Monday, March 25, 2013

IndyCar St. Pete GP

That was a satisfying race. A very satisfying race. How satisfying was it? And, for whom? What great questions, Internet. Let's look at the Bacchanalia of data:

Race Finish Driver Starting Position Laps Led Laps Completed Race Score
1 Hinchcliffe 4 26 110 67.82
2 Castroneves 5 42 110 73.09
3 Andretti 7 0 110 54.00
4 Kanaan 11 0 110 58.00
5 Dixon 20 0 110 72.00
6 de Silvestro 3 0 110 34.00
7 Viso 22 0 110 68.00
8 Sato 2 0 110 24.00
9 Wilson 13 0 110 42.00
10 Tagliani 17 0 110 46.00
11 Bourdais 21 0 110 50.00
12 Kimball 14 0 110 32.00
13 Rahal 15 0 110 30.00
14 Carpenter 23 0 110 42.00
15 Jakes 18 0 110 28.00
16 Power 1 26 107 2.31
17 Servia 12 16 104 15.42
18 Hunter-Reay 8 0 79 1.64
19 Hildebrand 24 0 78 21.09
20 Saavedra 9 0 72 -2.40
21 Vautier 6 0 69 -8.82
22 Beatriz 25 0 55 11.00
23 Newgarten 16 0 50 -0.36
24 Pagenaud 19 0 26 1.64
25 Franchitti 10 0 18 -2.91


As always (with IndyCar races) my numbers come from Indycar.com/stats.

So, observations on the scoring as well as general observations (this week, at least, done in award form):

- Race Score Podium 1) Castroneves 73.09; 2) Dixon 72.00; 3) Viso 68.00

- Now, that is a controversial three-spot! I don't disagree that the best and second best drives go to Helio and Dixon, but the algorithm and I might just have words about Viso edging Hinch. But, this is meant to spark thought. What do you think? Is moving up from P22 to P7 better than winning from the second row, while leading 26 laps? Let's ponder that.

- The "Flashes of Brilliance" Award for driver who impressed me, but whose Race Score fails to do so: Tristian Vautier. The IndyCar Rookie of the Year (Number 1 in a field of 1) looked VERY VERY good before retiring the car with mechanical problems. Keep your eyes on this guy. I don't think he'll finish in negative numbers very often.

- The "I Was Right" Award for thing I said that somehow wound up being true: I said Bourdais would look much better at St. Pete this year. (Not really a stretch, I know.) But, without the Lotus, and with an entire race completed, Bourdais jumped up from P21 to P11 and finished with a Race Score of 50.00.

- The "Robbed by Qualifying" Award goes to Simona de Silvestro. A good run for de Silvestro, holding on to P3 for a long period toward the end of the race before losing her tires in a nasty nasty way (trivia: What goes off faster, those Street Circuit Reds or the Pirelli "Hards" that were at Sepang for the F1 race?) at the end. Simona's finish in sixth spot is impressive, but her Race Score numbers would have been even better if she hadn't started inside the second row. Instead, she has a respectable Race Score of 34.00 and winds up just out of the points in...

The Race Score World Championship

R.S. Standings Driver Race Score RSWC Points
1 Castroneves 73.09 25
2 Dixon 72.00 18
3 Viso 68.00 15
4 Hinchcliffe 67.82 12
5 Kanaan 58.00 10
6 Andretti 54.00 8
7 Bourdais 50.00 6
8 Tagliani 46.00 4
9 Wilson 42.00 2
10 Carpenter 42.00 1


- Yes, that's right Ed Carpenter. Ed gets 1 RSWC point (Wilson wins the tiebreaker for finishing higher). I hereby go on the record saying that Carpenter finishes in top 10 in the RSWC this year.

- Remember what I said about Bourdais? I feel super-smart now.

- Tagliani picks up 4 points after not being seen much on television. Really nice livery, though! Really nice.

Well, that's that I've got for now. I might come back with an F1 Bonus Blog later this week, but don't be surprised if I don't. It's Holy Week. I'm busy. See you in Barber! (I'll surely blog before Barber, but that sounded good.)

-- Guido

Friday, March 22, 2013

The Ghost of St. Pete Past

Hello Internet! This week we look back at the IndyCar race at St. Petersburg last year, by breaking down the Race Scores from that race. We'll also look ahead to this week's season opener in St. Pete by looking at the differences between this year's race and last. Plus, I'll be unveiling the way that Scoring Indy will be quantifying season-long performance for the 2013 IndyCar season.

St. Pete 2012

I don't think I'll ever forget the picture of Helio slapping the "Dan Wheldon Way" sign. It went a long way toward healing for me. But, that's a lot of feelings, and this blog is about numbers, data, and performance. So, here is the carnival of data from last year's St. Pete Grand Prix:


Race Finish Driver Starting Position Laps Led Laps Completed Race Score
1 Castroneves 5 28 100 71.69
2 Dixon 6 37 100 74.27
3 Hunter-Reay 3 0 100 46.15
4 Hinchcliffe 4 0 100 44.23
5 Briscoe 2 9 100 41.04
6 Pagenaud 16 0 100 59.62
7 Power 1 11 100 32.42
8 Viso 12 0 100 44.23
9 Kimball 22 0 100 59.62
10 Wilson 15 0 100 42.31
11 Newgarten 19 0 100 46.15
12 Rahal 10 0 100 25.00
13 Franchitti 9 1 100 19.73
14 Andretti 7 0 99 11.67
15 Tagliani 17 0 99 26.88
16 Servia 23 0 99 34.48
17 Barichello 13 0 98 11.69
18 Carpenter 24 0 98 28.62
19 Hildebrand 18 3 96 14.98
20 Conway 11 0 75 0.48
21 Bourdais 26 0 73 18.56
22 Sato 14 11 73 2.40
23 Legge 25 0 59 9.96
24 de Silvestro 21 0 22 4.50
25 Kanaan 8 0 21 -3.02
26 Jakes 20 0 19 -0.27

Observations from last year's race:

- The Race Score Podium: 1) Dixon: 74.27; 2) Castroneves: 71.69; 3) Pagenaud 59.62

- Comments about those three drives. Dixon winds up on the first step of our podium, despite finishing last year's season opener in second. This is explained in the algorithm by his having gained the same number of positions as Helio and then leading nine more laps. Castroneves finishes just about two and a half points behind Dixon, and the discrepancy in laps led seems to be more of  a quirk of strategy than anything. Pagenaud started off his Rookie of the Year campaign by jumping up from P16 to P6, and snagging the third spot without even leading a lap.

- A quick comment about this system as a whole. Last year's St. Pete race is what made me keep at this little experiment of mine, when I was compiling all the data from last year's IndyCar series. The thought that the guy that finished second might have had a better drive than the guy who finished first and the idea that P6 actually performed better than the drivers in P3-P5 is just fascinating to me.

St. Pete 2013

Here's what will make things at this year's St. Pete GP slightly different:

- There are more laps (110 not 100), so each lap led will count for less.

- There is one less entry, so each finishing position and position gained or loss will count for more.

- There are no more Lotus engines, so a fifth of the field no longer has anchors on their cars. Expect a better day specifically for drivers whose names rhyme with Bebastien Soudais.

Race Score World Championship

So, here's what I'm going to be doing here at Scoring Indy for this year's IndyCar season. I'm going to track performance by drivers across the season with a system I'm calling the Race Score World Championship. Basically, I'm taking the top 10 Race Scores (Aggregate Race Scores for weeks with grid penalties) and awarding them Formula 1 style points. I'm doing this because I can, and also I want to do it because (looking at last year's Race Scores) I find that between 8-12 drivers have "good" races on any given week. So, look at the end of each IndyCar update for standings in the RSWC.

That's all I've got for now. Expect next week to be spotty, as it is Holy Week, and I won't have too much time for racing. Hopefully, I'll get St. Pete scores up as soon as I see a box score.

Let's go racing,

--Guido

PS The numbers in the above "carnival of data" come from indycar.com/stats

Wednesday, March 20, 2013

Bonus Blog -- F1 Australian GP

Hey there, Rockstars! The normal Blog Day (of Doom?) is Friday, but I carved out 10 seconds of free time (it's almost Holy Week, not a lot of free time for the pastor-types) to put together a little spreadsheet with the Race Scores for last weeks Formula 1 race in Melbourne.  Plus, I wanted to get some content out to all of you. Here's the festival of data:


Race Finish Driver Starting Position Laps Led Laps Completed Race Score against Start Race Score Standing
1 Raikkonen 7 27 58 86.91 1
2 Alonso 5 6 58 59.72 2
3 Vettel 1 6 58 46.08 4
4 Massa 4 3 58 45.77 5
5 Hamilton 3 4 58 39.81 6
6 Webber 2 0 58 29.55 11
7 Sutil 12 11 58 57.21 3
8 di Resta 9 0 58 36.36 7
9 Button 10 0 58 34.09 9
10 Grosjean 8 0 58 25.00 15
11 Perez 15 0 58 36.36 8
12 Vergne 13 0 58 27.27 13
13 Gutierrez 18 0 57 33.89 10
14 Bottas 16 0 57 24.92 16
15 Bianchi 19 0 57 27.12 14
16 Pic 22 0 56 29.08 12
17 Chilton 20 0 56 20.22 17
18 van der Garde 21 0 56 17.95 18
19 Riccardo 14 0 39 1.45 20
20 Rosberg 6 1 26 -7.06 22
21 Maldonado 17 0 24 0.78 21
22 Hulkenberg 11 0 0 2.27 19


Observations:

- Race Score Podium -- 1) Raikkonen: 86.91; 2) Alonso: 59.72; 3) Sutil: 57.21

- Obviously Kimi Raikkonen had the best drive, but watching the race I didn't think it would show up as THIS dominant (27.19 better than Alonso). Now, maybe that was because the Iceman's dominance made for bad TV, whereas the battles behind him were compelling.

- Very nice work from Adrian Sutil in his comeback race. From P12 on the grid to P7 in the box score thanks to some awesome tire (in F1 should I say "tyre?") management, which led to him leading the second-most laps. Put that man on the Race Score Podium!

- Finally, accolades go to Ferrari and Force India for being the only two constructors to have both drivers with an over 35 Race Score (35 is my working benchmark for "good drive"). More impressed by Force India here, since seeing cars number 15 and 16 with 35+ Race Scores is just flat out more shocking than cars number 3 and 4...

See you Friday for our regularly scheduled stuff: St. Pete IndyCar preview with last year's scores.

-- Guido

PS Data used in the spreadsheet above comes from both the official Formual 1 website as well as F1Fanatic.co.uk. 

Friday, March 15, 2013

The Method to my Madness


So, here’s me. I’m a born-and-bred Hoosier. I grew up outside Lafayette. When I was in the second grade, my class had “Indy 500 Week,” and I've been fascinated ever since. Now, I say that I was fascinated, but I grew up in a more stick-and-ball focused household, so I didn't really follow IndyCar racing until after Re-Unification. I went to IU, where I studied religious studies, and did sports radio at the student radio station, WIUX. I’m an ordained Lutheran pastor, and I live in Wisconsin.

I think that’s enough incidentals. The reason that I started this blog goes back to my radio days. I was fascinated by the different emphasis that different people would place on different statistics when trying to evaluate teams and players against each other. Think of the classical baseball example, which matters more: On Base Percentage or Runs Batted In? Does it matter more that a player gets on base, so that he can score, or that he gets the runners that are on base home? What value, in basketball do you place on a player’s defense? So, I wondered to myself: how, in racing, can you determine who had the “best” drive of any given race? So, I made a list of factors that I thought should matter:

1) Race Finish. You've had a whole race to sort yourself out. Where you end up matters in how well you drove.
2) Positions Gained. It’s easier to win from the front than the back, and a third-place podium finish from the pole (while impressive, getting to the podium is hard) is less impressive than one starting P25.
3) Leading Laps. Leading Laps matters. It means that you either have a fast race car or that you’re trying to make something happen with strategy. I say that you should get credit for that.

So, I decided to cook those numbers together. I arbitrarily determined that a win, leading every lap, from the pole was worth 100 points. 50 points are given based on finish (divided in a linear fashion). Then I cook the positions gained and lost in my mathematical black box (I’m not averse to sharing my formula, but I didn't assume that it would be interesting to all of you. If I’m wrong, I’ll put it in a future post.) I multiply that by 50 (now, this number can wind up being greater than 50 or less than zero) and then add it to the points for finish. The result is the driver’s “Race Score.” There is, however, one factor that I want to consider that I haven’t mentioned yet.

4) Finishing the Race. I don’t want to unfairly punish drivers for getting taken out of a race or having a mechanical problem. Such a problem can cause a driver to finish in a very low spot from a high starting spot on the grid. In my book, a driver who finishes P25 after starting P1, who blows a transmission on lap 5; did not have as bad of a day as a driver who falls to P25 from P1 and is still running at the end (thereby finishing on the lead or just 1 or 2 laps down.)

So, for this reason, the second term of the Race Score (the one based on gaining and losing positions and leading laps) gets multiplied by the fraction of laps completed over total laps. This way, the negative numbers don’t bite drivers so bad, when they lose major positions to mechanical or contact issues.

The final thing that I want to take into consideration in quantifying driver performance is:

5) Qualifying v. Starting Position. With the 10-grid penalties for unapproved engine changes, I think that it’s important to think about both where a driver starts the race and the spot where the car actually qualified. (Now, this doesn't matter at Indy, but the rest of the year, it’s a factor.)

So, to account for this in races where there’s a difference between starting and qualifying. I've decided that positions gained/lost should be calculated against both qualifying and starting position. Then, the two Race Scores are averaged for an Aggregate Race Score. In this way, drivers aren't overly rewarded for gaining positions, which are gained as a result of losing starting spots to an engine change penalty (The driver with the penalty, at least theoretically, based on qualifying, has a faster car than those around him/her at the start, making these positions “easier” to gain.) At the same time, credit has to be given for actually gaining the spots. I think that going half-and-half basing the score on qualifying and starting does a good job of this.

Now, is this system perfect? No, but I do like it. Ideally, I’d be able to do something with average running position, but I can’t find the statistics for that anywhere on the internet (I’d also have to completely re-work the formula, not that I’m averse to doing so.)

So, in working with what I have, this is where I've come down. So, as a teaser, here are the Top Three Race Scores from last season’s opener at St. Petersburg.

Driver Race Score Race Finish Race Start Laps Led
Dixon 74.27 2          6          37
Castroneves 71.69 1          5          28
Pagenaud            59.62 6          16          0

Coming next week, 2013 IndyCar preview including last year’s full St. Pete results.

Later,
--Guido