Google Groups no longer supports new Usenet posts or subscriptions. Historical content remains viewable.
Dismiss

Double when you are an underdog

0 views
Skip to first unread message

Dean Gay

unread,
Dec 18, 1997, 3:00:00 AM12/18/97
to

I posted this once before, but failed to include a subject description
and received no responses, so I thought I would try again:

Kit's answer to a question I asked concerning volatility and doubling
decisions got me wondering about something. Is it possible for a
situation to arise where it is correct to double when you are an
underdog (excluding post-Crawford doubles and situations where the
trailer redoubles because a loss at the current cube level would give
his opponent the match)?

I'm thinking of a situation in which you are very far behind in the
match and, though you have a slight negative equity, the volatility is
extremely high, so that if you were to roll well on your next roll you
could suddenly find yourself with extremely high equity and back in
the match. If such a situation does exist. Can anyone provide an
example?

Dean
(Chase)

Stein Kulseth

unread,
Dec 18, 1997, 3:00:00 AM12/18/97
to

In article <3498e371...@dnews.pacificnet.net>, d...@pacificnet.net (Dean Gay) writes:
|> I posted this once before, but failed to include a subject description
|> and received no responses, so I thought I would try again:
|>
|> Kit's answer to a question I asked concerning volatility and doubling
|> decisions got me wondering about something. Is it possible for a
|> situation to arise where it is correct to double when you are an
|> underdog (excluding post-Crawford doubles and situations where the
|> trailer redoubles because a loss at the current cube level would give
|> his opponent the match)?

Yes, it's not that uncommon either. Say you are trailing at 4-away, 3-away
holding a 2-cube. By Kit's match equity table losing puts you at 17%, while
winning puts you at 60%. Thus by doubling you risk 17% to gain 40% and given
sufficient market losers you can justify doubling with as low as 30% win
chance, like this for instance:

X leads O 3-away, 4-away
24 23 22 21 20 19 18 17 16 15 14 13
x
x


o o o o |2|
1 2 3 4 5 6 7 8 9 10 11 12

Also at this score the leader's 4-cube takepoint will be as high as 40%,
meaning that you will more often than not encounter situation in middle
game where you have significant market losers while being a small underdog.
You should be ready to redouble at this score as soon as you are close to
even with chances to gain an advantage.

Of course other more lopsided scores will offer even more clear-cut
examples of correct underdog doubles.

For fun here's an example of a correct underdog double from the one
leading the match, given that Kit's table applies to the two players
involved:

O leads X 2-away, 5-away
24 23 22 21 20 19 18 17 16 15 14 13
x
x
|1|

o o
1 2 3 4 5 6 7 8 9 10 11 12


O has only a 39% chance to win this game. Doubling gives a 39% chance to win
immediately + 61% * 60% to win later from 2-away 3-away, total 75.6%
Not doubling gives 39% * 85% chance to win from 1-away 5-away and a 61% * 68%
to win from 2-away 4-away, total 74.6%

Differences in skill levels should be taken into account, though, so that
if O is a weaker player, then the double is very clear, while if he is
the stronger player he should not double.
--
stein....@fou.telenor.no - http://www.fou.telenor.no/fou/ttkust


Dean Gay

unread,
Dec 18, 1997, 3:00:00 AM12/18/97
to

Thanks for your response. I follow the logic, but I notice that a
level 7 JellyFish evaluation disagrees with doubling in the second
example. Is this because JellyFish is using a different match equity
table, or is something else involved?

Chase


0 new messages