pgt:  In fact, it would be useful to be able to turn off the cube altogether, so you wouldn't have to choose between rolling the dice and doubling on each turn in situations where you know you won't be doubling.

You know that little grinning face on the button you click to get the smileys?  Well, on my display, when my opponent doubles, that little grin comes up right under the word "double"!  Boy is that annoying!

pgt:  I think I would support pgt's suggestion too, though there would be a small chance that one might miss a desirable double that way.

Andersp:  Oops, sorry, fixed now!

Andersp:  Thanks, friend :-)

joshi tm:  Well, I was tempted to suggest that, but I don't think autopass is appropriate where the player has a perfectly legal non-pass action available to him.  I think it's just the price we pay for the cube.

There are certain situations in which autopass is not appropriate even though a player will want to pass 999 out of 1000 times.  In particular, when a player is on the bar and has the opportunity to double, the theoretical possibility of a double means that the server must give him that opportunity even though he will almost never exercise it.

At present we have two choices in this situation -- "Roll dice" and "Offer double".  If the double were not available, the server would roll the dice itself and determine whether autopass should apply.  Since the double is available, however, the player must click on "Roll dice" and then submit.

Would it be difficult to add a "Pass double" button and program the server to treat this as a situation in which no double is available?  In other words, after a player clicked "Pass double", the server would roll the dice internally, and if no move was possible, immediately pass the game back to the player's opponent.

I know the second click is not a big deal, but it would improve the smooth operation of the game interface to eliminate it when it is not necessary. 

Thad: That's definitely bizarre.

AlliumCepa:  Well, not really, as a person wishing to resign could simply allow his match to time out.  The current situation actually involves timeouts (presumably unintentional) and not resignations.  See the link below (player #10)

Unrated Triple Gammon

nabla:  That would work.

playBunny:  Playing without this rule makes it pointless to have tournaments under this format, since it will almost always be the case that someone in such a large group will time out, resign too soon, etc.  Now that you've pointed it out, I'm going to cancel the triple gammon tournament I just created recently.

playBunny:  Now I understand, and I agree.

playBunny:  Are you saying that's the way it works here, or that it should work that way?

Because it certainly does not seem to be working that way.

fakar10:  There's some discussion here:  http://savannah.gnu.org/bugs/?19084  Don't know if it's helpful.

fakar10:  Fixing it in what way?

gambler104:  But does anyone know how it is actually set up here?

    I notice that it is possible to create a triple anti backgammon tournament.  I'm wondering how the scoring system works in such a case.

Dear everybody,



Please remember that gammons DO count in Triple Gammon tournaments.  Even though
each match is a single game, the scoring takes gammons and backgammons
into account.  Therefore, do not resign a game before you have borne
off a checker, if you have any chance of doing so.  Similarly, do not
resign if you still have checkers on the bar or in your opponent's home
table.  You will be giving your opponent points he or she may not have
earned if the game were played to conclusion.



(This just happened to me, an opponent resigned too soon.  I might have gotten the gammon, but it was not assured.)

If you belong to Dailygammon, check this tourney idea out.  The purpose is to minimize the effect of luck.

http://www.dailygammon.com/bg/forum2/main/read/18423#9

In any particular backgammon match, it's always possible to say that the outcome was determined by luck, since if the dice had been different the winner might have lost.  However, it's not possible to say that about 100 matches between the same players; the better player should win the majority of games, because the dice WILL more or less even out over that span.

Andersp:  I'm sure you are not stupid, just pretending to be ;-)

If I lose to a lower rated player, it may be that he or she played better than I on this occasion, or because the dice favored him or her.  Similarly, if I win a backgammon match, it may be because of the dice or because I played better.  In chess, it's almost always a matter of skill, though there can be occasions I am sure when a player with less overall chess playing ability outplays a better player. 

The point about backgammon is that the player is only partly in control, due to the random element.  Thus, a rating in backgammon is not so much a prediction about the outcome of a single match, as it is a prediction about the outcome of a large number of matches.  Given a sufficiently large sample, the luck factor will even out and the player with greater skill will win a majority of the games.

My BKR is the result of a mathematical formula being applied to the results of my games.  Some of those games were won or lost primarily on skill, others on luck.  I offer no conclusion as to the interpretation of the backgammon BKRs on this site.

I hope that clarifies it, in case you were not just pretending ;-)

Hrqls:  Thank you :-)

Andersp:  Assuming that a rating system can be properly designed, why do dice games differ in that respect from pure strategy games?  I don't mind losing games per se, I just would like the risks and rewards to be in balance.

To get off the soapbox, I thought it would be nice to have a discussion of Triple Backgammon strategy.  I have some thoughts and I'm sure others do to.

One observation that occurred to me is that, if you can imagine winning all your games in a tournament, you could do so and still not win the tournament.  This is fairly obvious if you think about the scoring system, but it makes the point fairly sharply that one should always be on the lookout for gammon opportunities, even at the risk of losing a single (or even a gammon).

 Andersp:  I disagree.  As I stated earlier, backgammon is a skill game with a random component.  A rating system is appropriate in backgammon to measure the players' relative skill.  It just takes more games to establish a reliable indicator of skill because of the random factor.  There are rating systems that work very well on other sites, mostly based the granddaddy of them all, FIBS.  This is not to say the rating system here doesn't work, but it does have serious defects.  The problems pointed out by nabla and Abigailll are the most significant ones. 

joshi tm:  The question is not whether ratings are important; of course, they are not in any real sense.  However, the question here is, if there are to be ratings, should they operate logically and consistently, or illogically and capriciously?

AbigailII:  Correct.  Of course, the fact that the system has one flaw is not a persuasive argument in favor of a different flaw. And yes, these are only significant to one who cares about having a rating system that works.  Those of us who do care are apparently in the minority, and not represented at all in the management of the site. 

Having started a TG tournament, I remember the other reason I don't like rated play.  The games are treated for BKR purposes just like any other game, but the strategy is different.  Whereas a gammon makes no difference in a standard one-point game, the whole point of TG is to play for gammons.  Therefore, a player will be willing to lose two games out of three if he can get a gammon in the third.  This will cause his BKR to decline, if he plays the optimal TG strategy.

My unrated TG tourney is still open for entries, and now is a single section of 20 players ;-)

nabla:

This is not a phenomenon that is limited to games with a luck factor, however.
Yes, that was exactly my point, but I never manage to say things in a concise way !

I see now that you said that, I was focused on backgammon the first time I read it :-)

nabla:  LOL -- correct!

Turning back to the BKR issue, I think also that the phenomenon you pointed out probably explains why ratings don't appear to tend to an average.  I play mostly multiple point games, so the BKR adjustments are closer to what they "should" be than if I played single games.

nabla:  You may be right.  Each match length has its own probability distribution.  If the match length is sufficiently high, then the probability of my winning might actually exceed the probability that is implicit in the BKR adjustments.  This is not a phenomenon that is limited to games with a luck factor, however.  

nabla:  I think that what you are describing is an independent phenomenon, which I did not mention for political reasons.    However, it actually should mitigate somewhat the phenomenon I described, because in a longer match the effect of luck is reduced.

Some of the mathematical discussion around BKR produces results that some may find counter-intuitive.  Although I believe strongly in intuition, some things we believe intuitively are just wrong.  This thought reminded me of a conversation I overheard on an airplane the other day.

One man was describing to another his experiences as a paratrooper.  He noted that he jumped with about 100 pounds of gear.  The other man exclaimed, "Well, you must have fallen really fast then!"

After a moment's reflection, I thought of Galileo's famous experiment dropping metal balls from the Tower of Pisa.  His great discovery was that the force of gravity produces a constant acceleration regardless of the mass of a falling object.  This led me to conclude that the listener's intuitive response was incorrect.

Then it occurred to me that my own intuition had failed me because I jumped (no pun intended) to the conclusion that a parachuter's rate of descent is independent of mass based on Galileo's experiment.  It turns out that the rate of descent is proportional to the square root of the mass of the parachute and its load. 

http://www.pcprg.com/rounddes.htm

I guess the lesson is that an ounce of experience is worth a pound of intuition and logic.

Moreover, if there were 63 games rather than 7, the problem could actually be worse.  This is because the BKR system as it presently is implemented will predictably drive my rating downward if I play a large number of games against lower rated players.

For example, suppose that on average, I should win 65% of my games and receive +4 points for a win, -12 points for a loss as described below (the actual win/loss disparity is probably greater).  If I play 100 games and win 65 of them, I will receive 65 x +4 points = +260.  However, I will lose 35 x -12 points or -420.  Therefore, simply by winning the predicted number of games I lose 160 points!

In actuality, I would lose fewer points, because as time went on my rating would be lower and the adjustments would moderate.  However, the principle is still the same, which is that as long as I and my opponents play at our respective skill levels, my rating will fall simply because I play lower rated players.  This should not happen.  If my average winning chances are 65%, then wins and losses should produce point adjustments in a ratio of approximately 1:2 (e.g. +4, -8).

The objection may be raised that, under my standard for a fair system, players would not be able to rise and fall significantly in the rankings.  The answer to that is that players should not rise and fall unless their skill level changes.  Everyone starts at the same level, and moves to a certain rating according to his or her skill level.  After that, changes in ranking should occur only if players get better or worse at the game.

It seems that, under the BKR system in place today, if everyone were constantly playing everyone else, all players should tend toward some average BKR.  I guess the only reasons this does not occur are that (i) new players are constantly being introduced to the system, (ii) old players leave, and (iii) players tend to select opponents closer to their own level.

Andersp:  There is no such thing as a "correct" BKR; it is the result of a calculation.

Andersp:  Unfortunately, there are only 8 players to a section. 

Andersp:  Backgammon is not a luck game; it is a skill game with a random component.  A large part of the skill in backgammon is understanding and working with the random component.  However, it frequently happens in a single game that luck overwhelms ths skill component, and a lower rated player wins because of having better dice rolls.  The chances of this happening diminish as the length of the match increases.  But in a single game, a very good player can expect to lose to an average player 25-35% of the time.  This would be unheard of in chess, on which the ratings system here is based.  

Hrqls:  OK, you win

Andersp:  That is precisely why I created the unrated tournament, so that I can play all comers for fun and not for ratings!

Andersp:  Absolutely.  For example, I went to the waiting games and picked out the highest rated opponent waiting to play backgammon.  His rating is less than 200 points lower than mine.  The predicted BKR change from playing this person is

win: 2425 (+4), ... loss: 2409 (-12)

In order for this to be fair, I would have to be 3 times as likely as the opponent to win this game; in other words, I would have to have a 75% chance of winning.  While this might be true in games without a random factor, it overstates my winning chances significantly in backgammon.   (They are  probably more like 55% than 75%).

As the ratings gap increases, the system becomes even more ridiculous.  It's not uncommon for me to receive a match invitation for which my upside potential is zero!

Yes, the ratings are important to me.  That is why I won't play against opponents whose ratings are much more than 100 points lower than mine.

I don't know how many high-rated players are reluctant, as I am, to join a single-game tournament against all comers.  Unfortunately, the BK rating system was not designed for backgammon and unfairly penalizes the higher rated player in a game between players with a large rating difference.  In order to permit higher rated players to enjoy a Triple Gammon tournament without worrying about ratings, I have created an unrated version.  All players are welcome.

Hrqls: No reaction.

AbigailII:  I made the same proposal on Dailygammon, which fell on deaf ears.

I'm at that point where I'm not starting any new games, and I've finished all the games I have with players who move at a reasonable pace ... all I have left is those players who move every couple of days.  I'm turning on email alerts and signing off. 

Oceans Apart:  According to the display, black has the cube 

KotDB:  Seems odd, but the person who timed out doesn't have much basis for complaint.  You should not feel bad about it.

IMHO a player who times out in a multigame match should forfeit the match, not just the current game. 

Does anyone have any thoughts about checker play in anti-backgammon?  Aside from the obvious -- fill your home with blots, hit the opponent as far from his home as possible, etc., I mean.  For example, how should the mid-board be handled?  I tend to focus entirely on my home board and my opponent's.  However, as I play against better opposition, I see that they use the midboard quite effectively at times.  I have seen players extend the blot-filling strategy to their own outer board to good effect.  And what about the endgame?  I believe that bearing off strategy is potentially even more important in the anti game than in the regular game.  Any thoughts about that?

I'm amazed that people can be so interested in a feature that will save time and trouble will waste so much of everyone's time arguing futilely about it. ;-)

Czuch Czuckers: Some recommendations from one who has been there and back:

1. Try to be more aware of your good fortune than you are of your bad.

2. Accept that good and bad dice are the yin and yang of backgammon, and so it is impossible to have one without the other.

3. Appreciate the game's unpredictability: isn't that why we play backgammon rather than chess?

4. Admire your opponent's success as if it were your own. It is.

Namaste!