. :blush:

The entries to both GP are the following (there are other drivers, like de Beaufort, for example, but, being outside the 25-best ranking, they don’t add points to the “quality” of the race:

MONACO
Jim Clark 6
Pete Arundell 5

Dan Gurney 6
Jack Brabham 5
John Surtees 6
Lorenzo Bandini 5

Graham Hill 6
Richie Ginther 4
Bruce McLaren 5
Phil Hill 4
Jo Siffert 4
Jo Bonnier 4

Innes Ireland 3
Trevor Taylor 2
Bob Anderson 3
Chris Amon 3
Mike Hailwood 2


Peter Revson 2

Maurice Trintignant 1
Total 76

ZANDVOORT
Jim Clark 6
Pete Arundell 5

Dan Gurney 6
Jack Brabham 5
John Surtees 6
Lorenzo Bandini 5

Graham Hill 6
Richie Ginther 4
Bruce McLaren 5
Phil Hill 4
Jo Siffert 4
Jo Bonnier 4



Bob Anderson 3
Chris Amon 3
Mike Hailwood 2
Tony Maggs 2
Giancarlo Baghetti 1



Total 71


So, both the 1964 Monaco and Dutch GPs are “high level” races. Anyway, your question could have been asked about other races. Why are some of them “high” level and others only “medium” level, when the respective entries were so similar? It is a problem of “boundaries” or “limits” between two levels. As you know, when one defines levels, a single point can make the difference.

It seems sometimes unjust that a little difference may promote or demote. But that’s what happens with levels. And, why 3 levels instead of, say, only 2 or maybe 4? After several trials, it seemed to me that 3 levels was the more balanced way to do it. But that’s just my opinion.
There were, of course, other possibilities. Instead of establishing levels, with the correspondent points, I could have used an immutable point system (for example, the one I’m using now for high level races) that I would then multiply by the numeric value of that particular race (or a decimal of that number). In the 1964 Dutch GP case, it would be something like this:

- Clark 45x71=3195
- Surtees 30x71=2130
- Hill 16x71=1136
- Arundell 11x71=781
- Gurney 6x71=426

It would certainly have been more “accurate”. Then, why don’t I do it like this? Mostly for simplicity’s sake. I felt it would be too complicated, that it would generate huge numbers, and that, in the end, it wouldn’t have changed all that much. It would have some weight on the total of points (and points per race, of course) but not in the added percentages criterion. In the 1970s, the initial “system” was designed in a much more complex way (it even considered points up to 5th placed drivers). As time went by, and after a series of experiences of trial and error, I have tried to make it as simple as possible without losing the essential. That was one of the reasons I focused on the first 3 drivers, only, and introduced levels (as it happens in some Rallyes championships).

At this point of our discussion, I think we can draw some conclusions.
First I would like to underline what we all agree on. I think all (or most of us) agree that “official stastistics” give us a distorted view of what went on (as far as performances/results are concerned).
And why?
- Because they do not consider a lot of sometimes highly participated races that were organized on an extra-championship basis.

- Because they seldom consider corrective factors, and they simply add very different things. It’s like if we were adding major currencies of different eras without converting them. Adding the number of laps a driver led makes no sense, because, as we know, a single lap in the Nurburgring is equivalent to 5 laps in Estoril; adding points gained during a career is also a distortion because those points changed throughout time; the same with victories or poles, when everyone knows that the 1955 Championship had 6 races only, and modern championships have between 16-19 races each year; and so on.

- And, of course, because they don’t consider what went on before 1950.

We all know this, of course, but the general public doesn’t and tends to accept this highly biased image, that enters his house via TV or motoring annuals with the sanctioned name of “statistics”.

“My system” tries to balance things a bit more, through the use of “corrective lens” (this may interest Doug Nye).
Is it a perfect system? Surely not, there is no such thing as perfection. But, is it a better one? I think it is. Imagine a kind of Mr. Magoo. The ideal for him was to find glasses that could give him a 20/20 vision accuracy, instead of those 6/20 he now has. Unfortunately, those glasses are impossible to get. Through perseverance, Mr. Magoo manages to get a pair of glasses that allow him a 15/20 vision accuracy. Wouldn’t they be better than the pair he’s using now?
My system tries to be Mr. Magoo new pair of glasses. And is it really?
It has been said here that the “system” has fundamental flaws, and that it is inaccurate. I think it is useful to make a clear distinction (what Don Capps called philosophical and technical issues) in order to separate problems with different natures and priorities.

The first questions, and the more important ones, are the following: is “my methodology” valid? Is the system well conceived? Is it coherent? are there inner contradictions in it?

This is the most important level of discussion. It’s like when one is planning or designing a house. It is important that the infra structure is well conceived, otherwise the building will come down, regardless of what we may put inside it.

If we think the system is invalid, there is no point in proceeding. But if we consider it to be valid, then it’s time to face and to discuss another kind of problems, precisely those that have to do with the application of the model.

It’s only here that the problems of “accuracy” or “ignorance” presents themselves. Could it be that I made some (or many) errors? Was I careful? Were my sources reliable? Is there too many question marks?

I suppose everyone knows what my sources are. And I also suppose that everyone knows, I admit there could be errors, and that there are gaps to be filled (because practice times in the 1930s and 1940s are, sometimes difficult or impossible to obtain).

But this has little to do with methodological issues. Errors can happen, they are detected and they may be corrected. In a built up phase they don’t have the importance some here think they do. If I can use my house example again, they are rotten bricks, that can be easily removed and replaced by sound ones. That will not be a menace to the house. That will ultimately depend on whether the building was well projected, or not.

Anyway, in my next message I will present the list of the things I do not know, of the gaps that have to be filled in my championship (1932-2005).

Regards

#41 Wolf

Wolf
  • Member

  • 7,883 posts
  • Joined: June 00

Posted 24 October 2005 - 12:18

Jpm2- what do You reckon about system involving 'rankings' I mentioned in my last post?

As for better evaluation of 'recovery' drives, maybe one could incorporate some points for number of overtaking manoevres during the race (it would be arbitrary to determine how much points- say 0.5 for one overtaking*)? E.g. try that on '67 Monza race and guesstimate how things look there...

#42 jpm2

jpm2
  • Member

  • 121 posts
  • Joined: December 04

Posted 24 October 2005 - 16:57

Sorry Wolf, yesterday I just couldn't find the time to reply to your message.
Here's some thoughts about it.

Originally posted by Wolf
Speaking of rankings, I just remembered a friend of mine has made a system that was (unfortunately briefly) employed in GPL community. It was similar to chess and FIFA rankings, and the idea might be of some use for purposes like this. With GP racing one could easily use WWI or WWII as starting point, because of preceeding breaks. But back to basic idea behind it...

This system is based on 'rankings' (at start everybody is awarded same amount of points, as are newcomers later on), and one earns or loses points according to whom he has beaten (in this case finished in front of) or lost to (finished behind). For example, Moss would earn certain amount of points for finishing in a race in front of Fangio (more than Fangio would earn by finishing in front of Moss), but say Trintignant would earn more points than Moss because he'd presumably be lower ranked than Moss and his feat would hence be greater. System has further features like that older scores 'fade' with time, so that e.g. beating Graham Hill in say '63 would earn one more points than beating him in '73...


It seems a good idea, but quite different from my own (philosophicaly speaking).

For me, one of the main objections is that it considers final positions only, and that could open the door to great "unfairness".
Other important objection is that the ranking seems to be based on fame or previous year performances and status (or, at least, I suppose it is). If it's so, it doesn't consider what we might call "falls from Grace". Take Andretti, for example. I imagine his ranking in 1979 would be very high, being the World Champion and all that. Nevertheless, in 1979 Andretti was not a force to be considered. The same can be said of Scheckter in 1980, and many many others. In "my system" Fangio is ranked BELOW Moss and several others in 1958. What would it be like in your friend's system?

I may be wrong but I felt your friend's system was based on the prestige of each driver.
As you know I have made another choice and my ranking refers to the season itself, and is a "tight" one. Moss and Fangio were top ranked drivers in 1955-57.

The idea of "scores fading with time" is curious but would introduce new grounds for subjectivity, and is wide open to critic because, as we know, directly relationed with aging. Drivers often go through periods of relative eclipse and rebirth. Take Brabham, for example. In 1964-65 it seemed he was decaying with age, but he come back strongly in 1966-67 and again in 1970.


Originally posted by Wolf
This method incorporates 'weighting' system for races similar to one jpm2 uses but is IMHO more objectively and 'up to date'.

P.S. He has also used 'weighing' factor for cars (I was involved with proposing and implementing it) and tracks*, which was relatively easy in terms of GPL and could pose a problem in 'real life'. But simpler and cruder method for cars could probably be used so that beating Fangio in Merc yields more points than beating him in P15.


Forgive my ignorance, but what is IMHO? :confused:
"weighing" factors for cars seems a good idea, :up: but why separate it from the ranking of driver? I never value the driver by himself. Always the binomium car/driver

Originally posted by Wolf
* we made coefficients based on World records at the time to evaluate those things- but the idea was that one would earn more points for e.g. driving a '67 Cooper and beating Eagle at Spa than at Monaco... [/B]


Depends on who was driving the Cooper and the Eagle...;)

Originally posted by Wolf
As for better evaluation of 'recovery' drives, maybe one could incorporate some points for number of overtaking manoevres during the race (it would be arbitrary to determine how much points- say 0.5 for one overtaking*)? E.g. try that on '67 Monza race and guesstimate how things look there...


Wolf, If I were you, I would drop that criterion. For several reasons. Just consider a dominant driver that takes the lead from the pole, he overtakes no one, he's so superior that he's never hard pressed, and he overtakes no one (Clark in 1965, for example). No points? :confused:
Then, please consider that overtaking manoevres would be impossible to determine in most GPs, even in more recent times. In a slipstrem race, for example, like Monza in the late 1960s or early 1970s.
And, finally, there's more to it than overtaking.

All the best

#43 Wolf

Wolf
  • Member

  • 7,883 posts
  • Joined: June 00

Posted 24 October 2005 - 19:15

Jpm2- thanks for reply. :) Regarding the prestige- as You call it, it's directly related to results, nothing arbitrary, so Your example of Andretti would soon in 1979 be reduced by loosing to other competitors. Because when some people earn points/prestige, it's at the expense of other drivers (in this case Andretti). And rank/prestige is adjusted after each race. As You say, the trick is finding the ballance beteen past results and current form, but that's pretty easily done*. E.g. I think FIFA algorithm gives a bit too much weight on previous results...

* the whole process is automated, it's only a matter of feeding the results into the program, and then results are listed after changing parameters

BTW, this system is not incompatible with Yours- it can be arranged that 'splits' are treated similar to final results (as desribed) but with smaller coefficient, as well as practice times...

Regarding my last suggestion, it's meant as possible addition to Your system, but it's also arbitrary in defining the parameters- quantifying whether say drive from 20th to 5th in a race is worth as say driving the whole race in 4th position. If so, give a points for 4th/15 for a spot gained and add to final score you obtained with other methods (quali, splits, final result). Obviously, I haven't thought it through ( :blush: ), because one would need penalize being overtaken or something to that effect. Also, I didn't mean every single overtaking because two drivers swapping places the whole race could earn a lot of points, but positions gained in longer intervals (say laps if lap charts are available, or between Your 'splits', or even from start to finish).

P.S: IMHO means 'in my humble opinion', it's an Internet abbreviation like BTW (by the way), IIANM (if I am not mistaken), IIRC (if I remember correctly), &c

#44 HDonaldCapps

HDonaldCapps
  • Member

  • 2,482 posts
  • Joined: April 05

Posted 24 October 2005 - 19:37

First. I must repond to this from Roger Valentine:

quote:
--------------------------------------------------------------------------------
Originally posted by HDonaldCapps
....which would mean Lee Petty, of course.

--------------------------------------------------------------------------------
No Don, it couldn't be him. He never took part in any important races!


Well, that is a matter of opinion with which I happen to disagree. Despite not having participated in what might be considered any "important races" -- which given the Eurocentrism of the forum does not require much expediture of gray matter to break the code as to what those "important races" might be -- I certainly have no problems with placing of Lee Petty high in the ranks of racers, Period. I have never understood the idea that the "The Best" could only come from the ranks of Grand Prix or Formula 1 drivers.


Second. I am begining to think that the last thing I want to do is provoke another epistle from Joao Pedro, but.....

.....I really find myself becoming more and more convinced that your championship could certainly never be mine. I could really care less about the validity of your selection criteria, the precison of your algorithms, the quality of the data you used, or even the acceptance of your results by the community.

Why? Because I simply like the old, muddled way of viewing the motor racing world. I think what you doing is fine for some, but an utter waste of time and effort for what I do and how I see the world. I am sick unto death of such schemes which reduce Jim Clark to "18,7171" or "148,722"; Stirling Moss to "135,522" or "14,407"; Tazio Nuvolari to "12,119" or "114,181"; Louis Chiron to "56,297"; or Dan Gurney to "67,812".

This system has no "soul." It is simply a ruthless crunching of numbers that spits out a list of results. We are inundated with lists and numbers and data today and we seem to be often not a lick wiser or more informed as to the past of motor racing.

Perhaps it is simply an instinctive reaction to yet another way to continue to depersonalize the past and reduce it to shadows in the form of numbers.

I am now going to bow out, sit back, and watch any further debate and discussion because I have said more than I intended already.

#45 jpm2

jpm2
  • Member

  • 121 posts
  • Joined: December 04

Posted 24 October 2005 - 20:23

Originally posted by HDonaldCapps

I am begining to think that the last thing I want to do is provoke another epistle from Joao Pedro, but.....

I simply like the old, muddled way of viewing the motor racing world. I think what you doing is fine for some, but an utter waste of time and effort for what I do and how I see the world. I am sick unto death of such schemes which reduce Jim Clark to ...

it is simply an instinctive reaction to yet another way to continue to depersonalize the past and reduce it to shadows in the form of numbers.


Yes, I know... it's hard to get rid of it (that's why it's called an "idée fixe"). :rolleyes:
So, and although the word reduce is still around, there will be no more epistles from me, Don.
I respect your views. :up:

Regards


PS: I think roger_valentine was trying to be funny :lol: with his "Lee Petty/important races" charade. He was trying to hit me, not you.

#46 Ruairidh

Ruairidh
  • Member

  • 1,074 posts
  • Joined: November 02

Posted 24 October 2005 - 20:41

[i]
Second. I am begining to think that the last thing I want to do is provoke another epistle from Joao Pedro, but.....

.....I really find myself becoming more and more convinced that your championship could certainly never be mine. I could really care less about the validity of your selection criteria, the precison of your algorithms, the quality of the data you used, or even the acceptance of your results by the community.

[/B]

Agreed on both counts, except the more I read the less I can accept this as anything more than a fun way to retrospectively award points for performance in a particular season. And that deserves our applause. But what's all this with the hyperbolic tag "Alternative Championship" and Fangio and Schumacher being the greatest ever. Hmmmmmm........

I'm a big fan of the work many sabermatricians have done to use the language of stats to reveal more of what we see with our eyes in Baseball, especially in the sense of comparative performance. That said I think this approach, or rather the overblowing of what this approach tells us, would be ripped to pieces by that community.

Why? Three reasons (a) way too much judgment on what data to use and how to weight it (b) too remote a connection between the performance data used and the persons objective at the time of performance and © unreconciled misalignments between the headline focus on individual drivers and the variables that comprise the car/team/driver package. And I don't see how you overcome (b) and haven't seen a real attempt to get to the bottom of ©.

And look, even in baseball, where the relationships lend themselves far more to a stat based approach (e.g. between individual and performance and intent at the time of performance) the simple fact is that the numbers do not tell the whole story - for example they inaccurately suggest that Derek Jeter is not the greatest shortstop of recent times ;)

#47 Wolf

Wolf
  • Member

  • 7,883 posts
  • Joined: June 00

Posted 24 October 2005 - 21:32

Don & Ruairidh- the systems we discuss are not even intended to be definitive on the subject of scores and points, let alone to provide insight into history...

I'd dearly like to have insight ito one race, let alone history from 1930-2005, or what not. Just consider what has to be put on record to paint the whole picture. No two cars are alike, and no two engines either... Is a Cooper-Climax identical to the one next to it in the pits? The difference migh not even be a type number but a Colotti gearbox that may cost the driver the whole championship. I remember reading Jack Brabhams reports from '65 about picking engines for him and Dan Gurney, trying to 'stay afloat'... No table with results I've seen has column for remarks, in which one could insert comment like 'nursing a sick engine from lap 5 onwards'. Your '61 series comes close to that, but even Web has it's limitations.

Methods like jpm2's only provide a glance of crtain things, from a certain point of view. Noone of us is fooling themselves that it is anything more than that. What we're trying to discuss is whether one way of looking at bunch of results is better than another- which of the number-crunching methods would produce a most balanced/fair view of that aspect. It does not mean we cannot (and indeed do not) appreciate e.g. the photo in 'Cruel Sport' of Jack Brabham's stare in pits at Nürburgring (IIRC) after his Brabham failed in first outing and what he must've thought back then...

#48 jpm2

jpm2
  • Member

  • 121 posts
  • Joined: December 04

Posted 25 October 2005 - 08:32

Originally posted by Wolf
Don & Ruairidh- the systems we discuss are not even intended to be definitive on the subject of scores and points, let alone to provide insight into history...
Methods like jpm2's only provide a glance of crtain things, from a certain point of view. Noone of us is fooling themselves that it is anything more than that. What we're trying to discuss is whether one way of looking at bunch of results is better than another- which of the number-crunching methods would produce a most balanced/fair view of that aspect. It does not mean we cannot (and indeed do not) appreciate e.g. the photo in 'Cruel Sport' of Jack Brabham's stare in pits at Nürburgring (IIRC) after his Brabham failed in first outing and what he must've thought back then...


Well said. :up:
Now, Wolf, let me thank you for your lesson in internet abbreviations. How stupid of me. :

Originally posted by Wolf
Regarding the prestige- as You call it, it's directly related to results, nothing arbitrary, so Your example of Andretti would soon in 1979 be reduced by loosing to other competitors. Because when some people earn points/prestige, it's at the expense of other drivers (in this case Andretti). And rank/prestige is adjusted after each race.


Your idea of adjusting rank after each race is good, but I think the concept of negative points can jeopardize the whole idea, especially because it is based on final results only. If you could have a criteria of “A beats B” instead of “A finishes in front of B”, it would be much better.
You see, Wolf, your ranking system may work reasonably well with nowadays races (and even with nowadays races we all know how a criterion based only on finishing positions can cause serious trouble because it determines the pecking order for next race practice). Anyway, for races of old, when mechanical failures were more frequent, that criterion would cause a see-saw effect on the ranking, don’t you think? Just imagine D. Gurney in 1965. He was the first to retire in the South African GP; he was 12th in the RoC; 9th in Goodwood; in didn’t show up in Silverstone, nor in Monaco; he finished 10th in the Belgian GP and he was the 4th driver to retire in the French GP, and he finally managed a 6th place finish in the British GP. Well, what would his ranking be by mid-season? Similar exercises with other drivers will show us similar “distorted” images. Take Clark in 1966, or Peterson in 1976, for example.

I would have to see how it works, and I would have to test it in the 1930s, 40s, 50s, 60s, and so on to see if it distorted them. Remember, the MAIN goal is to find a set of criteria that can be “fairer” than official statistics are and that can be applied to different epochs without MAJOR distortions in any of them.

Originally posted by Wolf
BTW, this system is not incompatible with Yours- it can be arranged that 'splits' are treated similar to final results (as desribed) but with smaller coefficient, as well as practice times...


You’re right. It is compatible with “my system” from a technical point of view. It could even make it better, although more complicated. But, the main philosophy is quite different, and philosophic “systems” should not be mixed; each one of them needs its own coherence.

I sincerely feel you should try to develop your friend’s system.

Regards

#49 jpm2

jpm2
  • Member

  • 121 posts
  • Joined: December 04

Posted 25 October 2005 - 10:29

Thank you for your comments on the methodology of “my system”. :up: That was precisely what I was looking for at this stage of discussion. Now let’s see it’s contents (I hope you’ll forgive this long reply and a bit of irony here and there)

Originally posted by Ruairidh
... the more I read the less I can accept this as anything more than a fun way to retrospectively award points for performance in a particular season. And that deserves our applause.


Good. “My system” is no big deal, certainly not one of the writings of the Fathers of the Church or something like that. It is retrospective (what is related with history, generally is :rolleyes: ) and won’t change our lives and our future. And thank you for the applause. :up:

Originally posted by Ruairidh
But what's all this with the hyperbolic tag "Alternative Championship" and Fangio and Schumacher being the greatest ever. Hmmmmmm........


Maybe you didn’t have the patience to read all the messages, and you let this thing fool you. As a matter of fact I had already explained that the expression “Alternative Championship” was not my choice. Here’s what I wrote in my 22nd October reply to Don Capps: “The expression “alternative championship” is not my own. I opened this thread with a different name “my championship”, “my system”, or something like that. Afterwards, that title was changed (by the editor, I suppose) into “alternative championship”. In my view it was not a good choice because it conveys the idea that “my system” is meant to substitute the actual championship as it exists. It is not, and it never was.

Originally posted by Ruairidh
I'm a big fan of the work many sabermatricians have done to use the language of stats to reveal more of what we see with our eyes in Baseball, especially in the sense of comparative performance. That said I think this approach, or rather the overblowing of what this approach tells us, would be ripped to pieces by that community.


What a great pity you are not yourself a member of that community of “sabermatricians”. If you were, you would certainly show us how my system would be “ripped to pieces”. As it is, I guess we’ll just have to keep on waiting for the arrival of those “sabermatrician” intellectual lions.

And why would those intelectual lions rip the system "to pieces"? For 3 reasons, you tell us:

Originally posted by Ruairidh
(a) way too much judgment on what data to use and how to weight it


I don’t think there can ever be "too much judgement" about anything. But, suppose there was. Who would be the judge of that dangerous “excess” of thinking? Kant, Hegel, or… maybe that terrible community of “sabermatricians” you told us about? :rolleyes:


Originally posted by Ruairidh
(b) too remote a connection between the performance data used and the persons objective at the time of performance


Wrong (in my opinion, of course). The connection is a very close one.
This is no baseball (no offense intended). This is no pedestrian race either, where one can guard himself to the last turn or to the last straight. This is not even team work in the general accepted sense of the expression (your team mate is precisely the first one you want to beat). There are no friends on the track (well, perhaps Collins and Hawthorn and a few more). This is “war”, with men surrounded by metal beasts, and this can kill you (well, at least it could, in not so recent days). We are entering here those aspects of Grand Prix racing that don’t show in the statistics, but Doug Nye is surely right when he used the word “war” in that context.
And this (i.e., Grand Prix racing) has its own spirit. The spirit to excel, to take the lead, to aim for fame and triumph. It is a very ancient spirit and we can find it in several human activities, particularly in war. In the Sixteenth Century Portuguese captains run madly in the open fields of India, sword in hand, and way in front of their men, because they wanted to be the first to meet the enemy, the first to escalate the walls of sieged cities, because they wanted their names registered for eternity in the Chronicles of the era. Most of them were easily butchered, but their courage enobled their families and their names remained with us. Grand Prix drivers also want to excel, they want to lead, they want to go faster. There are exceptions to this rule, of course, but this is the prevailing spirit. Take Fangio, for example, a man who was known to control his pace and to win by the minimum reasonable margin. Well, Fangio was up in front line of the starting grid and he was one of the top 3 at 50kms stage in most of his races
Take Clark, take Senna, take Schumacher, and so on, and you’ll find similar results.
In Grand Prix racing, the best drivers usually want to lead from the start, they want to beat everybody else, in the race and in practice too, they want to beat themselves, they want to go “faster”, as the late George Harrison sang in a famous song (precisely dedicated to Grand Prix racing).

The motivation will be the same in an extra-championship race, because drivers are intrinsically competitive persons, who are keen on taking risks, facing challenges, and crossing their limits. This is shown in a lot of examples of hard fights (take the example of Clark vs. Gurney in 1965 Roc), and even mortal accidents in non-championship races.

“The persons objective at the time of performance” is, usually, to lead, to be fastest in practice. And that “objective” is well connected with the performance data I use in “my system”.

Originally posted by Ruairidh
© unreconciled misalignments between the headline focus on individual drivers and the variables that comprise the car/team/driver package.


You have surely heard of centaurs. The driver as an individual, as a “person”, is not part of this statistic game (and perhaps that’s why several TNF members reacted so strongly). There is never the driver OUTSIDE what you call “variables”. It is always the binomium car/driver, the headline focus is not on the “driver”; it is always on the centaur. An entity that exists in the track and at a particular moment only.

Originally posted by Ruairidh
And I don't see how you overcome (b) and haven't seen a real attempt to get to the bottom of ©.


b) is overcomed (I think). As for c) I’m not sure I understood entirely your cryptic language. Maybe you could rephrase, in case my answer went out of target. From what I got, I think it is overcomed too. I don't want to get to what you call "the bottom" of that issue. that "bottom" is unattainable and the road to it would make the "system" too complicated.

Originally posted by Ruairidh
... the simple fact is that the numbers do not tell the whole story...


Of course not. No need to repeat the obvious. The “whole story” is God’s territory. Do you by any chance happen to know anyone who knows the “whole story” of the battle of Gettysburg? Or anyone who knows his own “whole story”, for that matter?
Let’s us leave false problems out of this, shall we?

Regards

#50 HDonaldCapps

HDonaldCapps
  • Member

  • 2,482 posts
  • Joined: April 05

Posted 25 October 2005 - 14:02

Originally posted by jpm2
quote:
--------------------------------------------------------------------------------
Originally posted by Ruairidh
... the simple fact is that the numbers do not tell the whole story...
--------------------------------------------------------------------------------

Of course not. No need to repeat the obvious. The “whole story” is God’s territory. Do you by any chance happen to know anyone who knows the “whole story” of the battle of Gettysburg? Or anyone who knows his own “whole story”, for that matter?
Let’s us leave false problems out of this, shall we?


Casting aside, for the moment, my earlier intention to butt out of this conversation, I damn near threw the computer on the floor and jumped up and down on it while wearing my protective vest (with all the plates) and kelvar to add as much weight as possible to my impact and saying very unpleasant things when I saw the accepted purpose of history being airily dismissed as a "false problem".......