Greatest Driver of all time statistical analysis

[andyps]

Has anyone else seen this - https://www.sheffield.ac.uk/news/nr/greatest-formu...

Makes for an interesting view but does throw up some anomalies. The top 10 apparently are:
1. Juan Manuel Fangio (ARG)
2. Alain Prost (FRA)
3. Fernando Alonso (SPA)
4. Jim Clark (SCO)
5. Ayrton Senna (BRA)
6. Jackie Stewart (GBR)
7. Nelson Piquet (BRA)
8. Emerson Fittipaldi (BRA)
9. Michael Schumacher (GER)
10. Sebastian Vettel (GER)

And Lauda doesn't even make the top 100 so, in my view, there are some flaws in the calculation!

[williamp]

Pointless comparison. How about if you remove the effect of your teammate? The driver who gives up his car so fangio can win? The driver who deliberately crashes, team orders etc

[Dr Z]

As mentioned, you can't account-for team orders which tend to be very pervasive in often very subtle ways in a model...it's a team sport after all. Trying to find the greatest driver in this like trying to find the proverbial needle in a haystack. However, I like what mathematician Dr. Phillips has done (see below)...in fact his blog has been cited in that paper by the Sheffield folk!

Who was the Greatest F1 Driver?

His top ten are:

1) Jim Clark
2) Jackie Stewart
3) Fernando Alonso
4) Michael Schumacher
5) Juan Manuel Fangio
6) James Hunt
7) Nico Rosberg
8) Sebastian Vettel
9) Alberto Ascari
10) Niki Lauda

He also recognises a lot of the limitations of such an exercise.

ETA: You ought to read the above alongside another article on the most dominant cars ever in F1:

https://f1metrics.wordpress.com/2015/10/08/the-mos...

The 2014 Mercedes W05 is second only to the legendary '61 Sharknose. Helps you to gauge the driver performance better.

Edited by Dr Z on Saturday 23 April 10:53

[R1 Indy]

Dr Z said:

As mentioned, you can't account-for team orders which tend to be very pervasive in often very subtle ways in a model...it's a team sport after all. Trying to find the greatest driver in this like trying to find the proverbial needle in a haystack. However, I like what mathematician Dr. Phillips has done (see below)...in fact his blog has been cited in that paper by the Sheffield folk!

Who was the Greatest F1 Driver?

His top ten are:

1) Jim Clark
2) Jackie Stewart
3) Fernando Alonso
4) Michael Schumacher
5) Juan Manuel Fangio
6) James Hunt
7) Nico Rosberg
8) Sebastian Vettel
9) Alberto Ascari
10) Niki Lauda

He also recognises a lot of the limitations of such an exercise.

ETA: You ought to read the above alongside another article on the most dominant cars ever in F1:

https://f1metrics.wordpress.com/2015/10/08/the-mos...

The 2014 Mercedes W05 is second only to the legendary '61 Sharknose. Helps you to gauge the driver performance better.

Edited by Dr Z on Saturday 23 April 10:53

Im sorry, Nico Rosberg No. 7 all time great..............

Edited by Big Al. on Thursday 28th April 18:12

[Dr Z]

R1 Indy said:

Im sorry, Nico Rosberg No. 7 all time great......

As mentioned in the article, if you compare his performance against one M. Schumacher and having fairly closely matched another L. Hamilton, who were all highly rated by the model led to this ranking. All current driver's status is in flux as their form changes, so it's best to ignore current drivers' ranking on that list.

ETA: why so many dots?

[Adam Ansel]

Gilles Villeneuve and Tazio Nuvolari. Both top 10. Both not on the list.
Ergo the list is a waste of electrons.

[andy_s]

R1 Indy said:

Im sorry, Nico Rosberg No. 7 all time great..................

You feck up the formatting doing that.

Edited by Big Al. on Thursday 28th April 18:11

[rsbmw]

Quoting it was a much better move

[EnglishTony]

If the point of publishing the survey was to attract funding for more serious work I can't help but feel that it has backfired.

[VladD]

Jim Clark (SCO), Jackie Stewart (GBR)?

[EnglishTony]

VladD said:

Jim Clark (SCO), Jackie Stewart (GBR)?

Sums it up nicely

[RichB]

EnglishTony said:

VladD said:

Jim Clark (SCO), Jackie Stewart (GBR)?

Sums it up nicely

[whatxd]

Dr Z said:

As mentioned, you can't account-for team orders which tend to be very pervasive in often very subtle ways in a model...it's a team sport after all. Trying to find the greatest driver in this like trying to find the proverbial needle in a haystack. However, I like what mathematician Dr. Phillips has done (see below)...in fact his blog has been cited in that paper by the Sheffield folk!

Who was the Greatest F1 Driver?

His top ten are:

1) Jim Clark
2) Jackie Stewart
3) Fernando Alonso
4) Michael Schumacher
5) Juan Manuel Fangio
6) James Hunt
7) Nico Rosberg
8) Sebastian Vettel
9) Alberto Ascari
10) Niki Lauda

He also recognises a lot of the limitations of such an exercise.

ETA: You ought to read the above alongside another article on the most dominant cars ever in F1:

https://f1metrics.wordpress.com/2015/10/08/the-mos...

The 2014 Mercedes W05 is second only to the legendary '61 Sharknose. Helps you to gauge the driver performance better.

Edited by Dr Z on Saturday 23 April 10:53

Mika Salo ahead of Mika Hakkinen?

Heinz-Harald Frentzen ahead of Ayrton Senna?

I think the algorithm he used has a few bugs, as do all of these other machine/mathematically produced lists that I've seen.

[Adam Ansel]

This looks far better.

1) Clark
2) Nuvolari
3) Villeneuve G
4) Fangio
5) Ascari
6) Schumacher
7) Prost
8) Stewart
9) Moss
10) = Lauda = Senna

[hairyben]

Dr Z said:

As mentioned, you can't account-for team orders which tend to be very pervasive in often very subtle ways in a model...it's a team sport after all. Trying to find the greatest driver in this like trying to find the proverbial needle in a haystack. However, I like what mathematician Dr. Phillips has done (see below)...in fact his blog has been cited in that paper by the Sheffield folk!

Who was the Greatest F1 Driver?

His top ten are:

1) Jim Clark
2) Jackie Stewart
3) Fernando Alonso
4) Michael Schumacher
5) Juan Manuel Fangio
6) James Hunt
7) Nico Rosberg
8) Sebastian Vettel
9) Alberto Ascari
10) Niki Lauda

He also recognises a lot of the limitations of such an exercise.

ETA: You ought to read the above alongside another article on the most dominant cars ever in F1:

https://f1metrics.wordpress.com/2015/10/08/the-mos...

The 2014 Mercedes W05 is second only to the legendary '61 Sharknose. Helps you to gauge the driver performance better.

Edited by Dr Z on Saturday 23 April 10:53

To be fair though, when your grand theory of unitfied driver superiority places alonso 3rd and rosberg 7th and hamilton, (who spanked them both on equal terms AND machinery), not on the list, then you have to screw it up and admit you've wasted your time.

Without accounting for the team politics/status it's always going to be flawed or what computer people call GIGO, and I struggle to understand why someone would put to much effort into it.

[Dr Z]

hairyben said:

To be fair though, when your grand theory of unitfied driver superiority places alonso 3rd and rosberg 7th and hamilton, (who spanked them both on equal terms AND machinery), not on the list, then you have to screw it up and admit you've wasted your time.

Without accounting for the team politics/status it's always going to be flawed or what computer people call GIGO, and I struggle to understand why someone would put to much effort into it.

I agree with the point about the fundamentally flawed nature of this exercise, in terms of actually finding the greatest F1 driver of all time. Welcome to sociology! (Or welcome to studying a human activity)

But, surely you agree that the model is self consistent and the results intuitive in many respects? I don't think either of the researchers are presenting these lists as the definitive list. But, it gives you a framework to look at driver performance in isolation. We can then interpret this in light of what we know about the politics of team orders. E.g. we know from history (some of you might have been there!) that the team politics in the 50s meant that Moss was always to yield to Fangio, so you interpret the races when they were team mates in light of that. It's nice to account for how dominant the car was, and how competitive the field was in a given season to put your favourite driver's performance in perspective.

I also get the sense that it is more about the methodology than the end result in the paper by the Sheffield researchers. These are papers by mathematicians trying to make sense of a sport that while in the basic sense is determined by the stop watch, also has a lot more to it. So, make of it what you will but cut them some slack.

As for the current driver ranking, if you'd have read his methodology, it depends on which 3-years of the career you use as a reference point. He has mentioned in a comment that Hamilton will have a new 3-year peak as per his model (2013-2015), so it will have to be updated. However, Hamilton didn't spank Rosberg or Alonso. You could say he spanked Button, but that's really pushing it. The stats don't bear that one out. But I notice, this subject is ripe for bringing out all the fanboys/girls, so I'll refrain from doing the [insert my favourite driver] whooped [insert your favourite driver]'s ass routine.

[EagleMoto4-2]

Adam Ansel said:

This looks far better.

1) Clark
2) Nuvolari
3) Villeneuve G
4) Fangio
5) Ascari
6) Schumacher
7) Prost
8) Stewart
9) Moss
10) = Lauda = Senna

Indeed, though my top 3 would be
1) Clark
2) Senna
3) Villeneuve G

[glazbagun]

Dr Z said:

As for the current driver ranking, if you'd have read his methodology, it depends on which 3-years of the career you use as a
reference point. He has mentioned in a comment that Hamilton will have a new 3-year peak as per his model (2013-2015), so it
will have to be updated. However, Hamilton didn't spank Rosberg or Alonso. You could say he spanked Button, but that's
really pushing it. The stats don't bear that one out. But I notice, this subject is ripe for bringing out all the
fanboys/girls, so I'll refrain from doing the [insert my favourite driver] whooped [insert your favourite driver]'s ass
routine.

It's been a while since I read the F1 metrics site, but from what I recall he pretty much said that if Hamilton had quit after
his debut year he'd be ranked much higher, it was his wilderness years that dragged him down. The converse happened with James
Hunt, IIRC, who outperformed almost everyone he was up against in the same car and got him above any 70's driver bar Jackie
Stewart.

Edited by glazbagun on Wednesday 27th April 20:32

[hairyben]

Dr Z said:

I agree with the point about the fundamentally flawed nature of this exercise, in terms of actually finding the greatest F1 driver of all time. Welcome to sociology! (Or welcome to studying a human activity)

But, surely you agree that the model is self consistent and the results intuitive in many respects? I don't think either of the researchers are presenting these lists as the definitive list. But, it gives you a framework to look at driver performance in isolation. We can then interpret this in light of what we know about the politics of team orders. E.g. we know from history (some of you might have been there!) that the team politics in the 50s meant that Moss was always to yield to Fangio, so you interpret the races when they were team mates in light of that. It's nice to account for how dominant the car was, and how competitive the field was in a given season to put your favourite driver's performance in perspective.

I also get the sense that it is more about the methodology than the end result in the paper by the Sheffield researchers. These are papers by mathematicians trying to make sense of a sport that while in the basic sense is determined by the stop watch, also has a lot more to it. So, make of it what you will but cut them some slack.

As for the current driver ranking, if you'd have read his methodology, it depends on which 3-years of the career you use as a reference point. He has mentioned in a comment that Hamilton will have a new 3-year peak as per his model (2013-2015), so it will have to be updated. However, Hamilton didn't spank Rosberg or Alonso. You could say he spanked Button, but that's really pushing it. The stats don't bear that one out. But I notice, this subject is ripe for bringing out all the fanboys/girls, so I'll refrain from doing the [insert my favourite driver] whooped [insert your favourite driver]'s ass routine.

You can't apply a logical or mathematical process to reinterpret statistics this way without accounting for any inaccuracy (bias, team orders) first though, not afterwards, as the innacuracy could be amplified by whatever calculation takes place, exponentially flawing the results.


As for hamilton, he's one of the best examples for being a multiple WDC whose been in as close as you'll get to equal terms/machinery with all his team mates, as a rookie he beat one of the most prolific drivers in his prime and then 3 years continuously beat a very quick and accomplished driver in rosberg also in his prime, to rank them both way over him makes a mockery of the instrument. hypothesise what you like about language and perceived fanboy allegiances I don't see the relevance, I'd certainly say this paragraph is more representative of facts than that table...

[Adam Ansel]

hairyben said:

As for hamilton, he's one of the best examples for being a multiple WDC whose been in as close as you'll get to equal terms/machinery with all his team mates, as a rookie he beat one of the most prolific drivers in his prime and then 3 years continuously beat a very quick and accomplished driver in rosberg also in his prime, to rank them both way over him makes a mockery of the instrument. hypothesise what you like about language and perceived fanboy allegiances I don't see the relevance, I'd certainly say this paragraph is more representative of facts than that table...

Hamilton was outscored by Button at McLaren.