View Single Post
  #63   Spotlight this post!  
Unread 14-02-2012, 16:07
Ian Curtis Ian Curtis is offline
Best Available Data
FRC #1778 (Chill Out!)
Team Role: Engineer
 
Join Date: Feb 2005
Rookie Year: 2004
Location: Puget Sound
Posts: 2,521
Ian Curtis has a reputation beyond reputeIan Curtis has a reputation beyond reputeIan Curtis has a reputation beyond reputeIan Curtis has a reputation beyond reputeIan Curtis has a reputation beyond reputeIan Curtis has a reputation beyond reputeIan Curtis has a reputation beyond reputeIan Curtis has a reputation beyond reputeIan Curtis has a reputation beyond reputeIan Curtis has a reputation beyond reputeIan Curtis has a reputation beyond repute
Re: The Ultimate Game-Breaker Robot: 2012 Edition

Quote:
Originally Posted by JesseK View Post
2 high goal auton shots + 3 high goal teleop shots. Dead reckon autonomous and have a preset delay so the shots don't interfere with partners. It's very doable.
FRC games are usually easy on paper, but as Bill Parcells said, "You are what your record says you are." Not your team in particular, but FRC as a whole. FRC as a whole has a fantastic track record of inspiration, but not a great one at scoring points.

I don't have access to unpenalized scoring data from before 2010, but looking at that does not paint a great picture of high scoring robots. In 2010 the mean robot scored 1.5 points per qualifying match. I think we can agree that the objectives that year were quite simple, if the details made them more challenging (the slope before the goal, and the fact that the goal did not extend all the way to the corner of the field).

Of course I can't find my files for 2011, but I believe the unpenalized score per robot per match was something on the order of 14 points. Hard to categorize this thanks to the minibot, but on average probably less than 2 "actions" for the mean robot.

Using a chart I made for Week 1, we see the mean score for 2006 qualifying matches was somewhere around 30. Again, penalties and the ramp make it hard to judge, but assuming the effects of penalties and the ramp balance each other out (this is a SWAG), then the average robot scored 3 and change balls. I'm not sure how you want to define an "action" for 2006, but this says if you got in position and shot all 10 starting balls, you still missed a two thirds of them! On a dump, you probably got all of them.

To underline this whole discussion, take a look at Jim Zondag's OPR distribution for 2011. The distribution is right skewed, meaning the mean robot scores more robots than the median robot. The median robot is more interesting to us, as it defines the 50% of the field, and at most events the 50% percentile is what gets you playing into Saturday. If you add up the percentages in Jim's graph, the median robot in 2011 scored between 0 and 5 points after penalties! Furthermore, since OPR is calculated using the sum of the alliance scores, it is likely that the real distribution is even more skewed than the OPR distribution.



When Woodie called FIRST "the hardest fun you'll ever have," he wasn't kidding. FRC is hard, really hard.

I think I got a little off topic, but back to to your original point. I think if you get two good robots on an alliance you can score those extra points, but I don't think the odds of drawing two good robots are that great. It would be interesting to do some further analysis of what score differentials have looked like in the past...
__________________
CHILL OUT! | Aero Stability & Control Engineer
Adam Savage's Obsessions (TED Talk) (Part 2)
It is much easier to call someone else a genius than admit to yourself that you are lazy. - Dave Gingery