Game #11
Open Division Final
San Mateo, California
20 July 1997
11 Point Match
Ron Karr (Blue) vs. Daniel Murphy (White)
Score: 7 - 6
Analysis by Beth Skillman and Richard McIntosh
A note to our readers:
In bringing these annotated matches to you,
BACKGAMMON BY THE BAY has been graced
by the largesse of numerous of backgammon's luminaries,
who have shared their time and efforts to bring light to darkness.
In order to present a varied outlook,
and for all those good things one can say about freshness,
we have recently been scouting around for participants
in our annotated games endeavors.
More than once folks whom we asked to be analysts expressed
reluctance to do so because they were "only an intermediate player."
In the belief that beginning players can learn a lot from intermediate players,
and that we intermediate players can learn a lot from each other,
we are putting our reputations where our mouths are:
BACKGAMMON BY THE BAY proudly presents commentary this month
by
Beth Skillman
and
Richard McIntosh.
BGBB's regular bimonthly analyst, Ron Karr,
continues his cooperative ways by playing in the finals
during a month he wasn't scheduled to be the game analyst.
Daniel Murphy bids adieu to San Francisco in style,
in his last California tourney before taking up Danish residency.
This match is a fine illustration of why Ron and Daniel led the pack
in the 1997 BGBB Brownie Points race
(although Ron started breathing a bit easier
when Daniel's move to Denmark limited his participation
in the Brownie Points quest).
|
BS:
|
Blue might consider breaking tradition, and his anchor,
had White escaped one of his back checkers.
Since that hasn't happened,
sanity should guide -- make the 4 point.
|
|
RM:
|
It's way too early to worry about extracting a back checker.
Blue should equalize closed points inside by making the 4.
Blue White
Candidate Plays Equity Win G/BG BG Win G/BG BG
8/4 6/4 -0.120 45.7% 11.8% 0.3% 54.3% 14.9% 0.5%
13/9 24/22 -0.195 43.2% 9.9% 0.3% 56.8% 15.6% 0.5%
|
|
BS:
|
It's interesting that JellyFish prefers the "instant gratification"
of making the bar point,
to the diversification of making the 10 and 3.
I've noticed the proclivity of the better bots to use double-2's
to make the 11 and 4 points early in the game
(versus "old school" adage
of making the opponent's instantly-gratifying 5 point).
The moral of this story must be, "A four-prime is a valuable thing."
|
|
RM:
|
Making another inside point threatens any blots produced by Blue's bad rolls,
but there may not be any immediately.
Putting builders on the 10 increases the chances of making the 4 as
well as the bar, or hitting the first escapee.
Making the bar creates a strong mini-prime,
potentially preventing escape altogether,
and there are still builders for the 4.
White creates the most pressure by closing the bar point.
White Blue
Candidate Plays Equity Win G/BG BG Win G/BG BG
13/7(2) +0.262 59.5% 17.8% 0.6% 40.5% 11.0% 0.3%
13/10(2) 6/3(2) +0.222 57.8% 17.5% 0.7% 42.2% 11.3% 0.3%
|
|
BS:
|
Blue must hit --
splitting (24 to 22) leaves no efficient 4, and the race demands a hit.
Following 13/11* up with 13/9 leaves fewer return shots from the bar
for White than the wimpier-seeming 11/7,
and leaves Blue a strong contender in this battle of the primes.
|
|
RM:
|
Putting another checker behind the blockade is essential,
and creating another builder for the 5 point is the best use of the 4.
Leaving the checker on 13 to pick up the other blot would be nice,
but it's not worth the risk of getting it sent back
behind White's stronger blockade.
Blue White
Candidate Plays Equity Win G/BG BG Win G/BG BG
13/9 13/11* -0.115 44.8% 14.4% 0.5% 55.2% 15.5% 0.6%
13/11*/7 -0.191 42.2% 12.8% 0.4% 57.8% 15.9% 0.6%
|
|
BS:
|
The time has come for Blue to put that potential into action.
Making the 5 point is the fruition of the priming efforts Blue has made
in bringing all those builders into action.
If not hit back,
Blue has 11 chances (any 1) to make an 8-to-4 five-prime;
and if White doesn't anchor on the 3 point,
another 13 chances to make a 7-to-3 five-prime.
--'Course that's not what JellyFish says. Never mind.
|
|
RM:
|
Improving the prime is a good play,
but Blue can't hold the prime for long if he doesn't escape soon.
The chance to escape a checker while White is on the bar
outweighs everything else.
Blue might even pick up that outfield blot.
Blue White
Candidate Plays Equity Win G/BG BG Win G/BG BG
24/15 +0.131 54.6% 17.3% 0.6% 45.4% 13.4% 0.6%
9/3 6/3 +0.089 50.7% 19.1% 0.7% 49.3% 12.0% 0.4%
7/1* 24/21 +0.084 50.6% 22.4% 0.7% 49.4% 15.2% 0.8%
11/5 8/5 +0.041 49.6% 18.7% 0.7% 50.4% 13.8% 0.6%
|
|
BS:
|
So Blue pays the price for listening to me.
Now, making the 3 point will give White good aces (to make an anchor)
and sixes (to hit),
and all kinds of combinations of 2s, 3s, 4s and 5s to extend his prime.
It's best to keep White off balance by hitting;
the 3 is best put to use by coming up to the edge of White's prime.
|
|
RM:
|
As so often happens,
Blue gets an immediate reminder that escape was correct last play.
He would like to escape now,
but with White getting a two-for-one hit-and-escape with thirteen numbers,
plus eleven different numbers to extend his prime inside,
Blue is forced to make a preemptive strike
and hope that his stronger board prevails.
Getting to the edge of the prime, poised to escape or cover,
is a critical play with the 3.
Blue White
Candidate Plays Equity Win G/BG BG Win G/BG BG
8/2* 24/21 +0.113 53.1% 19.5% 0.7% 46.9% 14.4% 0.8%
24/15 +0.023 51.4% 14.4% 0.5% 48.6% 14.6% 0.7%
|
|
BS:
|
Blue is now insignificantly ahead in the race (7 pips),
and can match White's four-strong prime with one of his own.
However, Blue stands poised to extend his prime,
and has bonus blot-hitting chances with an ace.
White's lack of anchor all too often makes for a gammon.
With such powerful entries in Blue's plus column,
a double is called for.
|
|
RM:
|
After the smoke clears,
it is obvious that Blue's stronger board paid dividends.
(Rolling 4s to get to the edge of White's prime didn't hurt either).
Blue has knock-out potential and lots of market losers,
but he could have problems extracting that back checker.
Double.
|
|
JF:
|
I would have doubled (level 7 evaluation).
Blue White
Level 7 Evaluation Equity Win G/BG BG Win G/BG BG
Cubeless +0.555 65.9% 30.5% 1.9% 34.1% 8.5% 0.2%
Level 5 Rollouts Equity Win G/BG BG Win G/BG BG
Cube centered +0.778 87.0% 3.9% 0.3% 13.0% 0.4% 0.0%
White owns cube +0.568 60.8% 33.4% 3.0% 39.2% 1.1% 0.1%
Cubeless +0.669 68.4% 34.7% 3.0% 31.6% 7.3% 0.4%
Level 6 Rollouts Equity Win G/BG BG Win G/BG BG
Cubeless +0.658 67.6% 34.6% 3.2% 32.4% 6.9% 0.4%
|
|
BS:
|
Facing such gammon potential,
a conservative response by White is in order.
Pass.
|
|
RM:
|
This smells like a pass.
Sixteen numbers hit twice,
and the cost of a gammon holding the cube is the match.
Blue still has to extract his last checker,
and there are a few horror sequences for him,
but White is looking down the barrel of a moderately large gun.
Using Kit Woolsey's
match equity table
and JellyFish's estimated outcomes
when White owns the cube,
we find
| Action |
Score |
Equity |
Probability |
Product |
Total Equity |
| Pass |
-3,-5 |
34% |
|
34.0% |
| Take, win gammon |
-4,-1 |
83% |
1.1% |
0.8% |
30.8% |
| Take, win regular |
-4,-3 |
59% |
39.2% |
23.1% |
| Take, lose regular |
-2,-5 |
25% |
27.4% |
6.9% |
| Take, lose gammon |
-0,-5 |
0% |
33.4% |
0.0% |
So White is giving up over 3% match equity by taking the cube.
Stepping away now leaves him very much in the battle,
with 34% match equity.
|
|
JF:
|
I would not have accepted (level 7 evaluation).
Blue White
Level 7 Evaluation Equity Win G/BG BG Win G/BG BG
Cubeless +0.555 65.9% 30.5% 1.9% 34.1% 8.5% 0.2%
Level 5 Rollouts Equity Win G/BG BG Win G/BG BG
Cube centered +0.778 87.0% 3.9% 0.3% 13.0% 0.4% 0.0%
White owns cube +0.568 60.8% 33.4% 3.0% 39.2% 1.1% 0.1%
Cubeless +0.669 68.4% 34.7% 3.0% 31.6% 7.3% 0.4%
Level 6 Rollouts Equity Win G/BG BG Win G/BG BG
Cubeless +0.658 67.6% 34.6% 3.2% 32.4% 6.9% 0.4%
|
|
BS:
|
Blue's minimal race lead,
superior board (three points made to White's two),
and the match score (Blue can win the match with a gammon here),
are each strong indications to play aggressively.
So, hitting 9/3* is called for, now What to do with the 2?
While Blue might be tempted to do "a little of this, a little of that,"
with an aggressive hit and a passive maximize builders
(6/4 or 7/5)
or the even more wimpish 1/3,
with so many elements calling for aggression,
an all-out blitz might pay off big time.
|
|
RM:
|
A gammon wins the match.
Get those checkers in the air.
Blue White
Candidate Plays Equity Win G/BG BG Win G/BG BG
9/3*/1* +0.519 64.3% 30.2% 1.8% 35.7% 8.4% 0.2%
9/3* 6/4 +0.467 64.4% 25.9% 1.5% 35.6% 9.3% 0.3%
9/3* 24/22 +0.450 64.4% 24.4% 1.2% 35.6% 9.2% 0.3%
9/3* 7/5 +0.429 62.8% 26.1% 1.6% 37.2% 10.0% 0.4%
24/16 +0.424 64.9% 19.9% 0.8% 35.1% 7.8% 0.2%
|
|
BS:
|
This is just the kind of roll you like to see
when you're sitting behind a four-prime.
Now, does Blue risk a seven by White after playing 4/14(2),
or safety the checker on the 15 point by playing 4/9(2), 15/5?
Again,
Blue must consider that a gammon pays dividends here,
and the more aggressive move is the better.
|
|
RM:
|
Timing is now against Blue.
It's best to swing those back checkers around and risk the indirect shot now
(16.6%)
than take chances over the next few sequences
with horror rolls that leave the trailers hanging out to dry.
Blue White
Candidate Plays Equity Win G/BG BG Win G/BG BG
21/11(2) +0.572 74.2% 11.2% 0.3% 25.8% 2.6% 0.1%
21/16(2) 15/5 +0.557 73.5% 11.2% 0.3% 26.5% 2.8% 0.1%
21/16(2) 15/10 6/1 +0.516 70.6% 14.2% 0.4% 29.4% 4.1% 0.1%
|
|
BS:
|
Being nearly 40 pips down in the race,
White must keep the two-point anchor.
White might gain ground in the race by hitting Blue's ace-point blot,
and has only the checker on the 17-point with which to do that,
so that checker should stay where it is.
That leaves 8/2 for the six.
Now,
while 17/11 still leaves "contact" in the outfield,
leaving the checker on the 17-point maximizes the contact.
8/4 is all that's left.
|
|
RM:
|
Timing and White's improving inner board make the blot on 17
a valuable asset for White and a threat against Blue.
For White, making inner points is key.
Make one, slot another.
White Blue
Candidate Plays Equity Win G/BG BG Win G/BG BG
8/2 8/4 -0.588 25.1% 3.3% 0.1% 74.9% 12.1% 0.4%
5/2 7/3 -0.630 23.5% 3.0% 0.1% 76.5% 12.7% 0.4%
17/11 8/4 -0.635 22.8% 2.0% 0.0% 77.2% 10.7% 0.3%
17/7 -0.641 20.4% 1.2% 0.0% 79.6% 5.9% 0.1%
8/2 17/13 -0.642 22.1% 1.8% 0.0% 77.9% 9.9% 0.3%
17/11 6/2 -0.654 21.6% 1.7% 0.0% 78.4% 10.2% 0.3%
8/2 6/2 -0.674 21.6% 2.3% 0.1% 78.4% 12.6% 0.3%
17/11 7/3 -0.676 20.8% 1.6% 0.0% 79.2% 10.5% 0.3%
|
|
BS:
|
Although Blue now has the chance for a super-aggressive play (11/8*),
he doesn't need to gammon,
it would just be nice to gammon White.
Blue can bide his time with this roll,
and White is still very much on the defensive.
|
|
RM:
|
With a blot inside,
the super-aggressive hit isn't worth the risk.
Since the outfield blot has to stay put to keep out of direct range,
almost anything with the inside checkers is beneficial.
6/4 6/5 is most flexible.
Blue White
Candidate Plays Equity Win G/BG BG Win G/BG BG
6/4 6/5 +0.538 73.0% 10.9% 0.3% 27.0% 3.3% 0.1%
7/4 +0.525 72.8% 9.8% 0.3% 27.2% 3.0% 0.1%
7/5 6/5 +0.505 71.9% 10.1% 0.3% 28.1% 3.5% 0.1%
11/8* +0.402 66.4% 18.5% 0.7% 33.6% 11.1% 0.6%
15/13 6/5 +0.395 66.6% 11.0% 0.3% 33.4% 4.9% 0.1%
|
|
BS:
|
White again has the opportunity to safety the checker on the 17-point,
and again (and for the very same reasons), should "maintain contact."
Making the 4-point and slotting the 3-point make this non-optimal roll
accomplish two good things.
|
|
RM:
|
Again, the blot on the 17 doesn't hurt; White wants inner points in
preparation for the (hoped for) hit.
Making the 3 point leaves more chances to cover the 4 next turn
(28 numbers)
than making the 4 and slotting the 3
(21 numbers).
White wants both points soon,
so flexibility is the rule.
White Blue
Candidate Plays Equity Win G/BG BG Win G/BG BG
8/3 7/3 -0.564 25.1% 3.6% 0.1% 74.9% 9.9% 0.4%
8/3 8/4 -0.572 24.4% 2.4% 0.1% 75.6% 8.2% 0.2%
17/12 8/4 -0.599 22.6% 1.8% 0.0% 77.4% 6.9% 0.2%
6/1 8/4 -0.602 23.4% 2.6% 0.1% 76.6% 9.3% 0.3%
7/2 8/4 -0.622 22.1% 2.2% 0.1% 77.9% 8.4% 0.2%
17/8 -0.629 20.5% 1.1% 0.0% 79.5% 4.9% 0.1%
17/12 7/3 -0.637 22.4% 2.1% 0.1% 77.6% 10.3% 0.4%
8/3 17/13 -0.639 22.7% 2.3% 0.1% 77.3% 11.1% 0.5%
|
|
BS:
|
White's board is now better than Blue's,
so Blue's aggressive days are in temporary abeyance.
Making the ace (4/1) may minimize Blue's danger
if he is unable yet again to safety his blot on the 15-point.
|
|
RM:
|
The hit last time would have paid off,
but Blue shouldn't be tempted.
White's board is the stronger now,
so hitting is even less advisable.
Blue can't bring his own blot to safety,
so covering the 1 point is the best play
to maintain his positional advantage.
Blue White
Candidate Plays Equity Win G/BG BG Win G/BG BG
4/1 +0.483 70.8% 9.6% 0.3% 29.2% 3.1% 0.1%
6/4 6/5 +0.413 68.4% 9.1% 0.3% 31.6% 4.6% 0.2%
|
|
BS:
|
White keeps gaining ground in the race by rolling big numbers;
here again he has the chance to put the checker
on the 17-point out of harm's way.
But is sixteen pips too much of a race deficit to overcome?
Now that Blue has made his ace-point,
the risk of a gammon looms large.
JellyFish's show down play of making an outfield anchor and the full prime
will make the outcome of this game, and probably the match, apparent in
the next two or three rolls. Running 17/8, although further down in
the rollout schedule, gives White the slim hope for a fly shot or the
slimmer hope for boxes. JellyFish, apparently, has some brass mixed in with
its silicon.
|
|
RM:
|
Making the full prime makes running more palatable.
White's now has two chances:
Hit something,
or
avoid being hit and roll large to get back into the race.
White Blue
Candidate Plays Equity Win G/BG BG Win G/BG BG
23/17 6/3 -0.517 28.9% 2.7% 0.1% 71.1% 12.1% 0.2%
17/8 -0.582 23.1% 1.3% 0.0% 76.9% 5.6% 0.1%
23/17 7/4 -0.585 25.4% 3.2% 0.1% 74.6% 12.4% 0.2%
|
|
BS:
|
With this roll, Blue is starting to think "gammon" again.
Unstacking the extra checkers from the 6-point,
making a five-point board and putting White on the bar to boot
make 6/2(2)* downright irresistable.
|
|
RM:
|
With a full prime across the table,
the risk of leaving a shot anywhere is substantial,
so a hit is mandatory to remove the immediate threat.
Shifting points from 4 to 2 has some appeal,
in order to use the other half to bring the outfield checkers closer to home
to provide a better landing spot for the trailer
or to bring it to safety now.
But a five point board
maximizes the pressure on White
and greatly improves the gammon potential for Blue.
Since a gammon wins the match,
that is the correct play.
Blue White
Candidate Plays Equity Win G/BG BG Win G/BG BG
6/2(2)* +0.782 79.8% 20.1% 0.2% 20.2% 1.7% 0.0%
11/9(2) 4/2(2)* +0.706 78.3% 15.5% 0.3% 21.7% 1.8% 0.1%
15/11 4/2(2)* +0.662 76.4% 14.6% 0.3% 23.6% 1.6% 0.0%
15/7 +0.661 77.9% 11.1% 0.2% 22.1% 1.0% 0.0%
|
|
BS:
|
White wants to get a shot.
If White gets a shot,
White wants to have the best possible board awaiting.
7/3 best maintains White's board.
|
|
RM:
|
Simple, straightforward 7/3,
and hope for a break.
White Blue
Candidate Plays Equity Win G/BG BG Win G/BG BG
7/3 -0.697 21.2% 2.5% 0.1% 78.8% 14.6% 0.2%
4/1 7/6 -0.720 19.9% 2.1% 0.0% 80.1% 13.9% 0.1%
7/4 2/1 -0.722 19.9% 2.2% 0.0% 80.1% 14.1% 0.2%
5/1 -0.728 20.1% 2.1% 0.0% 79.9% 15.1% 0.1%
4/1 4/3 -0.782 17.9% 1.4% 0.0% 82.1% 15.3% 0.1%
|
|
BS:
|
While 17/13 15/13 is best for preserving White's board,
it's best to minimize the gammon chances;
bring the back checker forward.
|
|
RM:
|
White is just waiting for a joker at this point.
The gammon risk is effectively nil.
The best play is a tossup between the two plays that make a crossover.
(The other play isn't that far away.)
White Blue
Candidate Plays Equity Win G/BG BG Win G/BG BG
17/11 -0.890 10.3% 0.1% 0.0% 89.7% 9.6% 0.0%
15/9 -0.891 10.4% 0.1% 0.0% 89.6% 9.8% 0.0%
17/13 15/13 -0.894 10.4% 0.1% 0.0% 89.6% 10.2% 0.0%
|
|
BS:
|
Crossovers help stave off the gammon.
Moving the back checker forward again gets the nod.
|
|
RM:
|
White's timing is adequate, so there is no reason to waste pips inside.
Moving either outfield checker is fine.
White Blue
Candidate Plays Equity Win G/BG BG Win G/BG BG
15/9 -0.733 14.1% 0.1% 0.0% 85.9% 1.7% 0.0%
11/5 -0.734 14.1% 0.2% 0.0% 85.9% 1.8% 0.0%
11/7 15/13 -0.739 13.9% 0.1% 0.0% 86.1% 1.8% 0.0%
15/11 5/3 -0.772 12.7% 0.2% 0.0% 87.3% 2.8% 0.0%
15/11 6/4 -0.774 12.5% 0.3% 0.0% 87.5% 2.8% 0.0%
11/7 6/4 -0.781 12.3% 0.2% 0.0% 87.7% 2.9% 0.0%
11/7 3/1 -0.782 12.3% 0.3% 0.0% 87.7% 3.1% 0.0%
6/2 15/13 -0.783 12.7% 0.3% 0.0% 87.3% 4.0% 0.0%
11/7 4/2 -0.784 12.3% 0.3% 0.0% 87.7% 3.3% 0.0%
11/7 5/3 -0.785 12.2% 0.3% 0.0% 87.8% 3.2% 0.0%
5/1 15/13 -0.787 12.5% 0.3% 0.0% 87.5% 3.9% 0.0%
15/11 4/2 -0.789 12.0% 0.2% 0.0% 88.0% 3.1% 0.0%
6/2 11/9 -0.789 12.3% 0.2% 0.0% 87.7% 3.8% 0.0%
15/11 3/1 -0.790 11.8% 0.2% 0.0% 88.2% 2.9% 0.0%
5/1 11/9 -0.795 12.0% 0.3% 0.0% 88.0% 3.7% 0.0%
|
|
BS:
|
This 42 can be played safely (6/4 6/2),
or it can be played aggressively (7/3*/1),
risking a 43 return shot by White and "repeater" sequences
where awkward large numbers are rolled (66, 55, 44, and 65).
JellyFish, ever the seeker of the gammon, prefers the hitting play.
|
|
RM:
|
Pick-and-pass offers better (although still small) gammon chances,
and only the 43 return shot immediately,
plus a direct shot after 66, 65, 55, or 44 next roll.
Since a gammon wins the match, it's worth the risk.
If the gammon didn't matter, playing safely would be better,
even with some ugly splits next roll
(64, 61, 54, or 51 leaving direct shots;
42 followed by 61 or 43 for White;
or 41 followed by 61 for White).
Blue White
Candidate Plays Equity Win G/BG BG Win G/BG BG
7/3*/1 +0.653 81.4% 2.7% 0.0% 18.6% 0.3% 0.0%
6/2 6/4 +0.646 82.1% 0.7% 0.0% 17.9% 0.3% 0.0%
|
|
BS:
|
Blue has no checkers off,
but White risks little to "stay back"
and cross his fingers that Blue's bear-off will not be smooth.
|
|
RM:
|
Either 6 from the outfield is fine.
White still hangs on by a thread,
waiting for a last-ditch shot
or large enough doubles to put him back in the race.
White Blue
Candidate Plays Equity Win G/BG BG Win G/BG BG
Bar/22 9/3 -0.735 13.5% 0.0% 0.0% 86.5% 0.6% 0.0%
Bar/22 11/5 -0.736 13.4% 0.0% 0.0% 86.6% 0.5% 0.0%
|
|
BS:
|
Here is a situation that requires Blue to think in the long term.
Blue would like to start taking checkers off already!
Because Blue has to assume that White will be around
for at least one more roll,
Blue must minimize his chances for leaving a shot.
White's game-winning closed board has to be uppermost in Blue's thinking now;
Blue will either have two checkers left on the five-point
(making 62, 52, 42, 32, 61, 51, 41 and 31 -- that's sixteen numbers --
potential game losers) or on the four-point
(making 61, 51, 41 and 31, that's half as many game losers).
Eight additional game losers isn't worth the minimal gain
that taking two checkers off realizes.
|
|
RM:
|
Blue got the doubles instead.
If he takes no checkers off,
only eight numbers leave a shot
(then eleven hit for White -- a 6.8% parlay).
The gammon is already a pipe dream,
so leaving eight more bad numbers in order to take off the first two checkers
isn't worth the risk of doubling White's chances.
Blue White
Candidate Plays Equity Win G/BG BG Win G/BG BG
6/2(2) 5/1(2) +0.816 90.8% 0.1% 0.0% 9.2% 0.0% 0.0%
6/2(2) 4/Off(2) +0.760 87.7% 0.5% 0.0% 12.3% 0.0% 0.0%
|
|
BS:
|
White hopes for one of Blue's eight bad numbers.
|
|
RM:
|
The thread supporting White is rather thin.
The next sequence is very likely the critical one,
so any play that doesn't break White's board is fine.
If White hits, he can cash immediately.
Even if he opens his board, he's back in the game.
If he doesn't hit, his racing chances are pretty meager.
White Blue
Candidate Plays Equity Win G/BG BG Win G/BG BG
9/4 5/3 -0.822 8.9% 0.0% 0.0% 91.1% 0.1% 0.0%
9/2 -0.823 8.9% 0.0% 0.0% 91.1% 0.1% 0.0%
9/4 3/1 -0.853 7.4% 0.0% 0.0% 92.6% 0.1% 0.0%
9/4 6/4 -0.869 6.6% 0.0% 0.0% 93.4% 0.1% 0.0%
6/1 9/7 -0.872 6.5% 0.0% 0.0% 93.5% 0.2% 0.0%
6/1 5/3 -0.874 6.4% 0.0% 0.0% 93.6% 0.3% 0.0%
6/1 6/4 -0.883 6.0% 0.0% 0.0% 94.0% 0.3% 0.0%
6/1 4/2 -0.895 5.5% 0.0% 0.0% 94.5% 0.5% 0.0%
6/1 3/1 -0.898 5.4% 0.0% 0.0% 94.6% 0.6% 0.0%
|
|
|
Blue went on to win the match, 11-6.
|
The game was recorded on tape and transcribed by Richard McIntosh.
Rollouts were made by Richard McIntosh,
using
JellyFish Analyzer 3.0.
Rollout results show equities
for the player on move.
Candidate plays were better than or
within 0.100 equity of the actual plays,
evaluated at level 7.
Parameter
values for rollouts on moves were:
- level 5
- 7776 games (36x216)
- horizon 7
- seed 720
Standard deviations of equity estimates were between 0.003 and 0.008,
typically 0.004.
Parameter values for level 5 rollouts on cube decisions were:
- level 5
- 23328 games (36x648)
- full game
- seed 720
- settlement limit 0.550
Parameter values for level 6 rollouts on cube decisions were:
- level 6
- 216 games (36x6)
- full game
- seed 720
Copyright © 1996-2010
BackGammon By the Bay
