Friday, July 17, 2015

Getting Mileage out of Milwaukee Part 2

Hey there Internet, it's Guido here with another Scoring Indy update. This week, we dive back in to the numbers from Milwaukee in order to build a better oval formula. Here we go!



Where We Left Off


Earlier this week, we looked at the relative importance of one of our mathematical terms: "on track" performance. This term takes two things into account: 1) positions gained over the course of the race, and 2) laps led. We left off by saying that this term "ought" to account for about a quarter of the total score.


This was problematic, however, because we were left with the issue of finishing position accounting for the other three quarters. We need another term: one that takes into account in-race performance (not just things that happen at the end), in order to give us an accurate picture in our Race Scores, which will then allow us to compare drivers. And, that leads us to:



Something New


Obviously, the best way to do this would be how I used to: by using Average Running Position. Unfortunately, there were 5,177 total laps completed during the IndyCar race in Milwaukee. To calculate Average Running Position, I would have to input 5,177 different data points into my spreadsheet. I don't know a better way to calculate this. If anyone knows a better way to do it, or of somewhere where Average Running Position is available online, let me know. I'd love to use it again.


But, for now, we don't have access to Average Running Position. It's just too much work for me. Instead, we're going to be using something that we're calling Representative Running Position. To calculate this, I input running position from 30 different laps into my spreadsheet. I picked 30 because it was approximately 1/8 of the race, and it was a nice round number. The 30 laps that I used were lap 1, each lap ending in zero (10, 20, 30, etc.), and four random laps. For this race it was: 29, 123, 167, and 245.


Now, instead of having to input over 5000 data points, I have to input less than 700. That's much more reasonable. And, I think that it does a good job of mirroring Average Running Position. So, if we divide up the remaining three quarters of the Race Score evenly among Finishing Position and Representative Running Position, we get a scorecard that looks like this:



Milwaukee Scorecard Redux


FinishDriverGridLedCompletedR.R.P.Race Score







1Bourdais111182502.9794.14
2Castroneves24025012.1378.96
3Rahal652506.6766.65
4Montoya802509.7760.78
5Newgarden11092503.7371.21
6Kanaan432504.6759.68
7Dixon10142506.5361.50
8Andretti902507.4755.00
9Pagenaud17025012.1353.44
10Carpenter22025016.1349.79
11Chaves12025011.3744.22
12Kimball7025012.9034.01
13Hunter-Reay16025010.9743.80
14Sato13025016.4029.58
15Munoz18025013.4336.82
16Vautier20024818.9727.55
17Hawksworth12022120.6213.95
18Wilson15121912.1225.92
19Karam301837.8216.83
20Coletti19015618.7213.28
21Briscoe201307.5610.13
22Power14013011.0611.68
23Jakes501138.386.38
24Mann2302724.001.62


Now, let's look at some of how this compares to how the Milwaukee scores looked when we left off earlier this week.

- BEFORE: Tristan Vautier had the highest score for anyone not completing 250 laps followed by Jack Hawksworth, and the scores weren't even close.

- NOW: Vautier still scores best (he completed almost all the race, after all), but now Justin Wilson is second, and Sage Karam is not out of the conversation. Also, the scores are more tightly clustered.

- BEFORE: The Top 5 finishers had the Top 5 Race Scores. In order.

- NOW: Montoya drops out of the Top 5 all together. Newgarden slots up in P3, and Scott Dixon has the fifth-best Race Score.

- BEFORE: Castroneves had a Race Score comparable to Sebastien Bourdais.

- NOW: Castroneves' number is more on the order of Newgarden and Rahal.


Finally

The last thing that we'll do here today is to compare how drivers stack up against each other using earlier-this-week's Two Term Formula, and today's Three Term formula that we will use going forward for ovals. Comparing drivers is, after all what we're all about here at Scoring Indy.

So, here's every driver ranked from 1 to 24 using each formula:




Three TermTwo Term



1BourdaisBourdais
2CastronevesCastroneves
3NewgardenRahal
4RahalMontoya
5DixonNewgarden
6MontoyaDixon
7KanaanCarpenter
8AndrettiPagenaud
9PagenaudKanaan
10CarpenterAndretti
11ChavesChaves
12Hunter-ReayHunter-Reay
13MunozKimball
14KimballMunoz
15SatoSato
16VautierVautier
17WilsonHawksworth
18KaramWilson
19HawksworthColetti
20ColettiKaram
21PowerPower
22BriscoeMann
23JakesBriscoe
24MannJakes


Personally, I think that our new Three Term Formula nails it. At least it nails it for Milwaukee. We'll see how it does at Iowa.


Stay Tuned

I'm excited for another set of data points and the always-thrilling racing of Iowa Speedway. So, buckle up for a wild ride. You can follow me on Twitter for blog updates: @ScoringIndy. That's all for now, I'll see you in a week with the scorecard from the Iowa Corn 300!

-- Guido

1 comment: