Interesting handicap theory question - Backgammon Forum - Discuss Backgammon Strategy, Clubs and Books

Two Plus Two Forums Sports and Games Chess and Other Board Games Backgammon Forum hosted by Bill Robertie.

Interesting handicap theory question

Post Reply Subscribe

...

10-26-2011 , 04:37 AM

SMUTTTT

Carpal \'Tunnel

Join Date: Jun 2009 Posts: 7,127

Hi guys. Sorry if this is too trivial or has been discussed before, discussed this subject briefly with a friend yesterday.

A couple of years ago he played an old backgammon pro. The pro likely plays close to optimally.
My friend is also really good though, but expects himself to make very marginal pass/take mistakes. He thinks he wins around 45-48% of the time vs him, over an infinite sample. I.e, there is no way he has an edge.

Before the match, the opponent offers my friend to lead by 1, and the option to choose whether to play to 4, or 5. Again, in both cases he leads with 1 point initially. Whats your choice?

Quote

10-26-2011 , 06:59 AM

mute

Carpal \'Tunnel

Join Date: Apr 2005 Posts: 11,815

Quote:

Originally Posted by SMUTTTT

The pro likely plays close to optimally.

Very doubtful.

Quote:

Originally Posted by SMUTTTT

Whats your choice?

Yes, it is actually slightly better to be 1-0 to 5 than 1-0 to 5. Needing an even amount of points is generally an advantage and especially when only needing three points it is difficult to cube effectively. Your opponent has tremendous recube power, because leaders take point on a recube is 40%.

Also because of the above, I think the leader is much more likely to make cube errors at 3-away 4-away, so this probably makes playing from 4-away 5-away even better.

And if the pro is really only 45-48% favorite then he is certainly taking the worst of it here.

Quote

10-26-2011 , 07:52 AM

SMUTTTT

Carpal \'Tunnel

Join Date: Jun 2009 Posts: 7,127

Thanks a lot. Good answer, Im on my phone so ill keep it brief for now. But just to be clear. Its the pro who offers him the choice who is supposedly a 52-56% fav to win, and my friend the slight dog.

Quote

10-26-2011 , 08:42 AM

mute

Carpal \'Tunnel

Join Date: Apr 2005 Posts: 11,815

Quote:

Originally Posted by SMUTTTT

Yep, I got that. Being down one playing to either four or five with two equal players the leader will win 57-58% of the time if you trust the match equity tables. So this doesn't look so great for the pro especially if he is only a 52% fave.

Of course there is chance, that your friend is overestimating his own abilities (most players do).

Quote

10-26-2011 , 08:44 AM

mute

Carpal \'Tunnel

Join Date: Apr 2005 Posts: 11,815

Quote:

Originally Posted by mute

Yes, it is actually slightly better to be 1-0 to 5 than 1-0 to 4.

FMP

Quote

10-26-2011 , 09:08 AM

uberkuber

veteran

Join Date: Jul 2009 Posts: 3,129

The difference is probably not that big anyway. Nice marketing by the pro though, making the underdog feeling more in control with a decision before the match has even started.

Quote

Post Reply Subscribe

...