Back to forum

Scott Nichols    (2009-04-05)

Hi Thibault. Most of us have no clue about what it takes to program. Something we think looks like an easy add-on may be a programming nightmare. It is VERY impressive all of the things you do on this great site. Just one question, is making the chess ratings happen instantly like the poker does one of those programming nightmares?

Thibault de Vassal    (2009-04-05 21:23:03)
Rating calculation : real time or not

Hi Scott, it would be not a nightmare for sure. But the way correspondence chess ratings are calculated is more a question of history IMO, just like FIDE WCH. Many players would probably think this is just a nonsense, while the others may find good reasons for this change. It seems to me that correspondence chess ratings have always been calculated every 2 or 6 months according to the organization. I thought about this question already, there are advantages in both solutions, so I'm not against the idea to open this (big) debate.

Nick Burrows    (2009-04-05 21:44:01)

What are the advantages of having 2 month installments?

Thibault de Vassal    (2009-04-06 02:31:08)
2 months advantages

Avoiding rating peaks and a clearer history are probably the most important points.

Scott Nichols    (2009-04-06 02:40:34)
A couple more thoughts

Maybe the history of updating ratings every 2-6 months was necessary with slow mail by letter or postcard games. I've played in the Golden Knights back in the 80's where games easily take over a year. But now with the instant moves, there might be cause for change. Also, (#1)towards the end of a rating cycle, the games noticably slow down because players do not want to resign and lose their chance to enter a particular tournament. #2. A player may achieve a rating milestone and want to enter a tourn. right away, but can't because his/her rating doesn't change for another few weeks. So he/her may delay resigning lost games and prolong others waiting for the change. If it had changed right away, he/her would enter the tournament and proceed with his/her other games at a normal pace. Just a couple of thought...would love to hear more opinions on this, :)

Don Groves    (2009-04-06 06:37:12)

Both Go and Poker ratings are changed after each game and doing the same for Chess would make all FICGS games consistent in this regard.

Also, I think Scott has a good point about having to wait to enter a tournament for which one is qualified based on "future rating" but must wait, perhaps weeks, until the next ratings update.

Thibault de Vassal    (2009-04-06 21:31:27)
Rating peaks

Among possible problems : It is likely that players could reach higher artificial ratings (peaks) this way, even if we change the complete system & the way tournaments are built. IMO instant ratings mean that games should start as soon as a player enter a tournament waiting list - gradually, like at IECG server - otherwise it would be even harder to predict your opponents tournament entry rating, by the way there is no more TER taken in account in Go rating calculation, that is an advantage in some ways but one of the main problems also]

As for me, the deep reason why I may prefer the 2 months system is this very special "moment" that FIDE players know when waiting for their next rating. The other system makes everything faster & faster, just like the world wide web but finally maybe the passion flies away faster also. My 2 cents :)

Scott Nichols    (2009-04-06 23:31:40)
Very good point

I agree, lets keep the passion in chess, what's a little wait. It took me forever to even switch to Algebraic notation. All of my old books are in the old notation. I was just going over the first Lasker-Steinitz match. Excellent fighting chess. Whatever you decide Thibault, it will be well received.

Thibault de Vassal    (2009-04-07 13:45:37)
Rating calculation

Thanks for encouragements, Scott :)

The question could be now : Why not to install the same system for Go, but it seems to me that instant ratings are justified in this case by the duration of the games : 1. There's no real difference of level in blitz & correspondence games, so only one rating list is best. 2. Due to the difference of ratings between strong & weak players, new players can find their right place quickly this way.

Now about Poker I'm not sure yet, maybe the 2 months system would be better. To be discussed in a few months.

Don Groves    (2009-04-08 00:04:54)
Two month Poker ratings?

I've got several Poker games still going that started over four months ago. To be comparable to Chess, the Poker ratings should be once per year ;-)

Thibault de Vassal    (2009-04-08 00:20:13)
Poker games duration

Yes, some games finish in less than a week, some may last... we don't know yet. I'm thinking about a way to reduce the thinking time in some cases looking like the "dead man defence" without changing the time control. Suggestions are welcome :)

Don Groves    (2009-04-08 06:58:42)
Time control

My suggestion is still the same -- have an intermediate time control for some minimum number of moves in a given period. For example, at least one move per week when not on vacation. I feel that if a player cannot make one move per week, they must have too many games for their available time.

For Poker, it should be even more moves per week since there isn't nearly as much to analyze at each move.