alanback: U serious?

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.

alanback: Try to play for fun, its only a dice game

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

alanback:   Still dont get it..how can rating in a luck game be so important?...if its a skill game you shouldnt be worried to play us "not good players"?

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.  

alanback:  But this tournament will probably have 64 players so a very good player like you shouldnt have bad luck in 63 games?

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

Andersp: yes .. its meant to be that way .. but unfortunaly thats not supported at the moment on brainking .. although fencer seems to have increased the possible maximum size of the sections

should i ask him to increase the number of players per sections ?

alanback: i know what you mean .. i mostly play tournaments without paying too much attention to my opponents ratings .. and i sometimes curse myself when i lose to a lower rated player .. i like the ratings :)

but i also like the game .. and to play against different opponents .. so everytime i still join new tournaments with new players instead of only the top players
(i do join the 2200+ tournaments as well though :))

i guess its even worse when you are 2400+ :)


but! to try a new tournament type on this site must be something special .. isnt it ? .. shouldnt that be worth a small risk in bkr ? ;)
(your win/loss ratio shows you win twice as many as you lose .. so thats more than 55% :))

come on ... join us for once .. : Triple Gammon

Hrqls:  OK, you win

alanback: thanks!!! a real sportsman!

would it better or worse when there are more players per section ?

alanback: You know, Fencer has been doing different things - like AutoPass which seemed like something that we would not see on this site.

Maybe it's time to see if Fencer is interesting in putting in place different rating "systems" for different types of games.

- - - - -

For those of you new, the rating system that BrainKing uses is a type based on Chess rating system - a 100% skill game. A game where a 2000 rated players should beat a 1500 rated player 99% of the time.

For Backgammon (and other games with luck factors), I would say Backgammon is around 60-75% skill game with 25-40% luck (others would say higher & lower - but that is a stupid debate to have - but everyone agrees that there is more luck to it then say Chess). Anyway, a 2000 rated backgammon players should beat a 1500 rated players probable around 75% of the time. Since bad dice can come into play. A good player can still win with bad dice (that is where skill come into play.), but enough bad dice can doom even the best gammon player.

So there are certain Gammon rating systems out there, which in my opinion should be used for Gammon - and other games with luck factors (Battleboats, Dice Poker, etc...)

Fencer - any chance of having different rating systems for different games? I'm not sure of the formulas off-hand, but if you were interested, I'm sure someone can help you get the correct formulas and such.

coan.net, fencer: http://www.fibs.com/ratings.html#explained

coan.net:  Do you think that "very good players" (2300+) who never play lower rated players have a correct BKR? 

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

coan.net: It is very true that the BKR system is not accurate when applied to backgammon, but it is a common mistake to think that it is because of the luck factor. In fact, as alanback stated, it is because of the multi-game matches (cubed or not) that are counted like if they were one game.

There is no such thing as an absolute measure for skill. What the rating system can calculate is the "winning expectation", and that defines the skill difference. If I win more than 50% of the games against someone, then I am defined as stronger. If I win 90% of the games, I am defined as much stronger.

Now, and in a non quantifiable way, in some games the weaker player has more chances to win ; but these are not necessarily the ones where pure luck (e.g. dices) are involved. These are usually the short games, and the games that have a lot of forced moves. Less moves to think about mean less occasion to make mistakes for the weaker player. Of course the luck factor plays a role too, but it is one factor amongst others.

What will happen in games which offer good chances to the lesser player is not that ratings will be inaccurate, but that the rating scale will be shrinked. For instance, instead of scaling for 400 BKR to 2400 BKR, it could scale from 1200 BKR to 1600 BKR. The only undesirable effect is that as one game still carries the same BKR change, there will be more variation of the positions of the players. E.g., if I lose 10 points, I will lose 10 positions in the ranking list, because the players are all very close to each other.

If wished, this can be corrected as there is a "rating scale" constant in the BKR formula. But you can see that it is definitely not the problem in backgammon, where top players do have a lot more than 1600 BKR !

The real problem in backgammon is that matches count the same as single games. But if player A beats player B 60% of the time in single games, he may beat him 90% of the time in a 5-point match. Or whatever, I didn't do the math :-) Hence, rating-wise the matches favour the better player and the single games favour the lesser player. This is a real distorsion in the rating system, and you can indeed see that good players who play a lot of single games are considerably underrated - AlliumCepa comes to my mind, although he plays matches as well. On the other hand, people who play only matches are overrated. For example, me (but not for rating reasons, I just don't like playing without the cube).

The good news is that one knows a very simple way to adjust the BKR formula in order to take the length of the match into account. Basically, in a n-length match, one should just multiply the rating difference by the square root of n. As you can notice, this changes nothing to the way single games are counted.

This is not a specific backgammon formula, because it is not perfect for taking cubed matches into account, but perfect for taking "first to n points" matches of any game into account. In fact it should be applied to all games. If I start a rated 10-games chess match against a lower rated player, I am just getting some undue rating points. Of course, this is much more often done in backgammon than in chess, because the doubling cube can be used only in matches.

The bad news is that I told Fencer about that quite some time ago and while he didn't say no, he did not sound too hot about updating the rating system. Indeed I understand pretty well there are more important issues to settle first. After all, rating are just numbers, what is important here are the games !

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.

alanback: I thought I was describing your very problem, that if you play single games against a lot of lower rated players you will lose a lot of BKR points. If you were to play 21-points matches against the same bunch of players, you would probably take #1 position in the BKR list :-)

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.  

alanback:
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 !

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 :-)

alanback: By the way, a heavier object does fall faster than a lighter one of the same shape, because the force caused by the air friction does not depend of the mass. Air friction is not something you want to factor out when you are a paratrooper ;-)

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: This is a very interesting discussion. I have recently felt that my BKR quite exaggerates my actual skill. Have usually have trouble with players rated above 2250 yet my rating has grown to >2300. I try to play multi-point matches with most players but often end up playing games with much lower ranked players (2000 - 2200). These games boost my BKR over time even though I'm not playing any better. Losses against similarly ranked player don't decrease my BKR significantly and so, my ranking continues to grow.

nabla: A perfect explanation of why cubed matches should be rated more than single games. There is another point. In a single game, a player is not concerned about losing a gammon. Thus, he may pile as many pieces as he can into your home squares, which makes winning quite difficult!!

tonyh: Why even bother to play without the cube? I don't.

grenv: I agree with you; It's just that it takes that much longer to play a 3 point match than a single game - but it is better backgammon. I am reverting to cubed matches.

tonyh: Well what this site needs is a Backgammon rating system in place - since the current rating system is based on Chess which is mostly a skill game - and does not work as well with games like Backgammon which does include some luck along with skill.

Of course that discussion has come up many times in the past, and Fencer had been a little interested, but not really enough to put it in place - but the hope that someday he will.

And remember - last time he did something to the rating system, the system looked back to game #1 and redid the ratings from the very start - so even though the rating system does not work too well now, my opinion is to not worry about it to much - and continue to play as well as you can - and you never know, Fencer might redo all the ratings with a better rating system and fix everything up to this point. (then again, maybe not - but we can hope.)

coan.net: Fencer might redo all the ratings with a better rating system and fix everything up to this point

That would be very interesting and I hope he does it that way. It would also be nice to have a preserved copy of the ranking tables just before the conversion. It would be fascinating to study how some players will have gone up in the rankings and others down because of the mix of match lengths and opponents that they play.