OBJECTIVE:
Players compete to correctly predict which teams will be invited to the 2019 Indiana Robotics Invitational.
SUBMISSION:
Indicate which teams you expect to be accepted to IRI using this Google form prior to 11:59:59 PM EDT on May 20th : https://forms.gle/49EVURPgqYeDrPEe9
Submissions after this deadline will not be scored.
Submissions that do not include Chief Delphi usernames will not be scored.
Only one submission (the most recent) will be scored for each participant.
SCORING:
All players will be ranked by their submissions. The first sort is number of correctly predicted acceptances and rejections, from greatest to fewest. Ties will be broken using the following criteria:
Number of correct predictions on the team (or teams, if there are multiple with the same) with the lowest correct prediction rate across all submissions.
Absolute value of the difference between predicted number of teams invited and actual number of teams invited (the closer you are to predicting how many teams are accepted, the better)
Disclaimer: my decision regarding the interpretation of any of these rules is final. I reserve the right to modify the rules and scoring criteria at any time.
Here are my lazily done predictions, methodology was as follows:
Explicitly invited if applied teams:
Last year winners: 2056, 195, 2168, 340
Hosts: 1024, 234, 45
Fill up to 70 more teams based on Max OPR from any event in the season
The cut off ended up being 31.04, with Team 2590 being the first team not invited
By sorting the IRI applicant using Caleb’s winning margin Elo, I got 47 applicants with Elo>1700. [Note: my team is below this somewhat reasonable cutoff Elo. We did not apply for IRI.]
Among Elo ranks 48-90, I find twelve more teams that clearly deserve an IRI invitation.
That leaves another dozen invites to complete the field, and I can only find three that don’t require a hard choice. So there are 8 to 10 invitees to select from group of 48, with no clear metric. I would need to video scout a lot of matches, and even if I took the time to do that, my own recognition bias would probably have more influence than objective factors.
So I leave the selection to the IRI committee. Whatever they decide to do, the event will be well worth the trip to see. I am also looking forward to corn-on-the-cob.
For me, the interesting thing will be to watch top level teams play defense. I firmly believe that Deep Space has to be played 2-Offense, 1-Defense at the highest level. So if the IRI selection committee only picks the best offensive bots for inclusion, many of them will be pressed into service on defense. Depending on the robot design, that could be disappointing, but with the calibre of drivers I expect, it could also be amazing!
We are derailing this thread, and I will stop posting when someone takes it back to the OP’s intent.
However, your comments about defense have me wondering if IRI selections will be based more on driver experience than on quantitative metrics. Getting the best defensive driver might be a better use for early picks, since there are so many top-tier scoring robots to choose among.
As someone who previously applied to IRI working with a team that made defense a big part of our gameplay back in 2014 when I was working with 4901… I don’t think strong defense gameplay alone is getting a team into IRI.
I agree. Teams that end up playing defense at IRI will have been selected by the committee based in large part on their offensive capability, since that is far easier to measure. Drivers who have experience putting up big offensive numbers AND playing smart defense will be the rare commodity.
6443 is hoping that defense gets some consideration. We primarily played defense this year but had one of the most unique and consistent hab 3 climbers in the world (successful climb in 80 of 85 matches). We also can score hatches (max 8-9 per match) including sandstorm but can prevent much more scoring than we can account for ourselves. With our strategy, we were alliance captains in 4 of 5 tournaments, winners in one, finalists in two (including Houston/Carver Division where we captained the #5 alliance past 330 and 364). Also won creativity awards at district, DCMP and Houston.Because of our style of play however, our offensive stats will never measure up if that is the crux of the decision criteria.
Here’s hoping anyway. We’d love a chance to play with and against many of our new friends from Houston…and make many new friends from Detroit Champs regions.
2928 is on the same boat with defense. Although we were perfectly capable of playing offense (with quite a few rockets to boot), we primarily played defense during the elimination rounds. We had a variety of 2 hatch sandstorms, and would then transition into defense, before coming back for a last second climb. We managed to pull many upsets with this strategy, almost taking down the #1 seed in Galileo (but then we died in the tiebreaker match ). Just like you guys, our offensive stats are pretty low.
Looks like we’re both gonna be crossing our fingers, :D.
Personally, I’m very curious how the committee will handle teams like 48, who won 3 regionals, but had a season ending ELO outside the top 500, never eclipsed 31 OPR at an event, and wasn’t on an alliance at Worlds.
I’d put them in if it were up to me, but I don’t know how much the committee will use these traditional ranking metrics to pick teams.
I’m also curious if they will use TrueSkill ranking to compare teams (as used on https://frc.jcharante.com/#/home/1). There 48 is much higher ranked.
). Just like you guys, our offensive stats are pretty low.