JHO2 : Bad start for Fransesca

[Johan Havegheer]

Geschrieben von: / Posted by: Johan Havegheer at 14 March 2000 20:46:12:
Hello,
Last night Fransesca plays Nimzo 2000b
Nimzo 2000b - Fransesca 0.78 : 4 - 0 !
Oeps, Francesca didn't have a change.
Johan

[Dann Corbit]

Geschrieben von: / Posted by: Dann Corbit at 14 March 2000 21:48:33:
Als Antwort auf: / As an answer to: JHO2 : Bad start for Fransesca geschrieben von: / posted by: Johan Havegheer at 14 March 2000 20:46:12:

Hello,
Last night Fransesca plays Nimzo 2000b
Nimzo 2000b - Fransesca 0.78 : 4 - 0 !
Oeps, Francesca didn't have a change.

Things like that are not at all uncommon, even for evenly matched chess engines. You see it all the time in partial SSDF results. The odds of getting either all heads or all tails in 4 coin tosses is a surprising 12.5% (6.25% + 6.25%)
since 0.5^4 = 0.0625, the odds of h-h-h-h is 6.25%
Similarly, the odds of t-t-t-t is 6.25%
Since for an OR we simply add the probabilities, it is more than 12% chance of completely lopsided victory.
Often, our intuition is very wrong about such things.
On the other hand, sometimes one program or the other really is much stronger. That's why the SSDF plays thousands of games.


My FTP site