Jump to content


Photo
* * * * * 1 votes

End of Season Prediction Graphs


  • Please log in to reply
108 replies to this topic

#1 turssi

turssi
  • Member

  • 3,368 posts
  • Joined: October 10

Posted 17 March 2013 - 18:27

Hi Guys and Gals,

Last year I did a funny "prediction thread" where I multiplied the points from the last race by the number of remaining races and added that to the current points totals. The original idea was to ridicule the old saying that the pilot is only as good as the last race, a saying that for most of us is ironical to start with, but in the heat of the season we tend to forget... Anyway, the thread got interesting when the end of the season came about and I decided to refine the formulas a little bit (thanks for the input forum user apoka).

This year I'm including the 03 pre-season tests as well, in order to get to posting sooner. Also I'm basing this on an Excel template and a graphical format to speed it up for me. So this years formula is: (points so far) + (points from the last 03 events divided by three) * (number of remaining races) = (end of season prediction).

----

Here is the WCC graph stating that Ferrari is the predicted champion after pre-season and after Australia. It's also showing 2nd place going to Merc and 3rd to Lotus. Merc is going upwards because its results are better when we disregard Jerez and include Australia. The opposite is true for Lotus.

Posted Image

For the WDC graph I decided to clean most of the drivers away, just leaving some more of the more interesting pilots in (the selection can change as the season proceeds):

Posted Image

To further explain what we are looking at here at the WDC graph I'll take Fernando as an example: He did real good during the pre-season, so he is right up there to start with. He did good in Australia as well and as the the non-appearance from Jerez test is cleaned out as well, he is now doing even better. So he is my top pick for the 2013 Champ after Australia!!!

As was last year, my numbers will have errors in them and my explanations may be lacking at times. So just ask for comments and justifications when there is something that does not seem to add up!

EDIT: Fixed the images to a smaller size.

Edited by turssi, 17 March 2013 - 20:50.


Advertisement

#2 turssi

turssi
  • Member

  • 3,368 posts
  • Joined: October 10

Posted 29 March 2013 - 22:53

Okey, so let's post the post-Malaysia predictions:

Posted Image

RBR has really improved from pre-season, but Mercedes has too and is now the nr1 prediction for taking the WCC!

Posted Image

Here is my prediction graphs with all pilots after Malaysia. Fernando came down to the level of mortals, Seb is the predicted champ with Lewis hot on his tail!

#3 turssi

turssi
  • Member

  • 3,368 posts
  • Joined: October 10

Posted 15 April 2013 - 15:26

OK, three races done and the prediction graphs should look like the current WDC standings (because my formula for this year is using the average of 03 races for the prediction).

Posted Image

Posted Image

And they do! Now let's keep watching, bring on Bahrain!!!

From the graphs so far we can see that Ferrari looked best in the pre-season and continues to shine. Only RBR is in front of them, showing that the Austrian team really did sandbag in the test sessions!

#4 Sakae

Sakae
  • Member

  • 19,256 posts
  • Joined: December 03

Posted 15 April 2013 - 15:33

Why you make things so complicated and unrealistic? Just plot predicted positions for top teams for full season, and then overlay actual results so we would know how good is your crystal ball. Simple.

#5 turssi

turssi
  • Member

  • 3,368 posts
  • Joined: October 10

Posted 15 April 2013 - 18:22

Why you make things so complicated and unrealistic? Just plot predicted positions for top teams for full season, and then overlay actual results so we would know how good is your crystal ball. Simple.


The actual results will be available only after season decider Brazil 2013.

#6 Ravenak

Ravenak
  • Member

  • 939 posts
  • Joined: September 12

Posted 15 April 2013 - 21:14

It's the 5th time I read one of these threads and I still don't understand how it works.

You gave us the exact calculation, but can you explain why you do it that way? What does it show and why? And how?

Edited by Ravenak, 15 April 2013 - 21:14.


#7 Shiroo

Shiroo
  • Member

  • 4,012 posts
  • Joined: October 12

Posted 15 April 2013 - 21:22

It's the 5th time I read one of these threads and I still don't understand how it works.

You gave us the exact calculation, but can you explain why you do it that way? What does it show and why? And how?

It is by using latest race and using it as current trend I asume.

#8 Ravenak

Ravenak
  • Member

  • 939 posts
  • Joined: September 12

Posted 15 April 2013 - 21:29

Yeah I get that, but can you extrapolate what it actually means?

What does each number do?

Basically, you take the last result and straighten the curb up or down for each driver? I'm that bad at maths :p

#9 turssi

turssi
  • Member

  • 3,368 posts
  • Joined: October 10

Posted 15 April 2013 - 23:40

@Ravenak: After each race I take the WDC points so far and extrapolate by multiplying the average points from the last three races by the number of remaining races. I do this for each driver separately and for the teams as well (for the teams I use the WCC standings).

About the graph I can use Fernando as an example: After the three pre-season sessions my formula put him at about 270 points, after Australia at about 390 points, after Malaysia at about 270 and now finally at about 320 after China.

So the Y-axis shows a prediction of the final championship points. On the X-axis I have all the 19 races plus a pre-season entry.

Of course all this is just for fun. Last year I did it with just a table and used only the result from the latest race when extrapolating. It might give some indication on who has been doing better than the rest when there is just five or four races remaining, but by that time we all know this just the same!

#10 superdelphinus

superdelphinus
  • Member

  • 3,175 posts
  • Joined: July 12

Posted 15 April 2013 - 23:46

Don't really get the point in this, apologies

#11 turssi

turssi
  • Member

  • 3,368 posts
  • Joined: October 10

Posted 15 April 2013 - 23:56

Don't really get the point in this, apologies


It just shows the total final points when one imagines that the pilot will keep scoring the same average for the all the remaining races.

The original point was to show that the old saying 'you're only as good as the last race' is pure bullocks. So in that I agree with you.

#12 Hanzo

Hanzo
  • Member

  • 899 posts
  • Joined: July 12

Posted 16 April 2013 - 00:56

The original point was to show that the old saying 'you're only as good as the last race' is pure bullocks.


I don't think is bollocks... I never took the phrase "You are only as good as your last... (match, race, game... etc)" literally... I take it as an irony about sports or life where people normally never take in consideration old records or merits, only your last result/performance. So that phrase is just a way of saying people don't have or don't want to have good memory or do proper analysis, they just look at your last result.

#13 turssi

turssi
  • Member

  • 3,368 posts
  • Joined: October 10

Posted 16 April 2013 - 01:17

@Hanzo: Yeah, it's true what you say and also last year someone pointed it out to me. It's a good message and we who forget and skip proper analysis need to be reminded about the fact at least yearly :-)

Edited by turssi, 16 April 2013 - 01:24.


#14 aditya-now

aditya-now
  • Member

  • 7,447 posts
  • Joined: June 02

Posted 16 April 2013 - 05:29

It just shows the total final points when one imagines that the pilot will keep scoring the same average for the all the remaining races.


It gives a certain perspective, when we think the driver will keep scoring the same average for all the remaining races - and it does give a realistic picture, IMHO.

Thanks for the graphs and thanks for the work! :up:


#15 Torsion

Torsion
  • Member

  • 627 posts
  • Joined: February 12

Posted 16 April 2013 - 06:30

It just shows the total final points when one imagines that the pilot will keep scoring the same average for the all the remaining races.

The original point was to show that the old saying 'you're only as good as the last race' is pure bullocks. So in that I agree with you.


I think it might be interesting if you also create an alternative graph where you add to the current points tally this year, to points scored by each driver/team last year for the remaining races.

Edited by Torsion, 16 April 2013 - 06:33.


#16 turssi

turssi
  • Member

  • 3,368 posts
  • Joined: October 10

Posted 16 April 2013 - 09:56

I think it might be interesting if you also create an alternative graph where you add to the current points tally this year, to points scored by each driver/team last year for the remaining races.


Thanks for the idea, I think I'll do this after Bahrain.

After all, in our hearts we know it to be true that there is always that one 'McLaren' or 'RBR' track where we expect our favourite adversary to do well just because their car was made for that track!

#17 Ravenak

Ravenak
  • Member

  • 939 posts
  • Joined: September 12

Posted 16 April 2013 - 17:14

Thanks for explaining, it kind of makes more sense now :)

#18 turssi

turssi
  • Member

  • 3,368 posts
  • Joined: October 10

Posted 28 April 2013 - 14:25

At this point of the season I'm dividing the top-4 in WDC and WCC to separate graphs to clean the view up a bit.

Ferrari's blunders are starting to hurt their predicted position in the WCC. No more tire or wing problems please!!!

Posted Image

On the low-end of things McLaren looks forward to recupering the pre-season form. The biggest surprise is predicting Toro Rosso to finish in front of both Sauber and Williams :-)

Posted Image

On the top-4 pilots we see that Fernando is suffering, Lewis has hit a roof, Kimi needs to get back to winning and that Sebatians rocket is taking him to the moon!

Posted Image

Best of the rest is Mark... or is it Romain? Too close to call, but none of these jokers is winning the WDC, that's for sure!

Posted Image

Edited by turssi, 28 April 2013 - 14:33.


#19 Afterburner

Afterburner
  • RC Forum Host

  • 9,173 posts
  • Joined: January 11

Posted 28 April 2013 - 14:40

The original point was to show that the old saying 'you're only as good as the last race' is pure bullocks. So in that I agree with you.

Hm, not too sure I agree with this. I think form fluctuates, so the most recent event in which any competitor participated seems to be a pretty good measure of current form, but at the same token, the idea that a competitor's form fluctuates makes this statement as invalid as it is valid because everyone is susceptible to a sudden fluctuation in form. :drunk:

I guess that's what you get when you type as you think. :p Either way, neat graphs--will keep an eye on them as the season progresses (as long as they stay on the front page, anyway :p). :up:

Advertisement

#20 apoka

apoka
  • Member

  • 5,878 posts
  • Joined: May 09

Posted 28 April 2013 - 15:18

The graphs start to beomce interesting now after we left the 3-race-prediction window! Thanks for posting.

#21 Sakae

Sakae
  • Member

  • 19,256 posts
  • Joined: December 03

Posted 28 April 2013 - 15:50

The actual results will be available only after season decider Brazil 2013.

Perhaps I didn't take enough time to think over what it is you are doing here, but at the face value, and forgive me for this, I am slightly lost why are you talking about end of season prediction, when there is no prediction of positions over full season. Periodical revisions to projections based on actual data-to-date as individual races are executed should be OK, I guess, but under normal circumstance I would be expecting for selected group of drivers (for simplicity sake) two lines. Predicted result for each race, and overlay of actual race finish. For my initial points count I would adjust it for 17, instead 19 races. (Based on historical performance, you may want to check on that, but I would expect probably one or two DNFs over course of a season). Anyway, it's good you are trying, and maybe you convince me at the end.  ;)

#22 turssi

turssi
  • Member

  • 3,368 posts
  • Joined: October 10

Posted 28 April 2013 - 16:45

@sakae: I have no prediction on individual race results. Only on the final WDC and WCC standings. That's why I call it 'end of season prediction'.

The prediction changes after each race and merges with the final results after Brazil. Check out the simple formula (in the original post) I use to confirm this. It has just 03 simple variables (points so far, average of last 03 races, number of races remaining).

Looking the latest graphs you can see the Y-axis has shown Ferrari around a predicted 550 WCC points up until Bahrain. At Bahrain this prediction fell to around 360 points because the prediction uses his current points after Bahrain + the average points scored from last 03 races (Their number one pilot Fernando retired in Malaysia, 1st in China and 8th in Bahrain, with Felipe scoring similar but lower points) for each of the remaining races. Since pre-season and up until Bahrain Ferrari average was great. It still is good, but RBR, Lotus and Mercedes are now in front when predicting with the formula I'm using.

Did I manage to explain?

edit reason: mixed up Fernando and Ferrari :-)

edit2: also it's called 'end of season prediction' because it tries to predict what the final results will be after all races of 2013 have been done. It's not because it was posted after 2012 was finished. In fact as it is currently it has nothing to do with last years historical results. Thinking about your post made me think that maybe we have mixed terminology in use!

Edited by turssi, 28 April 2013 - 19:01.


#23 Torsion

Torsion
  • Member

  • 627 posts
  • Joined: February 12

Posted 01 May 2013 - 10:30

@sakae: I have no prediction on individual race results. Only on the final WDC and WCC standings. That's why I call it 'end of season prediction'.

The prediction changes after each race and merges with the final results after Brazil. Check out the simple formula (in the original post) I use to confirm this. It has just 03 simple variables (points so far, average of last 03 races, number of races remaining).

Looking the latest graphs you can see the Y-axis has shown Ferrari around a predicted 550 WCC points up until Bahrain. At Bahrain this prediction fell to around 360 points because the prediction uses his current points after Bahrain + the average points scored from last 03 races (Their number one pilot Fernando retired in Malaysia, 1st in China and 8th in Bahrain, with Felipe scoring similar but lower points) for each of the remaining races. Since pre-season and up until Bahrain Ferrari average was great. It still is good, but RBR, Lotus and Mercedes are now in front when predicting with the formula I'm using.

Did I manage to explain?

edit reason: mixed up Fernando and Ferrari :-)

edit2: also it's called 'end of season prediction' because it tries to predict what the final results will be after all races of 2013 have been done. It's not because it was posted after 2012 was finished. In fact as it is currently it has nothing to do with last years historical results. Thinking about your post made me think that maybe we have mixed terminology in use!


I am struggling to understand why Fernando and Ferrari are starting at the top from Australia, when Kimi won the first race? are you bringing forward some sort of moving average from the last season?



#24 turssi

turssi
  • Member

  • 3,368 posts
  • Joined: October 10

Posted 01 May 2013 - 13:00

I am struggling to understand why Fernando and Ferrari are starting at the top from Australia, when Kimi won the first race? are you bringing forward some sort of moving average from the last season?


I gave them points from the pre-season so I would not have to wait for the first three races to start posting.

My pre-season prediction is based on the 25-18-15-12-10-8-6-4-2-1 point structure just like race Sundays. The order of the pilots is according to the fastest time of each pilot during the test days in Jerez one, Jerez two and was it Catalonia? So three sets already during the pre-season!

Anyway I used the conglomerate time sheet posts of Autosports front page instead of keeping track of who had gone and how fast :-) Those times put Ferrari on top, while RBR was still getting up to speed.

#25 Bean

Bean
  • New Member

  • 28 posts
  • Joined: April 03

Posted 01 May 2013 - 18:42

This is a fun diversion during these tough 3-week F1 breaks... Thanks for the work!



#26 turssi

turssi
  • Member

  • 3,368 posts
  • Joined: October 10

Posted 01 May 2013 - 22:47

@torsion:

Is this like something you suggested?

Posted Image

Don't know if I'll keep up with it, as it is not compatible with the final points my current graphs have, and because of this I don't know how to integrate them into one.

#27 Nustang70

Nustang70
  • Member

  • 2,439 posts
  • Joined: July 01

Posted 02 May 2013 - 08:03

Neat charts. Thanks.

#28 Torsion

Torsion
  • Member

  • 627 posts
  • Joined: February 12

Posted 02 May 2013 - 09:06

@torsion:

Is this like something you suggested?

Posted Image

Don't know if I'll keep up with it, as it is not compatible with the final points my current graphs have, and because of this I don't know how to integrate them into one.


Yes - thanks for the work.

Shows how strong Seb was in the second half of last season. If the same happens again he is clearly the favorite for the title this year.

Also its very interesting how the points for the top 4 diverged Monza onwards.

Edited by Torsion, 02 May 2013 - 09:10.


#29 turssi

turssi
  • Member

  • 3,368 posts
  • Joined: October 10

Posted 02 May 2013 - 13:04

Yeah, it's very clear that the championship has 3 segments, with major updates (or no updates, in case of team shifting focus to the next season) coming for Australia, Spain and Singapore.

#30 Beamer

Beamer
  • Member

  • 3,384 posts
  • Joined: September 00

Posted 03 May 2013 - 12:30

Nice! It's always fun to play with numbers, even if we all know it's got little real value ;) At what point in last year's season did your prediction begin to look like the actual end result?

#31 turssi

turssi
  • Member

  • 3,368 posts
  • Joined: October 10

Posted 03 May 2013 - 13:03

Nice! It's always fun to play with numbers, even if we all know it's got little real value ;) At what point in last year's season did your prediction begin to look like the actual end result?


Like, for the last three races, that's a few races into Vettel's final megaspurt! So very little real value :-)

#32 turssi

turssi
  • Member

  • 3,368 posts
  • Joined: October 10

Posted 13 May 2013 - 00:41

Here is the Torsion Special (named after forum user torsion) after Spain:

Posted Image

Still 14 races to go, so no comment!

#33 Ravenak

Ravenak
  • Member

  • 939 posts
  • Joined: September 12

Posted 13 May 2013 - 00:53

This cannot really be an even remotely meaningful graph, as Vettel's sudden domination starting in Korea gives him the edge in any case.

Not trying to be a buzzkill, but it will stay the same unless Alonso wins every single race now.

Edited by Ravenak, 13 May 2013 - 00:54.


#34 turssi

turssi
  • Member

  • 3,368 posts
  • Joined: October 10

Posted 13 May 2013 - 01:08

Looking at the top-4 pilots (below) we see Fernando bouncing back with an intent, BUT surprisingly KIMI is now the predicted champ! Better keep those 2nd places coming!!!

Posted Image

Top-4 constructors have no surprises: It's Ferrari for the win after Spain!

Posted Image

The lesser pilots' prediction says that Felipe, Paul and Jenson are finding some form. Good for them.

Posted Image

Finally to the not-WCC-material constructors. Here the McLaren and FI match-up continues to be too close to call. Good for FI, not so good for Mcl...

Posted Image

#35 turssi

turssi
  • Member

  • 3,368 posts
  • Joined: October 10

Posted 13 May 2013 - 01:10

This cannot really be an even remotely meaningful graph, as Vettel's sudden domination starting in Korea gives him the edge in any case.

Not trying to be a buzzkill, but it will stay the same unless Alonso wins every single race now.


I like the original idea (current points plus the average from the last three races for the future races) better. Uploaded the graph with Spain points being considered above.

#36 Torsion

Torsion
  • Member

  • 627 posts
  • Joined: February 12

Posted 13 May 2013 - 10:13

This cannot really be an even remotely meaningful graph, as Vettel's sudden domination starting in Korea gives him the edge in any case.

Not trying to be a buzzkill, but it will stay the same unless Alonso wins every single race now.


@turssi - hadn't realised you had named a graph after me :cat:

@Ravenak - personally I felt it was the only reasonable way to predict which teams and drivers would perform best in the upcoming races. I don't think any of the graphs here, nor any prediction graphs anywhere is meaningful if you simply take them by themselves, as each will have their own strengths and weaknesses. Its up to the person viewing them to figure out what they may or may not be saying about the championship.

For example, one thing I see is that it is necessary for Ferrari and Lotus to build a lead in the Championship over European races against RBR, as they (RBR) are historically strong in those fly away races where tyre deg is much less, and tracks are stop-and-go.

Edited by Torsion, 13 May 2013 - 10:22.


#37 turssi

turssi
  • Member

  • 3,368 posts
  • Joined: October 10

Posted 30 May 2013 - 18:36

Here's some more wood to the fire to keep the discussion going. I'm talking to you Ravenak and Torsion :-)

Posted Image

We got now six races before leaving Europe behind at Singapore. Now is the time for the pilots to keep their campaign on track by maximizing points gained and minimizing points lost. From the graph I see that catching up can be real difficult. I say that if Vettel does not do a zero points race soon, then the others can catch him only by getting 3 straight wins. Sounds like a hard task!

Let's keep an eye on the development and starting from Singapore the end game should be on between the top pilots.

#38 turssi

turssi
  • Member

  • 3,368 posts
  • Joined: October 10

Posted 30 May 2013 - 19:04

Looking at the WCC (predicted total points in the end of the season on the y-axis, while on the x-axis the situation after each race) the only team doing the right thing is RBR:

Posted Image

On the low end of things the situation is worse, no one is getting it right:

Posted Image

I won't name any names because this kind of lack of performance can't be blamed on just one person, or even four Pirelli tyres! What a shame.

#39 rasul

rasul
  • Member

  • 1,952 posts
  • Joined: October 12

Posted 30 May 2013 - 19:26

Thanks for the effort!

It is interesting that last year, after Monaco GP, Alonso and Hamilton had pretty much the same amount of points as this year. Vettel and Kimi(and Massa) have much more points than they had last year.
Last year:
1. Alonso - 76 points
2. Vettel - 73
3. Webber - 73
4. Hamilton - 63
5. Rosberg - 59
6. Kimi - 51
....
8. Grosjean - 35
14. Massa - 10

This year:
1. Vettel - 107 (+34)
2. Kimi - 86 (+35)
3. Alonso - 78 (+2)
4. Hamilton - 62 (-1)
5. Webber - 57 (-16)
6. Rosberg - 47 (-12)
7. Massa - 45 (+35)
.....
9. Grosjean - 26 (-9)

Advertisement

#40 turssi

turssi
  • Member

  • 3,368 posts
  • Joined: October 10

Posted 30 May 2013 - 19:28

Situation after Monaco on the WDC side of things:

Posted Image

I have included new pilots to the top class after good performances from NicoR and Mark and bad from Kimi and Fernando. So far the only pilot with no bad races is Seb and it shows above where the total is based on the points so far plus average of the last three races predicted for the remaining races.

Posted Image

Looking at the pilots who score less points the order is Paul, Felipe, Romain, Jenson, Adrian, Sergio, Jean-Eric, Daniel and NicoH followed by the 0-points club and its members. This is an interesting list where the FI pilots can be proud of themselves even if the team is not doing a good job. Felipe and Romain should be ashamed, as should Sauber and Williams! About McLaren and their pilots even I feel shamed!

Edited by turssi, 30 May 2013 - 19:36.


#41 turssi

turssi
  • Member

  • 3,368 posts
  • Joined: October 10

Posted 30 May 2013 - 19:41

@rasul: yeah, Sebastians 107 points after six races is crazy good. That's about 17.8 points from every outing. No mistakes by him nor the team and they still have the nerve the complain about the Pirelli's. Maybe they'll get some delaminations in Canada as Karma comes in :-)

#42 wrcva

wrcva
  • Member

  • 1,254 posts
  • Joined: January 10

Posted 31 May 2013 - 02:03

Here's some more wood to the fire to keep the discussion going. I'm talking to you Ravenak and Torsion :-)

Posted Image

We got now six races before leaving Europe behind at Singapore. Now is the time for the pilots to keep their campaign on track by maximizing points gained and minimizing points lost. From the graph I see that catching up can be real difficult. I say that if Vettel does not do a zero points race soon, then the others can catch him only by getting 3 straight wins. Sounds like a hard task!

Let's keep an eye on the development and starting from Singapore the end game should be on between the top pilots.


Nice perspective, thanks!

I wonder how this may look if you were to apply @rasul's anaysis as a correction factor to the projected results (2013 Canada onwards).
For example, if Seb scored, on average, 30% more points in completed 2013 races (2012 vs 2013) his projections for upcoming races (2012 points) are adjusted up, similarly Mark's projections will be adjusted down because he scored less points this year... it is a correction factor with respect to average current year performance to date taken as a proxy for how good/bad they might be doing in upcoming venues relative to 2012 type of thing. That will make Seb's slope even steeper...





#43 turssi

turssi
  • Member

  • 3,368 posts
  • Joined: October 10

Posted 31 May 2013 - 21:01

Nice perspective, thanks!

I wonder how this may look if you were to apply @rasul's anaysis as a correction factor to the projected results (2013 Canada onwards).
For example, if Seb scored, on average, 30% more points in completed 2013 races (2012 vs 2013) his projections for upcoming races (2012 points) are adjusted up, similarly Mark's projections will be adjusted down because he scored less points this year... it is a correction factor with respect to average current year performance to date taken as a proxy for how good/bad they might be doing in upcoming venues relative to 2012 type of thing. That will make Seb's slope even steeper...



Could make sense for sure, but what would you compare to get the correction factor? All races so far? That would be simple then: Σ(Points for the races so far in 2013)/(Points from the same races in 2012)*(Points from the race in question in 2012) for the dashed line races. I'll do both after Canada for you to compare.

The new one shall act as the base for my future white paper on F1-prediction models called "Applying wrcva-integration on rasul-corrected Torsion-graphs"!

#44 apoka

apoka
  • Member

  • 5,878 posts
  • Joined: May 09

Posted 31 May 2013 - 21:47

@rasul: yeah, Sebastians 107 points after six races is crazy good. That's about 17.8 points from every outing. No mistakes by him nor the team and they still have the nerve the complain about the Pirelli's. Maybe they'll get some delaminations in Canada as Karma comes in :-)

Well, in 2011 he had 143(!) points in the first 6 races averaging 23.8 points per race. That's crazy good.;)

In 2010, the WDC had 13.5 points per race after the end of the season, in 2011 20.6 points per race and in 2012 14.1 points per race. I'm just throwing in some numbers to put his points score into perspective.


#45 rasul

rasul
  • Member

  • 1,952 posts
  • Joined: October 12

Posted 31 May 2013 - 21:52

Could make sense for sure, but what would you compare to get the correction factor? All races so far? That would be simple then: Σ(Points for the races so far in 2013)/(Points from the same races in 2012)*(Points from the race in question in 2012) for the dashed line races. I'll do both after Canada for you to compare.

The new one shall act as the base for my future white paper on F1-prediction models called "Applying wrcva-integration on rasul-corrected Torsion-graphs"!

Heh. That's one hell of a name. :D

#46 turssi

turssi
  • Member

  • 3,368 posts
  • Joined: October 10

Posted 16 June 2013 - 16:05

OK, so here are two versions of the graph discussed above:

Posted Image

Here we just consider the points so far and complete the line using points from last year for 04 pilots, which gives us a statement that the season is already over :-) Expect no suprises!

Posted Image

The second one that adjusts the prediction by (2013 points so far) / (2012 points from the same races) and gives some crazy results (more than 25 points per race for Seb in the end season), but still manages to visualize that the Kimi/Lotus combo has had a better start to the season in 2013 than in 2012. So good in fact that it puts him into a similar expectative as the Fernando/Ferrari combo! Vettel's superiority is also underlined, while Lewis suffers heavily from the many non-points finishes in 2012.

#47 turssi

turssi
  • Member

  • 3,368 posts
  • Joined: October 10

Posted 16 June 2013 - 16:35

Back to my original style of predictions:

Posted Image

Mercedes and Mark/RBR are coming back to the game: Now we have 06 pilots with a predicted points total of over 200 points! Sebastian is leading and is stabilizing at over 450 points, Fernando is predicted to end about 70 behind and Nico another 70 points behind. Mark will be fourth while Lewis will pass Kimi, who is definetly crashing! That's what you get going from scoring 2nd places to scoring 2 points per race!!!

Posted Image

The second class pilots are all scoring below 200 predicted WDC points come end of the season. The pack is lead by Felipe and includes big recent performance crashes from Sergio and Romain, both of whom were not doing so well to start with....

#48 turssi

turssi
  • Member

  • 3,368 posts
  • Joined: October 10

Posted 16 June 2013 - 17:04

Finally looking at the constructors after Canada:

Posted Image

The Big-3 have finally arrived, I wonder if they are in the exact money spending order as well?

Posted Image

Lotus is still predicted to be the best of the rest, but their plight of no pilot scoring big recently is extremely visible! Also McLaren needs to get their act together if they don't want to fall behind STR in addition to FI!

Edited by turssi, 16 June 2013 - 17:17.


#49 apoka

apoka
  • Member

  • 5,878 posts
  • Joined: May 09

Posted 16 June 2013 - 17:08

Thanks for updating the graphs! :up:

#50 quaint

quaint
  • Member

  • 831 posts
  • Joined: June 13

Posted 17 June 2013 - 22:16

OK, so here are two versions of the graph discussed above:

Posted Image


Still can't help feeling like your graph is a bit off regarding Kimi and Fernando so far this season. After Spain RAI seems to be closer to ALO than to VET, even though the actual gaps were 15 points and 4 points, respectively. Furthermore, RAI and ALO seem to almost equal after Monaco, even though they really had between them a gap bigger than what they had after the season opener (but the graph certainly does not convey that).