[Bug-gnubg] Strange analysis result

[Lauri Salo]

Hi,

 

I just analysed my match with GNU and Snowie and the results were strange. Snowie rated me as expert (nearly world class) and my opponent as advanced, with average error rates 4.4 and 7.7 respectively. Checker play and doubling-wise it went: EU_FIN_Vrabec (checker play: 10 errors, no blunders; doubling: 2 errors, 1 blunder), StileBlue (checker play: 9 errors including 5 blunders; doubling: 2 errors, no blunders). So according to Snowie my checker play was superior and StileBlue's doubling play slightly better than mine.

 

GNU evaluated doubling play similarly, but gave a strange evaluation on checker play:

 

Player                                      StileBlue                      EU_FIN_Vrabec 

 

Checkerplay statistics:

 

Total moves:                            99                                98

Unforced moves:                      71                               68

 

Moves marked very good          0                                0

Moves marked good                 0                                 0

Moves marked interesting          0                                 0

Moves unmarked                     91                                90

Moves marked doubtful             1                                6

Moves marked bad                    5                                1

Moves marked very bad             2                               1

 

Error rate (total)               +1.152 (+12.137%)            +2.468 (+106.991%)

Error rate (pr. move)         +0.016 ( +0.171%)            +0.036 ( +1.573%)

 

Checker play rating:               Advanced                   Beginner 

 

Looking at the "moves marked doubtful/bad/very bad" GNU's analysis seems to agree with Snowie, with slight differences, but contradicting itself. I went through the errors I made and I'm positive that the "very bad move" I made couldn't possibly be so bad that it was a complete "match destroyer" as the analysis seems to suggest. I appreciate that the logic between the two bots is different and I often get slightly different analysis when I compare the two, but it's the first time something is this much off the mark. Any idea what happened?

 

Best regards,

 

Lauri // EU_FIN_Vrabec