Backgammon Luck versus Skill

Luck versus Skill

Why are stronger players luckier?

From:   Bob Sweeney
Address:   bobsweeney2@comcast.net
Date:   10 October 2002
Subject:   Is gnubg luck related to skill?
Forum:   rec.games.backgammon
Google:   cqgaqug9r7uliach297q8prrjctao1lqu9@4ax.com

Is gnubg luck related to skill? I ask because the bot is consistantly luckier than I am based on its own luck ratings. I just played twenty five games with the bot and it ended the game with a higher luck rating than me 16 times, I was higher on 4 and we had the same on 5. I know this is a tiny sample but I seem to consistantly get out-lucked by the bot and only did the 25 game test because of what seemed like a higher than average luck factor for gnubg over the last 4-5 months that I've been playing gnubg. Thanks, Bob Sweeney

Douglas Zare writes:

Suppose to start off each match, you make a blunder worth 20% mwc, but then you and your opponent play perfectly. Then you will win 30% of the time, and every time you win, your net luck will be +70%. Every time you lose, your net luck will be -30%. If you win 3 games in 10, then your total luck would be 210%-210%=0%. The total luck should average to 0 in the long run. However, the stronger player will be luckier most of the time. In a match between players who do not play at a novice level, the luckier player wins, so it is not useful to check who is luckier if you already know who won the match. What is useful is to see how much luck was used.

Did you find the information in this article useful?

Do you have any comments you'd like to add?