Hi there,

I (obviously ) think I'm a pretty good backgammon player but how would I go about getting a rating.....is it a FIBS rating I see talked about?

Do any programs rate you after you've played a number of games?

Sorry if it's a daft question, it would just be good to see exactly, and impartially, where I stack up against other players.

Thanks for reading.

The best way to measure how well you play is to get a copy of Extreme Gammon for desktops and laptops (there's a phone version, XGmobile, which isn't quite as good) and play a long series of games against it. It will measure what's called your error rate (ER or PR), which is roughly the average amount of equity per move that you lose through mistakes (measured in millipoints per non-forced move). A lower number is obviously better.

Here's a rough chart to assess your result:

2.00 to 3.50 -- World Class, a threat to win any tournament you enter.
3.50 to 5.00 -- strong Open player.
5.00 to 8.00 -- average to below-average Open player.
8.00 to 15.00 -- intermediate-level player.
15.00 to 30.00 -- beginner/novice player.

Thanks for that, after 100 games to 15 points (94-6 win ratio) I'm apparently at 3.57 so I'm guessing not too shabby.

I will try to improve although I suspect there is a ceiling to how good your skill is, everything else is down to the dice?

All due respect, the chances you played 100 15-point matches w.XG (or any bot capable of calculating a PR/ER) and won 94 and lost 6 are practically infinitesimal. There must be some misinformation here.

I don't think you could be saying you won 6 and lost 94 either... also virtually impossible.

Playing XG all the time also gives a better-than-normal PR, I have found, and I suspect it is due to seldom encountering unusual positions that you would against a human player. In addition, there is the tiniest of pauses even on a fast machine when cube decisions are close that alerts you you should be thinking of doubling. Once you are aware of it it is impossible to ignore, and definitely helps with cube decisions, although it has been a while now since I played against XG.

I probably play at a 9 against XG, and 10 against humans over quite a large sample size.

Yes, I've noticed this too. Your error rate against humans will be 0.5 to 1.0 higher than against XG, for the reason you mentioned -- more complex, non-standard positions, plus the lack of feedback in doubling situations.

Not to mention the fact that it never misses a cube so you can correctly take a much higher % of XG cubes than those of most humans.