"New" 2nd Week Scheduling Algorithm

[Lil' Lavery]

Originally Posted by GaryV1188

You are correct, the teams are put into pools A/B/C based on team number - and team number only.

At least in your regional, all the C list were rookies. Look at the GLR pools if you want to see the perverse conditions this creates. There were only 2 rookie teams, and 1114 was put in the C pool because 1114 was 2/3 the way down the list. This meant the two rookies could never be allied with 1114, but might have been opponents (it happened they weren't). 70 was a A, while 494 was a B, and indeed they were allies once and opponents once.

Given that pool setup, the assignment of "one from A, one from B, one from C" was randomized more than in the 1st week assignments. 1188 played with 3 teams twice, and played against 2 teams twice. There were no other duplicates. I understand other teams had more duplicates, but no one had a perpetual partner or opponent. In smaller regionals, to keep match timing somewhat separated, the duplicates are higher.

The A/B/C pools were also created in week 1. 116 faced our "A" opponent (122) every match. In addition, our "B" opponent became our "B" ally in our next match every single time. That may have resulted from VCU having 66 teams (and therefor divisible by 6), which also resulted in 116 playing exactly every 11th match (1, 12, 23, 34, 45, 56, 67, 78).
Even with the "pools" being controversial, the greatest extent of the problem has been fixed, and as it has already been shown, some people support the formation of pools. Perhaps an ideal would be a "weighted schedule", where the pools are not 100%, but teams are more likely to be playing with teams of different ages (although not guaranteed to be every single match). Or maybe having just 2 "pools", and each alliance having a member from each then one completely random team.