Back to "Advanced Angles" written by Steve Sax

New Kid on the Block

by Steve Sax
21 January 2010

Steve Sax

In September 2007 in Madison Wisconsin, I played a young first timer for the ABT, 23 year old Matt Cohn-Geier. He beat me on an 8 cube by rolling doubles on the last shake, in front of a packed audience (half a dozen or so) and thus the legend began.

Matt is from Evanston, Illinois and previous to his interest in backgammon, he was a competitive chess player who played at an expert level with a rating around 1950. He had played chess from about the age of 15 until around 21, when he found backgammon and abandoned chess for his new obsession. In his climb from obscurity to one of the world's best backgammon players, he lists several authors and influences which helped to mold his game to the level he has attained today.

The rest of this article (67.57 K) is premium content. Please subscribe below.

GammonVillage Subscriptions:

You must be signed in to rate articles.

Feedback about this article:

Subject: Re: New Kid on the Block

From: christoph

Date: 23 Jan 2010 03:26 EST

Excellent analysis, thanks. Is that entire match somewhere available for download?

Cheers, Christoph

Subject: Re: New Kid on the Block

From: Timothy075931

Date: 23 Jan 2010 17:45 EST

Excellent article as usual. I'm a bit puzzled, though, why the posted Snowie rollout of position 7 says "Proper cube action: Double, pass" since the equity of "Double, take" is listed as 0.979.

Subject: Re: New Kid on the Block

From: *Stick*

Date: 26 Jan 2010 11:36 EST

The match isn't available anywhere that I know of for download but if you email me I'll try to find it again and send it to you. My email is checkmugged@yahoo.com

The Snowie rollout just exports the proper cube action wrongly as you noted. Pay attention to the numbers, not the text.

Subject: Re: New Kid on the Block

From: PhilSimborg Member has a photo available

Date: 03 Feb 2010 03:53 EST

Great insights here, thanks. Both Stick and MCG are not as bad as I thought.

It would be fun to:

1. see how XG evaluates the match; 2. look at the biggest errors, roll them out, and see if there are any where Stick and MCG were right and the eval was wrong; 3. where the play is still an error, have experts take a closer look and see if they think the bot or the players truly got it right (even rollouts are not to be completely trusted) 4. Stick slightly outplayed MCG in this match. Is there any way to determine if this was because, overall, he had fewer tough decisions? I know there were other recorded matches where MCG outplayed Stick.

You must be signed in to post comments.