Previous: Cube statistics, Up: Statistics


8.4.3.4 Overall rating

The last section is the overall summary.

  1. Error rate (total): The accumulated checker play and cube decision errors reported by normalized and unnormalized.
  2. Error rate (per decision): The error rate per decision is the total error rate divided by the number of non-trivial decisions (i.e., the sum of unforced moves and close or actual cube decisions). Again, please note that is different from Snowie 4 that defines the error rate per decision as the total error rate divided by the total number of moves for both players. In general, your error rate per move will be lower in Snowie than in GNU Backgammon. An investigation of approximately 300 matches showed the on average the GNU Backgammon error rate with be 1.4 times higher than your Snowie 4 error rate.
  3. Equivalent Snowie error rate: For easy comparison GNU Backgammon will also print the total error rate calculated by the same formula as Snowie 4.
  4. Overall rating: Based on your normalized error rate per move GNU Backgammon will assign you a rating ranging from Awful! to Supernatural. See the table below for the thresholds.
  5. Actual result: The actual result of the game or match. For money game this is simply the number of points won or lost during the game or match. For match play the number is calculated as 50% added to the result of the game or match in MWC. For example, losing a match or winning a match corresponds to an actual result of 0% and 100%, respectively. Winning 1 point in the first game of 7 point match is worth 6% using Kit Woolsey's match equity table, hence the actual result is 56%.
  6. Luck adjusted result: The luck adjusted result is calculated as the actual result plus the total unnormalized luck rate. This is also called variance reduction of skill as described in Douglas Zare's excellent article Hedging Toward Skill. This should give an unbiased measure of the strengths of the players.
  7. MWC against current opponent: For match play GNU Backgammon will calculate your MWC against the current opponent. The number is calculated as 50% - your total unnormalized error rate + your opponent's total unnormalized error rate. If your opponent is really lucky but plays rather bad, this number can become larger than 100%, since he due to the extra luck has the possibly to give up even more MWC. However, GNU Backgammon will report this number as 100%, and 0% for the opposite situation. Note that this number is biased towards the analyzing bot, e.g., a 0-ply analysis a game between GNU Backgammon 2-ply and 0-ply will suggest that 0-ply is a favorite, which it is clearly not. For an unbiased measure use the luck adjusted result above.

Threshold for ratings:

Normalized total error rate per moveRating
0.000 - 0.002Supernatural
0.002 - 0.005World Class
0.005 - 0.008Expert
0.008 - 0.012Advanced
0.012 - 0.018Intermediate
0.018 - 0.026Casual Player
0.026 - 0.035Beginner
> 0.035Awful!

Please note: The Normalized total error rate per move is multiplied by 1000 in the default settings.