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

BGonline.org Forums

41S-54 4ply

Posted By: Timothy Chow
Date: Friday, 2 September 2011, at 2:41 p.m.

In Response To: 41S-54 4ply (Nack Ballard)

Nack wrote:

Did I assume correctly that the bot is telling us that if this rollout is extended to a googolplex trials, there is a 29.2% chance that S will top the rollout?

What the bot is saying is that if the plays are actually equally good, and if we use a Gaussian approximation, then there is a 29.2% chance that we would observe this much of a discrepancy (or worse) between the two plays.

In order to make a statement of the form that you made above, one has to take a Bayesian approach, which is not what the bots currently do.

Having said that, one can also show that under plausible assumptions, the two approaches will converge to the same result in the end, so you're not making much of an error (at least quantitatively) by interpreting the bot results in the way you're interpreting them.

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.