Page 12 of 13

Re: Best F1 driver ever?

Posted: Mon Aug 26, 2019 1:14 am
by bourbon19
tim3003 wrote:A lot of people make the point that success in F1 is car-dependent. Fair enough, but as I think I said, the best drivers tend to get the best cars, so isn't the playing field thus level again?

Maybe it's the statistican in me that wants to define numerically who is the best ! Maybe my way is not the correct way. But if not can we agree on another objective criteria that is? If not then all we have is subjective views, of which there are as many as there are drivers! Surely that's not good enough..
Yeah, I don't see that as possible - I have not read every post, but if it was not mentioned, you have a number of top drivers with unfinished careers, so you could never really have an accurate measure. In any case, what is wrong with "the most success with the opportunities given, whatever they may be" statistic? That you have done a bang up job measuring.

Re: Best F1 driver ever?

Posted: Mon Aug 26, 2019 9:21 am
by POBRatings
Zoue wrote:
tim3003 wrote:
Argentum wrote:
tim3003 wrote:My method explicitly ignores the evolutionary changes in the sport
Your method does no such thing. It arbitrarily gives WDC twice the points of a second place? why? What if someone else thinks the ratio should be 3:1? Or 7:2? Or any other figure?
Fair enough, but I have to start somewhere!
I don't think it's enough to say you have to start somewhere. The weightings you give have to have some science behind them otherwise they're just arbitrary and they call any results into question, as Argentum was trying to explain. You can't accurately classify "best" by rolling dice on the criteria.

Someone earlier referenced POBRatings and I'd agree that his system seems the most scientific I've seen so far. I still think it has a degree of subjectivity but it tries its hardest to take the car into account, which I don't think your system does, not enough anyway. And before you think I'm just criticising let me say I couldn't do any better myself and if you re-labelled it as an attempt to classify the most successful drivers I'd probably be more on board with it. But as it is I don't think there's enough to convince me or others that "best" and "most successful" are interchangeable.
POB's daughter Catherine here, using his log-in:

Some analysts profess to take the car into account but don't actually. As POB argues:

“Smith claims to have taken the quality of the car into account (“it is rare that a great driver can prosper without a great car, or vice versa” (p. 10)), yet some of his findings seem to be results-based data, bolstering the dominance of he drivers whom he deemed to be the ‘top three’. Smith neglects the fact that his nominated three drivers, Fangio, Clark and Schumacher, had superior cars for longer periods of time than most.”
[...]
“Unpalatably to many, race results – such as Smith appears to use – have always had far more to do with the lucky timing of driver-car combinations and circumstances than with the popularly perceived but seldom occurring direct, level-playing-field, head-to-head competition between drivers.

“It is therefore not possible to distinguish between ‘Top Drivers’ across eras with any degree of quantifiable and convincing certainty. Smith’s ultimate conclusions – that Fangio, Clark and Schumacher were the top three Formula One drivers – remain an illogical and unsubstantiated leap.”

From: https://grandprixratings.blogspot.com/2 ... ysing.html

By using race times and re-computing the rating figures when drivers changed teams/cars and got new team-mates, POB's system does in fact take the car into account:

“My rating figures for the cars was obtained by subtracting the driver-rating figure from the package-rating and then adding back the base 100.0 figure to obtain the car-rating. For example, in 1950 the Fangio/ Alfa Romeo 158 package-rated at 100.0 and Fangio driver-rated at 100.0, so the ‘subtraction’ leaves 0.0 (100.0 – 100.0 = 0.0), then adding back 100.0 [a convention of POB’s Rating System] gives the Alfa Romeo 158 the ultimate car-rating figure of 100.0. That same year the next-fastest package was the Ascari/ Ferrari 375 which package-rated at 100.5. If we subtract Ascari’s driver-rating of 100.1 from his package-rating of 100.5 = 0.4, then add back the base 100.0, the resulting figure is 100.4, representing the Ferrari Type 375’s car-rating of 100.4 (100.5 – 100.1 = 100.4). The Sommer/ Talbot-Lago package-rating was 103.0; subtracting Sommer’s driver-rating of 100.8 results in the Talbot-Lago’s car-rating of 102.2 (103.0 – 100.8 = 2.2 + 100.0 = 102.2).
[...]
"Establishing how fast the car is can only be achieved by removing the contribution made by the team’s driver(s). By using the speed of the team’s fastest driver, this process will most accurately produce the ultimate speed of any particular car."

From: http://grandprixratings.blogspot.com/20 ... lable.html

--
Follow POB posthumously on Instagram: https://www.instagram.com/pob_renaissanceman/

Re: Best F1 driver ever?

Posted: Mon Aug 26, 2019 10:20 am
by POBRatings
bourbon19 wrote:
tim3003 wrote:A lot of people make the point that success in F1 is car-dependent. Fair enough, but as I think I said, the best drivers tend to get the best cars, so isn't the playing field thus level again?

Maybe it's the statistican in me that wants to define numerically who is the best ! Maybe my way is not the correct way. But if not can we agree on another objective criteria that is? If not then all we have is subjective views, of which there are as many as there are drivers! Surely that's not good enough..
Yeah, I don't see that as possible - I have not read every post, but if it was not mentioned, you have a number of top drivers with unfinished careers, so you could never really have an accurate measure. In any case, what is wrong with "the most success with the opportunities given, whatever they may be" statistic? That you have done a bang up job measuring.
POB's daughter Catherine here, using his log-in:

POB found that, in order to define numerically who is the best, we cannot escape some subjective interpretation. This means that the need for some debate or reasoned argument is unavoidable. With subjective views only, the question would be a case of ships passing in the night, while with raw statistics only, many of the nuances and complexities of racing would be bulldozed. Critique should therefore focus not on 'which measure' should be used but rather on 'whether the two measures/perspectives - objective and subjective - have been kept in adequate balance with each other' - without polarising. Once the set of measures has been established, POB argued that debate/critique then boiled down to which races to take into account (the database).

Relevant excerpts from POB's 2016 'Explanatory Chapters':
From: http://grandprixratings.blogspot.com/20 ... lable.html

"Results-based systems favour the best cars while points-based systems are relatively arbitrary."

"Methodologies that use race-results or which are purely statistics-based are biased in favour of the drivers in the best car/teams. As the best teams are at the top so their drivers will also be found at the top, not necessarily the best drivers." [paraphrased]

"Top cars/teams usually get the best drivers and drivers perform better in top teams/cars. This is where some subjective judgement enters my Rating System. However, the margin of error is limited by actual race and or pre-race times or an aggregate of the two times."

"They have not necessarily rewarded the fastest or best drivers. One cannot use the official points scores for driver comparisons, based as they are on the combined driver-and-car ‘package’ performance, and not for driver or car separately."

"driver competitiveness is not consistent over a career, even for the best drivers. My driver-rating study shows that rookies tend to build up to their peak after three to five seasons, and older men’s speed tapers off towards career-end."

The strength of POB's methodology is that it identifies the less-than-great drivers who won championships when in superior cars, and the great drivers who failed to win due to inferior cars.

Keeping both objective and subjective measures in intelligent balance with each other arguably offers the best chance of correctly measuring and identifying (1) the unacknowledged top-rated drivers and (2) the lesser drivers who won mainly through having top-rated cars. However, depending on individual psychologies, some feel uncomfortable with the nebulousness of the subjective and will therefore throw in their lot entirely with statistics, while others favour the subjective and are unable to sidestep personalities or nationalities, buying into today’s media-driven cult of the hero and the celebrity.

--
Follow POB posthumously on Instagram: https://www.instagram.com/pob_renaissanceman/

Re: Best F1 driver ever?

Posted: Mon Aug 26, 2019 2:55 pm
by tim3003
POBRatings wrote:
Response by POB's daughter Catherine, using his log-in:

POB's Rating System incorporates both raw statistics and subjective opinion - but this is not unconstrained opinion in that if one is taking into account the same 'secondary measures' that he is considering, one would have a high chance of coming to the same conclusion as him. After the point at which raw statistics are no longer decisive, his ratings and rankings are based on reasoned argument, based on a range of predefined criteria (his 'secondary measures').

I know POB would have enjoyed discussing these 'numerous sources' with you. Having studied every race since 1894, he identified them as:
[1] Driver
[2] Car
[3] Team-mate
[4] Rivals
[5] Own-team
[6] Mishaps/luck

He termed these his "six criteria for driver-dominance", dominance being defined as 'winning a high number of races per season'.

POB's thesis was that no driver, no matter how great and talented, would be able to dominate without enjoying at least two of these six criteria. "Obviously one of these two must be driver talent. Dominance is not however possible due solely to driver ability."

Evidence for his assertions about these 6 criteria can be found here:
http://grandprixratings.blogspot.com/20 ... ating.html
This all sounds hugely and impressively detailed, but what are its conclusions? I can't find them on the Blog site..

Re: Best F1 driver ever?

Posted: Mon Aug 26, 2019 4:20 pm
by POBRatings
tim3003 wrote:
POBRatings wrote:
Response by POB's daughter Catherine, using his log-in:

POB's Rating System incorporates both raw statistics and subjective opinion - but this is not unconstrained opinion in that if one is taking into account the same 'secondary measures' that he is considering, one would have a high chance of coming to the same conclusion as him. After the point at which raw statistics are no longer decisive, his ratings and rankings are based on reasoned argument, based on a range of predefined criteria (his 'secondary measures').

I know POB would have enjoyed discussing these 'numerous sources' with you. Having studied every race since 1894, he identified them as:
[1] Driver
[2] Car
[3] Team-mate
[4] Rivals
[5] Own-team
[6] Mishaps/luck

He termed these his "six criteria for driver-dominance", dominance being defined as 'winning a high number of races per season'.

POB's thesis was that no driver, no matter how great and talented, would be able to dominate without enjoying at least two of these six criteria. "Obviously one of these two must be driver talent. Dominance is not however possible due solely to driver ability."

Evidence for his assertions about these 6 criteria can be found here:
http://grandprixratings.blogspot.com/20 ... ating.html
This all sounds hugely and impressively detailed, but what are its conclusions? I can't find them on the Blog site..
POB's daughter Catherine here, using his log-in:

Thanks for your interest. POB's reams of computations (and re-computations if something in the field changed mid-way through) were all done by hand. It was insane how many figures and variables he could hold in his mind at any one time. I can only conclude that he had one of those extreme systematising brains on the autistic spectrum, that could perceive a larger pattern built up from an infinite number of tiny details and nuances e.g., modifications to car models that caused his ratings to change. A poster on the F1 Fanatics forum characterised him as "...a most knowledgeable and charming man. I always thought he must be quite mad to be attempting to rate so many era's, drivers etc. something my brain could not even contemplate, definitely a touch of genius in him." However, I don't think he could have tackled (and critiqued) the question of 'which is the best F1 driver ever?' without his in-depth knowledge of motor-racing history. He could give an example from history to illustrate any theoretical point he was making in his Rating System, assuring me of its validity and rigour.

His conclusions are in his 17 published volumes, available on Lulu.com:

The keystone volume is 'Explanatory Chapters':
http://www.lulu.com/shop/patrick-obrien ... 88213.html

His specific conclusions about which drivers top-rated each season are given in 14 volumes, one for each season from 1894 until 2016. These too can be found on the Lulu.com page. Each of these contains 'Guidelines', outlining his methods and how to interpret his figures. In these books, he doesn't expect you to take his conclusions on faith; he presents his reasoning. Some of his blog posts also discuss his conclusions.

'Further Analysis' was the last book he wrote on his Rating System, right before he died in 2017 (he left me a note about where I could find it on his computer):
http://www.lulu.com/shop/http://www.lul ... 25565.html
This volume contains his '6 criteria for driver dominance'.

These 17 volumes are as much a distillation of GP history as they are the conclusions his System arrived at about the 'best drivers'.

Re: Best F1 driver ever?

Posted: Thu Aug 29, 2019 9:39 pm
by mcdo
I'm always late to these things. Anyway, Prost

Re: Best F1 driver ever?

Posted: Tue Sep 03, 2019 3:46 pm
by tim3003
mcdo wrote:I'm always late to these things. Anyway, Prost
That's an opinion. This thread is about how you prove it..

Re: Best F1 driver ever?

Posted: Tue Sep 03, 2019 4:37 pm
by Blake
tim3003 wrote:
mcdo wrote:I'm always late to these things. Anyway, Prost
That's an opinion. This thread is about how you prove it..
no, it isn't. mcdo is entitled to his opinion especially as there is no way to "prove" who the best ever driver in F1, no matter who one selects.

Re: Best F1 driver ever?

Posted: Tue Sep 03, 2019 4:44 pm
by DOLOMITE
tim3003 wrote:
mcdo wrote: I'm always late to these things. Anyway, Prost
That's an opinion. This thread is about how you prove it..
Well clearly you can't and if you think about the whole thread is about opinions. Sure you can come up with some stats to back up your opinions, but that's all they're doing. Any system, however complex is flawed from the outset. You can't do it. No disrespect to anyone who tries, but you have to be realistic about what you are trying to do. Comparing drivers who never raced each other? Or who never drove the same cars? Think about it for a minute. Form your opinion, do some research, back it up to the nth degree, but never lose sight of the fact that it is just an opinion.

I give you an overheard conversation between my 8 and 10 year old boys

Boy 1 "That's just your opinion"
Boy 2 "So, I'm still right"
Boy 1 "It's an OPINION - you can't prove you're right! Opinions can't be proved - else it would be a fact!"
..slight delay..

Boy 2 "Well then what's the point in having one?...."

Re: Best F1 driver ever?

Posted: Tue Sep 03, 2019 5:57 pm
by F1 MERCENARY
tim3003 wrote:
mcdo wrote:I'm always late to these things. Anyway, Prost
That's an opinion. This thread is about how you prove it..
The way Michael did by dedicating his life to F1 exclusively for a number of years, even after he was married. Living apart from Corinna for over 18 months in his obsession to become an even better version of himself. He logged thousands upon thousands of laps and miles, looking to see where he could do things better/differently/unconventionally that would yield better results. And that didn't necessarily pertain to complete laps as a whole, and often he'd run laps in order to work on a single corner or series of corners in a bid tidy it up as best and as much as possible, seeking perfection in every possible facet for every possible scenario.

Never before, nor since has any driver ever gone to such lengths, and never has a driver been THAT much better than the rest of the field. Clark was likely the closest to this ability, and Senna as well, but as great as Senna was, he "expected" to win out of sheer arrogance at times and it wasn't unlike him to bullrush his way through the field in dangerous fashion, when his supreme ability would have allowed him to overtake those drivers anyway.

As well, in Michael's era and ever since, the significant car advantage a Jim Clark and many before Michael's era enjoyed was considerably greater than any advantage Michael enjoyed, and if you compare Michael's teammates' results in those years when he was on a completely different level, they didn't enjoy anywhere near the gap Michael would while driving the same machinery. Rubens was on a similar level at times during 2 of the seasons as Michael's teammate which says a great deal about both Michael & Rubens.

Prost is the most cerebral F1 driver of all time with elite skill & ability and knew better than anyone in history how to get inside the heads of his rivals, teammate or not. And on track he was a calculating machine. His ONLY knock was his dislike and perhaps too great respect for driving in the wet, but I don't blame him for that. He was an excellent wet weather driver, but when conditions became torrential, he wasn't comfortable pushing the envelope the way Senna could and would. But Prost is my #2

My O.G. list is as follows

1. Schumacher
2. Prost
3. Senna
4. Clark
5. Fangio


My Current list is as follows

1. Schumacher
2. Hamilton
3. Prost
4. Senna
5. Clark
6. Fangio
7. Stewart
8. Piquet (one of the most underrated drivers of all time, and the man could work on all aspects of his cars, likely the toughest road to F1 of all time)
9. Alonso (Would like to slot him in closer to Hamilton but his comparative lack of further success doesn't allow me to do so)
10. Lauda

Re: Best F1 driver ever?

Posted: Tue Sep 03, 2019 6:11 pm
by Alienturnedhuman
DOLOMITE wrote:
tim3003 wrote:
mcdo wrote: I'm always late to these things. Anyway, Prost
That's an opinion. This thread is about how you prove it..
Well clearly you can't and if you think about the whole thread is about opinions. Sure you can come up with some stats to back up your opinions, but that's all they're doing. Any system, however complex is flawed from the outset. You can't do it. No disrespect to anyone who tries, but you have to be realistic about what you are trying to do. Comparing drivers who never raced each other? Or who never drove the same cars? Think about it for a minute. Form your opinion, do some research, back it up to the nth degree, but never lose sight of the fact that it is just an opinion.

I give you an overheard conversation between my 8 and 10 year old boys

Boy 1 "That's just your opinion"
Boy 2 "So, I'm still right"
Boy 1 "It's an OPINION - you can't prove you're right! Opinions can't be proved - else it would be a fact!"
..slight delay..

Boy 2 "Well then what's the point in having one?...."
I wouldn't go as far to say that it's impossible for a system to exist that would fairly rank the drivers in order of talent, and also that some opinions are more valid than others. It's more like there is a scale of accuracy to nonsense on both systems and opinions. Some opinions are accurate, sone are nonsense, some are somewhere inbetween.

The issue with a lot of systems that exist on the internet is that they come about it from the wrong direction. The creator of the system often has an idea in their head of who they think is the best, and suddenly notice a pattern that matches their opinion - or thinks of the qualities they deem important and construct a system - usually not realising they are manipulating it - that matches closely to their opinion and their confirmation bias means they think they have found the one true system.

If there is a one true system, then it will be ridiculously complex, far beyond any of the simple algebra formulas people are excited to announce to the world. There is no much nuance that affects the final data we end up with. There is also so much we don't know. We don't know who was favoured behind the scenes, it's impossible for us to properly assess dice roll races like Hockenheim this year. But just because a system couldn't be created by a fan doesn't mean that hypothetically one couldn't be created by some all knowing being.

Obviously, such a system is 'practically' impossible. It's like dividing by zero - but just as you get closer to zero, the output gets closer and closer to infinity - it would be possible to start getting closer and closer to an objective and accurate system with smaller and smaller error bars. But it would require a lot of research (vastly more than keyboard warrior is going to have to have put in with a couple of weekends browsing Wikipedia) and a great deal of statistical skill and analysis - it would probably take a PhD thesis to get close.

POB's system - as was the case with every other - was designed by a flawed flesh and blood system, but I am inclined to believe that it's closer to the accurate end of the scale rather than the nonsense one. Is it perfect? Far from it. But it's clear that a great deal of research was put in to it, and attempts to counteract a lot of the biases and rogue results that affect many of the mechanical mathematical models of the past.

Re: Best F1 driver ever?

Posted: Tue Sep 03, 2019 6:16 pm
by Alienturnedhuman
Blake wrote:
tim3003 wrote:
mcdo wrote:I'm always late to these things. Anyway, Prost
That's an opinion. This thread is about how you prove it..
no, it isn't. mcdo is entitled to his opinion especially as there is no way to "prove" who the best ever driver in F1, no matter who one selects.
This is entirely correct. You can't prove an opinion. You can - however - justify it and I suspect that is what tim3003 is driving at. Although I think that mcdo's post - even if it probably is his actual opinion - is meant to be read in a slightly facetious tone.

Re: Best F1 driver ever?

Posted: Tue Sep 03, 2019 10:08 pm
by pokerman
mcdo wrote:I'm always late to these things. Anyway, Prost
Cheers to you as well. :)

Re: Best F1 driver ever?

Posted: Wed Sep 04, 2019 9:45 am
by DOLOMITE
Alienturnedhuman wrote: I wouldn't go as far to say that it's impossible for a system to exist that would fairly rank the drivers in order of talent, and also that some opinions are more valid than others. It's more like there is a scale of accuracy to nonsense on both systems and opinions. Some opinions are accurate, sone are nonsense, some are somewhere inbetween.

The issue with a lot of systems that exist on the internet is that they come about it from the wrong direction. The creator of the system often has an idea in their head of who they think is the best, and suddenly notice a pattern that matches their opinion - or thinks of the qualities they deem important and construct a system - usually not realising they are manipulating it - that matches closely to their opinion and their confirmation bias means they think they have found the one true system.

If there is a one true system, then it will be ridiculously complex, far beyond any of the simple algebra formulas people are excited to announce to the world. There is no much nuance that affects the final data we end up with. There is also so much we don't know. We don't know who was favoured behind the scenes, it's impossible for us to properly assess dice roll races like Hockenheim this year. But just because a system couldn't be created by a fan doesn't mean that hypothetically one couldn't be created by some all knowing being.

Obviously, such a system is 'practically' impossible. It's like dividing by zero - but just as you get closer to zero, the output gets closer and closer to infinity - it would be possible to start getting closer and closer to an objective and accurate system with smaller and smaller error bars. But it would require a lot of research (vastly more than keyboard warrior is going to have to have put in with a couple of weekends browsing Wikipedia) and a great deal of statistical skill and analysis - it would probably take a PhD thesis to get close.

POB's system - as was the case with every other - was designed by a flawed flesh and blood system, but I am inclined to believe that it's closer to the accurate end of the scale rather than the nonsense one. Is it perfect? Far from it. But it's clear that a great deal of research was put in to it, and attempts to counteract a lot of the biases and rogue results that affect many of the mechanical mathematical models of the past.
Well I'll have to respectfully disagree! I can't see how it's possible. I totally agree that some produce more reliable results but ultimately you're going to have the same list of @20 drivers just with slight differences in order. But however clever the system, however many variables are factored in and weightings added, it's all just a bit of fun. What about careers that are cut short for example? Who know how good Bianchi was? What if Alonso had been in a better car most of his career? What if Schumacher had had stronger teammates? etc etc All these would have affected the outcomes of these system and yet the drivers would have been driving at the same level of ability either way. What about drivers who believe in "winning at the slowest speed possible" etc etc. There's a hundred reasons why the numbers will only tell us so much.
POBs system was (is?) incredibly meticulous but take this year as an example - Mercedes the "best" car is pretty much beyond doubt but what if their drivers had been Bottas and say Ocon and Hamilton had taken a sabbatical. What the the results stats suggest to us then?

Like I said, I love stats and lists, always have, I just think people need to be careful when citing them as "proof" of something.

Re: Best F1 driver ever?

Posted: Wed Sep 04, 2019 11:37 am
by tim3003
DOLOMITE wrote:
Well I'll have to respectfully disagree! I can't see how it's possible. I totally agree that some produce more reliable results but ultimately you're going to have the same list of @20 drivers just with slight differences in order. But however clever the system, however many variables are factored in and weightings added, it's all just a bit of fun. What about careers that are cut short for example? Who know how good Bianchi was? What if Alonso had been in a better car most of his career? What if Schumacher had had stronger teammates? etc etc All these would have affected the outcomes of these system and yet the drivers would have been driving at the same level of ability either way. What about drivers who believe in "winning at the slowest speed possible" etc etc. There's a hundred reasons why the numbers will only tell us so much.
POBs system was (is?) incredibly meticulous but take this year as an example - Mercedes the "best" car is pretty much beyond doubt but what if their drivers had been Bottas and say Ocon and Hamilton had taken a sabbatical. What the the results stats suggest to us then?

Like I said, I love stats and lists, always have, I just think people need to be careful when citing them as "proof" of something.
As I've said before in this thread: I agree with you in part, but I disagree too: Yes any system is going to have a human bias, and none can account for all the might-have-beens and fluke circumstances.

However my conclusion therefore is that the only way to rank drivers is via their results, which are not open to opinion. As in any other sport, it's trophies and titles that define greatness, not might-have-beens. Hopefully over the longer term luck tends to balance out but I'm not saying it always will. No this system is not perfect, but unless you just want to say there's no way to tell and all is opinion then I don't see any alternative. As I understand it the POB system only ranks drivers in each year, not across the years, so does not answer my original question.

Re: Best F1 driver ever?

Posted: Wed Sep 04, 2019 1:09 pm
by Herb
tim3003 wrote:
DOLOMITE wrote:
Well I'll have to respectfully disagree! I can't see how it's possible. I totally agree that some produce more reliable results but ultimately you're going to have the same list of @20 drivers just with slight differences in order. But however clever the system, however many variables are factored in and weightings added, it's all just a bit of fun. What about careers that are cut short for example? Who know how good Bianchi was? What if Alonso had been in a better car most of his career? What if Schumacher had had stronger teammates? etc etc All these would have affected the outcomes of these system and yet the drivers would have been driving at the same level of ability either way. What about drivers who believe in "winning at the slowest speed possible" etc etc. There's a hundred reasons why the numbers will only tell us so much.
POBs system was (is?) incredibly meticulous but take this year as an example - Mercedes the "best" car is pretty much beyond doubt but what if their drivers had been Bottas and say Ocon and Hamilton had taken a sabbatical. What the the results stats suggest to us then?

Like I said, I love stats and lists, always have, I just think people need to be careful when citing them as "proof" of something.
As I've said before in this thread: I agree with you in part, but I disagree too: Yes any system is going to have a human bias, and none can account for all the might-have-beens and fluke circumstances.

However my conclusion therefore is that the only way to rank drivers is via their results, which are not open to opinion. As in any other sport, it's trophies and titles that define greatness, not might-have-beens. Hopefully over the longer term luck tends to balance out but I'm not saying it always will. No this system is not perfect, but unless you just want to say there's no way to tell and all is opinion then I don't see any alternative. As I understand it the POB system only ranks drivers in each year, not across the years, so does not answer my original question.

The bit in bold is what I think everyone needs to realise. Every system I have seen has been flawed and open to bias. I'm fine with the fact that it is impossible to definitively rank drivers over generations.

Re: Best F1 driver ever?

Posted: Thu Sep 05, 2019 8:21 pm
by tim3003
Herb wrote:
tim3003 wrote:
No this system is not perfect, but unless you just want to say there's no way to tell and all is opinion then I don't see any alternative.

The bit in bold is what I think everyone needs to realise. Every system I have seen has been flawed and open to bias. I'm fine with the fact that it is impossible to definitively rank drivers over generations.
In what way is ordering drivers by the number of titles and wins they've achieved biassed? It just measures success. And surely any GP driver would agree that their goal is winning races and titles..

Re: Best F1 driver ever?

Posted: Thu Sep 05, 2019 8:24 pm
by Paolo_Lasardi
tim3003 wrote:
Herb wrote:
tim3003 wrote:
No this system is not perfect, but unless you just want to say there's no way to tell and all is opinion then I don't see any alternative.

The bit in bold is what I think everyone needs to realise. Every system I have seen has been flawed and open to bias. I'm fine with the fact that it is impossible to definitively rank drivers over generations.
In what way is ordering drivers by the number of titles and wins they've achieved biassed? It just measures success. And surely any GP driver would agree that their goal is winning races and titles..
Most successful does not equal best because other factors than driver quality (e.g. car quality, reliability, luck, etc. pp.) influence success in F1.

Re: Best F1 driver ever?

Posted: Thu Sep 05, 2019 8:28 pm
by Herb
tim3003 wrote:
Herb wrote:
tim3003 wrote:
No this system is not perfect, but unless you just want to say there's no way to tell and all is opinion then I don't see any alternative.

The bit in bold is what I think everyone needs to realise. Every system I have seen has been flawed and open to bias. I'm fine with the fact that it is impossible to definitively rank drivers over generations.
In what way is ordering drivers by the number of titles and wins they've achieved biassed? It just measures success. And surely any GP driver would agree that their goal is winning races and titles..
That data biased towards those that spend time in better equipment.

Re: Best F1 driver ever?

Posted: Thu Sep 05, 2019 8:37 pm
by Asphalt_World
Number of wins when compared drivers from different eras is pointless due to the massive differences in number of races per year in the sport.

Re: Best F1 driver ever?

Posted: Thu Sep 05, 2019 10:17 pm
by Blake
Paolo_Lasardi wrote:
tim3003 wrote:
Herb wrote:
tim3003 wrote:
No this system is not perfect, but unless you just want to say there's no way to tell and all is opinion then I don't see any alternative.

The bit in bold is what I think everyone needs to realise. Every system I have seen has been flawed and open to bias. I'm fine with the fact that it is impossible to definitively rank drivers over generations.
In what way is ordering drivers by the number of titles and wins they've achieved biassed? It just measures success. And surely any GP driver would agree that their goal is winning races and titles..
Most successful does not equal best because other factors than driver quality (e.g. car quality, reliability, luck, etc. pp.) influence success in F1.
+1, again.

Paolo, we have been agreeing way too much lately!

Re: Best F1 driver ever?

Posted: Mon Sep 09, 2019 9:02 am
by mcdo
tim3003 wrote:
mcdo wrote:I'm always late to these things. Anyway, Prost
That's an opinion. This thread is about how you prove it..
The record breaker of his time during what has often been considered the "Golden Age" of Formula 1. Genuine title contender across 9 different seasons with 4 different teams and 4 different engine marques. Won a title in a car that wasn't the fastest (1986). Bookended his career with points on debut, podium on retirement (as World Champion)

4 titles won. The drivers in the other car were:
- Niki Lauda
- Keke Rosberg
- Ayrton Senna
- Damon Hill

Other teammates beaten over a season:
- Nigel Mansell
- Jean Alesi
- Rene Arnoux
- Eddie Cheever
- Stefan Johansson

You won't find many lists better than that. I'm content that it justifies my opinion

Re: Best F1 driver ever?

Posted: Mon Sep 09, 2019 9:15 am
by Exediron
mcdo wrote:4 titles won. The drivers in the other car were:
- Niki Lauda
- Keke Rosberg
- Ayrton Senna
- Damon Hill
Prost is fairly unique in the quality of teammates he had when winning his titles. You won't find another driver who won four titles with world champions in the other car.

Re: Best F1 driver ever?

Posted: Mon Sep 09, 2019 9:40 am
by POBRatings
DOLOMITE wrote: Obviously, such a system is 'practically' impossible. It's like dividing by zero - but just as you get closer to zero, the output gets closer and closer to infinity - it would be possible to start getting closer and closer to an objective and accurate system with smaller and smaller error bars. But it would require a lot of research (vastly more than keyboard warrior is going to have to have put in with a couple of weekends browsing Wikipedia) and a great deal of statistical skill and analysis - it would probably take a PhD thesis to get close.

POB's system - as was the case with every other - was designed by a flawed flesh and blood system, but I am inclined to believe that it's closer to the accurate end of the scale rather than the nonsense one. Is it perfect? Far from it. But it's clear that a great deal of research was put in to it, and attempts to counteract a lot of the biases and rogue results that affect many of the mechanical mathematical models of the past.
POB's daughter Catherine here, using his log-in:

POB would agree – no system is perfect but some are less flawed than others. The rogue results regularly spewed out by mechanical mathematical models cast doubt on their methodology.

Some say that X driver was the best because he won more races. Some say it’s because of better percentages. Some because X driver brought passion to the sport. After decades of searching, POB realised that the only way to measure and separate Driver and Car from Package is through package-speed across a full season.

Using race-times as a primary measure, and comparing drivers in same-cars to infer relative speed, means that any subjective judgement is strictly constrained ('bounded'). Surely that is the most useful type of subjectivity – tethered tightly to objective measures? i.e., to reality.

IMHO, seeing the years of research and sheer volume of computation that went into POB’s Grand Prix Rating System, he earned a PhD many times over.

Re: Best F1 driver ever?

Posted: Mon Sep 09, 2019 9:47 am
by Jezza13
Exediron wrote:
mcdo wrote:4 titles won. The drivers in the other car were:
- Niki Lauda
- Keke Rosberg
- Ayrton Senna
- Damon Hill
Prost is fairly unique in the quality of teammates he had when winning his titles. You won't find another driver who won four titles with world champions in the other car.
Agree to a certain extent but to be fair, Lauda & Rosberg we're in their retirement year when Prost won those titles as their results in those years reflect.

Senna had atrocious reliability issues in 89 & of course there was Suzuka.

Hill was just in his 2nd year of F1, very much a #2 at Williams and a vastly inferior driver to Prost in 93.

None of the above though dampens my opinion of Prost. To me he's currently # 2 on my list of the best i've seen. He was sublime in 86, brilliant, and calmly drove quality races, clocking up points while Williams imploded to grab defeat from the jaws of victory.

He championed Senna joining McLaren & then didn't take a step back when it came time to fight fire with fire & he out drove Mansell at Ferrari in 1990

Re: Best F1 driver ever?

Posted: Mon Sep 09, 2019 9:56 am
by POBRatings
DOLOMITE wrote: Well I'll have to respectfully disagree! I can't see how it's possible. I totally agree that some produce more reliable results but ultimately you're going to have the same list of @20 drivers just with slight differences in order. But however clever the system, however many variables are factored in and weightings added, it's all just a bit of fun. What about careers that are cut short for example? Who know how good Bianchi was? What if Alonso had been in a better car most of his career? What if Schumacher had had stronger teammates? etc etc All these would have affected the outcomes of these system and yet the drivers would have been driving at the same level of ability either way. What about drivers who believe in "winning at the slowest speed possible" etc etc. There's a hundred reasons why the numbers will only tell us so much.
POBs system was (is?) incredibly meticulous but take this year as an example - Mercedes the "best" car is pretty much beyond doubt but what if their drivers had been Bottas and say Ocon and Hamilton had taken a sabbatical. What the the results stats suggest to us then?

Like I said, I love stats and lists, always have, I just think people need to be careful when citing them as "proof" of something.
POB's daughter Catherine here, using his log-in:

POB would agree that linear ranking lists are problematic. Debate about “who is the best driver ever” is more likely to generate dissension than arrive at an answer. This is because, as POB’s 17-volume thesis proved, the question is not coherent. It is not coherent to compare drivers that never faced each other. POB contended that it is impossible to nominate any single driver, or even a top five or ten, as ‘the greatest’. The most he could conclude is that from 1894-2013 there are 87 drivers who top driver-rated at the ultimate 100.0 according to his System. It is impossible to differentiate between them and identify just one or even three as “the greatest driver”; instead he places them all equal in the ‘top tier’. Those rating 100.1, 100.2, 100.3 etc would form the second, third and fourth tiers, with 100.5 being the slowest cut-off rating POB identified for being a potential race-winner. As POB writes:

“According to my System, the fastest drivers of each era rate equal (at 100.0) in the first tier: from Fangio, Ascari and Moss through to Prost, Schumacher and Alonso. I certainly would not be able to rank them linearly due to the incomparable variables. In sum, it is not possible to rank any of the top drivers across eras as greater or as the greatest. They never faced each other and cannot be directly compared.”

Who would want to be in an Elite List of 87 Drivers out of some 2,000-plus drivers since 1894?! The question of “who is the greatest driver” will likely endure because it taps into to our winner-takes-all culture, whereby the winner is disproportionately rewarded in terms of money, attention, publicity, congratulations, title and trophy, while many worthy runner-ups might have equal speed, skill, passion etc but just missed out on one or more of POB’s 6 criteria (listed above) coming together.

POB's Rating System factors in all the variables or potential influencers that you raise.

Re: Best F1 driver ever?

Posted: Sat Aug 28, 2021 2:43 pm
by POBRatings
POB's daughter Catherine here, using his log-in:

My background: little knowledge of F1 racing but an interest in my dad’s Grand Prix Rating System – some knowledge of his methodology but little knowledge of racing!

Continuing the debate about the many attempts to compute driver ratings/rankings, someone recently drew my attention to the following system devised try to answer to the question: “Who is the fastest driver?”

AWS/Formula1.com ratings

https://www.formula1.com/en/latest/arti ... iH1Cz.html

In 2020, Amazon Web Services and Formula One partnered to search for the fastest driver over one lap. Powered by AWS (Amazon Web Services), Formula1.com created a rating system based on machine learning. Amazon’s Machine Learning Solutions Lab took a full year to build an algorithm which could find the answer to the question: “Who is the fastest F1 driver over the past 40 years?”.

I understand that, whereas the POB System computes the fastest package (driver + car) based on race times (and pre-race times as a secondary measure), the Formula1.com system computes package times based on only a single qualifying lap.

Can anyone help with my questions about this?

(1) Does this methodology follow POB’s method in terms of trying to provide driver-ratings by combining time-speed gaps (differentials) between teammates?

(2) I understand that this system dates back only to 1983. How did they start with their initial comparison figures, if their system is based on a comparison of team-mate differentials? (Fernando Alonso was Lewis Hamilton’s teammate in 2007, Lewis Hamilton in turn has been Valtteri Bottas’s teammate since 2017, Valtteri Bottas was Felipe Massa’s teammate in 2014-2016, and Felipe Massa was Michael Schumacher’s teammate in 2006, etc)

(3) Does this system give any obviously anomalous ratings, for example more back-marker drivers coming out ahead of well-known fast drivers?

(4) Is it a problem that data is limited to qualifying times alone? Does this actually help to address “Who is the fastest F1 driver over the past 40 years?”. In what ways might the actual race time/speeds differ? Or is it more important that drivers are compared on the same (relatively narrow but consistent) measure – so they would all face the same biases/conditions?

(5) Out of interest, is this system widely respected and used? Are there any critiques of it?

Apologies for any ignorance shown in asking these questions! I'm keen to learn more and would welcome any corrections/ clarifications/ elucidations.

Re: Best F1 driver ever?

Posted: Sat Aug 28, 2021 4:12 pm
by schumilegend
My List
1) Schumacher
2) Senna
3) Fangio
4) Hamilton
5) Clark
6) Alonso
7) Prost
8) Stewart
9) Vettel
10) Lauda

Re: Best F1 driver ever?

Posted: Sun Aug 29, 2021 1:31 am
by Exediron
POBRatings wrote:
Sat Aug 28, 2021 2:43 pm
Can anyone help with my questions about this?

(1) Does this methodology follow POB’s method in terms of trying to provide driver-ratings by combining time-speed gaps (differentials) between teammates?

(2) I understand that this system dates back only to 1983. How did they start with their initial comparison figures, if their system is based on a comparison of team-mate differentials? (Fernando Alonso was Lewis Hamilton’s teammate in 2007, Lewis Hamilton in turn has been Valtteri Bottas’s teammate since 2017, Valtteri Bottas was Felipe Massa’s teammate in 2014-2016, and Felipe Massa was Michael Schumacher’s teammate in 2006, etc)

(3) Does this system give any obviously anomalous ratings, for example more back-marker drivers coming out ahead of well-known fast drivers?

(4) Is it a problem that data is limited to qualifying times alone? Does this actually help to address “Who is the fastest F1 driver over the past 40 years?”. In what ways might the actual race time/speeds differ? Or is it more important that drivers are compared on the same (relatively narrow but consistent) measure – so they would all face the same biases/conditions?

(5) Out of interest, is this system widely respected and used? Are there any critiques of it?

Apologies for any ignorance shown in asking these questions! I'm keen to learn more and would welcome any corrections/ clarifications/ elucidations.
I recall that this official ranking has come up on here before, although I couldn't find the discussion. I'll do my best to answer the questions you pose:

1) Although they haven't made the precise methodology public, my belief is that this is the core mechanism. However, the article also states that the machine learning algorithm weights the results based on things such as age and time together, and without knowing how much that weighting adjusts the final result it's difficult to determine how far the methodology strays from that employed by POB.

2) I believe (although I don't know) that one of the advantages of using machine learning is that they cross-compared every possible linkage between drivers, which likely gave enough data to indicate someone at #1 (Senna, apparently) without a human selecting a standard of comparison.

Whether this is a strength or not, is a matter of opinion. I know that in POB's system he would sometimes make an executive decision to fix a driver who was clearly operating at top-level speed for a season (such as Verstappen this season, I would imagine) to get the figures in line. I don't believe there's any such human oversight on the AWS results.

3) Many would say so. There are a few that don't agree at all with the comparative rankings generated by those on this forum (as seen in this thread, and elsewhere). For example:
  • They have Hamilton and Alonso at a smaller gap to one another than was observed when they were actually teammates, and also at a smaller gap than their shared partnership with Button would suggest.
  • Nico Hulkenberg is rated very highly, and ahead of at least one teammate who was definitively faster than him (Ricciardo)
  • Most people would consider Max to be significantly underrated
There is also a very sizable apparent recency bias, with a great many current drivers or recently retired drivers featuring.

4) The study was done specifically to determine the fastest qualifier, so in that regard I wouldn't say it's a problem -- as long as one accepts that the answer to who is the fastest driver all-around may not be the same.

For example, Senna being half a second quicker than Prost is likely accurate for qualifying, but was not an advantage he carried consistently into races (his margin over Prost was usually much smaller than that on race day).

5) I would say it isn't widely accepted, or particularly widely discussed -- which is perhaps a shame, because there's an interesting debate to be had on where the AWS findings differ from the human viewpoint.

For some further reading, I don't know if you've already come across this, but Amazon has a more in-depth explanation in this article, discussing some of the methodology in more detail:

https://aws.amazon.com/blogs/machine-le ... formula-1/

Re: Best F1 driver ever?

Posted: Sun Aug 29, 2021 2:01 am
by EPROM
Browsing the blog article, this AWS work sounds more like regular big data analytics, not machine learning.

Machine learning (particularly deep learning) usually involves training a model with sample data, and then using it to predict outcomes on a totally different data set. E.g., you give it a series of photos of monkeys and donkeys and train the model with which of them is a monkey vs. a donkey. You then show the model a new photo and have it predict whether it's a monkey or donkey. Dumb example, but the same idea.

I would think that the AWS on-screen display of remaining tire life would be a much better example of machine learning than this - which seems more like a sophisticated multiple regression model, just described by a popular buzzword. Not invalid - just a bit of a stretch to call it ML.

Re: Best F1 driver ever?

Posted: Sun Aug 29, 2021 3:47 pm
by POBRatings
EPROM wrote:
Sun Aug 29, 2021 2:01 am
Browsing the blog article, this AWS work sounds more like regular big data analytics, not machine learning.

Machine learning (particularly deep learning) usually involves training a model with sample data, and then using it to predict outcomes on a totally different data set. E.g., you give it a series of photos of monkeys and donkeys and train the model with which of them is a monkey vs. a donkey. You then show the model a new photo and have it predict whether it's a monkey or donkey. Dumb example, but the same idea.

I would think that the AWS on-screen display of remaining tire life would be a much better example of machine learning than this - which seems more like a sophisticated multiple regression model, just described by a popular buzzword. Not invalid - just a bit of a stretch to call it ML.
POB's daughter Catherine here, using his log-in:

I'm not a statistician or mathematician but your clear explanation persuades me - that 'machine learning' is not 100% accurately used here. Perhaps they confounded the name of their lab with the specific statistical tests they are running (regression analysis)!

"Fastest Driver
As part of F1’s 70th anniversary celebrations and to help fans better understand who are the fastest drivers in the sport’s history, F1 and the Amazon Machine Learning Solutions Lab teamed up to develop Fastest Driver, the latest F1 Insight powered by AWS.

"Fastest Driver uses AWS machine learning (ML) to rank drivers using their qualifying sessions lap times from F1’s Historic Data Repository going back to 1983. In this post, we demonstrate how by using Amazon SageMaker, a fully managed service to build, train, and deploy ML models, the Fastest Driver insight can objectively determine the fastest drivers in F1."

Source: https://aws.amazon.com/blogs/machine-le ... formula-1/

Re: Best F1 driver ever?

Posted: Mon Aug 30, 2021 10:25 am
by POBRatings
POB's daughter Catherine here, using his log-in:
Exediron wrote:
Sun Aug 29, 2021 1:31 am
I recall that this official ranking has come up on here before, although I couldn't find the discussion. I'll do my best to answer the questions you pose:

1) Although they haven't made the precise methodology public, my belief is that this is the core mechanism. However, the article also states that the machine learning algorithm weights the results based on things such as age and time together, and without knowing how much that weighting adjusts the final result it's difficult to determine how far the methodology strays from that employed by POB.

2) I believe (although I don't know) that one of the advantages of using machine learning is that they cross-compared every possible linkage between drivers, which likely gave enough data to indicate someone at #1 (Senna, apparently) without a human selecting a standard of comparison.

Whether this is a strength or not, is a matter of opinion. I know that in POB's system he would sometimes make an executive decision to fix a driver who was clearly operating at top-level speed for a season (such as Verstappen this season, I would imagine) to get the figures in line. I don't believe there's any such human oversight on the AWS results.

3) Many would say so. There are a few that don't agree at all with the comparative rankings generated by those on this forum (as seen in this thread, and elsewhere). For example:
  • They have Hamilton and Alonso at a smaller gap to one another than was observed when they were actually teammates, and also at a smaller gap than their shared partnership with Button would suggest.
  • Nico Hulkenberg is rated very highly, and ahead of at least one teammate who was definitively faster than him (Ricciardo)
  • Most people would consider Max to be significantly underrated
There is also a very sizable apparent recency bias, with a great many current drivers or recently retired drivers featuring.

4) The study was done specifically to determine the fastest qualifier, so in that regard I wouldn't say it's a problem -- as long as one accepts that the answer to who is the fastest driver all-around may not be the same.

For example, Senna being half a second quicker than Prost is likely accurate for qualifying, but was not an advantage he carried consistently into races (his margin over Prost was usually much smaller than that on race day).

5) I would say it isn't widely accepted, or particularly widely discussed -- which is perhaps a shame, because there's an interesting debate to be had on where the AWS findings differ from the human viewpoint.

For some further reading, I don't know if you've already come across this, but Amazon has a more in-depth explanation in this article, discussing some of the methodology in more detail:

https://aws.amazon.com/blogs/machine-le ... formula-1/
Thank you @Exediron for your extremely helpful response and pertinent information.

Perhaps they should revise their heading “The fastest driver in Formula 1” more modestly as “The fastest qualifier in Formula 1”. Reserve the word “driver” for the more complex/ stringent measurement, based on actual race times?

Even as a non-GP fan, I can see how your point 3) should generate lots of methodological delving and debate:

3) Many would say so. There are a few that don't agree at all with the comparative rankings generated by those on this forum (as seen in this thread, and elsewhere). For example:
They have Hamilton and Alonso at a smaller gap to one another than was observed when they were actually teammates, and also at a smaller gap than their shared partnership with Button would suggest.
Nico Hulkenberg is rated very highly, and ahead of at least one teammate who was definitively faster than him (Ricciardo)
Most people would consider Max to be significantly underrated
There is also a very sizable apparent recency bias, with a great many current drivers or recently retired drivers featuring.


I'm not so sure about your point here:

"Whether this is a strength or not, is a matter of opinion. I know that in POB's system he would sometimes [SIC] make an executive decision to fix a driver who was clearly operating at top-level speed for a season (such as Verstappen this season, I would imagine) to get the figures in line. I don't believe there's any such human oversight on the AWS results."

My understanding is that POB only did this once: when there were too few points of comparison in the early years of GP racing: "I therefore used Charron for 1898 as the first driver-rating benchmark. This rating was corroborated by contemporary sources, author Gerald Rose and driver Charles Jarrott, who reckoned Charron was the fastest" (Explanatory Chapter, 2016, p. 130). Hence the need to start his computations from Grand Prix racing's inception (1894).

Thereafter, he never needed to do this again; he used actual race times to establish arithmetically who attained the fastest rating (denoted 100.00 on his system, with 100.2, 100.3, 100.6, etc denoting slower ratings).

POB did refer to the opinions of experts such as team personnel, rivals, peers, observers and journalists, not to 'fix' his figures but rather to check that his computations tallied with what was going on on the track - to avoid the anomalous ratings so often seen in algorithm-generated ratings that try to by-pass human judgement and discernment (often unfairly denounced as 'subjectivity').

EDIT 1Sep21: In POB's own words: "In order to separate the driver from the package, the top-rated drivers as identified by my System are used as the base or benchmark for the driver-ratings each season. These are usually the drivers in the packages that score the fastest season-average time. Exceptions are when a top-rated driver is not in a top-rated car, such as Stirling Moss in 1961 or Alonso in 2008-2009 and in 2015-2016. Top cars/teams usually get the best drivers and drivers perform better in top teams/cars. This is where some subjective judgement enters my Rating System. However, the margin of error is limited by actual race and or pre-race times or an aggregate of the two times." (Explanatory Chapters, 2016, p. 117)

Re: Best F1 driver ever?

Posted: Mon Aug 30, 2021 10:33 am
by POBRatings
POB's daughter Catherine here, using his log-in:

I’ve had a read-through of what I will call ‘Smedley’s system’ (albeit it was co-authored) on Amazon Web Services (AWS):

The fastest driver in Formula 1
by Rob Smedley, Colby Wise, Delger Enkhbayar, George Price, Ryan Cheng, and Guang Yang | on 20 AUG 2020 | in Amazon SageMaker, Artificial Intelligence


Rob Smedley
https://aws.amazon.com/blogs/machine-le ... formula-1/

Here are a few things that jumped out at me while reading through it.

[1] They call their system “the first objective and data-driven model to determine who might be the fastest driver ever”. This strikes me as a deliberate lack of acknowledgement of POB’s work, and smacks of plagiarism, given the many points of conceptual similarity between the systems, and the fact that POB’s system has been in the public arena since 2011 (it was first publicised on Peter Windsor’s website): https://grandprixratings.blogspot.com/2 ... ricks.html.

The tagline of POB’s blog clearly established his work as “the first” and “the most objective”:

Patrick O’Brien’s Grand Prix Ratings
"He was a mine of accurate information and his book is a respected and valued part of my racing library." ~ Stirling Moss, OBE. *** “Patrick O’Brien’s system is the most objective I’ve seen to date.” ~ Peter Windsor. *** "He probably pushed F1 metrics forward further than anybody else has ever done and his contribution will not be forgotten." ~ PF1 forum. *** "His rating system [...] brought some kind of consistent view of F1 throughout the decades and offered me a lot of insight." ~ PF1 forum.
https://grandprixratings.blogspot.com/

The following strike me as evidence that they got a significant leg-up from POB’s rating system, which goes completely unacknowledged, as far as I can see:

[1a] The concept of inter-linkages between drivers who raced together on the same team providing a constant as a basis for comparison.

POB ('Explanatory Factors', 2016, p. 125): "Today with just two-driver teams, there are usually only about four or five interlinks. Even though there are fewer interlinks today than before 1961, there are more data points today because it is now compulsory for teams to compete in every race and all pre-race times are fully recorded."

POB ('Explanatory Factors', 2016, p. 118): "STEP 5: I now link this third driver’s (Coulthard’s) performances through Webber’s to Vettel’s. This I do on my working spreadsheets, developed from race-by-race tracking to arrive at season-by-season average time-based driver-ratings. All are measured proportionally against the fastest driver(s) within each season."

Whereas POB writes about his massive spreadsheet of driver-ratings going back to 1894, Smedley writes of “a network of teammate comparisons over the years”, going back to 1983. “For example, Sebastian Vettel and Max Verstappen have never been on the same team, so we compare them through their respective connections with Daniel Ricciardo at Red Bull.”

This ground-breaking concept of ‘the gap’ between team-mates was a method that POB cracked in 2002, aged 58, after a lifetime’s immersion in GP & F1 literature and race-viewing, and after publishing an analytical book on GP racing in 1994.

Smedley: “we compare qualifying data for drivers on the same race team (such as Aston Martin Red Bull Racing), where teammates have competed against each other in a minimum of five qualifying sessions. By holding the team constant, we get a direct performance comparison under the same race conditions while controlling for car effects.”

Instead of "team", POB referred to using the "car" as a constant. Both capture the same variables.

[1b] POB’s system is mindful to exclude outlier sessions – a nuance Smedley et al. incorporated in their system: “We identify and remove anomalous lap time outliers” ... e.g., "Vettel being penalized to comply with the 107% rule (which forced him to start from the pit lane)."

POB (Explanatory Chapter, 2016, p. 103): "But even in these ‘pure’ cases, a driver would very occasionally have trouble with his car or encounter traffic (even during one lap!) and set a time much slower than his norm. These I treated as outliers and excluded the time. Conversely, if race-time data were contaminated by car or driver trouble, causing a package’s time to deviate very obviously from its norm, I omitted the time and reverted to the pre-race times, as being more representative of how the package compared."

[1c] In ‘separating driver performance from car performance,” one of the aims of POB’s system was to identify fast drivers in slow cars and slow drivers in fast cars.

Similarly, Smedley’s model prides itself on recognising ‘unsung heroes’: “the model has ranked him [Kovalainen] so highly because of his consistent qualifying performances throughout his career. I, for one, am extremely happy to see Kovalainen get the data-driven recognition that he deserves for that raw talent that was always on display during qualifying”

= but would this translate into race performance? Or would some drivers excel in qualifying while others excel instead in the actual races? Surely the races require more stamina so that would be a more accurate measure (less dependent on car machinery) overall? Surely a point debate.

Smedley: “…Kovalainen doesn’t have the same number of World Championships as Hamilton, but his qualifying statistics speak for themselves—the model has ranked him high because of his consistent qualifying performance throughout his career.”

To address the question of whether some drivers differ in skill between qualifying and actual races, is there some statistical analysis of differentials between qualifying vs race performance?

POB appears to have done this analysis (from his 'Explanatory Chapters', 2016, p. 99): "Although packages are invariably slower in the races than in pre-race times, the gaps between team-mate packages and between different team packages are fairly constant whether pre-race or race-time. This is shown by an example of just two drivers below, although this pattern holds for the whole field throughout history.”

Do Smedley et al. make any effort to compare qualifying and race performance?

[1d] Smedley’s laying out of the problem (problematisation) bears an uncanny similarity to POB’s:

POB’s ‘introduction’ to his system, in the Guidelines section of all his Rating System books:

"The 2013 Formula One season was dominated by the Vettel/ Red Bull-Renault package, which won 13 of the 19 races. Many reckon that Vettel is undoubtedly one of the great drivers. Some however question this, arguing that Vettel was fortunate in having the fastest car, the Red Bull-Renault. Just how good was Vettel compared with his peers? Can his performance be separated from the performance of his car?
"The 2012 Formula One season was a close-fought, year-long battle between three packages: the Hamilton/ McLaren-Mercedes, the Vettel/ Red Bull Renault, and the Alonso/ Ferrari. It ended at the Brazilian Grand Prix finale on an exciting note: Red Bull-Renault driver Sebastian Vettel won the Drivers Championship title narrowly, by just three points from Ferrari’s Fernando Alonso. Vettel won 5 races from 7 poles, Hamilton won 4 races from 8 poles, while Alonso scored 3 wins from just 2 poles.
"However, if we consider that the front-running Hamilton/ McLaren-Mercedes package suffered several tardy pit-stops that curbed potential wins, and that the Alonso/ Ferrari package was clearly slower in both qualifying and the races throughout the season, one has to ask: ‘Who really was the fastest driver?’”


Smedley: “Formula 1 (F1) racing is the most complex sport in the world. It is the blended perfection of human and machine that create the winning formula. It is this blend that makes F1 racing, or more pertinently, the driver talent, so difficult to understand. How many races or Championships would Michael Schumacher really have won without the power of Benetton and later, Ferrari, and the collective technical genius that were behind those teams? Could we really have seen Lewis Hamilton win six World Championships if his career had taken a different turn and he was confined to back-of-the-grid machinery? Maybe these aren’t the best examples because they are two of the best drivers the world has ever seen. There are many examples, however, of drivers whose real talent has remained fairly well hidden throughout their career. Those that never got that “right place, right time” break into a winning car and, therefore, those that will be forever remembered as a midfield driver.”

[1e] Similarly, the ‘humble disclaimer’ about the accuracy/ objectivity/ validity/ veracity of the figures sounds like POB’s:

POB: (from his 'Explanatory Chapters', 2016, p. 38):

‘Although journalist and analyst Peter Windsor wrote in 2010 that my Rating System was “the most objective I’ve seen to date” (either from his 2010 blog and/or from one of his 2010 F1 Racing Japan magazine race reports), my System is not entirely objective nor perfect due to a number of factors which I will discuss below.’

[...]

“Peter Windsor, writing about my Rating System in December 2010, stated:

'… this system, like any other, is by definition imperfect. It is, though, about as near as you can get to the truth. Patrick did his own arithmetic – and guess what: the difference between Nico Rosberg and Michael Schumacher (0.3) is exactly the difference between the two drivers that the Mercedes F1 team themselves established by mid-season with their own methodology. Someone must be doing something right!'

Source: ‘Unique F1 Driver Ratings, 2010’ by Peter Windsor 阿拉蕾, 2010-12-10, Retrieved from https://m.hupu.com/bbs/1754685.html


POB ('Explanatory Factors', 2016, p. 177): "My own view is that Grand Prix racing is not an exact science and therefore does not lend itself entirely to black-and-white analysis. I suspect however that we will continue to argue for the superiority of one method over another, rather than appreciating that both contribute to the scientific analysis of Grand Prix racing performance and, most importantly, generate discernment and debate."

Smedley: “These rankings aren’t proposed as definitive, and there will no doubt be disagreement among fans. In fact, we encourage a healthy debate! Fastest Driver presents a scientific approach to driver ranking aimed at objectively assessing a driver’s performance controlling for car difference.”

[2] Where the systems differ:

[2a] In using “qualifying sessions lap times”, Smedley points out that their system does not take into account “racecraft or the ability to win races or drive at 200 mph while still having the bandwidth to understand everything going on around you…”.

In contrast, by using actual-race times as a primary measure and pre-race-times as a secondary measure when required, POB claims to take all this into account:

“During this work, my Rating System was criticised by a prominent Formula One journalist to the effect that it does not take into account a driver’s ‘management abilities’, that is, out-of-car skills and talents, such as Michael Schumacher had in ‘organising’ a team around himself to enhance success. This is not so. My System takes everything into account. Driver and car performance are measured on-track, and therefore include testing, preparation, qualifying, racing, life-path experience, and every capability a driver or anyone else may have brought to bear on performance during design.” ('Explanatory Chapters', 2016, p. 31)

[2b] Continuous updating of the figures:

Smedley: “the qualifying data consumed by the model is updated with fresh lap times after every race weekend”

In contrast, POB used to do the updating by hand – an incredible mass of figures to manipulate in one’s head.

Conclusion

It is clear that there are more than a few similarities between the two system, POB's devised from 2002 and publicised online since 2011; Smedley's account online dated 20 August 2020: https://aws.amazon.com/blogs/machine-le ... formula-1/

It’s a shame that Smedley et al. were unable to acknowledge any indebtedness to POB. The echoes of so many of POB’s analytical sentiments, nuances and conclusions are evident in Smedley’s blog alone. Why not simply acknowledge and critique POB’s contribution and then state how their system differs and how it develops it further e.g., using “Amazon SageMaker, a fully managed service to build, train, and deploy ML models”?

Even Isaac Newton showed more humility: “If I have seen further than others,” Newton wrote in a 1675 letter to fellow scientist Robert Hooke, “it is only because I have stood on the shoulders of giants who have come before me.”

In contrast, POB freely acknowledged the leg-up he got from Laurence Pomeroy’s (1949, 1954) system, and he critiqued it:

“[Pomeroy] devised a clever system of time-comparisons on ‘circuits that were used over more than one season’, in order to quantify and compare car-speeds, and identify progress (or a lack of progress) in car performance. Covering the seasons from 1906 to 1953, Pomeroy reduced the times to a simple numerical formula, the Pomeroy Index or ‘Py’ Index. Car performances were scored, measured and compared directly against his benchmark car, the 13.0-litre/ 793-cubic inch 1906 Renault AK, which won the 1906 French Grand Prix. He chose this car because it won this first French Grand Prix. He allocated it his benchmark figure of 100.0.
“Pomeroy’s was a ground-breaking and accurate system, based on fastest lap-times from every Grand Prix race for each season. Using interval data (fastest lap times), Pomeroy’s system gave cars a rating figure.”
[…]
“Pomeroy’s method inspired my Rating System, prompting me to use 100.0 as my base as well. However, as mentioned above, Pomeroy assumed that he was rating cars whereas he was in fact measuring packages (car-and-driver combined).”
('Explanatory Chapters', 2016, p. 45)

Standing on the shoulders of giants is a necessary part of creativity, innovation, and development; it doesn't make what you do less valuable!

An academic gave me the following advice:

If they have made no reference at all to the years of work done by POB then their public writing on it should be rejected until the omission has been made good.

I hope this prompts a reaction from the authors by way of an apology to POB plus a statement affirming his priority in this field.

Plagiarism is always a strong charge to level at anyone. But the definition of plagiarism is broad. The OED says it involves taking the work or the idea of someone else and passing it off as one's own. It makes no mention of it having to be verbatim.

From the evidence you present, it certainly looks like plagiarism to me. The fact, too, that it has been done so blatantly suggests that those involved seem to believe that anything published can simply be “lifted” and re-used under their own names, without any acknowledgement or attribution. And what is that if not plagiarism?

Re: Best F1 driver ever?

Posted: Mon Aug 30, 2021 1:12 pm
by pokerman
POBRatings wrote:
Mon Aug 30, 2021 10:25 am
POB's daughter Catherine here, using his log-in:
Exediron wrote:
Sun Aug 29, 2021 1:31 am
I recall that this official ranking has come up on here before, although I couldn't find the discussion. I'll do my best to answer the questions you pose:

1) Although they haven't made the precise methodology public, my belief is that this is the core mechanism. However, the article also states that the machine learning algorithm weights the results based on things such as age and time together, and without knowing how much that weighting adjusts the final result it's difficult to determine how far the methodology strays from that employed by POB.

2) I believe (although I don't know) that one of the advantages of using machine learning is that they cross-compared every possible linkage between drivers, which likely gave enough data to indicate someone at #1 (Senna, apparently) without a human selecting a standard of comparison.

Whether this is a strength or not, is a matter of opinion. I know that in POB's system he would sometimes make an executive decision to fix a driver who was clearly operating at top-level speed for a season (such as Verstappen this season, I would imagine) to get the figures in line. I don't believe there's any such human oversight on the AWS results.

3) Many would say so. There are a few that don't agree at all with the comparative rankings generated by those on this forum (as seen in this thread, and elsewhere). For example:
  • They have Hamilton and Alonso at a smaller gap to one another than was observed when they were actually teammates, and also at a smaller gap than their shared partnership with Button would suggest.
  • Nico Hulkenberg is rated very highly, and ahead of at least one teammate who was definitively faster than him (Ricciardo)
  • Most people would consider Max to be significantly underrated
There is also a very sizable apparent recency bias, with a great many current drivers or recently retired drivers featuring.

4) The study was done specifically to determine the fastest qualifier, so in that regard I wouldn't say it's a problem -- as long as one accepts that the answer to who is the fastest driver all-around may not be the same.

For example, Senna being half a second quicker than Prost is likely accurate for qualifying, but was not an advantage he carried consistently into races (his margin over Prost was usually much smaller than that on race day).

5) I would say it isn't widely accepted, or particularly widely discussed -- which is perhaps a shame, because there's an interesting debate to be had on where the AWS findings differ from the human viewpoint.

For some further reading, I don't know if you've already come across this, but Amazon has a more in-depth explanation in this article, discussing some of the methodology in more detail:

https://aws.amazon.com/blogs/machine-le ... formula-1/
Thank you @Exediron for your extremely helpful response and pertinent information.

Perhaps they should revise their heading “The fastest driver in Formula 1” more modestly as “The fastest qualifier in Formula 1”. Reserve the word “driver” for the more complex/ stringent measurement, based on actual race times?

Even as a non-GP fan, I can see how your point 3) should generate lots of methodological delving and debate:

3) Many would say so. There are a few that don't agree at all with the comparative rankings generated by those on this forum (as seen in this thread, and elsewhere). For example:
They have Hamilton and Alonso at a smaller gap to one another than was observed when they were actually teammates, and also at a smaller gap than their shared partnership with Button would suggest.
Nico Hulkenberg is rated very highly, and ahead of at least one teammate who was definitively faster than him (Ricciardo)
Most people would consider Max to be significantly underrated
There is also a very sizable apparent recency bias, with a great many current drivers or recently retired drivers featuring.


I'm not so sure about your point here:

"Whether this is a strength or not, is a matter of opinion. I know that in POB's system he would sometimes [SIC] make an executive decision to fix a driver who was clearly operating at top-level speed for a season (such as Verstappen this season, I would imagine) to get the figures in line. I don't believe there's any such human oversight on the AWS results."

My understanding is that POB only did this once: when there were too few points of comparison in the early years of GP racing: "I therefore used Charron for 1898 as the first driver-rating benchmark. This rating was corroborated by contemporary sources, author Gerald Rose and driver Charles Jarrott, who reckoned Charron was the fastest" (Explanatory Chapter, 2016, p. 130). Hence the need to start his computations from Grand Prix racing's inception (1894).

Thereafter, he never needed to do this again; he used actual race times to establish arithmetically who attained the fastest rating (denoted 100.00 on his system, with 100.2, 100.3, 100.6, etc denoting slower ratings).

POB did refer to the opinions of experts such as team personnel, rivals, peers, observers and journalists, not to 'fix' his figures but rather to check that his computations tallied with what was going on on the track - to avoid the anomalous ratings so often seen in algorithm-generated ratings that try to by-pass human judgement and discernment (often unfairly denounced as 'subjectivity').
The problem with using race times is that drivers are not always driving at full speed, the only time they are doing this is in qualifying and that's why qualifying is often used to determine who are the fastest drivers.

Like you say being fast over one lap is one thing and being fast in a race can be different again but unfortunately I don't believe you can gather accurate enough data from race pace.

I do believe that qualifying in itself can be used as a good guide, it's far more often the case that the better qualifier also performs better on race day, we see in your link the fastest drivers being Senna, Schumacher, Hamilton, Verstappen and Alonso based purely on qualifying stats so I think it shows it's a very good guide.

Re: Best F1 driver ever?

Posted: Mon Aug 30, 2021 2:06 pm
by POBRatings
POB's daughter Catherine here, using his log-in:
pokerman wrote:
Mon Aug 30, 2021 1:12 pm
The problem with using race times is that drivers are not always driving at full speed, the only time they are doing this is in qualifying and that's why qualifying is often used to determine who are the fastest drivers.

Like you say being fast over one lap is one thing and being fast in a race can be different again but unfortunately I don't believe you can gather accurate enough data from race pace.

I do believe that qualifying in itself can be used as a good guide, it's far more often the case that the better qualifier also performs better on race day, we see in your link the fastest drivers being Senna, Schumacher, Hamilton, Verstappen and Alonso based purely on qualifying stats so I think it shows it's a very good guide.
Wouldn't that be interesting to - to compare rankings based on qualifying vs. actual race-times?

POB appears to have done the analysis of qualifying vs race times (from his 'Explanatory Chapters', 2016, p. 99): "Although packages are invariably slower in the races than in pre-race times, the gaps between team-mate packages and between different team packages are fairly constant whether pre-race or race-time. This is shown by an example of just two drivers below, although this pattern holds for the whole field throughout history.”

The POB system and the Smedley system make different claims in respect of what information the speed data takes into account:

POB ('Explanatory Chapters', 2016, p. 31): "5. During this work, my Rating System was criticised by a prominent Formula One journalist to the effect that it does not take into account a driver’s ‘management abilities’, that is, out-of-car skills and talents, such as Michael Schumacher had in ‘organising’ a team around himself to enhance success. This is not so. My System takes everything into account. Driver and car performance are measured on-track, and therefore include testing, preparation, qualifying, racing, life-path experience, and every capability a driver or anyone else may have brought to bear on performance during design. These out-of-car skills can include attracting sponsorship, which improves budget and potentially enhances car and team performance. My Rating System is ‘neutral and ignorant’ of all that goes on outside of recorded speeds. Whatever went into making up the performance of a package, car, driver or team is inherently measured on-track by my Rating System."

Smedley: "The output focuses solely on one element of a driver’s vast armory—the pure speed that is most evident on a Saturday afternoon during the qualifying hour. It doesn’t focus on racecraft or the ability to win races or drive at 200 mph while still having the bandwidth to understand everything going on around you (displayed so well by the likes of Michael Schumacher or Fernando Alonso). This ability, which transgresses speed alone, allowed them both, on many an occasion, to operate as master tacticians. For someone like myself, who has had the honor of watching those very skills in action from the pitwall, I cannot emphasize enough how important those skills are—they are the difference between the good and the great. It is important to point out that these skills are not included in this insight. This is about raw speed only and the ability to push the car to its very limits over one lap."

Do you think it's an undue limitation of the Smedley system NOT to take racecraft etc into account?

Wouldn't measures based on qualifying times alone rely MORE heavily on car-machinery than do race-
times, which would surely be more of a test of driver skill, racecraft etc? Or do you think that focusing on qualifying times alone 'neutralises' the car somewhat so that any differences in driver skill comes to the fore?

This seems to me to enter the choppy waters of the 'Car:Driver ratio', which POB analysed as far as he could but ultimately left unresolved. He introduces this concept as follows:

"Assessments on the ratio of ‘Car contribution relative to Driver contribution’ to Package performance by engineers, drivers and commentators have ranged widely: from the commonly cited 50:50 to Peter Windsor’s 60:40 and Phillips’s (2015) average ratio of 61:39, others reckon it is 70:30, while Neubauer’s estimation is 90:10 when referring to the success of the car, claiming 90% resides in the car’s preparation" ('Explanatory Chapters', 2016, p. 76)

Perhaps they are simply to be viewed as two different but equally valid measures: qualifying times vs race times. But if the two measures conflict with each other for a particular driver, would we be able to discern why? If Driver A ranked highly in race times but ranked poorly based on qualifying, what would that indicate? If the two measures - qualifying and race-times - are measuring different skills, then they should surely not both be denoted under the heading 'Who is the fastest driver?"

Do you think modelling data has advantages over human-interpreted (but time-constrained) data? (POB could only arrive at ratings within the narrow limits of what the actual times worked out as). What would the disadvantages be?

Amid the calculus and statistical programmes, machine learning or not, the junctures at which human subjectivity (including reasoned but relatively arbitrary cut-offs) is brought to bear in the Smedley system should not be overlooked e.g.,

"It’s important to note that to quantify a driver’s ability, we need to observe a minimum number of interactions. To factor this in, we only include teammates who have competed against each other in at least five qualifying sessions. A number of parameters and considerations have been put in place as an effective means of identifying various conditions with unfair comparisons, such as crashes, failures, age, career breaks, or weather conditions changing over qualifying sessions."
https://aws.amazon.com/blogs/machine-le ... formula-1/

Also, if race-times are needed to establish which team-mates to compare, then surely race-times are an integral part of the Smedley system, despite their disclaimer that they measure only qualifying times? Surely which drivers would be paired with which other drivers would be determined by race performance/wins etc, not simply qualifying times?

Re: Best F1 driver ever?

Posted: Wed Sep 01, 2021 1:13 pm
by pokerman
POBRatings wrote:
Mon Aug 30, 2021 2:06 pm
POB's daughter Catherine here, using his log-in:
pokerman wrote:
Mon Aug 30, 2021 1:12 pm
The problem with using race times is that drivers are not always driving at full speed, the only time they are doing this is in qualifying and that's why qualifying is often used to determine who are the fastest drivers.

Like you say being fast over one lap is one thing and being fast in a race can be different again but unfortunately I don't believe you can gather accurate enough data from race pace.

I do believe that qualifying in itself can be used as a good guide, it's far more often the case that the better qualifier also performs better on race day, we see in your link the fastest drivers being Senna, Schumacher, Hamilton, Verstappen and Alonso based purely on qualifying stats so I think it shows it's a very good guide.
Wouldn't that be interesting to - to compare rankings based on qualifying vs. actual race-times?

POB appears to have done the analysis of qualifying vs race times (from his 'Explanatory Chapters', 2016, p. 99): "Although packages are invariably slower in the races than in pre-race times, the gaps between team-mate packages and between different team packages are fairly constant whether pre-race or race-time. This is shown by an example of just two drivers below, although this pattern holds for the whole field throughout history.”

The POB system and the Smedley system make different claims in respect of what information the speed data takes into account:

POB ('Explanatory Chapters', 2016, p. 31): "5. During this work, my Rating System was criticised by a prominent Formula One journalist to the effect that it does not take into account a driver’s ‘management abilities’, that is, out-of-car skills and talents, such as Michael Schumacher had in ‘organising’ a team around himself to enhance success. This is not so. My System takes everything into account. Driver and car performance are measured on-track, and therefore include testing, preparation, qualifying, racing, life-path experience, and every capability a driver or anyone else may have brought to bear on performance during design. These out-of-car skills can include attracting sponsorship, which improves budget and potentially enhances car and team performance. My Rating System is ‘neutral and ignorant’ of all that goes on outside of recorded speeds. Whatever went into making up the performance of a package, car, driver or team is inherently measured on-track by my Rating System."

Smedley: "The output focuses solely on one element of a driver’s vast armory—the pure speed that is most evident on a Saturday afternoon during the qualifying hour. It doesn’t focus on racecraft or the ability to win races or drive at 200 mph while still having the bandwidth to understand everything going on around you (displayed so well by the likes of Michael Schumacher or Fernando Alonso). This ability, which transgresses speed alone, allowed them both, on many an occasion, to operate as master tacticians. For someone like myself, who has had the honor of watching those very skills in action from the pitwall, I cannot emphasize enough how important those skills are—they are the difference between the good and the great. It is important to point out that these skills are not included in this insight. This is about raw speed only and the ability to push the car to its very limits over one lap."

Do you think it's an undue limitation of the Smedley system NOT to take racecraft etc into account?

Wouldn't measures based on qualifying times alone rely MORE heavily on car-machinery than do race-
times, which would surely be more of a test of driver skill, racecraft etc? Or do you think that focusing on qualifying times alone 'neutralises' the car somewhat so that any differences in driver skill comes to the fore?

This seems to me to enter the choppy waters of the 'Car:Driver ratio', which POB analysed as far as he could but ultimately left unresolved. He introduces this concept as follows:

"Assessments on the ratio of ‘Car contribution relative to Driver contribution’ to Package performance by engineers, drivers and commentators have ranged widely: from the commonly cited 50:50 to Peter Windsor’s 60:40 and Phillips’s (2015) average ratio of 61:39, others reckon it is 70:30, while Neubauer’s estimation is 90:10 when referring to the success of the car, claiming 90% resides in the car’s preparation" ('Explanatory Chapters', 2016, p. 76)

Perhaps they are simply to be viewed as two different but equally valid measures: qualifying times vs race times. But if the two measures conflict with each other for a particular driver, would we be able to discern why? If Driver A ranked highly in race times but ranked poorly based on qualifying, what would that indicate? If the two measures - qualifying and race-times - are measuring different skills, then they should surely not both be denoted under the heading 'Who is the fastest driver?"

Do you think modelling data has advantages over human-interpreted (but time-constrained) data? (POB could only arrive at ratings within the narrow limits of what the actual times worked out as). What would the disadvantages be?

Amid the calculus and statistical programmes, machine learning or not, the junctures at which human subjectivity (including reasoned but relatively arbitrary cut-offs) is brought to bear in the Smedley system should not be overlooked e.g.,

"It’s important to note that to quantify a driver’s ability, we need to observe a minimum number of interactions. To factor this in, we only include teammates who have competed against each other in at least five qualifying sessions. A number of parameters and considerations have been put in place as an effective means of identifying various conditions with unfair comparisons, such as crashes, failures, age, career breaks, or weather conditions changing over qualifying sessions."
https://aws.amazon.com/blogs/machine-le ... formula-1/

Also, if race-times are needed to establish which team-mates to compare, then surely race-times are an integral part of the Smedley system, despite their disclaimer that they measure only qualifying times? Surely which drivers would be paired with which other drivers would be determined by race performance/wins etc, not simply qualifying times?
I think we are looking at two things here, who are the fastest drivers and who are the best drivers, fastest drivers can be measured by lap times, best drivers you then include all the things that you mentioned, so one is data driven and the other perhaps somewhat subjective.

One is quite simplistic especially if using just qualifying times, the other far more complicated and I for one wouldn't know where to start, it takes someone like your Father to go into that kind of depth.

In my simplistic world Senna was faster therefore better than Prost, I believe we saw this in the races as well as qualifying, but for some others Prost brought more to the table that made him the better driver.

Re: Best F1 driver ever?

Posted: Thu Sep 02, 2021 8:23 am
by POBRatings
POB's daughter Catherine here, using his log-in:
pokerman wrote:
Wed Sep 01, 2021 1:13 pm

I think we are looking at two things here, who are the fastest drivers and who are the best drivers, fastest drivers can be measured by lap times, best drivers you then include all the things that you mentioned, so one is data driven and the other perhaps somewhat subjective.

One is quite simplistic especially if using just qualifying times, the other far more complicated and I for one wouldn't know where to start, it takes someone like your Father to go into that kind of depth.

In my simplistic world Senna was faster therefore better than Prost, I believe we saw this in the races as well as qualifying, but for some others Prost brought more to the table that made him the better driver.
Thank you for your response. Since 2017, so many things have arisen in Grand Prix racing that I'd love to have asked POB about so it's helpful to be able to ask on this forum instead.

I wonder if POB would even have continued with his ratings, with the technology becoming so computerised and the races being geared so much towards TV-viewing rather than driver skill, what with all the artificial obstacles I understand have been more recently introduced. (This is where my knowledge breaks off and I have to confess to not having watched a race; I would of course take an avid interest in watching a race on TV with POB now though). What would you say are the biggest things in racing that have changed since say 2017, if any?

Your distinction between "best" and "fastest" driver makes me think "driver" should be qualified in titles such as "Who is the fastest F1 driver?" - in terms of "Who is the fastest qualifier?" and "Who is the fastest racer?" - albeit likely only people in the field would know what the distinction was and why it matters.

I actually asked POB about what "best driver" meant and he clarified that it meant "the fastest". I guess he's taking ***race*** speed as the litmus test of fastest driver, factoring in racecraft, race pace etc. Also over a longer period of time (a whole GP season?) so that wins, pole positions, fastest laps set, podiums, and pre-times can all be factored in. Surely a more robust, more holistic and less superficial measure than one-shot 'fastest lap'?

As it's in qualifying that the fastest times are usually achieved, due to fresher tyres, lower fuel loads, and engine qualifying modes, POB might consider that taking qualifying results only, as a reference, is likely to be a misleading indicator of the real performance of the cars on race day.

Didn't teams in the 1980s use turbo engines with special specifications for qualifying? ie different equipment between qualifying and races? Do they use the exact same equipment between qualifying and races today?

As you say "we are looking at two things here, who are the fastest drivers and who are the best drivers", so it reduces to an epistemological matter of what COUNTS as "fastest" or "best".

Re: Best F1 driver ever?

Posted: Thu Sep 02, 2021 9:52 am
by POBRatings
POB's daughter Catherine here, using his log-in:

I've just had a look at Autosport.com's 'supertimes':
https://www.autosport.com/f1/news/top-f ... 3/5283233/

Autosport.com’s Supertimes bear a noteworthy resemblance to the POB system in terms of the following:
• It uses teammates as a benchmark to measure the ‘gap’ between drivers.
• It uses the ‘100.00’ arithmetic figure as representative of the fastest time.

An early draft of POB's book on this subject was in fact titled:

THE GAP: FORMULA ONE PERFORMANCE RATINGS 1950 -2007
Patrick O’Brien
December 2007

"Racing performance has been analysed, measured and quantified by direct comparison as the time-speed differential between competitors. Continued analyses subsequent to my 1994 study found that all that matters in assessing competitiveness in F1 racing is 'the gap'; that is the time intervals between competitors. Hence this book’s title. This is the essence of racing."


And this concept was of course written about in his later work, published on Lulu.com:

"My Rating System uses actual on-track times to quantify and compare the time differences between competitors. In 2002, having realised that all that matters in Grand Prix racing is ‘the gap’, I devised my Rating System. ‘The gap’ refers to the time differentials between competitors. Racing is all about speed and ‘by how much’ competitor A beat competitor B." (Explanatory Chapters, 2016, p. 3).

Cf. Autosport.com writes about: "My trigger is more the gaps between the top teams, the percentage" - Scott Mitchell, Jan 28, 2019. They present a table of percentages resembling POB's notation.

As in the Smedley/AWS/Formula1.com system, it appears that POB's innovation and contribution to this field have been blatantly lifted, unacknowledged and unattributed.

This is something of a dagger through the heart for me, seeing how many years POB worked on this and how hard he tried to introduce his ground-breaking research to the motorsport press, while being told that his ratings-figures would likely only cause controversy. Now others appear to be riding on his back, standing on his shoulders, blithely discussing his figures as 'supertimes' in the mainstream motor-sport media.

Can I just check - does anyone know:
1. When did Autosport.com introduce these 'supertimes'?
2. Had such percentage figures, in the 100.00 notation that POB devised, been used before? ie prior to say 2017?
3. Who is the author of these supertimes? Is it Scott Mitchell?
4. Did they review the literature? POB reviewed the literature on this question in his 'Explanatory Chapters' (2016) so he could say with authority what is new, what was wrongheaded. He did not seek to pass others' ideas off as his own, but incorporated what worked from the past and acknowledged this, and then made it clear what his own innovation was; what he was contributing to the field.
5. Are these ratings discussed on the Autosport.com forum?
https://forums.autosport.com/forum/2-racing-comments/
Is anyone here also on the Autosport.com forum?
6. Are these supertimes widely used and respected?

As with the Smedley/AWS/Formul1.com online publications, I would like to ask that Autosport.com reference all the years of work done by Patrick O'Brien on this question, and acknowledge his priority in this field. If anyone can help with this, it would be much appreciated.

Re: Best F1 driver ever?

Posted: Thu Sep 30, 2021 1:01 pm
by POBRatings
POB's daughter Catherine here, using his log-in:

Can anyone help me with labelling 6 pictures of Formula One cars? I tried uploading the pictures with Photobucket but it seems one has to pay for 'hosting bandwidth'. and I don't think I'd use it often enough. I therefore got the idea to post the pictures on POB's website, in the hope that the moderators would allow this just for the purposes of getting names & model numbers for these cars. I understand one shouldn't use the forum to generate traffic to one's site (like advertising) so to circumvent this, I can delete this post immediately afterwards. The 6 cars are:

https://grandprixratings.blogspot.com/2 ... ovide.html

EDIT: All numbered "44", I assume they're the same car, so hopefully it's just one make and model you'd need to supply!

Re: Best F1 driver ever?

Posted: Thu Sep 30, 2021 1:41 pm
by Jezza13
POBRatings wrote:
Thu Sep 30, 2021 1:01 pm
POB's daughter Catherine here, using his log-in:

Can anyone help me with labelling 6 pictures of Formula One cars? I tried uploading the pictures with Photobucket but it seems one has to pay for 'hosting bandwidth'. and I don't think I'd use it often enough. I therefore got the idea to post the pictures on POB's website, in the hope that the moderators would allow this just for the purposes of getting names & model numbers for these cars. I understand one shouldn't use the forum to generate traffic to one's site (like advertising) so to circumvent this, I can delete this post immediately afterwards. The 6 cars are:

https://grandprixratings.blogspot.com/2 ... ovide.html

EDIT: All numbered "44", I assume they're the same car, so hopefully it's just one make and model you'd need to supply!
Hi Catherine

I'm happy for others to correct me if i'm wrong but they all look to be the 2018 Mercedes F1 WO9.