Match Play at 2-away/2-away

 Minimum game winning chances to double

 From: Walter Trice Address: wgt@world.std.com Date: 4 March 1999 Subject: Re: What's wrong with 2 point matches? Forum: rec.games.backgammon Google: F82rpK.IHJ@world.std.com

```Tom Penney wrote:
> Julian is correct. When both players are 2 points away from wining a
> match it is correct to double with anything over 50% chance of winning ...

A common misconception. Actually you don't need to be a favorite in the
game to double. Consider the following bearoff position:

- - - - - -|
O|
|
<- off   |
|
X   X|
- - - - - -|

X's cubeless probabiity of winning is 39.9%, so that is his probability of
winning the match if he doubles.

If he doesn't double, he gets a 70% match winning probability if he rolls
one of the 8 numbers that bear off, and a 30% chance if he rolls one of the
other 28 (because he should drop his opponent's cube, having only a 25% or
less cpw.)

So his total mwp.  would be 8/36*70% + 28/36*30% = 38.9%, which is less
than what he would get by doubling.

This situation turns out to be fairly typical. In a race it is usually
right to be doubling with a cpw around 40%. In a LONG race, you may need as
much as 41% or 42% to double.

-- Walter Trice
```

Match Play at 2-away/2-away

Basic strategy  (Darse Billings, Feb 1995)
Counterexample?  (Jim Williams+, Mar 1998)
Do you need an advantage to cube?  (Keene Marin+, Feb 2006)
Double immediately?  (Chuck Bower, Oct 1998)
Ever too good to double?  (Kit Woolsey, July 1995)
Minimum game winning chances to double  (Walter Trice, Mar 1999)
Practical strategy  (Walter Trice, July 1995)
Practical strategy  (Albert Steg+, Feb 1995)
Proof for doubling immediately  (Robert Koca+, May 1994)
Proof of doubling with market losers  (Walter Trice+, July 2001)
Sample game  (Ron Karr, Dec 1996)