Can I beat the best backgammon player in the world?

Factual Questions
1

I can’t play chess to save my life. But I’m a decent backgammon player. It’s a fairly simple game. The strategy is easy to learn. And as good as you may be, there’s always a luck factor due to the dice.

Now I play one version of the game online, and in this system, players are ranked. While I have played some truly awful players who leave endless strings of single chips on the board, for the most part I think that once you know how to really play, there’s not much difference between two players. That is to say, someone ranked at number 500 who just doesn’t play that much but understands the game could, IMO, beat the person ranked #1 almost 50% of the time.

Am I right in this? Or is there a depth to backgammon playing that I simply will never understand without playing for years at a time, and therefore the top player would easily beat me hands down time and time again…?

2

There’s a lot of luck in backgammon. I don’t know exactly how much, but enough that you can easily defeat a superior player in any given game or even match of several games. A large amount of the skill in backgammon comes from its gambling nature: realizing when to double and when to accept/decline a double.

In this way, it is similar to various forms of draw poker, in which there is a modicum of skill involved in determining which cards to discard, but most of the skill comes from deciding whether to bet and how much.

3

Opus is right. Backgammon is a gambling game. If you are “pretty good” you could certainly beat the world’s best player in a one game deal. Which I imagine would be impossible in chess (but I’m not a chess expert).

4

Unless you are somewhere in the same realm of ability as clearly better ranked player will beat you every time. No luck in chess…just skill (or lack of it). Someone like Bobby Fischer or Gary Kasparov will beat all comers save their immediate peers and that is a small club.

5

Backgammon, like poker is a game of positive, long term expectations.
Meaning… A world class player could lose a match to a novice player at any one time, but if they play “X” number of matches, the wc player will win more than 50% of the matches every time “X” being big enough of a number to overcome the short term luck factor of the novice.
I don’t know ‘exactly’ what number “X” would have to be because this really is the $64,000 question, but the bottom line is the better skilled you are at the game, the higher your winning percentage will be.

Imagine if you own a casino that has a roulette table.
Some lucky guy walks in and starts betting big, he cleans you out that night because he bucks the odds and hits every number under the sun. Can he do this every night? The answer is no.
He can get lucky once in a while, but he will be unlucky far more times than lucky, and this is why you own the casino!

Backgammon is a very deceptive game on the surface.
It can take an afternoon to learn, yet a lifetime to master.
This is why it is such an action game when it comes to money play.
Amateurs and intermediate players develop a false sense of security when they first learn to play because it takes no time at all to learn the “basics”, and this leads to a confidence in their abilities which is much higher than it should be.

Software exists today which can tell you with a reasonable level of accuracy, your skill level, as opposed to your opponent’s skill level. By utilising this knowledge, you would be able to calculate your future expected win/loss rate to within + or – 10%. If you play online, this knowledge is very valuable because it essentially tells you which players are stronger than you, and which are weaker. It takes away the guess work, and now you just play the fish (if you play for money).

Suppose you and I play 100 matches (7 point matches), and we decide to play each match for $20.
We enter our saved match file into the program; the program evaluates our play and assigns an error rate to each of us.
Based on our error rates, it then calculates with extreme accuracy what % we are to win.
For example if I am a stronger player than you, the program could tell us.

 Details for Crazyfoo after 100 analysed matches:
Expected win rate: 66% (This is how many matches I “should” win based on my skill level compared to yours.

Effective win rate: 55%: This is the number of matches I “Actually” won.
In the short run the ‘effective’ rate can have some wild swings.
I’ve had sessions where my expected win rate was 80%, but because of horrible luck, my
“Effective” rate was 0% because I lost 5 matches out of 5, but I know if all things were equal, I should win 4 out of 5.
In the long term, the Effective and Expected become fairly close.

One night I lost 13 matches in a row, and I know based on my skill level that I should’ve won 10/13.
You don’t quit playing, just like the casino doesn’t close its doars because it had one unlucky night and some guy made a pile on roulette.

So getting back to if we played 100 matches.

Based on the above information, I know that I will win 66% of the matches that you and I play.
Remember, there is no guaranteed this will happen, but after thousands of files in my match history database, the more you play, the more accurate the number is.

So armed with this info, I know that I will win $20 sixty-six times (66x$20 = $1,320)
However I will lose 34 times (34 x $20 = $680)

$1,320
-$680
$640 profit after 100 matches.
This way I know that on average I will win $6.40 each time we play.
If on average a 7 point match take one hour to play, then I know what to expect from you for my long term “expectation” on an hourly basis.

So to answer your question, yes, an amateur can win on any given occasion, but to say the expert and novice are close in the long run is not correct, the difference between the 2 are night and day.

Hope this helps.

If you’re interested in looking at the software, it’s over at www.bgsnowie.com
I bought it 6 years ago, and it has paid for itself 100x over.
It is a neuronet software and plays better than the best players in the world.

6

You TANTALIZE me with a great post, get me interested in Snowie…

and BAM. PC only. No Mac version from what I can tell. Sigh.

Thanks anyway, though.

7

Don’t despair yet my backgammon friend.
GNU software has been in development for quite some time, and from what I hear from my “cheap” backgammon buddies who didn;t want to fork out the $380 US for Snowie 4, they say GNU is as good or better than Snowie.
I don’t know what theyb base their claim on, but I have also heard from Snowie owners that GNU is a good solid software.

Only reason I never bothered is because I’m used to Snowie, so I never had a reason to look elsewhere.

I’ve never tried it, I don’t know, but the word on the street is this is a good program and does what it’s supposed to.

Don’t quote me, but I believe it is avaialble for you MAC types.

http://www.gnubg.org/

Best of all it’s free.

If you’re really starting to take an interest in the game, there’s a fantastic online bg magazine which my friend is editor of. It has a plethoria of valuable information for all types of students of the game, regardless of your skill level.
www.gammonvillage.com

Feel free to email me if I can be of any help to your pursuit of knowledge.

Regards and good luck

8

It’s the offering of the doubling cube that makes the game go to the best player. The strategies are fairly simple. What is not is looking ar the board and telling whether the odds are 50/50 or 52/48. Given an advantage, the better player will double his bet, and win more often that not. The poorer player will be less likely to recognize the advantage until it is obvious to his opponent, who has the option to decline the double.

9

What’s more, it is not necessarily obvious when you should double, and when you should accept or decline. Intuitively, you might think that you should double as soon as your chances of winning are much better than 50% - but surely your opponent should decline such a double, since they will probably lose the doubled game?

In fact, they can correctly accept the double when their chance of winning is as low as 25%, sometimes even lower. And it is wise to wait until your chance of winning is more like 70% before doubling, because by doing so you lose the advantage of having control of the doubling cube.

10

I’ll agree with the sentiment that you could, but it’s probably not very likely. Skill in backgammon is more complicated than it seems… the ‘competency’ level is just a set of rules of thumb that are pretty easy to learn and apply, but ‘expert’ status involves the application of all kinds of probability math to figure out when it’s okay to break the rules of thumb, in order to get a shot at a particular advantage. Also, as has been noted, the strategy for the doubling cube can get quite intricate.

AFAIK, chapter two of “juxtaposition” (piers Anthony) has a good depiction of a high-stakes match between a semi-expert backgammon player, and a lucky mook :smiley: Including a description of some good tricks that the semi-expert uses to try and minimize the effect of bad luck against him… the best of which are psychological.