Blitz cube 6
Tim Chow
X:Player 2 O:Player 1
Score is X:0 O:0. Money session, Jacoby
+13-14-15-16-17-18------19-20-21-22-23-24-+
| X O O | | O X X |
| X O O | | O |
| | | O |
| | | O |
| | | O |
| |BAR| |
| | O | |
| | | |
| O | | X |
| O X O | | X X X |
| O X X X O | | X X X |
+12-11-10--9--8--7-------6--5--4--3--2--1-+
Pipcount X: 136 O: 160 X-O: 0-0
Cube : 1
X on roll, cube action
---
Tim Chow
badgolferman
If this is a 5 or 7 point match I would say double/take. O's position
on the X's bar point makes this a take in my opinion. However I do not
quite understand the Jacoby and money session rules so I can't offer a
guess in that regard.
Walt
If X makes the four point he'll lose his market (62 64 65 52 54 42 44 22
- that's 14/36) , unless O gets a really nice roll. So that makes me
want to double. X has better position, is way ahead in the race, and
threatens to close O out with 11 checkers in the zone.
That's three of three PRT, so the question becomes is it too good. With
Jacoby, it can't be.
What about the take? Usually PRT is drop, but O has some assets here.
If she enters and scrambles to the bar point she'll have a nearly equal
holding game. The three point blockade with two checkers stuck behind
it is worth something too. X has four blots, although with O's minimal
board that's not so much of a factor. Add 'em up, and I think I see a
narrow take.
D/T.
//Walt
Paul
>
> - Show quoted text -
I also vote for a double take -- I agree completely with Walt's
analysis. (But I don't follow his practice of never reading other
posts before attempting to answer for myself.)
Paul Epstein
Tim Chow
wrote:
> If this is a 5 or 7 point match I would say double/take. O's position
> on the X's bar point makes this a take in my opinion. However I do not
> quite understand the Jacoby and money session rules so I can't offer a
> guess in that regard.
99% of the time, you can think of money with Jacoby and a centered
cube as equivalent to 7-away/7-away with a centered cube. 0.9% of the
time, your cube decision at 7-away/7-away is "too good to double" but
with the Jacoby rule you simply cash. The remaining 0.1% of the time,
you're not too good, but the possibility of being too good is in the
air and affects the cube decision or (rarely, perhaps 0.01% of the
time) the checker-play decision.
---
Tim Chow
Walt
> On Sep 1, 11:55 am, "badgolferman" <REMOVETHISbadgolfer...@gmail.com>
> wrote:
>> If this is a 5 or 7 point match I would say double/take. O's position
>> on the X's bar point makes this a take in my opinion. However I do not
>> quite understand the Jacoby and money session rules so I can't offer a
>> guess in that regard.
>
> 99% of the time, you can think of money with Jacoby and a centered
> cube as equivalent to 7-away/7-away with a centered cube. 0.9% of the
> time, your cube decision at 7-away/7-away is "too good to double" but
> with the Jacoby rule you simply cash.
And for the take decision, Jacoby doesn't matter at all. (I think I
understand this much...) But the checker play leading up to an initial
cube *might* be different, right? You can make plays that carry big
gammon risks more readily if Jacoby is in effect.
> The remaining 0.1% of the time,
> you're not too good, but the possibility of being too good is in the
> air and affects the cube decision or (rarely, perhaps 0.01% of the
> time) the checker-play decision.
I'm not sure I understand this.
//Walt
Tim Chow
> And for the take decision, Jacoby doesn't matter at all. (I think I
> understand this much...) But the checker play leading up to an initial
> cube *might* be different, right? You can make plays that carry big
> gammon risks more readily if Jacoby is in effect.
It *might* be different, yes. But it is very rare. Your second
statement, that plays that carry big gammon risks are more plausible
when Jacoby is in effect, is largely a misconception.
Let me try to explain it this way. Ultimately, the *only difference*
that the Jacoby rule makes (assuming strong players) is that *you
can't be too good* when the cube is centered. The way the Jacoby rule
is phrased, of course, is that *you can't win a gammon* when the cube
is centered. This is technically correct, but it causes a lot of
confusion. The point is that when the cube is handled correctly, the
vast majority of gammons occur when *the cube is already turned*. The
only exceptions are when a player goes from being not good enough to
double to being too good to double on the *very next turn*, and then
stays too good (or, rarely, flip-flops between being not good enough
and being too good) to the very end of the game, without ever getting
to a D/T or D/P situation along the way. I'll call this "gammoning
the hard way." With accurate cube handling, this sort of situation
happens quite rarely.
What usually happens when you get gammoned is that you get the cube
somewhere along the way to the denouement. And as soon as the cube
gets turned, the Jacoby rule vanishes into thin air. So let's think
about the situation you brought up, where you're contemplating a
checker play with a lot of gammon risk. You might be thinking, "The
cube is centered now, so I can't get gammoned now." Well, *of course*
you can't get gammoned *now*, as in *this turn*, because your opponent
hasn't borne off 11 checkers yet! But that's a trivial, silly point.
The real question is, what are the chances that I will get "gammoned
the hard way"? If all of my gammon losses are going to happen with a
cube turn along the way, then the Jacoby rule is irrelevant.
Let's put it yet another way. Say you take a big risk now. What is
your plan for avoiding a gammon when things go south? Are you going
to drop the cube when it's offered, so that you won't get gammoned?
But the drop decision is independent of the Jacoby rule, so you
shouldn't be planning to drop cubes that you wouldn't drop otherwise.
But if you take the cube then the Jacoby rule offers you no
protection. The only gammons you avoid are the "hard" ones, when the
cube *never comes at you*.
Basically, unless there's a significant chance that you or your
opponent will be too good to double on the very next turn, you should
just ignore the Jacoby rule.
---
Tim Chow
Bradley K. Sherman
> +13-14-15-16-17-18------19-20-21-22-23-24-+
> | X O O | | O X X |
> | X O O | | O |
> | | | O |
> | | | O |
> | | | O |
> | |BAR| |
> | | O | |
> | | | |
> | O | | X |
> | O X O | | X X X |
> | O X X X O | | X X X |
> +12-11-10--9--8--7-------6--5--4--3--2--1-+
>Pipcount X: 136 O: 160 X-O: 0-0
>Cube : 1
>X on roll, cube action
No double/take.
--bks
Tim Chow
I think Walt analyzed this one pretty well. The rollout indicates a
double and a take. One tricky point is evaluating the relevance of
O's bar-point anchor. It doesn't offer any direct protection against
getting closed out (though it does provide some indirect return shots
from the bar and interferes somewhat with X's attempts to marshall his
forces). Its main value is that O has a chance of escaping to the
anchor by rolling a 7 from the bar, converting to a holding game.
Here O has reasonable chances of avoiding the closeout, and the
question is how good O's holding game will be if it comes to that.
The answer to that question is quite sensitive to the location of X's
back checkers. In the diagrammed position, X still has two checkers
back behind X's three-prime. That may not seem like much, but it
seems to be enough to allow X to take.
As usual I've included some variants with XGR+ evaluations. They
illustrate what happens if we move X's rear checkers forward a bit.
=================================================================
Analyzed in Rollout
No Double
Player Winning Chances: 67.37% (G: 25.17% B: 0.56%)
Opponent Winning Chances: 32.63% (G: 5.67% B: 0.24%)
Double/Take
Player Winning Chances: 67.29% (G: 26.19% B: 0.58%)
Opponent Winning Chances: 32.71% (G: 5.72% B: 0.25%)
Cubeless Equities: No Double=+0.347, Double=+1.108
Cubeful Equities:
No Double: +0.763 (-0.087)
Double/Take: +0.849
Double/Drop: +1.000 (+0.151)
Best Cube action: Double / Take
Rollout
1296 Games rolled with Variance Reduction.
Moves and cube decisions: 3 ply
Confidence No Double: ±0.018 (+0.745<E<+0.780)
Confidence Double: ±0.024 (+0.825<E<+0.874)
eXtreme Gammon Version: 1.21
=================================================================
XGR+ evaluation of original position, for calibration:
Analyzed in XG Roller+
Player Winning Chances: 67.08% (G: 24.80% B: 0.57%)
Opponent Winning Chances: 32.92% (G: 5.62% B: 0.20%)
Cubeless Equities: No Double=+0.342, Double=+1.065
Cubeful Equities:
No Double: +0.760 (-0.080)
Double/Take: +0.840
Double/Drop: +1.000 (+0.160)
Best Cube action: Double / Take
=================================================================
XGID=-----bEBB---bC-aabBcb-b--A:0:0:-1:00:0:0:1:0:10
X:Player 1 O:Player 2
Score is X:0 O:0. Money session, Jacoby
+13-14-15-16-17-18------19-20-21-22-23-24-+
| X O O | | O X |
| X O O | | O X |
| | | O |
| | | O |
| | | O |
| |BAR| |
| | O | |
| | | |
| O | | X |
| O X O | | X X X |
| O X X X O | | X X X |
+12-11-10--9--8--7-------6--5--4--3--2--1-+
Pipcount X: 136 O: 160 X-O: 0-0
Cube : 1
X on roll, cube action
Analyzed in XG Roller+
Player Winning Chances: 68.29% (G: 25.12% B: 0.46%)
Opponent Winning Chances: 31.71% (G: 4.88% B: 0.14%)
Cubeless Equities: No Double=+0.366, Double=+1.143
Cubeful Equities:
No Double: +0.797 (-0.128)
Double/Take: +0.925
Double/Drop: +1.000 (+0.075)
Best Cube action: Double / Take
=================================================================
XGID=-----aEBBa--bC-aabBcb-b--A:0:0:-1:00:0:0:1:0:10
X:Player 1 O:Player 2
Score is X:0 O:0. Money session, Jacoby
+13-14-15-16-17-18------19-20-21-22-23-24-+
| X X O O | | O X |
| X O O | | O |
| | | O |
| | | O |
| | | O |
| |BAR| |
| | O | |
| | | |
| O | | X |
| O X O | | X X X |
| O X X X O | | X X X |
+12-11-10--9--8--7-------6--5--4--3--2--1-+
Pipcount X: 136 O: 160 X-O: 0-0
Cube : 1
X on roll, cube action
Analyzed in XG Roller+
Player Winning Chances: 71.08% (G: 21.41% B: 0.45%)
Opponent Winning Chances: 28.92% (G: 4.36% B: 0.13%)
Cubeless Equities: No Double=+0.422, Double=+1.224
Cubeful Equities:
No Double: +0.856 (-0.144)
Double/Take: +1.022 (+0.022)
Double/Drop: +1.000
Best Cube action: Double / Drop
---
Tim Chow