I think worrying about "perching" this is not the right approach. As long as the player is not cheating, it`s not fair to "control" this thing.
Chess is a mater of the "day". Sometimes I`m playing bad but I usually keep playing anyway. I know that some other day I`ll get my rating points back. But it`s only up to me if I stop or not before the level of my game drops even more. No one needs to control it. In the other hand, If my rating goes up the roof I also have the right to stop when I want, for whatever reason.
I could`t agree more that the worse possible reason to stop playing is "pearching", tough. I never do it, but again, it`s not cheating, so there´s no need to create rules to fight it.
It doesn't matter if it's cheating or not. What matters is that it lowers the integrity and accuracy of the ratings, which is one of the major benefits of auto-pairing pools in the first place.
Perhaps there could be a pool setting, selected by participants?
1) Play anyone available asap
2) Anyone available +/- 200 (Longer Wait)
3) Anyone available +/- 100 ( You probably have time to make a cuppa)
I don't see that it would affect the integrity of having randomly selected opponents....but as has been stated elsewhere....larger the pool, the less of an issue it is.
Take care that splitting pools into more specialized ones leads to less players in each pool.
I would simply replace pools by a single button after each game done in the list : "rematch", or "same cadency match"
"list matches" already offer black/white autoselection and opponent rating range selection.
litchess would aumatically compute the elo of the cadency for which the criteria are met.
pools are not necessary to rematch and have a elo according to the rating.
sorry last sentence : i meant "a rating according to the cadency"
I thought maybe showing average rating of pool would be good and also building up an hour view of this of the week - estimating average pool rating for that hour based on historical data! ( yes sorry i like ideas that are wacky! )
Then people could find an hour spot during the week which they prefer
It would also be useful for the higher rated who might see on calendar view prediction that the average pool rating is high say on monday uk 6 pm to 7pm
Useful idea ?!
I don't know if any ideas are being looked at by Thibault and co., but it seems the pools are becoming more exclusive rather than inclusive. That may not be the intent, but the lowest ranked player out of 18 in the 1 min pool just now was 1611.
The pools are highly competitive for sure, but players who are lower-ranked are not joining in. This has potential long-term implications for lichess's growth, and it'd be a shame if lichess didn't feel like a welcome place for everyone.
With pools we have entered the dark period of Lichess.
One step deeper, now we are explained why it's good to wait.
I'll bet for a fork before end of this year.
You can view it that way if you want. Pools have been out less than 2 weeks so it seems only natural to start writing your eulogies.
The fact of the matter is that the feature isn't even complete yet.
Also, NoTimewasters, I don't understand why you keep beating that drum of 'exclusivity'. Do you really expect a low level player to be proficient in bullet? Or to even want to play bullet?
Comparatively the 5+0 pool's lowest rated player (as of writing) is less than 1300. This seems only natural because a low level player is far more capable of playing 5+0 than 1+0.
I really like the recent changes to the pools, with the waves and the displayed timer until the next wave.
Nice work Clarkey!
