Sell on Monday, Win on Sunday [Finished]

[Der Bayer]

Race 1.1 Results

Here are the results of the first race: https://youtu.be/JXLuG1ELiHc
The reverse grid race on the Automation Test Track will be released tomorrow.

[strop]

That was very middle of the pack for me, especially that getting passed left and right at the start xD Very much feeling the lack of POWAAAAAA.

Let's just hope that some of the tracks are very technical and use lots of chained curves with harsh elevation changes and not many straights

[RobtheFiend]

Holy ****, the fuel consumption down Daffy Flyer and Sonory Roar for Der Bayer and biafo131. I think an afterburner would use less fuel, and be more efficient.

[Packbat]

I did not predict that nearly as well as I thought I would.

Also, I feel like I'm losing positions because I don't have the top-speed of my immediate competition.

[torts]

Third place! Can't believe it. The race car was actually good!
And yes, the reason for such ludicrous fuel consumption figures is very much a biiiiiiig, biiiiiig turbo with lots of BOOOOOST!

[NormanVauxhall]

Znopresk.
Sell on monday, sell on sunday. (no win involved in our sales strategy)

[Packbat]

Mm, going back to my predictions:

• To overperform: #1, #2, #6, #14, #18, #20, #22, #23
• To underperform: #3, #4, #5, #9, #10, #13, #15, #21

Ignoring drops, my measurements based on qualifying and their finishes in the first heat say we should see:

  Code:

 #2: 1st place.
#23: 2nd or better.
#20: 3rd or better.
#22: 4th or better.
#14: 5th or better.
 #1: 7th or better.
#18: 8th or better.
 #6: 10th or better.
 #4: 13th or worse.
#15: 14th or worse.
#21: 17th or worse.
#13: 18th or worse.
#10: 19th or worse.
 #5: 20th or worse.
 #9: 21st or worse.
 #3: Last place.

...which, well, basically translates to:

• #11 - me, Packbat - should get freight-trained even worse in the reverse start, and finish behind:
• #1 - pHanta - and #18 - TrackpadUser - and possibly also #6 - Gooles.
• #12 - Pyrlix - and possibly #17 - Chipskate - might pick up a couple positions from:
• #4 - NiuYorqCiti - and #15 - 8bs.
• #16 - titleguy1 - may (although looking at his tyre wear, probably not) pick up a couple positions from:
• #9 - NormanVauxhall - #5 - VicVictory - and #10 - Lordred.

Going back to my original data, #20 - biafo131 - and #22 - Leonardo9613 - might actually outpace #23 - Der Bayer: Der Bayer's car is toward the bottom end of my straight-line speedster list and the former two cars are near the top. Likewise, #5 - VicVictory - has a reasonably good chance to outperform #10 - Lordred - as his is one of the fastest in the slow group and Lordred's is one of the slowest.

I guess we'll see if any of these predictions bear out. Yay science!

[strop]

If more than 60% of your predictions come true, you should, if you are not already, become a meteorologist

[Packbat]

Heh - I'll take that under advisement.

Honestly, thinking it through, I'm betting there will be a lot more luck than I allowed for, here. So I'm going to be a bad scientist and decide how to analyze the data after the race.

[Der Bayer]

Race 2 Results

Here are the results of the reverse grid race at the Automation Test Track: https://youtu.be/XAdzK0EXyoU

[strop]

Well, that was very consistent on my part, at least. I'm too tired to look through the full list to compare your predictions, Packbat, but there's one thing I just noticed...

Monza and Laguna Seca are on the track list.

I'm so screwed.

[Packbat]

Aaaagh, Gooles, I almost had you! I was faster all up the Sonory Roar, but just couldn't make the pass coming into the Bavarian Bend.

And yes, prediction time. <Angle brackets> for failed predictions, {curly braces} for successful, [square brackets] for ambiguous.

  Code:

 #2: 1st place. <3rd place>
#23: 2nd or better. <4th place>
#20: 3rd or better. <6th place>
#22: 4th or better. {2nd place}
#14: 5th or better. <9th place>
 #1: 7th or better. {1st place}
#18: 8th or better. [21st place - 79.94% damage]
 #6: 10th or better. {7th place}
 #4: 13th or worse. [dropped, engine failure]
#15: 14th or worse. {15th place}
#21: 17th or worse. {16th place; behind #4's Heat 1 position, but was ahead of #4 at time of engine failure, but #4 had damage at that point; but also behind #18's Heat 1 position}
#13: 18th or worse. <13th place>
#10: 19th or worse. [17th place; same comment as #21]
 #5: 20th or worse. [18th place; same comment as #21]
 #9: 21st or worse. {20th place; same comment as #21}
 #3: Last place. {22nd place; was behind #4 at time of engine failure}

And also:

• #11 finished behind #1 and #6, but ahead of #18 (who had damage).
• #12 and #17 both finished ahead of #4 and #15 (and were both ahead of #4 at time of engine failure).
• #16 finished ahead of #9, #5, and #10.
• #22 finished ahead of #23, but #20 was behind.
• #5 finished behind #10.

(Note: feel free to argue with any or all of these scorings. As I said, I am being bad scientist.)

I count seven wrong predictions, eighteen? correct predictions, and fourish ambiguous.

...so apparently I should be a meteorologist?? ?

I feel as though the theory wasn't well borne out, but wasn't refuted either. I suspect there were several cars that got held up in the first heat and were clear to run in the second, which would explain some of the anomalies - if you look at places where the fastest lap times were out of order, you find:

• #20 - biafo131 - was slower than the four next cars in race, and finished behind two of them;
• #1 - pHanta - was faster than the four cars prior, and finished ahead of all of them;
• #8 - strop - was faster than #6 - Gooles - but still finished behind;
• #4 - NiuYorqCiti - was faster than #7 - HighOctaneLove - but had bad accidents and an engine failure;
• #12 - Pyrlix - and #17 - Chipskate - were both faster than #15 - 8bs - and both finished substantially ahead;
• #13 - Razyx - was faster than #21 - WoarLord - and finished several places ahead;
• #9 - NormanVauxhall - was faster than #5 - VicVictory - but finished behind with severe damage; and
• #16 - titleguy1 - was faster than #9 - NormanVauxhall - and finished way up the field.

...so "compare fastest lap times" seems to also have a 2/3rds success rate.

I don't know. I'll let you know when I figure anything out.

Edit: I guess TL;DR of the above is, "straight line speed seems to help, and keep your eye on cars that finished below their fastest-lap position."

[strop]

Looking at the trends in your data, I'd say you've done quite well. Thus I agree with your overall hypothesis that straight line speed seems to help, though of course this is also something that I had informally concluded earlier, thus subject to confirmation bias, and my reading of this is strictly a post hoc analysis, thus I am also being a terrible scientist At least here, this is good enough for me to continue looking to your analysis to help me improve my BROBOT performance. It has really created quite a different playing field, and some have taken to it better than others!

[Riso]

titleguy1 totally ruined my race with his terrible driver deciding to play bumper cars!!1

[RobtheFiend]

What i found most telling; the guy making the software was caught out in a random event(driver error) and lost about 7 seconds.

No hidden code to make his cars overperform.