Jump to content


Photo
* * * * - 8 votes

Qualifying Averages Between Team Mates


  • Please log in to reply
356 replies to this topic

#1 tyker

tyker
  • Member

  • 209 posts
  • Joined: April 11

Posted 01 May 2011 - 13:11

1. Liuzzi 0.86s
2. Vettel 0.55s
3. Rosberg 0.4s
4. Kobayashi 0.38s
5. Alonso 0.37s
6. Glock 0.24s
7. Kovalainen 0.2193s
8. Hamilton 0.2187s
9. Petrov 0.17s
10.Buemi 0.16s
11.Barrichello 0.12s
12.Sutil 0.03s

Advertisement

#2 Andy865

Andy865
  • Member

  • 2,447 posts
  • Joined: January 10

Posted 01 May 2011 - 13:12

Oh good.

#3 Dunder

Dunder
  • Member

  • 6,784 posts
  • Joined: April 01

Posted 01 May 2011 - 13:34

Always enjoy these threads! :D

Just a question for the OP. How have you (or will you) account for anomalies? For example Webber going out in Q1 in China.

#4 Zava

Zava
  • Member

  • 4,965 posts
  • Joined: September 10

Posted 01 May 2011 - 13:38

Always enjoy these threads! :D

Just a question for the OP. How have you (or will you) account for anomalies? For example Webber going out in Q1 in China.

I guess he compared it to Vettel's Q1 time.
at least that's what I'd do :) compare the last session's times they both competed in.

#5 UprightRacer

UprightRacer
  • Member

  • 93 posts
  • Joined: July 09

Posted 01 May 2011 - 14:16

12.Sutil 0.03s


hasn't he been beaten by his teammate so far?

#6 Dunder

Dunder
  • Member

  • 6,784 posts
  • Joined: April 01

Posted 01 May 2011 - 14:24

hasn't he been beaten by his teammate so far?


Yeah, that is kinda what I was talking about by anomalies.
In Australia for example, Sutil was 1s faster in Q1 but Di Resta ws 5s faster in Q2 (due to Sutil losing it at the the last corner).

I am not sure how the OP has accounted for that but given that Di Resta qualified higher than Sutil in Malaysia and in China, there can be no (sensible) way you can have Sutil ahead for the season to date.


#7 DarthWillie

DarthWillie
  • Member

  • 1,690 posts
  • Joined: November 07

Posted 01 May 2011 - 15:05

I think these comparisons can stop from the next race onwards, probably some drivers will start to sacrifice their Q to save tyres. Webber might have set a trend last race

#8 tyker

tyker
  • Member

  • 209 posts
  • Joined: April 11

Posted 01 May 2011 - 18:40

Always enjoy these threads! :D

Just a question for the OP. How have you (or will you) account for anomalies? For example Webber going out in Q1 in China.

I took their relative Q1 times plus i gave Webber the benefit of 3 tenths because he didnt have KERS

hasn't he been beaten by his teammate so far?

In Australia Sutil went about 0.5s quicker in Q1 than what di Resta did in Q2, i'm looking at fastest laps

Edited by tyker, 01 May 2011 - 18:42.


#9 tyker

tyker
  • Member

  • 209 posts
  • Joined: April 11

Posted 02 May 2011 - 11:26

I think these comparisons can stop from the next race onwards, probably some drivers will start to sacrifice their Q to save tyres. Webber might have set a trend last race

I still think its important to qualify well and then start well, you could say that Vettel lost the last race in part to his bad start, also Alonso never recovered from his bad start relative to Massa. As for Webber i think driving a rocketship helps somewhat as well and would Webber have been as close to Vettel had he been on a 3 stopper which seemed quicker and Vettel not been held up on his first stint after starting bad from pole.

#10 Dauemannen

Dauemannen
  • Member

  • 50 posts
  • Joined: March 11

Posted 02 May 2011 - 13:13

So you're comparing the fastest times across the parts of qualifying they were both in, correct?

i gave Webber the benefit of 3 tenths because he didnt have KERS

If you're to give Webber 3 tenths for missing KERS, sholdn't you also give Schumi 4 tenths or whatever each of the two times his rear wing failed? There are probably a lot more car failures that could have attributed a few tenths here and there, and if you're compensating for some of them you should compensate for all of them. Or else your table is biased.

#11 carlb5253

carlb5253
  • Member

  • 507 posts
  • Joined: March 11

Posted 02 May 2011 - 15:58

This is 7 shades of silly.

Could have been good if you was accurate.


#12 tyker

tyker
  • Member

  • 209 posts
  • Joined: April 11

Posted 02 May 2011 - 16:44

Maybe i shouldnt bother with Schumacher then he always seems to be having problems of one kind or another

#13 Reinmuster

Reinmuster
  • Member

  • 644 posts
  • Joined: September 10

Posted 03 May 2011 - 06:19

Maybe the topic starter can shows us how he got those figures?

Just to clear things up?




#14 tyker

tyker
  • Member

  • 209 posts
  • Joined: April 11

Posted 03 May 2011 - 13:14

Maybe the topic starter can shows us how he got those figures?

Just to clear things up?

Fastest times achieved in qualifying upto the point where neither driver was eliminated

#15 Buttoneer

Buttoneer
  • RC Forum Admin

  • 16,240 posts
  • Joined: May 04

Posted 03 May 2011 - 22:11

Fastest times achieved in qualifying upto the point where neither driver was eliminated

Plus a bit of random tweakery to account for a failure here and there, of course. While this might be a great topic for discussion, the list of numbers you provided without any qualification or explanation, is next to useless.

Add some meat to the bare bones of the initial post to give us the full details of how you came up with these numbers and, if the basis is reasonable, this should be a decent thread.

#16 Frans

Frans
  • Member

  • 7,701 posts
  • Joined: April 01

Posted 03 May 2011 - 22:27

I already like this thread very much! Give us the numbers and all will be happy! :):)

#17 Buttoneer

Buttoneer
  • RC Forum Admin

  • 16,240 posts
  • Joined: May 04

Posted 03 May 2011 - 22:29

I already like this thread very much! Give us the numbers and all will be happy! :):)

Be honest now, there's only one number you like :lol:

#18 tyker

tyker
  • Member

  • 209 posts
  • Joined: April 11

Posted 03 May 2011 - 23:52

Plus a bit of random tweakery to account for a failure here and there, of course. While this might be a great topic for discussion, the list of numbers you provided without any qualification or explanation, is next to useless.

Add some meat to the bare bones of the initial post to give us the full details of how you came up with these numbers and, if the basis is reasonable, this should be a decent thread.

Ok but its late now i'll crunch the numbers tomorrow

#19 simplyfast

simplyfast
  • Member

  • 867 posts
  • Joined: July 10

Posted 04 May 2011 - 00:11

IMHO unless you discount (ignore) any qually session where a driver has a mechanical problem, is blocked by a back marker or any other non-driver related problem (EG starts raining before one of the drivers gets out on track), as well as any qually where the two drivers are not using the same tyres, you have no hope of getting any meaningful data (but i am certain some fans of drivers however will be very happy).
Besides with only being 3 races into the season, its far far too early (too small a sample so any errors are magnified out of all real proportion) for any meaningful conclusions to be drawn yet, unless you only included those pairings who have had no problems and been on the same strategies.

IMHO i think we need to be at-least 1/2 way into the season before we can hope to have enough data to show any real differences between the pairings, with any degree of confidence.
But at-least it does give us something to talk about :lol:

Advertisement

#20 StefanArak

StefanArak
  • Member

  • 191 posts
  • Joined: April 11

Posted 04 May 2011 - 00:20

Worthless stats due to the way you've done them. I'll make a new thread soon with the actual averages and keep updating them as the season progresses.

Edited by StefanArak, 04 May 2011 - 00:21.


#21 Dunder

Dunder
  • Member

  • 6,784 posts
  • Joined: April 01

Posted 04 May 2011 - 01:04

Below is the raw data in terms of the differentials so far.

Methodology is as follows:

Each driver is credited with his fastest time regardless of whether it was set in Q1, Q2 or Q3.
No results are amended.

I would intend to account for anomalies as follows:
After 5 races the worst result for each driver will be excluded from the mean average (which will therefore be the average of 3 sessions at that stage)
After 10 races the worst 2 results for each driver will be excluded from the mean average (which will therefore be the average of 6 sessions at that stage)
After 15 races the worst 3 results for each driver will be excluded from the mean average (which will therefore be the average of 9 sessions at that stage).

I am open to suggestions/ideas but based on what we have seen so far, I think it is necessary to have the fairly large number of excluded results and to take the best result regardless of session simply due to tyre availability (or lack thereof).

Posted Image

#22 tyker

tyker
  • Member

  • 209 posts
  • Joined: April 11

Posted 04 May 2011 - 01:07

I'll do one before i got to bed

Hamilton ----------- Button
Q3 - 1:24.309 Q3 - 1:24.779 = 0.472
Q3 - 1:34.974 Q3 - 1:35.200 = 0.226
Q3 - 1:34.463 Q3 - 1:34.421 = -0.042
Total divided by 3 = 0.2186667

#23 tyker

tyker
  • Member

  • 209 posts
  • Joined: April 11

Posted 04 May 2011 - 07:14

Below is the raw data in terms of the differentials so far.

Methodology is as follows:

Each driver is credited with his fastest time regardless of whether it was set in Q1, Q2 or Q3.
No results are amended.

I would intend to account for anomalies as follows:
After 5 races the worst result for each driver will be excluded from the mean average (which will therefore be the average of 3 sessions at that stage)
After 10 races the worst 2 results for each driver will be excluded from the mean average (which will therefore be the average of 6 sessions at that stage)
After 15 races the worst 3 results for each driver will be excluded from the mean average (which will therefore be the average of 9 sessions at that stage).

I am open to suggestions/ideas but based on what we have seen so far, I think it is necessary to have the fairly large number of excluded results and to take the best result regardless of session simply due to tyre availability (or lack thereof).

Posted Image

Only difference i did was to discount times when one driver has been knocked out and i discounted the Heidfeld/Petrov time in Australia, looking at that chart 8 out of the 12 averages are the same as mine

Edited by tyker, 04 May 2011 - 07:18.


#24 marcoferrari

marcoferrari
  • Member

  • 2,004 posts
  • Joined: June 10

Posted 04 May 2011 - 08:21

Only difference i did was to discount times when one driver has been knocked out and i discounted the Heidfeld/Petrov time in Australia, looking at that chart 8 out of the 12 averages are the same as mine


You have there a mistake with Alguersuari as he was faster then Buemi in China...  ;)
Maybe it will be better to put all three top sector times of each qualy together?
Adrian Sutil was over 6 tenths quicker over Di Resta in Melbourne, but he made a mistake in last corner...

#25 Buttoneer

Buttoneer
  • RC Forum Admin

  • 16,240 posts
  • Joined: May 04

Posted 04 May 2011 - 08:41

Maybe it will be better to put all three top sector times of each qualy together?

Yeah, sounds like a great idea. Lets see what you can come up with.

#26 Dunder

Dunder
  • Member

  • 6,784 posts
  • Joined: April 01

Posted 04 May 2011 - 09:45

You have there a mistake with Alguersuari as he was faster then Buemi in China... ;)
Maybe it will be better to put all three top sector times of each qualy together?
Adrian Sutil was over 6 tenths quicker over Di Resta in Melbourne, but he made a mistake in last corner...


He was in Q3 but both drivers were quicker in Q2 than in Q3.
Fastest sectors might be an idea but I, personally have always found the "optimum lap" idea unsatisfying.


#27 Dunder

Dunder
  • Member

  • 6,784 posts
  • Joined: April 01

Posted 04 May 2011 - 09:50

Only difference i did was to discount times when one driver has been knocked out and i discounted the Heidfeld/Petrov time in Australia, looking at that chart 8 out of the 12 averages are the same as mine


I always assumed that your data was correct. It is the subjective and unexplained discounts/adjustments that were always going to cause "trouble".
In previous years these threads have always had lively 'debate' when one or other driver had issues outside their control or made a mistake. My view is that by excluding each driver's worst x results to account for such anomalies just based on the raw, unadjusted data you get a very good picture of the year as a whole.


#28 marcoferrari

marcoferrari
  • Member

  • 2,004 posts
  • Joined: June 10

Posted 04 May 2011 - 11:28

He was in Q3 but both drivers were quicker in Q2 than in Q3.
Fastest sectors might be an idea but I, personally have always found the "optimum lap" idea unsatisfying.


Yes, but the deciding gap is important, not the one in Q1 or Q2, if they entered also Q3...

#29 Watkins74

Watkins74
  • Member

  • 5,684 posts
  • Joined: February 10

Posted 04 May 2011 - 12:13

Yes, but the deciding gap is important, not the one in Q1 or Q2, if they entered also Q3...


:up: I agree.

#30 Dunder

Dunder
  • Member

  • 6,784 posts
  • Joined: April 01

Posted 04 May 2011 - 12:22

Yes, but the deciding gap is important, not the one in Q1 or Q2, if they entered also Q3...


In previous years, I would have fully agreed with you.

This year though we are seeing midfield cars (especially) getting into Q3 and going slower than in Q2 because they have no fresh soft tyres left. I think this was the case for both STR drivers in China but there will be (and have been) instances where insisting that the lap times used are from the latest session means that one driver is on fresh tyres and the other is on used rubber which opens up more 'cans of worms'.


#31 tyker

tyker
  • Member

  • 209 posts
  • Joined: April 11

Posted 04 May 2011 - 17:49

I always assumed that your data was correct. It is the subjective and unexplained discounts/adjustments that were always going to cause "trouble".
In previous years these threads have always had lively 'debate' when one or other driver had issues outside their control or made a mistake. My view is that by excluding each driver's worst x results to account for such anomalies just based on the raw, unadjusted data you get a very good picture of the year as a whole.

Yes i see your point there it probably would produce a slightly more accurate result

#32 tyker

tyker
  • Member

  • 209 posts
  • Joined: April 11

Posted 04 May 2011 - 17:53

Yes, but the deciding gap is important, not the one in Q1 or Q2, if they entered also Q3...

The basis of my post was to compare the fastest outright lap acheived in qualifying when both drivers were on track

#33 race addicted

race addicted
  • Member

  • 19,476 posts
  • Joined: June 01

Posted 04 May 2011 - 18:22

IMHO unless you discount (ignore) any qually session where a driver has a mechanical problem, is blocked by a back marker or any other non-driver related problem (EG starts raining before one of the drivers gets out on track), as well as any qually where the two drivers are not using the same tyres, you have no hope of getting any meaningful data (but i am certain some fans of drivers however will be very happy).
Besides with only being 3 races into the season, its far far too early (too small a sample so any errors are magnified out of all real proportion) for any meaningful conclusions to be drawn yet, unless you only included those pairings who have had no problems and been on the same strategies.


Agreed. I used to calculate these numbers, and post a thread to discuss them, and indeed I ignored making a comparison when and if one driver suffered a mishap of some sort. Otherwise it's absolutely meaningless, and the mean average table posted above, is useless too. Making comparisons across the three sessions is extremely deceiving.


#34 marcoferrari

marcoferrari
  • Member

  • 2,004 posts
  • Joined: June 10

Posted 04 May 2011 - 18:44

In previous years, I would have fully agreed with you.

This year though we are seeing midfield cars (especially) getting into Q3 and going slower than in Q2 because they have no fresh soft tyres left. I think this was the case for both STR drivers in China but there will be (and have been) instances where insisting that the lap times used are from the latest session means that one driver is on fresh tyres and the other is on used rubber which opens up more 'cans of worms'.


Yes, but at the end, Alguersuari was 7th on the grid and Buemi only 9th... Q3 was the most important part of the quali for they two, even if they were generally slower then in Q2... In the last 10 minute session both were on track at the same time and both with used soft tyres... This doesn t really show the reality, but nevermind it is your calculation...;)

Edited by marcoferrari, 04 May 2011 - 18:46.


#35 tyker

tyker
  • Member

  • 209 posts
  • Joined: April 11

Posted 04 May 2011 - 23:13

Yes, but at the end, Alguersuari was 7th on the grid and Buemi only 9th... Q3 was the most important part of the quali for they two, even if they were generally slower then in Q2... In the last 10 minute session both were on track at the same time and both with used soft tyres... This doesn t really show the reality, but nevermind it is your calculation...;)

You miss the point its not about who qualifies highest on the grid its just to see who can post the fastest time

#36 H2H

H2H
  • Member

  • 2,891 posts
  • Joined: June 09

Posted 05 May 2011 - 07:41

Agreed. I used to calculate these numbers, and post a thread to discuss them, and indeed I ignored making a comparison when and if one driver suffered a mishap of some sort. Otherwise it's absolutely meaningless, and the mean average table posted above, is useless too. Making comparisons across the three sessions is extremely deceiving.


That is the point, the comparision should be relevant and the methodology clear. So a marker should be added to a time driven with tangible external impediments, for example the lack of KERS for Mark or the lack of time and space for a proper lap in China for Heidfeld and others. Usually such times are also the first to be elimated by taking out the 1-2-3 worst gaps, but it would be much more clear that way.

The ability to string the best sectors togheter would be also interesting.

Edited by H2H, 05 May 2011 - 07:43.


#37 Dunder

Dunder
  • Member

  • 6,784 posts
  • Joined: April 01

Posted 05 May 2011 - 16:24

That is the point, the comparision should be relevant and the methodology clear. So a marker should be added to a time driven with tangible external impediments, for example the lack of KERS for Mark or the lack of time and space for a proper lap in China for Heidfeld and others. Usually such times are also the first to be elimated by taking out the 1-2-3 worst gaps, but it would be much more clear that way.

The ability to string the best sectors togheter would be also interesting.


That is my thinking too.
The main benefit to my mind is that the only subjective assessment needed is how many sessions to exclude over the season as opposed to having to decide whether driver A was hindered by enough to invalidate a result.

In my initial post, I mentioned that I would start to exclude times/sessions (one for each driver) after 5 races. I will bring this forward and post the results with one excluded result per driver after qualifying on Saturday. I can definitely see both sides of the argument regarding comparing times from different sessions.


#38 tyker

tyker
  • Member

  • 209 posts
  • Joined: April 11

Posted 05 May 2011 - 19:20

That is my thinking too.
The main benefit to my mind is that the only subjective assessment needed is how many sessions to exclude over the season as opposed to having to decide whether driver A was hindered by enough to invalidate a result.

In my initial post, I mentioned that I would start to exclude times/sessions (one for each driver) after 5 races. I will bring this forward and post the results with one excluded result per driver after qualifying on Saturday. I can definitely see both sides of the argument regarding comparing times from different sessions.

It will be interesting to compare the results with mine

#39 Dunder

Dunder
  • Member

  • 6,784 posts
  • Joined: April 01

Posted 05 May 2011 - 19:44

Just as a matter of interest, below is the relative performance of the CARS so far.

Methodology is as follows:
The fastest time set by either driver for each team is included regardless of whether it is set it Q1, Q2 or Q3.
I have excluded the single worst result for each team in the "Mean Average" result. At this stage, I only did this to check that my formulae calculate correctly but over time I will use the same number of exclusions over the season as is done in the drivers table.

Posted Image

Advertisement

#40 tyker

tyker
  • Member

  • 209 posts
  • Joined: April 11

Posted 05 May 2011 - 19:56

Just as a matter of interest, below is the relative performance of the CARS so far.

Methodology is as follows:
The fastest time set by either driver for each team is included regardless of whether it is set it Q1, Q2 or Q3.
I have excluded the single worst result for each team in the "Mean Average" result. At this stage, I only did this to check that my formulae calculate correctly but over time I will use the same number of exclusions over the season as is done in the drivers table.

Posted Image

Autosport have just released the numbers as well:-

1. Red Bull
2. McLaren + 0.5s
3. Ferrari + 1.2s
4. Mercedes + 1.4s
5. Renault + 1.7s
6. Sauber + 2.1s
7. Toro Rosso + 2.2s
8. Williams + 2.6s
9. Force India + 2.7s
10.Lotus + 4.7s
11.Virgin + 6.1s
12.HRT + 6.4s

#41 thesham01

thesham01
  • Member

  • 1,196 posts
  • Joined: August 10

Posted 05 May 2011 - 21:11

I think that the times should only be from Q3, as that is what counts. That is when the pressure is on and we see how fast someone really is. After all that is how we decide who is on pole, not finding the fastest time from any of the qualifying sessions. If you are slower in Q3 then thats that, you are not as fast when it counts, and I think that is all that matters. Why not bring in practice times otherwise.

As for anomalies like Sutil in Aus, mistakes like that should be discounted, huge errors. Small errors are part and parcel of the game.

And the idea of discounting the worst one in 5 is a good idea, keep it tight.

But I really do disagree with the taking the fastest time from all sessions.

#42 Dunder

Dunder
  • Member

  • 6,784 posts
  • Joined: April 01

Posted 05 May 2011 - 22:03

I think that the times should only be from Q3, as that is what counts. That is when the pressure is on and we see how fast someone really is. After all that is how we decide who is on pole, not finding the fastest time from any of the qualifying sessions. If you are slower in Q3 then thats that, you are not as fast when it counts, and I think that is all that matters. Why not bring in practice times otherwise.

As for anomalies like Sutil in Aus, mistakes like that should be discounted, huge errors. Small errors are part and parcel of the game.

And the idea of discounting the worst one in 5 is a good idea, keep it tight.

But I really do disagree with the taking the fastest time from all sessions.


OK. If there is a majority in favour (as there appears to be), I will make changes as follows:

1) Differentials between drivers will be based on the last qualifying session in which both drivers set a time.
2) If both drivers compete in a session and one fails to set a time (or makes a mistake rendering a time set unrepresentative) he will be given a time a nominal 1 second slower than his team mate.
3) Each driver's three worst relative performances of the season will be excluded from the results. These exclusions will be added one at a time after the 4th, 9th and 14th qualifying sessions of the season.

I will keep the CAR performance as it is.


#43 Buttoneer

Buttoneer
  • RC Forum Admin

  • 16,240 posts
  • Joined: May 04

Posted 05 May 2011 - 22:56

Absolutely disagree with thesham01. If we only take Q3 then we have anomalous entries where one driver of the team makes it through and the other not. You start making exceptions for that sort of thing and the whole calculation process falls apart. What he says is true, only for those drivers with a realistic chance of pole, so it's quite a limited view based on an opinion that only the drivers at the front matter. For a team such as Force India, actually making it into Q3 is the pressure point, and for Lotus, getting into Q2. All sessions are important. Adding amounts to make up for the difference just adds artificiality.

If we only take Q3, then we only need ever look at the final result.

We are not dealing with a system where the cars change between sessions any more, with the addition of fuel. The change is only a more rubbered-in circuit and possibly environmental conditions.

Regarding exclusions etc, I agree your approach seems quite fair.

#44 oetzi

oetzi
  • Member

  • 2,924 posts
  • Joined: April 10

Posted 05 May 2011 - 23:03

The way quali is structured now means that there's not likely to be more than a few sessions all year when a team has both drivers either going for pole or trying like hell to make the next phase will give it their all, surely? So how do you compare?

Surely the laps that matter are more than ever the laps that matter. The others go to the wind.

#45 pyrot

pyrot
  • New Member

  • 9 posts
  • Joined: May 11

Posted 06 May 2011 - 21:33

OK. If there is a majority in favour (as there appears to be), I will make changes as follows:

1) Differentials between drivers will be based on the last qualifying session in which both drivers set a time.



There is a problem with that. A driver could set an amazing time in Q2, the fastest of the weekend, but then for whatever reason set a much slower lap in Q3 and using your method his raw speed for that weekend will be ignored and will be misrepresented because you ignored his fastest lap of the previous session. The point of this process is to compare the pure speed of team mates, so their fastest laps should be used, what session they did them in should not matter, because this process is not determining grid positions. So I think its best to use the fastest lap based on last session both drivers set a time. A slight modification of yours.

If both drivers compete in a session and one fails to set a time (or makes a mistake rendering a time set unrepresentative) he will be given a time a nominal 1 second slower than his team mate.


This is totally arbitary and could heavily scew the final average. With my change, this would not be needed. Just use the fastest lap in qualifying.

Edited by pyrot, 06 May 2011 - 22:46.


#46 tyker

tyker
  • Member

  • 209 posts
  • Joined: April 11

Posted 07 May 2011 - 08:00

There is a problem with that. A driver could set an amazing time in Q2, the fastest of the weekend, but then for whatever reason set a much slower lap in Q3 and using your method his raw speed for that weekend will be ignored and will be misrepresented because you ignored his fastest lap of the previous session. The point of this process is to compare the pure speed of team mates, so their fastest laps should be used, what session they did them in should not matter, because this process is not determining grid positions. So I think its best to use the fastest lap based on last session both drivers set a time. A slight modification of yours.



This is totally arbitary and could heavily scew the final average. With my change, this would not be needed. Just use the fastest lap in qualifying.

:up:

#47 Dunder

Dunder
  • Member

  • 6,784 posts
  • Joined: April 01

Posted 07 May 2011 - 08:03

There is a problem with that. A driver could set an amazing time in Q2, the fastest of the weekend, but then for whatever reason set a much slower lap in Q3 and using your method his raw speed for that weekend will be ignored and will be misrepresented because you ignored his fastest lap of the previous session. The point of this process is to compare the pure speed of team mates, so their fastest laps should be used, what session they did them in should not matter, because this process is not determining grid positions. So I think its best to use the fastest lap based on last session both drivers set a time. A slight modification of yours.


This is actually how it is calculated at the moment and how I think/suggested we should proceed.
As you can see the majority of posters who have commented (even though it is a small sample) appear to disagree.

#48 Dunder

Dunder
  • Member

  • 6,784 posts
  • Joined: April 01

Posted 07 May 2011 - 10:59

OK before quali in Turkey gets under way I have decided to maintain my original method of calculation i.e. the fastest lap achieved regardless if it is set in Q1, Q2 or Q3.

Anyone who prefers a different methodology is, of course, free to post their calculations for comparison.

#49 Dunder

Dunder
  • Member

  • 6,784 posts
  • Joined: April 01

Posted 07 May 2011 - 14:02

As promised, I have updated the figures according to the original methodology:

Drivers
Posted Image

I realise it is early days but the biggest surprises to me are the gap between Vettel and Webber and the d'Ambrosio is matching Glock.

Cars
Posted Image



#50 Kohque

Kohque
  • Member

  • 359 posts
  • Joined: April 11

Posted 07 May 2011 - 14:05

As promised, I have updated the figures according to the original methodology:

Drivers
Posted Image

I realise it is early days but the biggest surprises to me are the gap between Vettel and Webber and the d'Ambrosio is matching Glock.

Cars
Posted Image


Cool stats, thanks! :up: