The Great Big 2021 A > B > C Thread

Forum rules
Please read the forum rules
User avatar
Exediron
Posts: 8904
Joined: Sun Feb 23, 2014 3:53 am
Location: Michigan, USA
Contact:

Re: The Great Big 2021 A > B > C Thread

Post by Exediron »

Another thing to add is that, over the course of a season, the median will smooth out the effect of temporary slumps or the adjustment period for a driver coming into a team and initially struggling. Let's imagine the following data set, taking the Ocon / Alonso pairing as a base:

R1: Alonso > Ocon -0.938%
R2: Ocon > Alonso -0.629%
R3: Ocon > Alonso -1.094%
R4: Alonso > Ocon -0.231%
R5: Alonso > Ocon -0.430%
R6: Alonso > Ocon -0.475%
R7: Alonso > Ocon -0.327%
R8: Alonso > Ocon -0.253%
R9: Alonso > Ocon -0.070%
R10: Alonso > Ocon -0.280%
R11: Alonso > Ocon -0.022%
R12: Alonso > Ocon -0.171%
R13: Alonso > Ocon -0.329%
R14: Alonso > Ocon -0.911%
R15: Alonso > Ocon -0.002%
R16: Alonso > Ocon -0.229%
R17: Ocon > Alonso -0.048%

For the purposes of constructing this example, I've used the first three races from this season and then input Ricciardo's actual qualifying results from 2020 for the remainder. The data set reflects a season where Alonso struggled badly to get to grips with the car early on, then settled in and showed a consistent and strong advantage over Ocon.

Using a mean method, we get a gap of -0.170% in Alonso's favor, a result that sits at about the 37% point on the season; roughly a third of results were below that for Alonso, and two thirds were above. By contrast, if you exclude the first three races the mean average shifts dramatically to 0.262%, indicating almost a whole tenth faster for Alonso.

The median gives us 0.231% for the whole season, with only a tiny change to 0.242% by excluding the first three races. It has successfully pointed to a range of common results, blunting the impact of the outliers -- in this case, the start of the season where one driver struggled.

If the goal of this exercise is to determine the fair gap between drivers, I can't see any advantage for allowing outlier results to color the final analysis. Rather than waiting for a second outlier from the other driver to cancel out the first (which may not happen), it's better to use a method that sets them aside entirely.
PICK 10 COMPETITION (7 wins, 21 podiums): 3rd in 2016
TOP THREE CHAMPIONSHIP (No Limit Excedrin Racing): Champions in 2015 & 2018 | 2nd in 2017 & 2019
AUTOSPORT GP PREDICTOR: 2017 USA & P-F1 Champion

pokerman
Posts: 37597
Joined: Wed Dec 19, 2012 1:30 pm

Re: The Great Big 2021 A > B > C Thread

Post by pokerman »

Exediron wrote:
Mon May 03, 2021 10:27 pm
pokerman wrote:
Mon May 03, 2021 2:13 pm
Another point is that I would be guessing in the median system that whoever comes out on top in the head to head in qualifying then automatically comes out on top with the average gap so it's not really about averages as such?
That is correct. In the unlikely event that one driver was generally quicker than the other but lost the qualifying battle anyway, it will focus on the smallest advantage from the winning driver. Normally, this will be a ~0.050 or less result, indicating a nearly even comparison. I will note that in my time compiling these, I have not come across an instance where the quicker driver lost the qualifying head-to-head.

To the second part of your post: It is still an average -- it is simply a different sort of average. An average is about finding a number that expresses the middle of a data set, and mean and median are two different ways of arriving at that conclusion. Both are subject of potential flaws, but I prefer the median because it is much less sensitive to being influenced by abnormal data points. In a mean average you have to remove those data points entirely or allow them to dominate the data, whereas with a median you can keep them and they still express influence over the result without dominating it.
Right I thought that was the case, with my system I vet all the data so it's not necessarily compromised by big swings, to that end I also disallow wet qualifying sessions which you include, for my part I'm looking for basically the quickest driver in the dry and wet driving is a different skill set including that seems like it would corrupt what we're trying to find.
Lewis Hamilton #44

World Drivers Titles: 7 (1st)
Grand Prix Wins: 99 (1st)
Pole Positions: 101 (1st)
Podiums: 172 (1st)


PF1 Pick 10 Competition
2014: Champion

mikeyg123
Posts: 19425
Joined: Sat Jan 28, 2012 4:13 pm

Re: The Great Big 2021 A > B > C Thread

Post by mikeyg123 »

pokerman wrote:
Tue May 04, 2021 12:48 pm
Exediron wrote:
Mon May 03, 2021 10:27 pm
pokerman wrote:
Mon May 03, 2021 2:13 pm
Another point is that I would be guessing in the median system that whoever comes out on top in the head to head in qualifying then automatically comes out on top with the average gap so it's not really about averages as such?
That is correct. In the unlikely event that one driver was generally quicker than the other but lost the qualifying battle anyway, it will focus on the smallest advantage from the winning driver. Normally, this will be a ~0.050 or less result, indicating a nearly even comparison. I will note that in my time compiling these, I have not come across an instance where the quicker driver lost the qualifying head-to-head.

To the second part of your post: It is still an average -- it is simply a different sort of average. An average is about finding a number that expresses the middle of a data set, and mean and median are two different ways of arriving at that conclusion. Both are subject of potential flaws, but I prefer the median because it is much less sensitive to being influenced by abnormal data points. In a mean average you have to remove those data points entirely or allow them to dominate the data, whereas with a median you can keep them and they still express influence over the result without dominating it.
Right I thought that was the case, with my system I vet all the data so it's not necessarily compromised by big swings, to that end I also disallow wet qualifying sessions which you include, for my part I'm looking for basically the quickest driver in the dry and wet driving is a different skill set including that seems like it would corrupt what we're trying to find.
Doesn't using the median factor those big swings out for you? Thus preventing the creep of personal bias.

pokerman
Posts: 37597
Joined: Wed Dec 19, 2012 1:30 pm

Re: The Great Big 2021 A > B > C Thread

Post by pokerman »

Exediron wrote:
Mon May 03, 2021 10:50 pm
Another thing to add is that, over the course of a season, the median will smooth out the effect of temporary slumps or the adjustment period for a driver coming into a team and initially struggling. Let's imagine the following data set, taking the Ocon / Alonso pairing as a base:

R1: Alonso > Ocon -0.938%
R2: Ocon > Alonso -0.629%
R3: Ocon > Alonso -1.094%
R4: Alonso > Ocon -0.231%
R5: Alonso > Ocon -0.430%
R6: Alonso > Ocon -0.475%
R7: Alonso > Ocon -0.327%
R8: Alonso > Ocon -0.253%
R9: Alonso > Ocon -0.070%
R10: Alonso > Ocon -0.280%
R11: Alonso > Ocon -0.022%
R12: Alonso > Ocon -0.171%
R13: Alonso > Ocon -0.329%
R14: Alonso > Ocon -0.911%
R15: Alonso > Ocon -0.002%
R16: Alonso > Ocon -0.229%
R17: Ocon > Alonso -0.048%

For the purposes of constructing this example, I've used the first three races from this season and then input Ricciardo's actual qualifying results from 2020 for the remainder. The data set reflects a season where Alonso struggled badly to get to grips with the car early on, then settled in and showed a consistent and strong advantage over Ocon.

Using a mean method, we get a gap of -0.170% in Alonso's favor, a result that sits at about the 37% point on the season; roughly a third of results were below that for Alonso, and two thirds were above. By contrast, if you exclude the first three races the mean average shifts dramatically to 0.262%, indicating almost a whole tenth faster for Alonso.

The median gives us 0.231% for the whole season, with only a tiny change to 0.242% by excluding the first three races. It has successfully pointed to a range of common results, blunting the impact of the outliers -- in this case, the start of the season where one driver struggled.

If the goal of this exercise is to determine the fair gap between drivers, I can't see any advantage for allowing outlier results to color the final analysis. Rather than waiting for a second outlier from the other driver to cancel out the first (which may not happen), it's better to use a method that sets them aside entirely.
I'm not sure of your example being a practical working model, it was said to take Ocon nearly all season to get up to speed after a year out of F1 and joining a new team, and here we are proposing it takes Alonso just 3 races with him being 2 years out of F1 then being on the same level of Ricciardo.

For my part I have had a situation were one driver wins the head to head but the other driver ends up with the better average also I could use an impractical example of how that could happen. This season we have 23 races, driver A wins the head to head 12-11 each time he's 0.1s ahead thus the median has him as being 0.1s faster.

However when driver B was quicker he was always 0.5s quicker thus we have this:-

11 x 0.5s = 5.5s -
12 x 0.1s = 1.2s =
4.3s / 23 =
0.19s

With this example we have 0.3s between the 2 results, quite a difference but obviously quite an extreme example. Like we've already said with the median whoever wins the head to head also is given the better average, now I'm sure in the scientific world in different applications that's perfectly fine but in this sporting context I don't think that's what I'm looking for as in what I would call a true average.
Lewis Hamilton #44

World Drivers Titles: 7 (1st)
Grand Prix Wins: 99 (1st)
Pole Positions: 101 (1st)
Podiums: 172 (1st)


PF1 Pick 10 Competition
2014: Champion

pokerman
Posts: 37597
Joined: Wed Dec 19, 2012 1:30 pm

Re: The Great Big 2021 A > B > C Thread

Post by pokerman »

mikeyg123 wrote:
Tue May 04, 2021 1:18 pm
pokerman wrote:
Tue May 04, 2021 12:48 pm
Exediron wrote:
Mon May 03, 2021 10:27 pm
pokerman wrote:
Mon May 03, 2021 2:13 pm
Another point is that I would be guessing in the median system that whoever comes out on top in the head to head in qualifying then automatically comes out on top with the average gap so it's not really about averages as such?
That is correct. In the unlikely event that one driver was generally quicker than the other but lost the qualifying battle anyway, it will focus on the smallest advantage from the winning driver. Normally, this will be a ~0.050 or less result, indicating a nearly even comparison. I will note that in my time compiling these, I have not come across an instance where the quicker driver lost the qualifying head-to-head.

To the second part of your post: It is still an average -- it is simply a different sort of average. An average is about finding a number that expresses the middle of a data set, and mean and median are two different ways of arriving at that conclusion. Both are subject of potential flaws, but I prefer the median because it is much less sensitive to being influenced by abnormal data points. In a mean average you have to remove those data points entirely or allow them to dominate the data, whereas with a median you can keep them and they still express influence over the result without dominating it.
Right I thought that was the case, with my system I vet all the data so it's not necessarily compromised by big swings, to that end I also disallow wet qualifying sessions which you include, for my part I'm looking for basically the quickest driver in the dry and wet driving is a different skill set including that seems like it would corrupt what we're trying to find.
Doesn't using the median factor those big swings out for you? Thus preventing the creep of personal bias.
It does take away any personal bias but I just question the practicality of the system in what I would call finding a true average.
Lewis Hamilton #44

World Drivers Titles: 7 (1st)
Grand Prix Wins: 99 (1st)
Pole Positions: 101 (1st)
Podiums: 172 (1st)


PF1 Pick 10 Competition
2014: Champion

User avatar
Exediron
Posts: 8904
Joined: Sun Feb 23, 2014 3:53 am
Location: Michigan, USA
Contact:

Re: The Great Big 2021 A > B > C Thread

Post by Exediron »

pokerman wrote:
Tue May 04, 2021 1:21 pm
For my part I have had a situation were one driver wins the head to head but the other driver ends up with the better average also I could use an impractical example of how that could happen. This season we have 23 races, driver A wins the head to head 12-11 each time he's 0.1s ahead thus the median has him as being 0.1s faster.

However when driver B was quicker he was always 0.5s quicker thus we have this:-

11 x 0.5s = 5.5s -
12 x 0.1s = 1.2s =
4.3s / 23 =
0.19s

With this example we have 0.3s between the 2 results, quite a difference but obviously quite an extreme example. Like we've already said with the median whoever wins the head to head also is given the better average, now I'm sure in the scientific world in different applications that's perfectly fine but in this sporting context I don't think that's what I'm looking for as in what I would call a true average.
I think it's worth some debate if a driver who is less often quicker, but more ahead when he is quicker (weird sentence, I know) is actually the quicker driver. Obviously your example is an extreme one, as you say, but how far does it need to slide before you consider the driver who qualified ahead more often to be the quicker one? What if we had instead:

5 x 0.5s = 2.5s -
18 x 0.1s = 1.8s =
0.7s / 23 =
0.030s

In that data set, would you think it's fair to call the driver who only qualified ahead five times out of a twenty-three race season quicker, even by a small amount?
PICK 10 COMPETITION (7 wins, 21 podiums): 3rd in 2016
TOP THREE CHAMPIONSHIP (No Limit Excedrin Racing): Champions in 2015 & 2018 | 2nd in 2017 & 2019
AUTOSPORT GP PREDICTOR: 2017 USA & P-F1 Champion

User avatar
Johnson
Posts: 1810
Joined: Sat Feb 09, 2013 1:20 am

Re: The Great Big 2021 A > B > C Thread

Post by Johnson »

A.J. wrote:
Mon May 03, 2021 11:04 am
A median is almost always a better data point to have than the mean (average) - just imagine if Ricciardo (hypothetically) beats Norris by a tenth in ever single qualifying session except 2, where Norris has a 1s advantage over him for whatever reason (such as in Portugal). The median will tell you that Ricciardo beat Norris by a tenth over the course of the season (which is accurate), while the average between them over a 20 race season will show Norris ahead by a bit, even though he was beaten 18 times out of 20 in this example.

Same reason why America has such issues with poverty, despite the mean income being amongst the highest in the world.
Median can also be equally inaccurate in extreme examples. Personally I prefer a mixture of both. If you have 20 qualifying sessions, remove the 3 most extremes from either end. That leaves 14 qualifying sessions from which you create a mean. This helps remove both of the negative aspects of both mean and median.

pokerman
Posts: 37597
Joined: Wed Dec 19, 2012 1:30 pm

Re: The Great Big 2021 A > B > C Thread

Post by pokerman »

Exediron wrote:
Tue May 04, 2021 5:43 pm
pokerman wrote:
Tue May 04, 2021 1:21 pm
For my part I have had a situation were one driver wins the head to head but the other driver ends up with the better average also I could use an impractical example of how that could happen. This season we have 23 races, driver A wins the head to head 12-11 each time he's 0.1s ahead thus the median has him as being 0.1s faster.

However when driver B was quicker he was always 0.5s quicker thus we have this:-

11 x 0.5s = 5.5s -
12 x 0.1s = 1.2s =
4.3s / 23 =
0.19s

With this example we have 0.3s between the 2 results, quite a difference but obviously quite an extreme example. Like we've already said with the median whoever wins the head to head also is given the better average, now I'm sure in the scientific world in different applications that's perfectly fine but in this sporting context I don't think that's what I'm looking for as in what I would call a true average.
I think it's worth some debate if a driver who is less often quicker, but more ahead when he is quicker (weird sentence, I know) is actually the quicker driver. Obviously your example is an extreme one, as you say, but how far does it need to slide before you consider the driver who qualified ahead more often to be the quicker one? What if we had instead:

5 x 0.5s = 2.5s -
18 x 0.1s = 1.8s =
0.7s / 23 =
0.030s

In that data set, would you think it's fair to call the driver who only qualified ahead five times out of a twenty-three race season quicker, even by a small amount?
The end result though is not that extreme, only a 0.1s discrepancy, so you're going from drivers being reasonably close to bang on equal, we both know that in real life this doesn't happen like that, also the way I do it is that extreme gaps often get discounted at source because sometimes there are extenuating circumstances that disadvantages the slower driver or it may be a wet session were we often see big gaps.

From what I see the sites that do this use a mean average rather than a median average but for some I still complain because they use all the data without any kind of context which again includes wet qualifying, when you do this then obviously a median average works better.

For me I just find in such an emotive sport it's such an unemotive system to use the median average, it makes little difference if a driver out qualifies another driver on merit by 8 tenths rather than 4 tenths because both gaps have exactly the same weight in a driver pairing with an average less than 4 tenths, out and out excellence basically gets ignored.
Lewis Hamilton #44

World Drivers Titles: 7 (1st)
Grand Prix Wins: 99 (1st)
Pole Positions: 101 (1st)
Podiums: 172 (1st)


PF1 Pick 10 Competition
2014: Champion

User avatar
Exediron
Posts: 8904
Joined: Sun Feb 23, 2014 3:53 am
Location: Michigan, USA
Contact:

Re: The Great Big 2021 A > B > C Thread

Post by Exediron »

pokerman wrote:
Wed May 05, 2021 4:00 pm
For me I just find in such an emotive sport it's such an unemotive system to use the median average, it makes little difference if a driver out qualifies another driver on merit by 8 tenths rather than 4 tenths because both gaps have exactly the same weight in a driver pairing with an average less than 4 tenths, out and out excellence basically gets ignored.
It sounds like we agree on the difference, but disagree on which one is desirable. In my opinion, a system that lets rare excellent results outweigh more common results is a flawed system -- too emotional, if you will. It produces something that may be closer to the feeling of the viewer, but not necessarily closer to the truth.

At the end of the 2021 season, people will remember that Norris out-qualified Ricciardo by over a second that one time in Portugal. But it's very unlikely to become a pattern, and in the final tally the fact that he managed to absolutely smash RIC in a single qualifying session means precisely that -- nothing more.
PICK 10 COMPETITION (7 wins, 21 podiums): 3rd in 2016
TOP THREE CHAMPIONSHIP (No Limit Excedrin Racing): Champions in 2015 & 2018 | 2nd in 2017 & 2019
AUTOSPORT GP PREDICTOR: 2017 USA & P-F1 Champion

User avatar
EPROM
Posts: 96
Joined: Thu Feb 04, 2021 12:59 am
Location: USA - Maryland

Re: The Great Big 2021 A > B > C Thread

Post by EPROM »

Another (similar) approach is to eliminate the "outliers"/"sports", and then take a true mean. That is arguably more accurate but may require some human analysis to prune the data sets. For fully automated analysis, the median is generally less susceptible to outliers - unless you have sophisticated outlier rejection analytics.

This is what I do for a living - and have done 40 years ago (little has really changed...). Management still tends to resist such techniques. Some things never change.

pokerman
Posts: 37597
Joined: Wed Dec 19, 2012 1:30 pm

Re: The Great Big 2021 A > B > C Thread

Post by pokerman »

EPROM wrote:
Thu May 06, 2021 2:25 am
Another (similar) approach is to eliminate the "outliers"/"sports", and then take a true mean. That is arguably more accurate but may require some human analysis to prune the data sets. For fully automated analysis, the median is generally less susceptible to outliers - unless you have sophisticated outlier rejection analytics.

This is what I do for a living - and have done 40 years ago (little has really changed...). Management still tends to resist such techniques. Some things never change.
Well that's what I try to do but I've never in my life been called sophisticated so maybe I fall short. :lol:
Lewis Hamilton #44

World Drivers Titles: 7 (1st)
Grand Prix Wins: 99 (1st)
Pole Positions: 101 (1st)
Podiums: 172 (1st)


PF1 Pick 10 Competition
2014: Champion

pokerman
Posts: 37597
Joined: Wed Dec 19, 2012 1:30 pm

Re: The Great Big 2021 A > B > C Thread

Post by pokerman »

Exediron wrote:
Wed May 05, 2021 7:30 pm
pokerman wrote:
Wed May 05, 2021 4:00 pm
For me I just find in such an emotive sport it's such an unemotive system to use the median average, it makes little difference if a driver out qualifies another driver on merit by 8 tenths rather than 4 tenths because both gaps have exactly the same weight in a driver pairing with an average less than 4 tenths, out and out excellence basically gets ignored.
It sounds like we agree on the difference, but disagree on which one is desirable. In my opinion, a system that lets rare excellent results outweigh more common results is a flawed system -- too emotional, if you will. It produces something that may be closer to the feeling of the viewer, but not necessarily closer to the truth.

At the end of the 2021 season, people will remember that Norris out-qualified Ricciardo by over a second that one time in Portugal. But it's very unlikely to become a pattern, and in the final tally the fact that he managed to absolutely smash RIC in a single qualifying session means precisely that -- nothing more.
It's just different idealisms, it just doesn't sit right with me that we ignore Norris' excellent lap which basically puts Norris were he should be at this moment in time, he's been so much quicker than Ricciardo after the first 3 weekends.
Lewis Hamilton #44

World Drivers Titles: 7 (1st)
Grand Prix Wins: 99 (1st)
Pole Positions: 101 (1st)
Podiums: 172 (1st)


PF1 Pick 10 Competition
2014: Champion

WHoff78
Posts: 918
Joined: Wed Dec 11, 2013 11:01 pm

Re: The Great Big 2021 A > B > C Thread

Post by WHoff78 »

I see the value in both approaches. For me, if you can have some fairly neutral rules for rejecting outliers then I would look more towards the mean. As mentioned previously, in the really close calls where a driver wins out say 10-9 over the course of the season, I do not believe that a median that will simply point to the smallest margin that the winning driver had over the season is a true comparison of there qualifying speed. And you can be fairly confident that out of 10 qualifying wins the median will be a low number. I think the mean would provide a more meaningful comparison under those circumstances. Generally speaking both help to paint the picture. What's really most important is that you disclose what method has been used so that we understand the limitations of each particular method.

pokerman
Posts: 37597
Joined: Wed Dec 19, 2012 1:30 pm

Re: The Great Big 2021 A > B > C Thread

Post by pokerman »

Right I said I would give my numbers after this qualifying session, in my early posts I did say I've factored in that new drivers will take some time to get up to full speed however I never envisaged how big a struggle it might be for some drivers.

I also said I would use percentages and then convert back to seconds because taking into account circuit length surely must give a more accurate result and I was surprised that in the main it made little to no difference, maybe because the tracks used so far have been reasonably similar in length.
pokerman wrote:
Sat Dec 19, 2020 9:17 pm
Right here goes, looking at Leclerc there's not many data points for him, this year he was either excellent or Vettel was sub par, nevertheless at this moment in time I can only go with what happened so:-
Leclerc > Sainz 0.47s
Leclerc > Sainz 0.22s

I'm really surprised by Sainz's strong start, in respect to Vettel we perhaps already can answer my question, in 2020 Vettel was sub par.


Onto the Max v Perez matchup plenty of data points here to make a more valid judgement with a bit of variance for Perez settling into his new team, so:-
Max > Perez 0.31s - 0.36s
Max > Perez 0.34s

This has to seen as a good start by Perez helped in part by Max messing up 2 qualifying laps, I would say a similasr situation to McLaren the big gap today puts us about were the gap should be.


Next the Ricciardo/Norris matchup, we have very recent data for this again to make a solid predication, again a bit of variance for Ricciardo settling into his new team, so:-
Ricciardo > Norris 0.21s - 0.26s.
Norris > Ricciardo 0.26s

On one hand not surprised to see the gap because when Ricciardo first joined Renault the Hulk should have been 0.3s up after 3 races if not for issues with his car, by the 4th race Ricciardo qualified in front on merit as he did today. The bigger surprise is how much Norris has improved, given his age and inexperience I guess there was always a chance it would happen.


Moving onto Vettel and Stroll, if he wants to recover his reputation then really needs to improve on this prediction:-
Vettel > Stroll 0.2s
Stroll > Vettel 0.02s

This actually is not a bad start by Vettel when you compare to some other new matchups, I would say that he's probably on course for my prediction


Alonso against Ocon, can Alonso be back at his best, he needs to because even at his best I would be predicting this:-
Alonso > Ocon 0.15s
Ocon > Alonso 0.31s

It has to been seen as no surprise for a 40 year old Alonso 2 years out of F1 to make a slow start, I think some Alonso fans perhaps underestimated what was in front of him. When I made my prediction I said Alonso at his best, it might be very difficult for him to beat Ocon.

Next is something we can't really know, Gasly against Tsunoda, I see Gasly as being a little better than average while Tsunoda is talented and a fast learner, I think he can contain Gasly to no more than this prediction:-
Gasly > Tsunoda 0.1s
Gasly > Tsunoda 0.3s

No surprise to see Gasly in front so far, my prediction was very much weighted to the second half of the season in respect to Tsunoda in part because he's a fast learner, for now though I have to wonder if he's struggling a bit at this level.

Finally two rookies Schumacher and Mazepin, both slow learners so who has impressed me more as juniors and that would be Schumacher, however I don't see Schumacher as being super fast so I predict:-
Schumacher > Mazepin 0.1s
Schumacher > Mazepin 0.54s

I did say that both drivers are slow learners but at this point it feels like Schumacher is doing a bit of a George Russell.
Lewis Hamilton #44

World Drivers Titles: 7 (1st)
Grand Prix Wins: 99 (1st)
Pole Positions: 101 (1st)
Podiums: 172 (1st)


PF1 Pick 10 Competition
2014: Champion

User avatar
Exediron
Posts: 8904
Joined: Sun Feb 23, 2014 3:53 am
Location: Michigan, USA
Contact:

Re: The Great Big 2021 A > B > C Thread

Post by Exediron »

And here is my own update after Spain:

LECLERC / SAINZ
Projected: 0.386% to 0.542% (Leclerc)
Current Result: 0.330% (Leclerc)
Data (click to show)

VERSTAPPEN / PEREZ
Projected: 0.385% - 0.544% (Verstappen)
Current Result: 0.279% (Verstappen)
Data (click to show)

NORRIS / RICCIARDO
Projected: 0.262% - 0.403% (Ricciardo)
Current Result: 0.058% (Ricciardo)
Data (click to show)

STROLL / VETTEL
Projected: 0.262% - 0.316% (Vettel)
Current Result: 0.236% (Stroll)
Data (click to show)

OCON / ALONSO
Projected: 0.158% - 0.426% (Alonso)
Current Result: 0.677% (Ocon)
Data (click to show)

Since there is an even number of results, the median result is currently the average of the two middle numbers.

Spain didn't change a lot in terms of results. Leclerc shifts a little lower owing to a small qualifying advantage over Sainz; Verstappen goes up by about a tenth, although his Spanish result is an extreme and doesn't directly factor into it. Norris/Ricciardo hardly changes, because the middle two results are both slight Ricciardo wins. Stroll loses about a tenth over Vettel, while Ocon more or less holds station over Alonso.

Shifting to my subjective analysis of the results, I would say that the Ferrari and Red Bull pairs are tracking close to expectations, with Sainz looking stronger than anticipated and Perez potentially weaker. The McLaren pair has been a surprise, but we may be seeing a correction in which Ricciardo gets on top of the car and becomes the faster driver as was projected. Similar story at Aston Martin, with Vettel definitely picking up form lately compared to the start of the season.

The major outlier is at Alpine, where Alonso is currently making a strong case for age and car familiarity being significant factors in driver performance. After this season ends, I intend to take a look at comparing Alonso's 2021 season with Schumacher's 2010. The fact that Alonso is much closer to Ocon on race pace than qualifying implies to me that he will get on top of the problem and close the gap, but I'll be interested to see how this season continues to develop.
PICK 10 COMPETITION (7 wins, 21 podiums): 3rd in 2016
TOP THREE CHAMPIONSHIP (No Limit Excedrin Racing): Champions in 2015 & 2018 | 2nd in 2017 & 2019
AUTOSPORT GP PREDICTOR: 2017 USA & P-F1 Champion

JN23
Posts: 3637
Joined: Mon Aug 08, 2016 11:14 am

Re: The Great Big 2021 A > B > C Thread

Post by JN23 »

pokerman wrote:
Sat May 08, 2021 7:17 pm
Right I said I would give my numbers after this qualifying session, in my early posts I did say I've factored in that new drivers will take some time to get up to full speed however I never envisaged how big a struggle it might be for some drivers.

I also said I would use percentages and then convert back to seconds because taking into account circuit length surely must give a more accurate result and I was surprised that in the main it made little to no difference, maybe because the tracks used so far have been reasonably similar in length.
pokerman wrote:
Sat Dec 19, 2020 9:17 pm
Right here goes, looking at Leclerc there's not many data points for him, this year he was either excellent or Vettel was sub par, nevertheless at this moment in time I can only go with what happened so:-
Leclerc > Sainz 0.47s
Leclerc > Sainz 0.22s

I'm really surprised by Sainz's strong start, in respect to Vettel we perhaps already can answer my question, in 2020 Vettel was sub par.


Onto the Max v Perez matchup plenty of data points here to make a more valid judgement with a bit of variance for Perez settling into his new team, so:-
Max > Perez 0.31s - 0.36s
Max > Perez 0.34s

This has to seen as a good start by Perez helped in part by Max messing up 2 qualifying laps, I would say a similasr situation to McLaren the big gap today puts us about were the gap should be.


Next the Ricciardo/Norris matchup, we have very recent data for this again to make a solid predication, again a bit of variance for Ricciardo settling into his new team, so:-
Ricciardo > Norris 0.21s - 0.26s.
Norris > Ricciardo 0.26s

On one hand not surprised to see the gap because when Ricciardo first joined Renault the Hulk should have been 0.3s up after 3 races if not for issues with his car, by the 4th race Ricciardo qualified in front on merit as he did today. The bigger surprise is how much Norris has improved, given his age and inexperience I guess there was always a chance it would happen.


Moving onto Vettel and Stroll, if he wants to recover his reputation then really needs to improve on this prediction:-
Vettel > Stroll 0.2s
Stroll > Vettel 0.02s

This actually is not a bad start by Vettel when you compare to some other new matchups, I would say that he's probably on course for my prediction


Alonso against Ocon, can Alonso be back at his best, he needs to because even at his best I would be predicting this:-
Alonso > Ocon 0.15s
Ocon > Alonso 0.31s

It has to been seen as no surprise for a 40 year old Alonso 2 years out of F1 to make a slow start, I think some Alonso fans perhaps underestimated what was in front of him. When I made my prediction I said Alonso at his best, it might be very difficult for him to beat Ocon.

Next is something we can't really know, Gasly against Tsunoda, I see Gasly as being a little better than average while Tsunoda is talented and a fast learner, I think he can contain Gasly to no more than this prediction:-
Gasly > Tsunoda 0.1s
Gasly > Tsunoda 0.3s

No surprise to see Gasly in front so far, my prediction was very much weighted to the second half of the season in respect to Tsunoda in part because he's a fast learner, for now though I have to wonder if he's struggling a bit at this level.

Finally two rookies Schumacher and Mazepin, both slow learners so who has impressed me more as juniors and that would be Schumacher, however I don't see Schumacher as being super fast so I predict:-
Schumacher > Mazepin 0.1s
Schumacher > Mazepin 0.54s

I did say that both drivers are slow learners but at this point it feels like Schumacher is doing a bit of a George Russell.
How are you converting your percentages back to seconds poker?

I've calculated some of the gaps using seconds and don't get the same figures as you so guessing you're using a certain lap time to calculate the gap?

Eg. the gap I get for Verstappen and Perez is 0.32, Stroll and Vettel 0.05.

Paolo_Lasardi
Posts: 3042
Joined: Sat May 26, 2012 2:04 pm

Re: The Great Big 2021 A > B > C Thread

Post by Paolo_Lasardi »

Exediron, you claim in the Indycar thread that Grosjean was generally faster than Magnussen at Haas. Can you back this up with numbers?

pokerman
Posts: 37597
Joined: Wed Dec 19, 2012 1:30 pm

Re: The Great Big 2021 A > B > C Thread

Post by pokerman »

JN23 wrote:
Thu May 13, 2021 8:22 pm
pokerman wrote:
Sat May 08, 2021 7:17 pm
Right I said I would give my numbers after this qualifying session, in my early posts I did say I've factored in that new drivers will take some time to get up to full speed however I never envisaged how big a struggle it might be for some drivers.

I also said I would use percentages and then convert back to seconds because taking into account circuit length surely must give a more accurate result and I was surprised that in the main it made little to no difference, maybe because the tracks used so far have been reasonably similar in length.
pokerman wrote:
Sat Dec 19, 2020 9:17 pm
Right here goes, looking at Leclerc there's not many data points for him, this year he was either excellent or Vettel was sub par, nevertheless at this moment in time I can only go with what happened so:-
Leclerc > Sainz 0.47s
Leclerc > Sainz 0.22s

I'm really surprised by Sainz's strong start, in respect to Vettel we perhaps already can answer my question, in 2020 Vettel was sub par.


Onto the Max v Perez matchup plenty of data points here to make a more valid judgement with a bit of variance for Perez settling into his new team, so:-
Max > Perez 0.31s - 0.36s
Max > Perez 0.34s

This has to seen as a good start by Perez helped in part by Max messing up 2 qualifying laps, I would say a similasr situation to McLaren the big gap today puts us about were the gap should be.


Next the Ricciardo/Norris matchup, we have very recent data for this again to make a solid predication, again a bit of variance for Ricciardo settling into his new team, so:-
Ricciardo > Norris 0.21s - 0.26s.
Norris > Ricciardo 0.26s

On one hand not surprised to see the gap because when Ricciardo first joined Renault the Hulk should have been 0.3s up after 3 races if not for issues with his car, by the 4th race Ricciardo qualified in front on merit as he did today. The bigger surprise is how much Norris has improved, given his age and inexperience I guess there was always a chance it would happen.


Moving onto Vettel and Stroll, if he wants to recover his reputation then really needs to improve on this prediction:-
Vettel > Stroll 0.2s
Stroll > Vettel 0.02s

This actually is not a bad start by Vettel when you compare to some other new matchups, I would say that he's probably on course for my prediction


Alonso against Ocon, can Alonso be back at his best, he needs to because even at his best I would be predicting this:-
Alonso > Ocon 0.15s
Ocon > Alonso 0.31s

It has to been seen as no surprise for a 40 year old Alonso 2 years out of F1 to make a slow start, I think some Alonso fans perhaps underestimated what was in front of him. When I made my prediction I said Alonso at his best, it might be very difficult for him to beat Ocon.

Next is something we can't really know, Gasly against Tsunoda, I see Gasly as being a little better than average while Tsunoda is talented and a fast learner, I think he can contain Gasly to no more than this prediction:-
Gasly > Tsunoda 0.1s
Gasly > Tsunoda 0.3s

No surprise to see Gasly in front so far, my prediction was very much weighted to the second half of the season in respect to Tsunoda in part because he's a fast learner, for now though I have to wonder if he's struggling a bit at this level.

Finally two rookies Schumacher and Mazepin, both slow learners so who has impressed me more as juniors and that would be Schumacher, however I don't see Schumacher as being super fast so I predict:-
Schumacher > Mazepin 0.1s
Schumacher > Mazepin 0.54s

I did say that both drivers are slow learners but at this point it feels like Schumacher is doing a bit of a George Russell.
How are you converting your percentages back to seconds poker?

I've calculated some of the gaps using seconds and don't get the same figures as you so guessing you're using a certain lap time to calculate the gap?

Eg. the gap I get for Verstappen and Perez is 0.32, Stroll and Vettel 0.05.
I guess so but even then our numbers are still incredible close not too be a problem I would have thought?
Lewis Hamilton #44

World Drivers Titles: 7 (1st)
Grand Prix Wins: 99 (1st)
Pole Positions: 101 (1st)
Podiums: 172 (1st)


PF1 Pick 10 Competition
2014: Champion

User avatar
Exediron
Posts: 8904
Joined: Sun Feb 23, 2014 3:53 am
Location: Michigan, USA
Contact:

Re: The Great Big 2021 A > B > C Thread

Post by Exediron »

Paolo_Lasardi wrote:
Sat May 15, 2021 11:00 am
Exediron, you claim in the Indycar thread that Grosjean was generally faster than Magnussen at Haas. Can you back this up with numbers?
I'll admit, I've never actually run the numbers and am just going off my impression. But I will try to find time to do so later this weekend and come up with an answer!
PICK 10 COMPETITION (7 wins, 21 podiums): 3rd in 2016
TOP THREE CHAMPIONSHIP (No Limit Excedrin Racing): Champions in 2015 & 2018 | 2nd in 2017 & 2019
AUTOSPORT GP PREDICTOR: 2017 USA & P-F1 Champion

User avatar
Exediron
Posts: 8904
Joined: Sun Feb 23, 2014 3:53 am
Location: Michigan, USA
Contact:

Re: The Great Big 2021 A > B > C Thread

Post by Exediron »

As requested by Paolo_Lasardi, I've fed the 2017-2020 Haas pairing of Grosjean/Magnussen into the formula. This is one of the larger sample sizes I've used so far, so the resulting answer should have a higher confidence factor than normal.

And without any further ado, here are the numbers:

2017: Grosjean > Magnussen: 0.134%
Data (click to show)
2018: Grosjean > Magnussen: 0.269%
Data (click to show)
2019: Magnussen > Grosjean: 0.046%
Data (click to show)
2020: Magnussen > Grosjean: 0.001%
Data (click to show)

Grosjean starts off the clear stronger driver in the partnership, peaking in 2018 when he shaded Magnussen quite firmly, finishing with a run of 8 qualifying wins from the last 9 races. The form switched in 2019 and 2020, with Magnussen posting a slight advantage -- the smallest advantage my system has ever indicated in 2020.

In total, RoGro and KMag combined for 72 usable data points during their four years together:

Overall: Grosjean > Magnussen : 0.020%
Data (click to show)

The final median falls between two numbers, with an indicated average between the two of 0.020% in Grosjean's favor. This truly was a rather close partnership; out of 72 usable results, 13 were under 0.100% difference.

Conclusion: Grosjean was the faster on average, but by a very small margin. I think it would be fair to state that Grosjean = Magnussen as a shorthand, since 0.020% is below what I consider to be the margin of error of this method.
PICK 10 COMPETITION (7 wins, 21 podiums): 3rd in 2016
TOP THREE CHAMPIONSHIP (No Limit Excedrin Racing): Champions in 2015 & 2018 | 2nd in 2017 & 2019
AUTOSPORT GP PREDICTOR: 2017 USA & P-F1 Champion

JN23
Posts: 3637
Joined: Mon Aug 08, 2016 11:14 am

Re: The Great Big 2021 A > B > C Thread

Post by JN23 »

pokerman wrote:
Sat May 15, 2021 4:23 pm
JN23 wrote:
Thu May 13, 2021 8:22 pm
pokerman wrote:
Sat May 08, 2021 7:17 pm
Right I said I would give my numbers after this qualifying session, in my early posts I did say I've factored in that new drivers will take some time to get up to full speed however I never envisaged how big a struggle it might be for some drivers.

I also said I would use percentages and then convert back to seconds because taking into account circuit length surely must give a more accurate result and I was surprised that in the main it made little to no difference, maybe because the tracks used so far have been reasonably similar in length.
pokerman wrote:
Sat Dec 19, 2020 9:17 pm
Right here goes, looking at Leclerc there's not many data points for him, this year he was either excellent or Vettel was sub par, nevertheless at this moment in time I can only go with what happened so:-
Leclerc > Sainz 0.47s
Leclerc > Sainz 0.22s

I'm really surprised by Sainz's strong start, in respect to Vettel we perhaps already can answer my question, in 2020 Vettel was sub par.


Onto the Max v Perez matchup plenty of data points here to make a more valid judgement with a bit of variance for Perez settling into his new team, so:-
Max > Perez 0.31s - 0.36s
Max > Perez 0.34s

This has to seen as a good start by Perez helped in part by Max messing up 2 qualifying laps, I would say a similasr situation to McLaren the big gap today puts us about were the gap should be.


Next the Ricciardo/Norris matchup, we have very recent data for this again to make a solid predication, again a bit of variance for Ricciardo settling into his new team, so:-
Ricciardo > Norris 0.21s - 0.26s.
Norris > Ricciardo 0.26s

On one hand not surprised to see the gap because when Ricciardo first joined Renault the Hulk should have been 0.3s up after 3 races if not for issues with his car, by the 4th race Ricciardo qualified in front on merit as he did today. The bigger surprise is how much Norris has improved, given his age and inexperience I guess there was always a chance it would happen.


Moving onto Vettel and Stroll, if he wants to recover his reputation then really needs to improve on this prediction:-
Vettel > Stroll 0.2s
Stroll > Vettel 0.02s

This actually is not a bad start by Vettel when you compare to some other new matchups, I would say that he's probably on course for my prediction


Alonso against Ocon, can Alonso be back at his best, he needs to because even at his best I would be predicting this:-
Alonso > Ocon 0.15s
Ocon > Alonso 0.31s

It has to been seen as no surprise for a 40 year old Alonso 2 years out of F1 to make a slow start, I think some Alonso fans perhaps underestimated what was in front of him. When I made my prediction I said Alonso at his best, it might be very difficult for him to beat Ocon.

Next is something we can't really know, Gasly against Tsunoda, I see Gasly as being a little better than average while Tsunoda is talented and a fast learner, I think he can contain Gasly to no more than this prediction:-
Gasly > Tsunoda 0.1s
Gasly > Tsunoda 0.3s

No surprise to see Gasly in front so far, my prediction was very much weighted to the second half of the season in respect to Tsunoda in part because he's a fast learner, for now though I have to wonder if he's struggling a bit at this level.

Finally two rookies Schumacher and Mazepin, both slow learners so who has impressed me more as juniors and that would be Schumacher, however I don't see Schumacher as being super fast so I predict:-
Schumacher > Mazepin 0.1s
Schumacher > Mazepin 0.54s

I did say that both drivers are slow learners but at this point it feels like Schumacher is doing a bit of a George Russell.
How are you converting your percentages back to seconds poker?

I've calculated some of the gaps using seconds and don't get the same figures as you so guessing you're using a certain lap time to calculate the gap?

Eg. the gap I get for Verstappen and Perez is 0.32, Stroll and Vettel 0.05.
I guess so but even then our numbers are still incredible close not too be a problem I would have thought?
Yeah you're right, was just curious!

Paolo_Lasardi
Posts: 3042
Joined: Sat May 26, 2012 2:04 pm

Re: The Great Big 2021 A > B > C Thread

Post by Paolo_Lasardi »

Exediron wrote:
Sun May 16, 2021 5:11 am
As requested by Paolo_Lasardi, I've fed the 2017-2020 Haas pairing of Grosjean/Magnussen into the formula. This is one of the larger sample sizes I've used so far, so the resulting answer should have a higher confidence factor than normal.

And without any further ado, here are the numbers:

2017: Grosjean > Magnussen: 0.134%
Data (click to show)
2018: Grosjean > Magnussen: 0.269%
Data (click to show)
2019: Magnussen > Grosjean: 0.046%
Data (click to show)
2020: Magnussen > Grosjean: 0.001%
Data (click to show)

Grosjean starts off the clear stronger driver in the partnership, peaking in 2018 when he shaded Magnussen quite firmly, finishing with a run of 8 qualifying wins from the last 9 races. The form switched in 2019 and 2020, with Magnussen posting a slight advantage -- the smallest advantage my system has ever indicated in 2020.

In total, RoGro and KMag combined for 72 usable data points during their four years together:

Overall: Grosjean > Magnussen : 0.020%
Data (click to show)

The final median falls between two numbers, with an indicated average between the two of 0.020% in Grosjean's favor. This truly was a rather close partnership; out of 72 usable results, 13 were under 0.100% difference.

Conclusion: Grosjean was the faster on average, but by a very small margin. I think it would be fair to state that Grosjean = Magnussen as a shorthand, since 0.020% is below what I consider to be the margin of error of this method.
:thumbup:

Thank you very much, very interesting!

pokerman
Posts: 37597
Joined: Wed Dec 19, 2012 1:30 pm

Re: The Great Big 2021 A > B > C Thread

Post by pokerman »

JN23 wrote:
Sun May 16, 2021 9:12 am
pokerman wrote:
Sat May 15, 2021 4:23 pm
JN23 wrote:
Thu May 13, 2021 8:22 pm
pokerman wrote:
Sat May 08, 2021 7:17 pm
Right I said I would give my numbers after this qualifying session, in my early posts I did say I've factored in that new drivers will take some time to get up to full speed however I never envisaged how big a struggle it might be for some drivers.

I also said I would use percentages and then convert back to seconds because taking into account circuit length surely must give a more accurate result and I was surprised that in the main it made little to no difference, maybe because the tracks used so far have been reasonably similar in length.
pokerman wrote:
Sat Dec 19, 2020 9:17 pm
Right here goes, looking at Leclerc there's not many data points for him, this year he was either excellent or Vettel was sub par, nevertheless at this moment in time I can only go with what happened so:-
Leclerc > Sainz 0.47s
Leclerc > Sainz 0.22s

I'm really surprised by Sainz's strong start, in respect to Vettel we perhaps already can answer my question, in 2020 Vettel was sub par.


Onto the Max v Perez matchup plenty of data points here to make a more valid judgement with a bit of variance for Perez settling into his new team, so:-
Max > Perez 0.31s - 0.36s
Max > Perez 0.34s

This has to seen as a good start by Perez helped in part by Max messing up 2 qualifying laps, I would say a similasr situation to McLaren the big gap today puts us about were the gap should be.


Next the Ricciardo/Norris matchup, we have very recent data for this again to make a solid predication, again a bit of variance for Ricciardo settling into his new team, so:-
Ricciardo > Norris 0.21s - 0.26s.
Norris > Ricciardo 0.26s

On one hand not surprised to see the gap because when Ricciardo first joined Renault the Hulk should have been 0.3s up after 3 races if not for issues with his car, by the 4th race Ricciardo qualified in front on merit as he did today. The bigger surprise is how much Norris has improved, given his age and inexperience I guess there was always a chance it would happen.


Moving onto Vettel and Stroll, if he wants to recover his reputation then really needs to improve on this prediction:-
Vettel > Stroll 0.2s
Stroll > Vettel 0.02s

This actually is not a bad start by Vettel when you compare to some other new matchups, I would say that he's probably on course for my prediction


Alonso against Ocon, can Alonso be back at his best, he needs to because even at his best I would be predicting this:-
Alonso > Ocon 0.15s
Ocon > Alonso 0.31s

It has to been seen as no surprise for a 40 year old Alonso 2 years out of F1 to make a slow start, I think some Alonso fans perhaps underestimated what was in front of him. When I made my prediction I said Alonso at his best, it might be very difficult for him to beat Ocon.

Next is something we can't really know, Gasly against Tsunoda, I see Gasly as being a little better than average while Tsunoda is talented and a fast learner, I think he can contain Gasly to no more than this prediction:-
Gasly > Tsunoda 0.1s
Gasly > Tsunoda 0.3s

No surprise to see Gasly in front so far, my prediction was very much weighted to the second half of the season in respect to Tsunoda in part because he's a fast learner, for now though I have to wonder if he's struggling a bit at this level.

Finally two rookies Schumacher and Mazepin, both slow learners so who has impressed me more as juniors and that would be Schumacher, however I don't see Schumacher as being super fast so I predict:-
Schumacher > Mazepin 0.1s
Schumacher > Mazepin 0.54s

I did say that both drivers are slow learners but at this point it feels like Schumacher is doing a bit of a George Russell.
How are you converting your percentages back to seconds poker?

I've calculated some of the gaps using seconds and don't get the same figures as you so guessing you're using a certain lap time to calculate the gap?

Eg. the gap I get for Verstappen and Perez is 0.32, Stroll and Vettel 0.05.
I guess so but even then our numbers are still incredible close not too be a problem I would have thought?
Yeah you're right, was just curious!
If you show me your numbers I could see what I might have done differently.
Lewis Hamilton #44

World Drivers Titles: 7 (1st)
Grand Prix Wins: 99 (1st)
Pole Positions: 101 (1st)
Podiums: 172 (1st)


PF1 Pick 10 Competition
2014: Champion

User avatar
Exediron
Posts: 8904
Joined: Sun Feb 23, 2014 3:53 am
Location: Michigan, USA
Contact:

Re: The Great Big 2021 A > B > C Thread

Post by Exediron »

Post-Monaco update:

LECLERC / SAINZ
Projected: 0.386% to 0.542% (Leclerc)
Current Result: 0.375% (Leclerc)
Data (click to show)

VERSTAPPEN / PEREZ
Projected: 0.385% - 0.544% (Verstappen)
Current Result: 0.376% (Verstappen)
Data (click to show)

NORRIS / RICCIARDO
Projected: 0.262% - 0.403% (Ricciardo)
Current Result: 0.052% (Ricciardo)
Data (click to show)

STROLL / VETTEL
Projected: 0.262% - 0.316% (Vettel)
Current Result: 0.134% (Stroll)
Data (click to show)

OCON / ALONSO
Projected: 0.158% - 0.426% (Alonso)
Current Result: 0.643% (Ocon)
Data (click to show)

Leclerc and Verstappen both continue to settle into something close to their projected range, although in the case of Verstappen by a series of very up-and-down data points.

Norris and Vettel are both still trailing their teammate, but trending in the direction of a reversal; for Norris, another result in his favor will likely swing the median heavily his way.

Finally, very little change at the Ocon/Alonso pairing, where the result continues to be strongly against expectations.
PICK 10 COMPETITION (7 wins, 21 podiums): 3rd in 2016
TOP THREE CHAMPIONSHIP (No Limit Excedrin Racing): Champions in 2015 & 2018 | 2nd in 2017 & 2019
AUTOSPORT GP PREDICTOR: 2017 USA & P-F1 Champion

pokerman
Posts: 37597
Joined: Wed Dec 19, 2012 1:30 pm

Re: The Great Big 2021 A > B > C Thread

Post by pokerman »

I plan to update every 3 races, this may show up any marked differences as drivers get more use to their new teams.
Lewis Hamilton #44

World Drivers Titles: 7 (1st)
Grand Prix Wins: 99 (1st)
Pole Positions: 101 (1st)
Podiums: 172 (1st)


PF1 Pick 10 Competition
2014: Champion

User avatar
Exediron
Posts: 8904
Joined: Sun Feb 23, 2014 3:53 am
Location: Michigan, USA
Contact:

Re: The Great Big 2021 A > B > C Thread

Post by Exediron »

pokerman wrote:
Sun May 23, 2021 12:51 am
I plan to update every 3 races, this may show up any marked differences as drivers get more use to their new teams.
That's not a bad call. I think I might switch to doing every other, since the median result only properly updates on odd-numbered races. :thumbup:
PICK 10 COMPETITION (7 wins, 21 podiums): 3rd in 2016
TOP THREE CHAMPIONSHIP (No Limit Excedrin Racing): Champions in 2015 & 2018 | 2nd in 2017 & 2019
AUTOSPORT GP PREDICTOR: 2017 USA & P-F1 Champion

pokerman
Posts: 37597
Joined: Wed Dec 19, 2012 1:30 pm

Re: The Great Big 2021 A > B > C Thread

Post by pokerman »

Exediron wrote:
Sun May 23, 2021 1:16 am
pokerman wrote:
Sun May 23, 2021 12:51 am
I plan to update every 3 races, this may show up any marked differences as drivers get more use to their new teams.
That's not a bad call. I think I might switch to doing every other, since the median result only properly updates on odd-numbered races. :thumbup:
A 23 race season comes in handy then. :)
Lewis Hamilton #44

World Drivers Titles: 7 (1st)
Grand Prix Wins: 99 (1st)
Pole Positions: 101 (1st)
Podiums: 172 (1st)


PF1 Pick 10 Competition
2014: Champion

Battle Far
Posts: 802
Joined: Mon Jun 09, 2014 8:26 pm

Re: The Great Big 2021 A > B > C Thread

Post by Battle Far »

Absolutely not a dig but subjectively after 4 & 1/2 races

Sainz is much closer to Le Clerc, not anticipated
Norris is much better than Ricciardo, not anticipated
Ocon is better than Alonso, not anticipated

All numerical rating systems did not suggest this prior to the season start, given the obvious predictive failure, why are you bothering?

pokerman
Posts: 37597
Joined: Wed Dec 19, 2012 1:30 pm

Re: The Great Big 2021 A > B > C Thread

Post by pokerman »

Battle Far wrote:
Sun May 23, 2021 12:19 pm
Absolutely not a dig but subjectively after 4 & 1/2 races

Sainz is much closer to Le Clerc, not anticipated
Norris is much better than Ricciardo, not anticipated
Ocon is better than Alonso, not anticipated

All numerical rating systems did not suggest this prior to the season start, given the obvious predictive failure, why are you bothering?
If you're looking to judge just after 5 races then I guess you don't understand how these things work, it's already been highlighted in mainstream media how much new drivers are struggling in their new teams, I think in part you can look at the 3 day winter test when in the near past it was 8 days.

Also you're not factoring in rookie performance, Norris, drivers getting old, Alonso, plus he's not driven in F1 for 2 years.

In respect to Sainz it was already being questioned whether Vettel was making Leclerc look better then he really was, Vettel was that poor, each season gives you more information.

Also in respect to Ricciardo he was getting beat by the Hulk in the early season races at Renault before he started to turn things around, the year after Ocon clearly struggled with being a year out of F1 and joining a new team, I even factored that into my calculations after I recognised that in recent years that had been a problem for new drivers, and like I say with the restricted winter testing this has become an even bigger factor.
Lewis Hamilton #44

World Drivers Titles: 7 (1st)
Grand Prix Wins: 99 (1st)
Pole Positions: 101 (1st)
Podiums: 172 (1st)


PF1 Pick 10 Competition
2014: Champion

User avatar
Exediron
Posts: 8904
Joined: Sun Feb 23, 2014 3:53 am
Location: Michigan, USA
Contact:

Re: The Great Big 2021 A > B > C Thread

Post by Exediron »

pokerman wrote:
Sun May 23, 2021 12:37 pm
In respect to Sainz it was already being questioned whether Vettel was making Leclerc look better then he really was, Vettel was that poor, each season gives you more information.
Sainz is actually pretty much dead on inside the projected range so far. A little towards the low end, but the low end is where you end up if you assume the 2020 season was slightly anomalous in terms of Vettel's pace.

Ricciardo being so far off Norris is indeed a surprise. But it's a surprise to everyone who eyeballed the pairing as well, not just an analytics surprise. I don't believe I ever heard anyone suggest that Ricciarco would get absolutely dominated by Norris, and frankly the fact that Sainz is about where we expected against Leclerc while Norris is dominating Ricciardo is a strong indicator that Ricciardo is not performing at his actual level.
PICK 10 COMPETITION (7 wins, 21 podiums): 3rd in 2016
TOP THREE CHAMPIONSHIP (No Limit Excedrin Racing): Champions in 2015 & 2018 | 2nd in 2017 & 2019
AUTOSPORT GP PREDICTOR: 2017 USA & P-F1 Champion

KingVoid
Posts: 3315
Joined: Thu Jan 08, 2015 7:54 am

Re: The Great Big 2021 A > B > C Thread

Post by KingVoid »

What this season so far has taught me is that Sainz might have been the most underrated driver on this forum, prior to this season.

His performance against Leclerc is very good, not a complete match but he’s doing damn well for himself.

Norris destroying Ricciardo is also making Sainz’ performance against Norris at McLaren look better.

I actually think that his struggles at Renault against Hulkenberg might have been an outlier in his career.

Quality driver.

User avatar
Exediron
Posts: 8904
Joined: Sun Feb 23, 2014 3:53 am
Location: Michigan, USA
Contact:

Re: The Great Big 2021 A > B > C Thread

Post by Exediron »

KingVoid wrote:
Mon May 24, 2021 2:55 am
I actually think that his struggles at Renault against Hulkenberg might have been an outlier in his career.
I've said this for years. I'm feeling moderately vindicated so far this season!
PICK 10 COMPETITION (7 wins, 21 podiums): 3rd in 2016
TOP THREE CHAMPIONSHIP (No Limit Excedrin Racing): Champions in 2015 & 2018 | 2nd in 2017 & 2019
AUTOSPORT GP PREDICTOR: 2017 USA & P-F1 Champion

pokerman
Posts: 37597
Joined: Wed Dec 19, 2012 1:30 pm

Re: The Great Big 2021 A > B > C Thread

Post by pokerman »

Exediron wrote:
Mon May 24, 2021 1:18 am
pokerman wrote:
Sun May 23, 2021 12:37 pm
In respect to Sainz it was already being questioned whether Vettel was making Leclerc look better then he really was, Vettel was that poor, each season gives you more information.
Sainz is actually pretty much dead on inside the projected range so far. A little towards the low end, but the low end is where you end up if you assume the 2020 season was slightly anomalous in terms of Vettel's pace.

Ricciardo being so far off Norris is indeed a surprise. But it's a surprise to everyone who eyeballed the pairing as well, not just an analytics surprise. I don't believe I ever heard anyone suggest that Ricciarco would get absolutely dominated by Norris, and frankly the fact that Sainz is about where we expected against Leclerc while Norris is dominating Ricciardo is a strong indicator that Ricciardo is not performing at his actual level.
I think it's quite obvious that Ricciardo is not performing to his normal level, this is definitely new team blues and we have to acknowledge that it can affect some drivers, then we have Norris definitely raising his game, I'm sure I said this is something Norris needed to do, so given his inexperience I never defined that Norris couldn't improve on what we had seen so far.

As an aside your numbers don't yet show just how dominant Norris has been I believe?
Lewis Hamilton #44

World Drivers Titles: 7 (1st)
Grand Prix Wins: 99 (1st)
Pole Positions: 101 (1st)
Podiums: 172 (1st)


PF1 Pick 10 Competition
2014: Champion

pokerman
Posts: 37597
Joined: Wed Dec 19, 2012 1:30 pm

Re: The Great Big 2021 A > B > C Thread

Post by pokerman »

Exediron wrote:
Mon May 24, 2021 3:10 am
KingVoid wrote:
Mon May 24, 2021 2:55 am
I actually think that his struggles at Renault against Hulkenberg might have been an outlier in his career.
I've said this for years. I'm feeling moderately vindicated so far this season!
I think we have to accept that using 1 year comparisons is never ideal.
Lewis Hamilton #44

World Drivers Titles: 7 (1st)
Grand Prix Wins: 99 (1st)
Pole Positions: 101 (1st)
Podiums: 172 (1st)


PF1 Pick 10 Competition
2014: Champion

User avatar
Exediron
Posts: 8904
Joined: Sun Feb 23, 2014 3:53 am
Location: Michigan, USA
Contact:

Re: The Great Big 2021 A > B > C Thread

Post by Exediron »

pokerman wrote:
Mon May 24, 2021 9:23 pm
As an aside your numbers don't yet show just how dominant Norris has been I believe?
No, they certainly don't. But that's more because of Ricciardo somehow managing to out-qualify Norris in Bahrain and Imola (despite being slower) than anything to do with my system.
PICK 10 COMPETITION (7 wins, 21 podiums): 3rd in 2016
TOP THREE CHAMPIONSHIP (No Limit Excedrin Racing): Champions in 2015 & 2018 | 2nd in 2017 & 2019
AUTOSPORT GP PREDICTOR: 2017 USA & P-F1 Champion

User avatar
Covalent
Posts: 10334
Joined: Fri Oct 06, 2006 12:07 pm

Re: The Great Big 2021 A > B > C Thread

Post by Covalent »

KingVoid wrote:
Mon May 24, 2021 2:55 am
What this season so far has taught me is that Sainz might have been the most underrated driver on this forum, prior to this season.

His performance against Leclerc is very good, not a complete match but he’s doing damn well for himself.

Norris destroying Ricciardo is also making Sainz’ performance against Norris at McLaren look better.

I actually think that his struggles at Renault against Hulkenberg might have been an outlier in his career.

Quality driver.
Could it be that Leclerc is slightly overrated?

User avatar
Exediron
Posts: 8904
Joined: Sun Feb 23, 2014 3:53 am
Location: Michigan, USA
Contact:

Re: The Great Big 2021 A > B > C Thread

Post by Exediron »

Covalent wrote:
Tue May 25, 2021 6:37 am
KingVoid wrote:
Mon May 24, 2021 2:55 am
What this season so far has taught me is that Sainz might have been the most underrated driver on this forum, prior to this season.

His performance against Leclerc is very good, not a complete match but he’s doing damn well for himself.

Norris destroying Ricciardo is also making Sainz’ performance against Norris at McLaren look better.

I actually think that his struggles at Renault against Hulkenberg might have been an outlier in his career.

Quality driver.
Could it be that Leclerc is slightly overrated?
It could be. But I don't see any particular reason to think so.

Both Sainz and Norris are doing better than anticipated this season. Occam's Razor says that by far the simplest explanation -- since their major source of comparison was each other -- is that both were previously underrated.
PICK 10 COMPETITION (7 wins, 21 podiums): 3rd in 2016
TOP THREE CHAMPIONSHIP (No Limit Excedrin Racing): Champions in 2015 & 2018 | 2nd in 2017 & 2019
AUTOSPORT GP PREDICTOR: 2017 USA & P-F1 Champion

mikeyg123
Posts: 19425
Joined: Sat Jan 28, 2012 4:13 pm

Re: The Great Big 2021 A > B > C Thread

Post by mikeyg123 »

Covalent wrote:
Tue May 25, 2021 6:37 am
KingVoid wrote:
Mon May 24, 2021 2:55 am
What this season so far has taught me is that Sainz might have been the most underrated driver on this forum, prior to this season.

His performance against Leclerc is very good, not a complete match but he’s doing damn well for himself.

Norris destroying Ricciardo is also making Sainz’ performance against Norris at McLaren look better.

I actually think that his struggles at Renault against Hulkenberg might have been an outlier in his career.

Quality driver.
Could it be that Leclerc is slightly overrated?
It's possible but would be a lot more likely if Ricciardo was monstering Norris.

User avatar
Covalent
Posts: 10334
Joined: Fri Oct 06, 2006 12:07 pm

Re: The Great Big 2021 A > B > C Thread

Post by Covalent »

mikeyg123 wrote:
Tue May 25, 2021 7:07 am
Covalent wrote:
Tue May 25, 2021 6:37 am
KingVoid wrote:
Mon May 24, 2021 2:55 am
What this season so far has taught me is that Sainz might have been the most underrated driver on this forum, prior to this season.

His performance against Leclerc is very good, not a complete match but he’s doing damn well for himself.

Norris destroying Ricciardo is also making Sainz’ performance against Norris at McLaren look better.

I actually think that his struggles at Renault against Hulkenberg might have been an outlier in his career.

Quality driver.
Could it be that Leclerc is slightly overrated?
It's possible but would be a lot more likely if Ricciardo was monstering Norris.
Yes good point.

User avatar
Exediron
Posts: 8904
Joined: Sun Feb 23, 2014 3:53 am
Location: Michigan, USA
Contact:

Re: The Great Big 2021 A > B > C Thread

Post by Exediron »

French GP update:

LECLERC / SAINZ
Projected: 0.386% to 0.542% (Leclerc)
Current Result: 0.352% (Leclerc)
Data (click to show)

VERSTAPPEN / PEREZ
Projected: 0.385% - 0.544% (Verstappen)
Current Result: 0.376% (Verstappen)
Data (click to show)

NORRIS / RICCIARDO
Projected: 0.262% - 0.403% (Ricciardo)
Current Result: 0.142% (Norris)
Data (click to show)

STROLL / VETTEL
Projected: 0.262% - 0.316% (Vettel)
Current Result: 0.134% (Stroll)
Data (click to show)

OCON / ALONSO
Projected: 0.158% - 0.426% (Alonso)
Current Result: 0.629% (Ocon)
Data (click to show)

We have some movement this time, although not universally:

Once again, the Ferrari and Red Bull pairings hold more-or-less steady as they are. Verstappen went 2-0 over the last two whereas Leclerc drew his pair with Sainz, but in both cases the new numbers didn't bump the median out of position.

A different story at McLaren, as Norris has finally amassed enough qualifying wins to move the needle in his direction. Somewhat unfairly to his early season form, his indicated advantage of 0.142% actually seems fairly accurate for where the head-to-head stands at present.

Speaking of unfair results, Vettel and Stroll haven't changed at all: two consecutive N/A results for Stroll means Vettel hasn't been able to bank what would likely have been two results above his current average.

And finally, the Alpine pairing hasn't moved much, but is poised on the brink of a significant slide in Alonso's favor. Two consecutive qualifying wins for the older driver have left the head-to-head at 4-3, and Ocon's early (and huge) qualifying advantages may soon disappear from the median if the trend continues.
PICK 10 COMPETITION (7 wins, 21 podiums): 3rd in 2016
TOP THREE CHAMPIONSHIP (No Limit Excedrin Racing): Champions in 2015 & 2018 | 2nd in 2017 & 2019
AUTOSPORT GP PREDICTOR: 2017 USA & P-F1 Champion

Post Reply