[ View Thread ] [ Post Response ] [ Return to Index ] [ Read Prev Msg ] [ Read Next Msg ]

BGonline.org Forums

unusual match

Posted By: Henrik Bukkjaer
Date: Thursday, 14 July 2011, at 12:38 p.m.

In Response To: unusual match (Timothy Chow)

Not considering variance reduction but thinking of CMC's observation.

Doesn't the gap of 2.5 percent tell us something (given enough matches). The bot plays at it's best level and analyze at the same level. It says "I should win x%" but it only wins y%. That difference must be the net worth of what the bot doesn't understand correctly? Why else would there be a difference? The actual result is a fact (again, given enough games) that's the inescapable truth about what happens when "skill/style x" plays "skill/style y". If the bot over evaluates it's own side, then there must be room for improvement!

This could be interesting: See such data broken into different sets, playing different levels of opponents and playing different styles/game types.

Messages In This Thread

 

Post Response

Your Name:
Your E-Mail Address:
Subject:
Message:

If necessary, enter your password below:

Password:

 

 

[ View Thread ] [ Post Response ] [ Return to Index ] [ Read Prev Msg ] [ Read Next Msg ]

BGonline.org Forums is maintained by Stick with WebBBS 5.12.