[CromMacBasho]

An interesting look at MotoGP that doesn't take race wins into consideration...

(Stolen from AMS) and restolen by me from STN


I was watching the MotoGP quite closely and I am kind of disappointed by
the point system. How shallow it is, how little information it gives, it
promotes 'point gains' and most over -all say that Rossi would win if
not bad luck or Yamaha.

So I pulled the PDF data from the motogp.com, massaged it with Emacs,
regexp's and wrote simple calculation on top of that in Python.

Everything to calculate the best rider based on the time not points and
give a chance those who crashed here and there. Here are rules I used:
-for each race I found the best time and speed
-for each rider I calculate the percentage of it
-if rider id make at least ten races, he is out
-finally I calculate the average
-for interested sources are here:
http://no-choice.com/motogp/motogppy.txt

As you can see it is normalized to the best rider on each race so
crashing or DNF really does not matter here.

And results. I expected Vale to win, but wait a second, he is 3rd or
even 4th .... And first is KFC with a big margin.

But the most important for me, that all of them are just GOOD. The
differences of sub-% on 1000 miles races - they just do not matter.

--------
69 Nicky HAYDEN : raced 16 times with 99.744% of max speed
26 Dani PEDROSA : raced 15 times with 99.635% of max speed
46 Valentino ROSSI: raced 14 times with 99.587% of max speed
27 Casey STONER : raced 10 times with 99.587% of max speed
33 Marco MELANDRI : raced 16 times with 99.565% of max speed
65 Loris CAPIROSSI: raced 16 times with 99.328% of max speed
15 Sete GIBERNAU : raced 10 times with 99.260% of max speed
10 Kenny ROBERTS : raced 15 times with 99.220% of max speed
24 Toni ELIAS : raced 13 times with 99.186% of max speed
56 Shinya NAKANO : raced 11 times with 99.127% of max speed
5 Colin EDWARDS : raced 16 times with 99.067% of max speed
71 Chris VERMEULEN: raced 14 times with 98.865% of max speed
21 John HOPKINS : raced 16 times with 98.817% of max speed
6 Makoto TAMADA : raced 16 times with 98.629% of max speed
7 Carlos CHECA : raced 16 times with 98.481% of max speed
17 Randy DE : raced 10 times with 98.239% of max speed
30 Jose Luis : raced 10 times with 97.841% of max speed
77 James ELLISON : raced 16 times with 97.431% of max speed
66 Alex HOFMANN : raced 10 times with 97.118% of max speed
--------
69 Nicky HAYDEN : raced 16 times with 100.258% of best time
26 Dani PEDROSA : raced 15 times with 100.368% of best time
27 Casey STONER : raced 10 times with 100.415% of best time
46 Valentino ROSSI: raced 14 times with 100.419% of best time
33 Marco MELANDRI : raced 16 times with 100.438% of best time
65 Loris CAPIROSSI: raced 16 times with 100.683% of best time
15 Sete GIBERNAU : raced 10 times with 100.750% of best time
10 Kenny ROBERTS : raced 15 times with 100.789% of best time
24 Toni ELIAS : raced 13 times with 100.823% of best time
56 Shinya NAKANO : raced 11 times with 100.882% of best time
5 Colin EDWARDS : raced 16 times with 100.944% of best time
71 Chris VERMEULEN: raced 14 times with 101.149% of best time
21 John HOPKINS : raced 16 times with 101.205% of best time
6 Makoto TAMADA : raced 16 times with 101.394% of best time
7 Carlos CHECA : raced 16 times with 101.544% of best time
17 Randy DE : raced 10 times with 101.796% of best time
30 Jose Luis : raced 10 times with 102.230% of best time
77 James ELLISON : raced 16 times with 102.641% of best time
66 Alex HOFMANN : raced 10 times with 102.970% of best time




--------
1 track: GRAN PREMIO DE ESPA

 

[derby]

i'd be curious how this calculations came out for previous seasons.

 

[spidey]

Wait a sec now, buckeroo . . . while I can do all kinds of algebur and multiplicimication and stuff, that post confused the hell out of me. someone 'splain to me, mang?

 

[gm2]

i realize that in this system you can't get a good score unless you raced well... but are time and speed really the only measurements? traffic and track layout play a big part in both. sometimes the #2 and #3 riders hold eachother up, then one of them ends up winning.

this is totally oversimplifying it but if you find a way to make crashing statistically irrelevant you've just obviated the need to race.

still, interesting. bravo hayden.

 

[CromMacBasho]

I agree with you gm2. It makes MotoGP a TT with no crash penalty. It does, however show that Hayden is able to run as fast or faster than anyone on the grid. A lot of people say he doesn't deserve the championship. Stats say he does, I think.

As the author pointed out though, all of them are GOOD. Really good.

This info makes me glad Stoner is on a Duc next season. I got my USA fix this year. Next year I hope it's all Ducati.

 

[vindingo]

When you say max speed, is that top speed or max avg. speed for lap? Max top speed would be irrelivant. The way you set it up is kind of like how they do time trials in SCCA. They do it because it is safer to race against the clock, but as stated previously there are many other that make "racing" racing. This does prove that Nicky was consistently good for all the haters out there.

(thread jack) who is this Jose Luis Cardoso? he either dnf or is in last place every single race in the official standings. Why does he crash so much, or is it bike failure? (thread jack)

 

[gm2]

d'Antin Ducati rider. he and james ellison were battling for last place all year.

 

[derby]

When you say max speed, is that top speed or max avg. speed for lap? Max top speed would be irrelivant. The way you set it up is kind of like how they do time trials in SCCA. They do it because it is safer to race against the clock, but as stated previously there are many other that make "racing" racing. This does prove that Nicky was consistently good for all the haters out there.

(thread jack) who is this Jose Luis Cardoso? he either dnf or is in last place every single race in the official standings. Why does he crash so much, or is it bike failure? (thread jack)

it's not max speed. it's the season average for each riders' percentage of each race's fastest lap.

 

[test-reg]

Lies, damn lies and statistics.

I have to be honest and say I don't understand it either but the fact that Dani is up there with Nicky and he finished 5th makes very little sense to me compared to Nicky's 1st in the Championship. Try calculating for last year and the year before, if Rossi is still 3rd or 4th then we know that these stats, however interesting, have at best a passing correlation to the actual results.

 

[pennyrobber]

There are always ways to tweak rules to make results change. I was actually having a similar discussion with the misses the other day after the GP race. We were talking about how the points system could be waited differently. The main thing to remember is that each rider knows exactly what they need to do when they go out racing so other measure are really irrelevant. Take drag racing for instance. You can actually cover the quarter mile faster than your competitor and still lose the race if he gets a better jump off the line. You can argue that the faster car should win but that isn't how the rules are set up. Similar thing applies here.

 

[derby]

how would you prefer to see them weighted? and why?

here is are the current points for AMA and FIM championships:

POS	AMA	FIM (motogp & worldsbk)
1	36	25
2	32	20
3	29	16
4	27	13
5	26	11
6	25	10
7	24	9
8	23	8
9	22	7
10	21	6
11	20	5
12	19	4
13	18	3
14	17	2
15	16   1
16	15
17	14
18	13
19	12
20	11
21	10
22	9
23	8
24	7
25	6
26	5
27	4
28	3
29	2
30	1

note: AMA gives a 1pt bonus for both pole and "most laps lead"

as you can see, the AMA system rewards consistancy and the FIM system rewards wins.

hell, formula 1 only "pays" 8 positions:

POS	FIA
1	10
2	8
3	6
4	5
5	4
6	3
7	2
8	1

 

[ducpainter]

Jeez, now my head hurts.

 

[test-reg]

FIM scoring goes down to 15th place with one point, just ask Gary McCoy.

I don't have a problem with scoring by the way, not in MGP, F1 is another story. :

14th = 2 points
15th = 1 point

 

[derby]

FIM scoring goes down to 15th place with one point, just ask Gary McCoy.

..

14th = 2 points
15th = 1 point

yup, i fudged my table... i'll correct it.

I don't have a problem with scoring by the way, not in MGP, F1 is another story. :

what is your issue with f1 points? not enough spread between p1-3? they recently changed f1 points, including paying two more positions. they used to only award points down to p6.

 

[test-reg]

Yeah, that's pretty much it, I guess it comes down to the same reasoning, a win should be rewarded more than it currently is IMO. I've seen interesting comparisons between the old and new scoring systems in F1, it doesn't always turn out the way I'd expect. I've been following F1 since before the points change, I think I preferred the old system but I like that they score down to 8th, so, I'm not sure how they'd resolve that.

edit: I think the worst is the AMA, you end up with someone winning all the races except a couple of DNFs and they're just clinging to the points lead. I can certainly see where people are coming from with regards to rewarding consistency but I personally feel that a winning rider on a defective bike (if that's the cause) shouldn't be penalized as much for a mechanical failure, there should be some way to transfer that to constructors. The constructors points table should perhaps be much more closely dispersed in terms of the differences in points and the riders should be more widely dispersed. The laps lead and qualifying points, while they can give a better idea of performance over a race weekend, just end up complicating matters IMO. Whether someone led the most laps or qualified first really shouldn't be a defining factor in a championship which obviously promotes winning. Anyway... </rant>