View Single Post
  #224   Spotlight this post!  
Unread 02-07-2012, 09:22
Jared Russell's Avatar
Jared Russell Jared Russell is offline
Taking a year (mostly) off
FRC #0254 (The Cheesy Poofs), FRC #0341 (Miss Daisy)
Team Role: Engineer
 
Join Date: Nov 2002
Rookie Year: 2001
Location: San Francisco, CA
Posts: 3,082
Jared Russell has a reputation beyond reputeJared Russell has a reputation beyond reputeJared Russell has a reputation beyond reputeJared Russell has a reputation beyond reputeJared Russell has a reputation beyond reputeJared Russell has a reputation beyond reputeJared Russell has a reputation beyond reputeJared Russell has a reputation beyond reputeJared Russell has a reputation beyond reputeJared Russell has a reputation beyond reputeJared Russell has a reputation beyond repute
Re: 2012 IRI Predictions

Here's a thought...

Getting into IRI is largely (but not entirely) due to on-field results. Since triple balancing is a significant factor in this game, it stands to reason that robots that have triple balanced reliably in competition had an advantage when it came to getting invited.

However, no OPR stats reflect the ability to triple balance (since there is no 40 point bonus in qualification rounds, and in elims there is no such thing as OPR).

In light of this, I tested the hypothesis that, on average, the long robots invited to IRI have a higher OPR than the non-longs. (I used the invitation list since I'm not clear whether the "final" acceptance list has been posted)

What I found (based on the post-Championship Ed Law scouting database; all OPR numbers are adjusted for the co-op bridge):

Best event OPR:
Average for longs: 28.5
Average for non-longs: 26.7

Latest event OPR:
Average for longs: 25.7
Average for non-longs: 24.0

"World rank" weighted average OPR:
Average for longs: 23.4
Average for non-longs: 21.3

Hypothesis confirmed. On average, long bots at IRI are 1.7 - 2.2 OPR points better than wides. An alliance with 3 longs has a projected total score anywhere from 5.2 to 6.5 points higher than an alliance with 3 non-longs (depending on your preferred metric).

Between the on-paper advantage of ~2 more balls in the high goal per match for an all-long alliance, and the fact that triple balancing necessarily means that everyone must stop shooting early and carries no guarantee of success, maybe things aren't so bad for longs after all

Last edited by Jared Russell : 02-07-2012 at 09:40. Reason: Had 2481 listed as a long, incorrectly - fixing calculations
Reply With Quote