I have not proposed anything. I am against the move to 2 minutes. Please stop trying to push this onto me: it's not my decision at all. All I'm doing is acknowledging and ensuring that the metrics don't change halfway through the test.
The evidence is that a majority of people seem to want it. I've made my own opinion of polls, their use, and this poll in particular, clear to the admin team, but if it is good enough for them then it's their call to make. Not mine.
Longer-lived teams, different reasons to play (e.g. no playoff), more concessions (and therefore noise - ~22% vs ~45% COL vs CCL last I looked), fewer AFKs (because people just concede).
But hey, if you think there is something useful you can mention from the data available in COL then please, let the admins know!
Why would I want to propose any metric to measure a change that has no justification at all to begin with?
Because the trial is happening and you want it to change back. You want there to be more evidence for the 2 min option than for the 3 min option. I'm not arguing for the reasons for making the change, I am arguing for the reasons to say "3 minutes is better than 2". 3 minutes is the default, but if there are better grounds for going to 2 minutes than staying at 3 then that is the direction the admins will take it.
COL is not a good "lab". People compared console with PC instead to show that the max match length is shorter.
the criteria is not made public and can be manipulated to suit whatever the organizers want
It won't be. I'm aware of the criteria and am against the change. There will be no variation away from those by the admin team. That said, if people have ideas for suitable metrics then please be forthcoming.