Question for the statistically minded.
Posted 16 October 2001 - 03:09
Anyway, for the last couple of years I have been collecting statistics for the season. However there are a couple of statistics that I am unsure about.
One is finishes, if a driver retires, say five laps from the end of the race, but is classified as a finisher, would you put down that the driver had a retirement or a finish (or both)?
Another one is at the start of the race. If a driver stalls on the grid and must start from the back, would you record his original qualifying position or that he started from the back?
I think there are some more stats I'm confused about but for the moment I can't remember them!
Posted 16 October 2001 - 05:01
Posted 16 October 2001 - 18:11
As for qualifying positions and starting positions, these are two seperate items in my book.
Posted 16 October 2001 - 21:06
The Left-footed Baptists versus the Right-footed Baptists in the Piedmont of South Carolina many years ago: During the footwashing ritual that was a part of their services, one group thought the left foot should be washed first, as had been the case for many years. Another group later determined that it should be the right foot which should be washed first. After a period, those who believed that the right foot being washed first split off and created their own church. Although the two groups remained cordial, shared many beliefs, and continued to work many community services together, this one relatively small issue became a dividing point between the two groups.
And so it is with statistics such as those collected in conjunction with motor racing. There are many examples to use in building your statistical database: the Paul Sheldon format for Grand Prix and Voiturette racing; the Phil Harms format for US the National Championship series; the Greg Fielden format for the NASCAR Grand National series; and many, many others.
One important aspect of doing your own statistics for motor racing is the tactile manipulation of the data. My view has always been that you really have to format your data to fit the modality which suits you -- that is, what works for you.
I lean towards something more akin to the Sheldon model, but essentially breaking it down into what I what to see when I peruse race data:
-- circuit and event data: event name, circuit data, race distance, date and other items such as the purse or whatever
-- qualifying or grid data: ordinal formation of qualifying or practice times, with times of those who DNS but recorded times during practice or qualifying and a notation
-- results (an area which I have modified over the years back to where it was decades ago): the 1st line has position (1st, 2nd...14th, 15th), driver(s), entrant, sponsor or car name, marque (chassis no.); 2nd line has the laps completed, time data or reason retired
-- notes: exactly that, notes on the event that are relevant to the "story" of the race with absolutely no consistency from event to event... ... and covers the name of the promoter or organizing club if known, additional information on a specific individual such as a race fatality or etc. or if there is question about a reason for retirement, chassis or whatever, essentially neat tidbits about the event
-- if there is a championship involved, championship standing
-- then references for the data
It must be noted that the various ways in which individual clubs or sanctioning bodies have determined "finishers" or those "classified" is a thread in and of itself. The lack of a common standard applied commonly or in a standarized way will trip you up severely if you aren't careful.
One small but significant problem with Paul Sheldon & crew was the use of contemporary means of "classification" to events which were using other means of determining finishing orders or how the results were arranged or expressed. These conscious "errors" were blindly copied by the multitudes and have often passed into general usage in this day and age. Forix and others have accepted these and so be it to generation after generation it seems.
Although statistics are important, in my view they serve little purpose unless it is as a means to "tell a part of the story" or provide a record of an event for others to either build upon or correct.
At any rate, Ossiefan, unless you are using the building of the database to maximize your familiarization with the information -- assuming it to be GP or "F1", your time would be better spent peering into other series or formulae if possible since the fascination with "F1" statistics has all but blotted out that of other series or forms of racing. (Gosh, that was an awkward sentence!) However, that is far easier said than done as many here will tell you.
The current 90% rule came into being in F1 during the 1966 season. In 1962/1963, the CSI applied a 2/3's rule moreorless uniformly for GP events. The AC de Monaco had its own 50% rule for years prior to the CSI getting involved. Anyway, I am sure that many others can offer their sage advice as well.
Posted 17 October 2001 - 02:51
[Quote]"At any rate, Ossiefan, unless you are using the building of the database to maximize your familiarization with the information -- assuming it to be GP or "F1", your time would be better spent peering into other series or formulae if possible since the fascination with "F1" statistics has all but blotted out that of other series or forms of racing. (Gosh, that was an awkward sentence!) However, that is far easier said than done as many here will tell you."[UnQuote]
What do you mean exactly?
Posted 17 October 2001 - 03:16
if Schumacher had 99 poles (it might happen you know! ) , qualifies 1st to get 100, but stalls on the grid and must start from the back, would you say he has 100 poles?
Definitively yes ! For me a pole is the best time set during the qualifying session. So, except if the driver is disqualified after the session, his pole is counted through the records once the session is over. In F1, if he stalls before the warm-up lap he will start last, but no one will take his place on the number spot on the grid. It is not the same at Indy for instance. When the poor Scott Brayton killed himseld some days after having clinched the pole, Tony Stewart who had second best time started from the pole and Danny Ongais who substitued for Brayton started last. Still, I guess that even there no pole was recorded for Stewart and one was for Brayton.
But all this is for drivers career stats. When I do race-by-race summaries I would record the effective starting place of each, eventually marking by an asterisk drivers who did not start on the spot they had qualified at (which happened quite often in motorsport history). As said Fines :
As for qualifying positions and starting positions, these are two seperate items in my book.
Posted 17 October 2001 - 18:20
I really don't care very much for all the ju-ju driver statistics that so many are ga-ga over. I am more a context person so doomed to being over in the far corner with the very few others who always say, "...but, let's put that in its proper context..." and are pounded to mush by the If It's a Number, Its Important School.
While numbers do have an importance, Mr. Clemens did put that concept into its proper perspective.
As the late and lamented Roseanna Roseannadanna always said, "Whatever."
Oh, the Scott Brayton case for who was on the pole or the Jacques Laffite case of how many starts he made are two good examples for how the statistical game can lose the gist of what happened: the pole winner was not the polesitter on race day -- to me Brayton was the fastest qualifier but a DNS and the notes would reflect that; and although that start wasn't a "real" start that was a real broken leg -- to me it was a real start.
Posted 17 October 2001 - 18:43
By the same token, I always credit Mike Thackwell with starting the 1980 Canadian GP. I don't care that it was subsequently ruled that the first start didn't take place. It did. I saw it on TV. If that wasn't a start what was everyone doing? All the competitors, officials and spectators knew the race had started. Someone's subsequent ruling that it didn't happen can't alter the fact that it did!
Only in a contrived "official" listing can it be said Thackwell did not start that race.
Posted 17 October 2001 - 20:02
Another excellent example of statistical frivolity overwhelming reality: like you, to me that was a real, genuine start for Mike Thackwell.
Unless the stats are tied to the story, they are essentially just numbers. Motor racing is a volitile environment where the "truth" changes a great deal. Stats as usually rended rarely convey much about the event itself, unlike one of the few sports where the stats actually have meaning -- American baseball.
Not long ago, I found the scoring charts I did for games (Giants v Cardinals, Dodgers v Phillies, Yankees v Indians) I did while we visited New York (and our old neighborhood) when I was about 10 years old. Using the charts, I could replay each game for my son and daughter. They were amazed at how easy it really was once they understood the information. We all scored a minor game later and we re-played later -- and proved that we were all at the same game!
One of the great ways I found to get "involved" when I first started going to races in the mid-50s was to run a lap chart and make my notations on the chart. Many others here have done the same I am sure. It really makes a difference in what the stats show: sometimes a "minor" placing is the result of a herculean effort -- Hill with a chassis almost literally broken in two at the 68 Canadian GP or Clark at the 62 German GP coming through the field -- or how the real battle was down in the field or that so-and-so was winning/running well when the car broke. Stats in motor racing rarely capture that well, especially the ju-ju stats.
Also, some of the CSI/FIA statistical notions in later years border on the silly -- the Lafitte and Thackwell cases being just two examples. My own take is that stats have become more "important" as those involved have become....well -- less.
Posted 17 October 2001 - 21:16
Having said that, I do find statistical analysis great fun, but only - only - as an adjunct to reading about a race/season, never as a substitute