1 (edited by m610 2012-04-09 02:44 PM)

Topic: Pointless Race Analysis

So what does a more or less retired engineer/scientist do after the race is over? Build a massive spreadsheet and analyze the data to death.


So here are some of the results from the recent Pointless race:


First, I start with us.


Tinyvette Results
Laps Completed:     220
That sore of sounds like a lot, but the winner completed 281 laps. Normally, or rather recently, the winner is only about 30 laps ahead of us at the end of the race.


Miles Raced:     523.6 miles
The track distance was not include in the MyLaps or Specialty Timing reports so I just added 0.1 mile to the distance given for last fall's the Skankaway race.


Overall Position: 73, 42.69%
This sure isn't in the range we had been doing lately, but at least it is not our worse finish. Hey, it's a silver lining, even if it is tarnished.


Class: B
Cars in Class: 57
Position in Class: 19, 33.33%


Number of Cars that Passed Us: 2001, or 9.10 per lap
It felt more like 30-40 cars per lap passing us. To get this number I just added up the number of laps more cars finishing ahead of us completed.


Number of Cars We Passed: 7085, or 32.20 per lap
I like that number. Unfortunately, most of our passing was of cars that were in the pits or on the trailer. To get this number I just added up the number of laps more cars finishing behind of us completed. This does not include cars with laps less than 0 laps. Maybe I should redo this one using results from cars that finished the race, but then it would probably drop to around 0.5 per lap.


General Analysis
Cars in the Race: 171
This includes cars that never completed one lap.


Hours Racing: 14.41
This was averaged from the total time raced by the top 5 cars.


Best Lap Time: 123.137 seconds, 2.03.137
Eyesore. Who did you expect? wink


Worse Lap Time: 206.748 seconds, 3.26.748
Pink Lloyd, and this probably included a tow, or push, or coasting half a lap. The back half of Infineon is slightly down hill. I can't wait to see the Lloyd running after its teething problems are taken care of.


Average Best Lap Time: 138.885 seconds, 2.18.885
We were slower than this, again, damn it.


Best Lap Speed: 69.58 mph (Eyesore)


Slowest Best Lap Speed: 41.44 mph (Lloyd)


Average Lap Speed: 62.07 mph


Attrition
Cars Taking the Checkered: 118
This was estimated based on total race time plus time for two more laps.


Cars Not Taking the Checkered: 53
That's a failure rate of 31%


Distance
Total Laps Raced: 32536
I just added up everyone's laps.


Average Laps Raced: 194.83
At least we were above average in this category.


Total Miles Raced: 77435.68
Holy crap!


Heap Profiles
Oldest Car: 1958, 54 years old, the Lloyd.


Newest Car: 2004, only 8 years old


Typical Age of Car: 1985 at  26.59 years of age
It's time to start shopping for the next model year LeMon, 1986!


Number of cars over 40 years old: 8
The Tinyvette is in his category.


Number of 40 or older C Class Cars: 4
It seems like all 40 year old cars would be in class C, just because.


Class Results
Class    Number of Cars    Best Lap Time    Average Lap Time    Laps Completed        Average Laps Completed        Penalty Laps
X        4                123.137            147.54                803                    200.75                        2
A        88                124.962            133.45                17676                200.86                        227
B        57                128.103            136.44                10571                185.46                        32, plus 2.5 billion
C        22                138.683            152.74                3486                158.45                        2
Total:    171                 -                 -                    32536                190.27                        263
(The tabs/columns looked OK when composing this, but get screwed up in preview. Sorry.)


Interpretation
Distance traveled: 3.11 times the circumference of the Earth and nearly 1/3 the way to the moon. Not bad. But only 0.08% of the way to the sun, but, 10 billionths of a light year.


Estimated mpg: 12.41 mpg
This estimate is based on a 5 gallon/hour usage and average lap times.


Estimate Fuel Usage: 6237.76 gallons
About a tanker truck full, but not a small truck.


Estimate Carbon Footprint: 61.13 tons
I know you really care about this one, but it is still a fun number. 60 tons is equivalent to about 1/3 the weight of all of the cars in this race, assuming 2300 pounds per car.



More to come.

Re: Pointless Race Analysis

Very interesting. Feels good to see that we were above avg on a lot of those data points, older then average as well @ 34 yrs old. Just bellow the avg best lap time.. ours was 2:20.38 which still impressed the hell out of me.

We are improving overall and having fun.. awesome.

5x loser - and just all around not very good
My Crappy Blog
Former Sex Pistons Craptain
And it feels so funny, like bathing in honey...

Re: Pointless Race Analysis

Introducing the IoDSR
I like to try to mine the numbers to see if I can discover some new relationship between winning or just finishing and other factors, or just find numbers that back up the popular wisdom. In an earlier post I introduced the RTOT, relative time on track, and it correlated well with finishing position. It also showed how the field fell into four groups, the leaders, the determined, the ever hopeful, and the seriously struggling. But more on that later, because I now have the IoDSR.


IoDSR = Index of Doing Something Right. It's mathematically very simple. Rank each car by best lap time then divide that by the team's finishing position. Ideally everyone would get a 100%, the fastest car finishing first, the slowest last, and so on. And yes, Eyesore turned the fastest lap (2:03) and took first place, grabbing an IoDSR of 100%. And sure enough, the slowest car, the Lloyd, with a best lap of 3:26, 169th, finished 167th, scoring an IoDSR of 101%. (I love that car. I'm not picking on you guys. I swear.)


Here's the chart.
http://a3.sphotos.ak.fbcdn.net/hphotos-ak-ash4/475379_428169943863835_192201637460668_1887623_264551350_o.jpg


And here is what I see in that chart. The peak is at 100%, meaning that a relatively high proportion of cars were scoring close to the 100% mark, more of less meeting expectations. But also note that the groups that fall below this are also pretty strong. In fact, 43% of the field failed to meet expectations.


Just past 100% the distribution drops off quickly, tailing out to around 400% before it dissipates into groups of one. The Tinyvette comes in at around 180%, not bad, but we also had a bad weekend. If we had finished in the 20-30th range as we have been recently we'd have a score of around 400-500%, territory occupied in this race by B210 Racing, the Hasselhoffs, Bernal Dads, and Eco Challenged, while up around 700-800% is the Oldsmobile and Clueless Viking territory. We generally do a lot better than our lap times would indicate.  Maybe next time.


Then there is that one datum out there at 1460%. !!!! It was the Crown Vic (Sierra Auto Recycling) that finished in fifth place with a best lap time of 2:14, a lap time ranking of 73rd. I'm guessing astronaut diapers and mid-air refueling. However they did it, great job!

Re: Pointless Race Analysis

That's awesome data.

I need to do some analysis of our data as well, as the one thing I discovered coming in 5th in B class (and 34th overall): the competition is stiff up front.

It takes a sorted car, good driving, minimal time off track, etc. That damn 969 Crown Vic seemed impossible to catch, even though our best lap was over 5 seconds better!

5 (edited by m610 2012-04-09 10:50 PM)

Re: Pointless Race Analysis

We found that out too. Once you get into the top 30 small mistakes can cost you a bunch of positions. A fast pit stop can set you back 10 positions!


I'm going to analyze the time slip numbers for that Crown Vic, along with ours, to see what they were doing. I've done that in the past for other cars to see if there was anything we could do to close the gap on them. That led us to try three drivers per day at BW, which would get us 11 laps over the weekend, plus we found nearly a minute we could cut off our pit stops. In the end, mistakes and bad luck on track killed up, but we still got 25th overall.


We don't do the 3-driver rotation anymore. We'd rather that everyone get to drive. Chasing a win is one thing, but having fun is more important. That said, we define having fun as doing everything well. If we do that then a good finish is a bonus.

Re: Pointless Race Analysis

I hate to bug you, but hey, you already did the work...

Where did the Billy Beer Car #80 end up in the IoDTR? We seemed to have our shit together in the pits, but still have really slow fuel jugs meaning we can't get in and out in less than 10 minutes with fueling. Just wondering how many positions we can get back in the pits if we can half that. I know if we can cut 3 minutes out of each one that would be about 8 extra laps at Sears and we'd be pretty close to the Sex Pistons.

Constructor/Owner/Driver - Billy Beer Ford Futura

7 (edited by m610 2012-04-09 11:57 PM)

Re: Pointless Race Analysis

I'll see if I can clean up the spreadsheet enough that others can figure it out, then I'll make it available for downloads.


Billy Beer, 153rd fastest with a 2:28, 63rd overall, IoDSR= 248%


We were 134th fastest with a 2:22, finished 73rd, IoDSR = 183%


But I think the most impressive statistic is that combined we raced enough miles to circle the Earth over 3 times.


Edit: Here's a printout. It's a PDF file so you can zoom in to see the small text.

Link: http://www.kstreetstudio.com/files/tiny … /IoDSR.pdf

Re: Pointless Race Analysis

Cool data! I'm probably missing something, but can you calculate the lap distance by the best lap mph and best lap time? If you do that, it's pretty close to your .1 mile adjustment for 7a.  We were around 272% IoDSR, but in our case we were probably doing something wrong as our best lap time probably could have been better.  Still, the number is a reflection of us staying out on the track.


I'm actually loading all of the lap data for all teams.  I wanted to see how many positions we lost each time we came off the track, how much time we spent off the track compared to others, etc.  I'm about a third of the way through. So far there are no surprises.  By far, The Most Interesting Car in the World had the least amount of pit lap time (28 minutes).  Only two in the 40-50 minute range and 4 in the 50-60 minute range.  I don't get a couple of things about the timecards, though.  I think each team has one lap number listed twice.  I assume this is the last lap on Saturday and the first lap on Sunday.  I'm not sure how those laps are treated for scoring.  The other thing is that you can't just add up the total time and total laps to come up with the end of race results.  Now that I'm thinking about it, it is probably due to the Sunday re-grid.  If the #1 and #2 cars are on the same lap at the end of the day Saturday, whatever margin there was gets erased by the Sunday start.

Re: Pointless Race Analysis

The data from Specialty Timing did not include distance or best lap speeds. They'd need to know the distance to be able to get the speed.

I'm doing the same for our team and a few others, just to see how all this shakes out. It is well worth doing at least once. You get to see what matters most.

I just finished Eyesore's data. Their actual RTOT is 94%, and they run a really long first stint, almost 5 hours on Saturday and about 3:20 on Sunday, only 5 pit stops all weekend.  Their first pit stop was fairly long, for them, but normally they were in the 9-10 minute range. They spent a total of 50 minutes in the pits, which for them would be equal to 20 of their average laps.

I should really write an app for this.

10 (edited by ppressle 2012-04-10 10:43 AM)

Re: Pointless Race Analysis

Shhhh!   Staying out of the pits is our secret weapon. 

On Saturday we happened to pit when the black flag came out, so that pit stop may be hard to deduce from the timing data.

Very interesting data though.  Thanks for sharing it.

Have you considered doing the same chart against the average of each teams best 10 or n laps?  Getting a clear lap is such a sporadic thing in Lemons.

Pete

11 (edited by blinded 2012-04-10 11:11 AM)

Re: Pointless Race Analysis

m610 wrote:

The data from Specialty Timing did not include distance or best lap speeds...


Huh, that's weird.  I have 69.58 mph as a best for Eyesore, but now I have no idea where I got it.  I must have grabbed it from the wrong race or something because I sure don't see it now.

EDIT: Lol, it was from your original post, which I'm guessing is based on your estimated distance and not from the timecards:

m610 wrote:

Best Lap Speed: 69.58 mph (Eyesore)

Re: Pointless Race Analysis

We came in at exactly 100%.  Coincidentally the car broke on the lap following our fastest lap and went straight into the pits.
Anyway, glad to fatten the middle of your standard distribution.

13X losers (or is 14 now?) refusing to learn from our failures.
Organizer's Choice!  Trophy should have a bottle opener on it.

13 (edited by Lemons 300ZX 2012-04-10 02:11 PM)

Re: Pointless Race Analysis

Wow this is great information see the things Opel owners are able to come up with. big_smile We are genius in our own world most of the time but you seem to be an actual genius. HAHA

Re: Pointless Race Analysis

Seriously cool data there Mike!

Somehow we got a 177.27% IoDSR, in spite of some repair time in the pits.

I'll be studying this for quite a while trying come up with some kind of secret weapon...

"Real ZomBees prefer Bacon"
IOE(x2) MGB/SAAB 96, Judge's Choice, Class C Win, & 2011 Hooniverse Car of the Year!
MRolla, Stick Figure/Animal House, Free Range MR2, SAAB Sonett, "The Death Flip"
2008 Exoskeleton Jag Fiasco, Concours d Lemons - Rue Britannia, worse British car.

Re: Pointless Race Analysis

Very cool data, thanks for putting it together.  I feel okay about our 211%, and I know our 45 minute repair on day one hurt us pretty bad. 

-Matt

30 Time Loser with Class C, Index of Effluency, I got Screwed and Heroic Fix Trophies who usually races, according to Car and Driver, The Greatest Road Racing Cherokee in History.

Check out our FB Page and follow our various exploits.

Re: Pointless Race Analysis

I'm still working on this analysis and may have a few revisions, but I think they will be minor. All will be revealed and explained in a paper we are submitting to the Journal d'Lemons. wink

Let me clean up the spreadsheet a bit and I will post it so everyone can see their numbers or do whatever they want with them.

Re: Pointless Race Analysis

m610 wrote:

I'm still working on this analysis and may have a few revisions, but I think they will be minor. All will be revealed and explained in a paper we are submitting to the Journal d'Lemons. wink

Let me clean up the spreadsheet a bit and I will post it so everyone can see their numbers or do whatever they want with them.

Just out of curiousity, how did you extract the data from the pdf?

13X losers (or is 14 now?) refusing to learn from our failures.
Organizer's Choice!  Trophy should have a bottle opener on it.

Re: Pointless Race Analysis

Copy/paste.

For the hourly time sheets I just CTRL-A to select all, the CTRL-C to copy, and in the spreadsheet, CTRL-V to paste. I think I recall that it pastes as a single line of numbers and it is fairly easy to get all that moved around. Then the fun of parsing everything begins.

For the time card PDF file I just copy/paste the columns I need. These do come in as a single column of numbers, so just grab the different sections and move of copy/paste to get them into three columns. Two complications, one lap will be listed twice. That is at the end of Saturday/start of Sunday. Also, one entry in the third column will be blank, the one where your car turned its fastest lap. Just find that then shift everything down one row, or delete the whole column because it is pretty trivial. You can recalculate it.

19 (edited by m610 2012-04-13 08:59 AM)

Re: Pointless Race Analysis

I've decided that the spreadsheet will never be done/perfect, and even if it was, it might not be as easy for others to use as it is for me. So I went ahead and uploaded what I have now.

You can get my spreadsheet here: http://www.kstreetstudio.com/files/tiny … tless2012/

I use Quattro Pro and that file is there, but I also exported it to the Excel format for the rest of the world. Try both to see which works better for you. I checked the Excel version in OpenOffice and it imports OK, the numbers are there but the charts and some formatting are not right. But you are going to make your own charts anyway, right?

The raw and parsed data that were imported are in the last couple of pages, the EOR_Import (end of race) and MyLaps_Import. The "Whole Field" page contains the hourly data for the whole race.

Enjoy.

20 (edited by m610 2012-04-13 01:02 PM)

Re: Pointless Race Analysis

A few more graphs from our FB page.


The judged do a pretty good job sorting the cars by class. Class C is grouped at the end and while A and B overlap a lot, there are no group B cars in the top 10 or so. The A/B overlap is pretty broad, but that might just be to motivate a couple of B class teams to beat the A class teams.
http://a5.sphotos.ak.fbcdn.net/hphotos-ak-snc7/461241_431149843565845_192201637460668_1899625_1066621348_o.jpg


Here's our weekend, plotted along with our friends and Eyesore. Unscheduled trip to the pits will kill you.
http://a5.sphotos.ak.fbcdn.net/hphotos-ak-ash4/461006_431018796912283_192201637460668_1899336_2076716335_o.jpg


Here's the IoDSR by class and finishing position. Slowish cars that do well can easily score a high IoDSR. Ratios are like that. Also note how every C class cars meet or exceed expectations (IoDSR ≥ 100%). Then again, expectations for them were never that high to begin with, and all it takes is a few A and B class cars to drop out or do poorly to promote the whole field of C's.
http://a8.sphotos.ak.fbcdn.net/hphotos-ak-ash4/472262_431135960233900_192201637460668_1899558_1278501626_o.jpg

Re: Pointless Race Analysis

Lets see if I can get this graphic up right...
our Head of Rooster Juice Racing Dynamic Data Analysis publishes a piece of data that we like to review
Its the laps "behind the leader" chart whereby he charts how many laps we trail the race leader.
So it nearly always trends up, except in the case where the leader is in the pits and we're on track.
Basically the goal is to lose as little ground (keep it flat) as possible - we chart where our pits stops are and you can see clearly the number of laps lost due to each broken car part, driver with weak bladder, fogged up windshield, or black flag

lapsbehind

Paluck
Rooster Juice Racing
Bay Area, CA

Re: Pointless Race Analysis

It looks like Brook was keeping pace with Eyesore on Sunday.

Re: Pointless Race Analysis

yeah - he's a good driver, but even our quick drivers (and certainly not the car) can't honestly keep up with Eyesore, so I'm guessing there must have been a quick eyesore pitstop or two in there and probably a fair amount of yellow flag - keeping the field even

Paluck
Rooster Juice Racing
Bay Area, CA

Re: Pointless Race Analysis

Here's my first stab.  I think it actually captures how we did fairly well.  FYI, this does not account for BS or penalty laps.  Any suggestions on other data points that a 2X loser should be looking at?

http://img220.imageshack.us/img220/2778/sp2012results.jpg

Re: Pointless Race Analysis

It looks like you recovered nicely from a start far back in the field when the green flag dropped. Then there was some trouble at around lap 53 that set you back, after which you slowly worked your way back through the field.

Lap times were better on Sunday than Saturday, which is no doubt true for every team, but then your started dropping off in the end. Mechanical? Lucky to even finish?

It's tough to recover from long periods of time off the track. I'm still looking at our numbers and everything points to the importance of staying out on track, unless you are in the top 10, then staying out on track AND quick laps are important.