planetf1.com

It is currently Sun Oct 20, 2019 4:13 am

All times are UTC


Forum rules


Please read the forum rules



Post new topic Reply to topic
Author Message
PostPosted: Thu Sep 19, 2019 10:32 pm 
Offline
User avatar

Joined: Sat Mar 06, 2004 8:07 am
Posts: 1233
to settle a discussion with a friend he asked me to remodel the current season as if all drivers had had no retirements and instead finsihed those races in their average finish position for the year. He was convinced Ricciardo would be way higher in the WDC. Anyway , just for fun folks:

1 (-0)  Lewis Hamilton (284)
2 (-0)  Valtteri Bottas (238)
3 (+1)  Charles Leclerc (212)
4 (-1)  Max Verstappen (199)
5 (-0)  Sebastian Vettel (169)
6 (+1)  Carlos Sainz Jr. (74)
7 (-1)  Pierre Gasly (70)
8 (-0)  Daniel Ricciardo (43)
9 (+1)  Daniil Kvyat (42)
10 (-1)  Alexander Albon (37)
11 (-0)  Nico Hülkenberg (36)
12 (+2)  Lando Norris (32)
13 (-1)  Kimi Räikkönen (31)
14 (-1)  Sergio Pérez (29)
15 (+1)  Kevin Magnussen (21)
16 (-1)  Lance Stroll (20)
17 (-0)  Romain Grosjean (14)
18 (-0)  Antonio Giovinazzi (3)
19 (-0)  Robert Kubica (1)
20 (-0)  George Russell (0)

So actually Norris is the biggest "loser" but overall the difference is negligible.

_________________
"I'd rather lose a race going fast enough to win it, than win one going slow enough to lose it".
-Stirling Moss


Top
 Profile  
 
PostPosted: Fri Sep 20, 2019 8:32 am 
Offline
User avatar

Joined: Thu Feb 13, 2003 9:39 pm
Posts: 3587
Does this just add points to the driver in question or does it readjust that race result? And if so are those readjustments factored into the drivers' average position?


Top
 Profile  
 
PostPosted: Fri Sep 20, 2019 11:01 am 
Offline
User avatar

Joined: Thu Apr 22, 2010 11:24 am
Posts: 2272
I’m not entirely sure how you’ve calculated this but your methodology seems flawed to me. For example Norris on 32 points is only 7 more than he has in real life. He lost 10 just on the last lap at Spa.

_________________
I remember when this website was all fields.


Top
 Profile  
 
PostPosted: Fri Sep 20, 2019 11:54 am 
Offline

Joined: Tue Sep 26, 2006 9:21 am
Posts: 3853
Banana Man wrote:
I’m not entirely sure how you’ve calculated this but your methodology seems flawed to me. For example Norris on 32 points is only 7 more than he has in real life. He lost 10 just on the last lap at Spa.

I expect that's happened because he was classified in Spa, having compeleted more than 90% of the race.

_________________
Forum rules: viewtopic.php?f=3&t=14979

Please report forum problems to us, via PM/Feedback Thread. Screenshots will also help.


Top
 Profile  
 
PostPosted: Fri Sep 20, 2019 12:39 pm 
Offline
User avatar

Joined: Sat Mar 06, 2004 8:07 am
Posts: 1233
no race re-modelling as that's just conjecture. Literally as follows

Drivers total points / races completed = av/pts finish
Av pts/finish * 14

So for Norris has 25 pts from 11 finishes (as per Mod's comment earlier)

That's an av of 2.7 pts/race. So 2.7 * 14 = 32

_________________
"I'd rather lose a race going fast enough to win it, than win one going slow enough to lose it".
-Stirling Moss


Top
 Profile  
 
PostPosted: Fri Sep 20, 2019 3:21 pm 
Offline

Joined: Tue Oct 21, 2003 11:06 am
Posts: 7812
Location: Belgium
DOLOMITE wrote:
no race re-modelling as that's just conjecture. Literally as follows

Drivers total points / races completed = av/pts finish
Av pts/finish * 14

So for Norris has 25 pts from 11 finishes (as per Mod's comment earlier)

That's an av of 2.7 pts/race. So 2.7 * 14 = 32
I've probably not understood anything about your method, but 25/11=2.27. With that number, your final calculation is correct. With an average of 2.7, he would have had 38 points. A sticky computer key?

_________________
Use every man after his desert, and who should scape whipping? Use them after your own honour and dignity.

Maria de Villota - Jules Bianchi


Top
 Profile  
 
PostPosted: Fri Sep 20, 2019 8:42 pm 
Offline
User avatar

Joined: Sat Mar 06, 2004 8:07 am
Posts: 1233
Fiki wrote:
DOLOMITE wrote:
no race re-modelling as that's just conjecture. Literally as follows

Drivers total points / races completed = av/pts finish
Av pts/finish * 14

So for Norris has 25 pts from 11 finishes (as per Mod's comment earlier)

That's an av of 2.7 pts/race. So 2.7 * 14 = 32
I've probably not understood anything about your method, but 25/11=2.27. With that number, your final calculation is correct. With an average of 2.7, he would have had 38 points. A sticky computer key?


No, just the perils of updating PF1 whilst sat on a conference call! Meant to type 2.27. Which is why the final numbers are correct.

_________________
"I'd rather lose a race going fast enough to win it, than win one going slow enough to lose it".
-Stirling Moss


Top
 Profile  
 
PostPosted: Fri Sep 20, 2019 10:07 pm 
Offline
User avatar

Joined: Thu Apr 22, 2010 11:24 am
Posts: 2272
Okay understood but the methodology is too flawed to really mean anything.

For example if Norris had broken down 4 laps earlier, he would be better off as it would count as a DNF and his points per finish would be higher, because you remove an 11th place from his average.

Likewise LeClerc in Bahrain, Norris in France, Stroll in Italy etc. The driver is penalised, rather than corrected for the points lost.

_________________
I remember when this website was all fields.


Top
 Profile  
 
PostPosted: Sat Sep 21, 2019 7:45 am 
Offline
User avatar

Joined: Thu Feb 13, 2003 9:39 pm
Posts: 3587
There is a huge flaw with this system, as you would expect it to not change the positions much. If you are just adding in their season average score by the number of retirements then it's only going to make any difference to drivers who had disproportionate number of retirements. If every driver had the same number of retirements it would make zero difference to their score. If they have retired from 1 out of 11 races it will increase their score by 10%. Given that for the lower ranked drivers, the percentage difference in points is high (ie, +100% from 10th to 9th, +100% from 9th to 8th, +50% from 8th to 7th) the loss of points in the midfield from one failed result is more statistically singificant as well - as it's not just the swing from the driver who lost it, it's a gain to the drivers who inherited them, with whom they are in competition with. Norris losing his 5th place in Spa is the perfect example of this. Had he not retired he had that 5th place nailed on, there is no speculation there, it's not like if he had retired with 20 laps left where anything could happen. He lost 10 points from that (and Albon, Kyvat, Hulkenberg and Perez all gained 2 points) all drivers he would have leapfrogged. That 12 point delta is 50% of his 25 points, yet if this is counted as a retirement he only gets 2 points back, and his rivals keep their advantage.


Top
 Profile  
 
PostPosted: Sat Sep 21, 2019 2:38 pm 
Offline

Joined: Mon Nov 26, 2018 12:53 am
Posts: 554
Wow, what a crazy thread with a wild methodology!


Top
 Profile  
 
PostPosted: Sun Sep 22, 2019 8:41 pm 
Offline
User avatar

Joined: Sat Mar 06, 2004 8:07 am
Posts: 1233
Alienturnedhuman wrote:
There is a huge flaw with this system, as you would expect it to not change the positions much. If you are just adding in their season average score by the number of retirements then it's only going to make any difference to drivers who had disproportionate number of retirements. If every driver had the same number of retirements it would make zero difference to their score. If they have retired from 1 out of 11 races it will increase their score by 10%. Given that for the lower ranked drivers, the percentage difference in points is high (ie, +100% from 10th to 9th, +100% from 9th to 8th, +50% from 8th to 7th) the loss of points in the midfield from one failed result is more statistically singificant as well - as it's not just the swing from the driver who lost it, it's a gain to the drivers who inherited them, with whom they are in competition with. Norris losing his 5th place in Spa is the perfect example of this. Had he not retired he had that 5th place nailed on, there is no speculation there, it's not like if he had retired with 20 laps left where anything could happen. He lost 10 points from that (and Albon, Kyvat, Hulkenberg and Perez all gained 2 points) all drivers he would have leapfrogged. That 12 point delta is 50% of his 25 points, yet if this is counted as a retirement he only gets 2 points back, and his rivals keep their advantage.


Alien, this is all correct but not sure why you felt the need to post it. I never claimed the system was anything - just said what I'd done. Certainly didn't say it was a perfect system. I'm well aware of what it does/doesn't do. Norris losing his 5th place is actually the perfect example of why I DID take this approach. Lando nearly scored the points...but didn't. That's all you're left with. Coulda shoulda woulda... What if if he'd retired 2 laps earlier than he did - still say it was "nailed on", what if it was 3 laps? 4?.. You say it's 50% of his 25 points but with your approach you'd have to go back and work out the points you think he *should* have had - not what he actually had. And while you're doing that don't forget to remove points he picked up due to attrition...

It's a system of averages and I made that very clear so pointing out what it isn't seems a little unnecessary.



If you feel you can re-model the race results according to what you "know" would have happened then I'm sure we'd all love to see that.

_________________
"I'd rather lose a race going fast enough to win it, than win one going slow enough to lose it".
-Stirling Moss


Top
 Profile  
 
PostPosted: Sun Sep 29, 2019 3:42 am 
Offline

Joined: Mon Nov 26, 2018 12:53 am
Posts: 554
DOLOMITE wrote:
Alienturnedhuman wrote:
There is a huge flaw with this system, as you would expect it to not change the positions much. If you are just adding in their season average score by the number of retirements then it's only going to make any difference to drivers who had disproportionate number of retirements. If every driver had the same number of retirements it would make zero difference to their score. If they have retired from 1 out of 11 races it will increase their score by 10%. Given that for the lower ranked drivers, the percentage difference in points is high (ie, +100% from 10th to 9th, +100% from 9th to 8th, +50% from 8th to 7th) the loss of points in the midfield from one failed result is more statistically singificant as well - as it's not just the swing from the driver who lost it, it's a gain to the drivers who inherited them, with whom they are in competition with. Norris losing his 5th place in Spa is the perfect example of this. Had he not retired he had that 5th place nailed on, there is no speculation there, it's not like if he had retired with 20 laps left where anything could happen. He lost 10 points from that (and Albon, Kyvat, Hulkenberg and Perez all gained 2 points) all drivers he would have leapfrogged. That 12 point delta is 50% of his 25 points, yet if this is counted as a retirement he only gets 2 points back, and his rivals keep their advantage.


Alien, this is all correct but not sure why you felt the need to post it. I never claimed the system was anything - just said what I'd done. Certainly didn't say it was a perfect system. I'm well aware of what it does/doesn't do. Norris losing his 5th place is actually the perfect example of why I DID take this approach. Lando nearly scored the points...but didn't. That's all you're left with. Coulda shoulda woulda... What if if he'd retired 2 laps earlier than he did - still say it was "nailed on", what if it was 3 laps? 4?.. You say it's 50% of his 25 points but with your approach you'd have to go back and work out the points you think he *should* have had - not what he actually had. And while you're doing that don't forget to remove points he picked up due to attrition...

It's a system of averages and I made that very clear so pointing out what it isn't seems a little unnecessary.



If you feel you can re-model the race results according to what you "know" would have happened then I'm sure we'd all love to see that.


I think you need to accept that your methodology was terrible and take your medicine. He was right to call you out on it.


Top
 Profile  
 
PostPosted: Sun Sep 29, 2019 4:07 am 
Offline
User avatar

Joined: Thu Feb 13, 2003 9:39 pm
Posts: 3587
DOLOMITE wrote:
Alienturnedhuman wrote:
There is a huge flaw with this system, as you would expect it to not change the positions much. If you are just adding in their season average score by the number of retirements then it's only going to make any difference to drivers who had disproportionate number of retirements. If every driver had the same number of retirements it would make zero difference to their score. If they have retired from 1 out of 11 races it will increase their score by 10%. Given that for the lower ranked drivers, the percentage difference in points is high (ie, +100% from 10th to 9th, +100% from 9th to 8th, +50% from 8th to 7th) the loss of points in the midfield from one failed result is more statistically singificant as well - as it's not just the swing from the driver who lost it, it's a gain to the drivers who inherited them, with whom they are in competition with. Norris losing his 5th place in Spa is the perfect example of this. Had he not retired he had that 5th place nailed on, there is no speculation there, it's not like if he had retired with 20 laps left where anything could happen. He lost 10 points from that (and Albon, Kyvat, Hulkenberg and Perez all gained 2 points) all drivers he would have leapfrogged. That 12 point delta is 50% of his 25 points, yet if this is counted as a retirement he only gets 2 points back, and his rivals keep their advantage.


Alien, this is all correct but not sure why you felt the need to post it. I never claimed the system was anything - just said what I'd done. Certainly didn't say it was a perfect system. I'm well aware of what it does/doesn't do. Norris losing his 5th place is actually the perfect example of why I DID take this approach. Lando nearly scored the points...but didn't. That's all you're left with. Coulda shoulda woulda... What if if he'd retired 2 laps earlier than he did - still say it was "nailed on", what if it was 3 laps? 4?.. You say it's 50% of his 25 points but with your approach you'd have to go back and work out the points you think he *should* have had - not what he actually had. And while you're doing that don't forget to remove points he picked up due to attrition...

It's a system of averages and I made that very clear so pointing out what it isn't seems a little unnecessary.



If you feel you can re-model the race results according to what you "know" would have happened then I'm sure we'd all love to see that.

I wasn't intending to be confrontational, I have the utmost respect for you with all the stats and analysis that you provide to the forum. That's ultimately why I was surprised that you presented it as mathematically it's highly unlikely to make any significant changes - however after rereading the original post I misunderstood the situation. I thought your friend was convinced that Ricciardo would be much higher if not for retirements and you came up with this system to test his hypothesis, I didn't realise it was your friend also suggested this method.


Top
 Profile  
 
Display posts from previous:  Sort by  
Post new topic Reply to topic

All times are UTC


Who is online

Users browsing this forum: LBET and 1 guest


You cannot post new topics in this forum
You cannot reply to topics in this forum
You cannot edit your posts in this forum
You cannot delete your posts in this forum

Search for:
Jump to:  
Powered by phpBB® Forum Software © phpBB Group