BIG BAD WOLF: Second forfeit recorded, 5 to go :)

fismoluni: I appreciate your points, but I think that the rankings on this site should reflect players who are active on this site. We don't include all the world's chess grandmasters in our chess rankings, nor all the backgammon champions in the world in our backgammon rankings. Just being a good player shouldn't be enough -- the player should have earned the ranking in games played on this site, and should be required to maintain a certain minimum level of activity on this site in order to retain that ranking. This is not disrespectful of anyone's ability. It just reflects the fair principle that a high ranking player should defend his status regularly.

fismoluni: I finish many games every week; I think Fencer is extremely generous to the occasional player who climbs to the top of the ratings and then smugly sits there playing as little as possible to protect his/her rating. I don't think my ratings should be compared to those of a player who can't manage to finish one game a month!

I just noticed that a time-out in a multiple-game match does not forfeit the match, but only the current game. This is arguably a good thing, since it reduces the penalty for inadvertently allowing the clock to run out. However, it may impose a hardship on other players in situations where the person timing out has no intention of continuing the match. The situation just came up in the first game of a 21 point backgammon match; assuming my opponent has left the site or just lost interest in backgammon, is it really necessary for me to wait for him to time out 21 times before recording a win? Maybe this should be a user-settable option or, at least, one that tournament directors can modify.

Oh, wait; I just did some further investigating. The system awarded me a backgammon when he timed out! So this only has to happen 7 times in order for me to win the match ;-)

I notice he has lost quite a few matches in this way lately.

NightHawk1958: I don't think you need to apologize for being religious (r nonreligious!), it's enough that you didn't mean to offend anyone!

Marfitalu: Oh, Fencer is no doubt the greatest player on BK ;-)

Agreed. But I think I've said enough.

Fencer: I can accept that. However, I respectfully suggest that Fwiffo's intolerance was tangible in its expression.

Fwiffo: That's absurd. Discard your intolerance and read NightHawk's posting for its meaning; acknowledge the reverence of human liberty.

nobleheart: That would be an improvement, but right now all the messages in a game are displayed at the bottom of the page. I would like to see message retained throughout a multiple game match.

MagicDragon: About $33.14 according to the Yahoo calculator.

plaintiger: To expand upon that suggestion, when I am playing a multiple game match, it can be disconcerting to have the message box cleared after each game. I would like to see the message box contain all the messages sent during the match, not just the current game.

Duh. If you read what I posted here before, ignore it.

Oh, never mind.

wellywales: It's on my graphs. If it isn't shown, it may be because your current rating is the highest you have achieved.

I notice that in a match where one player has doubled, the cube appears with a "2" even when reviewing moves that preceded the double. The cube should be displayed as a "1" when reviewing a point in the match where the cube had not yet been turned.

TwoThouSevHun: It's a waste of time if your goal is to achieve a rating several hundred points higher than anyone else. If you goal is to meet people and have fun playing games with them, and to test your abilities against theirs, then I would say the change in the ratings system is irrelevant.

TwoThouSevHun: If the system had been correct from the beginning, you would never have had a lead of several hundred points. Period. In other words, that lead was due to a flaw in the system and not to your abilities. Get over it!

BuilderQ: And of course "correct" means "according to design", which begs the question whether the design is correct for all games; there is considerable sympathy among backgammon players for the view that the BKR system isn't well designed for backgammon.

nobleheart: It only sucks if you didn't earn your high rating on merit. Anyone with a high ranking should defend it regularly, or lose it.

In the midst of all this turmoil, I have achieved an adjusted all-time high in my personal Nackgammon BKR (i.e., highest on my current BK ratings chart). This is not an all-time high in adjusted ratings, however, as nstre apparently rated 2036 after his first game (compared to my current 2033 after 375 games).

nobleheart: Just a totally unfair dig at furbster's spelling :-)

furbster: Hm, Scrabble variant perhaps? ;-)

Hrqls: old BKR = 1859, new BKR = 1895 (+36)
alanback: old BKR = 2001, new BKR = 1980 (-21)
I have 390 counted games in Nackgammon -- still getting adjustments as high as 21 points

Obviously it is the method for establishing an initial rating that is the most absurd in backgammon. It probably is fine for games of pure skill, but makes no sense at all in backgammon where the worst player can beat the best player by pure luck. Even in a game of skill, it could be distortive if a highly rated player happened to lose for some reason other than being outplayed, e.g. by timing out. But in backgammon, distortion is almost guaranteed in a signficant percentage of cases.

It doesn't seem to be possible to create a tournament of multiple-point matches. That is, one in which each player would play the other a 3-point match, for example. Am I missing something?

playBunny:

>>The new system is that the 16 points are now apportioned according to the rating difference.

I'm seeing adjustments in excess of 16 points in some games -- why should that be?

Let us not forget how much we get from Fencer for little or no investment on our part!

Pythagoras: The algorithm used by the system previously to compute BKR had a bug -- if the players' ratings were less than 400 points apart, then the winner's BKR adjustment was always +8 and the loser's was -8 regardless of which player won. The adjustments should be smaller if the higher-rated player wins, and larger if the lower-rated player wins.

If the ratings difference was larger than 400 points, the system formerly assigned negligible adjustments if the higher rated player won, and relatively large adjustments if the lower rated player won (I'm not sure this has changed!).

Now, the ratings adjustment is always larger if the lower-rated player wins than if the higher-rated player wins.

Formerly, if a high-rated player was careful to play only opponents whose ratings were within 400 points of his own, he was pretty much guaranteed that his rating would continue to rise as long as he won more than half his games. Now, it is very difficult to even maintain a high rating.

pauloaguia: LOL! Actually I'm relieved that the recomputation didn't leave me down in the middle of the pack somewhere

What does it indicate about the system, that 4 out of the top 5 backgammon players are pawns with low experience?

To me it indicates that the minimum experience for the definitive ratings is too low.

Well, the new ratings are out and they are about what I expected them to be. With the ratings no longer inflated, the point differentials between winning and losing are also not as great as they were before. I think we can live with this.

playBunny: Your examples do not take into account the ratings adjustments that would occur after each match -- presumably the point loss would be smaller each time your rating decreased (and the opponent's rating increased). However, I'm not convinced that your conclusion isn't right; we will have to see what comes out of the process. Since Fencer will always have the database of past results, it will always be possible to introduce a different formula if this one proves unworkable.

I always thought the ELO formula was unncessarily complex, though complexity counts for little in a computerized world. What about a formula that awarded a winner a number of points equal to C x (r/R), where C is a constant (such as 4), r is the loser's rating, and R is the winner's rating? The loser would subtract the same number of points. There could be a minimum adjustment (such as 1) and a maximum (such as 10).