Tournament Stats

[Atorrante]

okay, good suggestions.

[李建澔2]

I think that I would be good to have stats because sometimes you are just lucky with opponents and win the game so we can see who actually played the best.

[DVbS78rkR7NVe]

Okay then, let’s narrow the question: which stats are interesting and realistically can be done?

We know how to measure time to make a move so the speediest player can be found. We know number of moves so the can find longest game. And shortest game that went into counting. Close in score games, if they’re rare. Biggest score difference. I guess most dead things after scoring can be done.

Sandbagger category for winning a lot of games but not taking the first place or something.

Perhaps something like improving most rating points during the tournament, but that’s not easy.

[Lys]

@S_Alexander

Perhaps something like improving most rating points during the tournament, but that’s not easy.

Api’s tournament players has rank, so we can sample it during the tournament and make stats.
I just have to understand it’s format. It’s just numbers, so I must compare it to ranks in order to understand conversion rule.

UPDATE: variable “playerranking” spans from -13 to 39 in the set of players for “Through the years” tournament.
Here is my translation:

• 5 and less: 25k
• 6 to 29: 24k to 1k
• 30 to 38: 1d to 9d
• 39 and more: 9d+

I made a chart and posted it in the other thread:

Through the Years: Long Correspondence

Some rank stats. Data from august 20th (yesterday). Players vs ranks [image] Detail for 25k players: [image] Variable “playerranking” spans from -13 to 39 in this set of players. Here is my translation: 5 and less: 25k 6 to 29: 24k to 1k 30 to 38: 1d to 9d 39 and more: 9d+ It looks like even 35k (-5) could be meaningful, accordig to the curve.