Match turnaround

[Brandon_L]

Has anyone else noticed how little time between matches we have at these district events? I'm all for the 12 matches/team but it got to the point where we were told to que for our next match while we were on the field for our previous match. Twice. Meanwhile having a nonfunctioning bot, we had no time to work on it.

Any thoughts on this? Was it just us with an unlucky schedule?

We did have very helpful volunteers that did the best they could to give us the most possible time (they knew of our issues all day), but they could only do so much.

Have we reached the point where we can run matches too efficiently?

[Steven Donow]

It's one of the consequences of the large amount of matches. In all 3 of the district events I've participated in, we've always had at least one or two "3 match turnarounds" where we're getting called to queue before we're even back at our pits. But then at the same time, we've also almost always have one or two match turnarounds that are longer than an hour.

It might be advantageous (if possible) for FIRST to look into trying to redo/modify the match generator to try and space out these matches more, especially with all the areas that could possibly be moving to a District system over the next 3 or 4 years.

[GaryVoshol]

The District FTA's and scorekeepers have played with match separation in the schedule-generating algorithm. When they set the minimum match separation to > 3, you end up playing with and against the same robots too much, and miss playing with other robots. So for the sake of diversity in matches, you have to take a short turnaround or two.

[Foster]

Look at it like practice for the eliminations, once you make semis, they are pretty much back to back matches.

[p00rleno]

This is the writeup on the matchmaking algorithm, and you might find it to be an interesting read; http://www.idleloop.com/matchmaker/

As TCNJ's scorekeeper, I can say that there's very little control over the particulars of the schedule. It picks what it thinks is 'best' and that's what goes, unless there are glaring problems in which we can choose to regenerate the schedule. The only solution to this problem is either to increase the time at the event available for matches, or to decrease the number of matches played. The scheduled 8 minute match turnaround was picked almost entirely by the algorithm based on the fact that we input 12 matches and the amount of time available.

As Gary said, if we try to increase the match separation, the number of distinct allies/opponents drops sharply. If I remember correctly, every team saw at least 22 different alliance partners during quals, which is pretty good.

We did our best to note what teams were having particular issues during matches and dispatched volunteers to help and get them resolved.