Rematch button is hit-and-miss

"But in atomic even if opponent is good, he cal [can] lose in 5 moves"

No, this is an unjust statement. Atomic has clear opening theory developed by every different player. If a player doesn't know their mainlines well enough, they cannot be considered a good player to begin with. It's natural for players to lose in 5 moves but it is directly influenced by the player's lack of skill and / or experience.

We should be optimizing for pearls, not sand. Therefore, while your point will be considered, with all due respect I don't think making an exception for lightning atomic will justify itself.

While at it, try to find a game of rated lightning atomic I've played that I've lost in under five moves. lichess.org/@/Unihedron/search?perf=14

http://ru.lichess.org/nR79BrVZ/black
http://ru.lichess.org/LRXRieU0/black
http://ru.lichess.org/rIflC6ib
http://ru.lichess.org/0VcgVIMa/black
http://ru.lichess.org/k7uDeLZL
http://ru.lichess.org/vGiNZmqF
http://ru.lichess.org/ZrpFk28B/black
It's still not all games, rated, where you lose within 5 moves.

Chesstroll_Ingot

#24

And more. It's you have so much, who's rating is high.
But when I create a game 0+1 atomic. Usually opponents are under 1900 rated, so they lose more more often. And it's only rated games I wrote, how much casual where you lose in 5 moves, or even in 3.
Yes, there are a lot where you llose in 3.

Chesstroll_Ingot

#25

Do you understand now, how it's uncomfortable, when people lose in 5 moves?
Taking in the account that most of them are worse players than you, 1800-1900 rated (I don't play under 1750).
And taking in the account also, that is's so difficult to find an opponent. You need to wait 5 minutes (literally). If you create 0+1.

Chesstroll_Ingot

#26

And you need to wait 20 sec at least to find an opponent of atomic (0+1), who will lose very rare in under 5 moves.
This my displeasure is reasoned.

Chesstroll_Ingot

#27

20 minutes* or even 30, or even more.

Arcticpenguin

#28

I think high rated Atomic players like Unihedron probably don't fall into opening traps very often. In practice, however, 25% of rated Atomic games (65,592 out of 264,204) end before move 5, at least according to the Advanced search. Some of these probably are boosters, but it's hard to imagine that it is most of them. For comparison, 5.5% of KotH, 1.2% of Classical and 0.6% of Blitz games end before move 5.

Chesstroll_Ingot

#29

Yes. And more:
Did you take all atomic games?
If you take only within 1 minute, then
26,155 of 77,000
It's every third game! And it's including 1+0!
But we talk only about 0+1 where this percentage is even higher!

Arcticpenguin

#30

Yes, that was all rated Atomic games not just Bullet. If the Lichess developers end up deciding it is worth make an exception for Atomic, I see no reason to limit it to Bullet Atomic when such a high percentage of all Atomic games end so quickly.

Perhaps one thing to try would be to remove the button only after ~3 rematches in a row with short games. That could get the number of false positives down to a more manageable percentage in line with the other variants. Otherwise, I guess you'll just have to make do without rematches sometimes.

This topic has been archived and can no longer be replied to.