Problem of the Week #70: August 1


Cash game, center cube.




Should Black double? Should White take if doubled?


Note: All ‘cash game’ problems assume the Jacoby Rule is in effect. That is, you can’t win a gammon unless the cube has been turned.

Race is even (if I counted right)... other features of the position are roughly even, without going into great detail, although Black's more compact home board probably gives him a slight edge based on the static features of the position... so Black's edge (and a possible double) would be based mostly on immediate attacking chances.

Volatility is high, and there are some double-hit numbers, but the take is crystal-clear, so IDK, I just don't see myself reaching for the cube. There don't seem to be many huge market-losing sequences, and I'd just as soon cash or offer an efficient double if things went my way.

No double/take.

No double/beaver

Double here seems very marginal. Black is a bit ahead in the race and has a few threats. However, a gammon seems very unlikely. White has a very easy take here.

Answer - Double/Take

I really dont like black position. Race praticalluy even, 0 flexibility, if he doesnt hit will be in serious trouble. I would take white actually...
NO DOUBLE / BEAVER

I agree that white's take is pretty clear here. However, I find it hard to believe that white is better than 50% with black on roll here. Black has lots of really good rolls - 61, 11, 33, 55 ,66, 52, 32, 51, 53 come to mind. As white, I am not beavering.

So that leaves double/take or no double/take. Looking at blacks good moves, there are a handful of market losers when black hits then white dances. Overall though, black still has a long way to go to get home and I would wait for a more efficient double.

No double/ take.

Well, cube ownership is pretty huge in this type of position, so White can be a dog cubeless and still have a correct beaver (which you may have taken into account, it wasn't clear to me from your post).

That said, I agree that we aren't particularly close to beaver territory, since Black's immediate initiative is a larger factor than cube ownership IMO.

I see this somewhat like ferrengi wrote, except that, with 11 men in the attack zone, if things go his way, Black will win relatively many gammons. Hence pretty clear double.

Double / Take

good clarification- I find myself agreeing with you for alot of these problems

Gammons seem really asymmetric here. It's easy to see black losing (although I can't believe he's anything but a mild favorite now), but it's far easier to see him landing a gammon that white. I have no idea what the right play is here- there's a lot of game to be played, certainly a lot of volatility if black doesn't crush fast. If gammons were live (like 17-away/17-away), I wouldn't be inclined to double because I could already cash in somewhat. As is, I don't think it's a trivial ND/take (and I'll **** myself if it's actually ND beaver.. or double/drop for that matter). The gammons and the immediate crushes make a real case for doubling.

IDK/take.

"Who are you kidding?!" was my initial reaction to Problem 70. Unless this had been posed as a problem, I never would have considered doubling! Black is certainly ahead, but is now out of time, and badly stretched. His best chance is to launch an attack. When it works, there will be some gammons in his bag.

To get a better feel for the position, I layed out all twenty-one rolls, and determined the most aggressive attacking play for each. Not all the rolls were best used for blitzing, but there were still many more that I first suspected.

Very Strong Plays
Hit Once and Make Five-Point Board
55 -- 13/3*, 8/3, 6/1

Hit Once and Cover Inside
52 -- 5/3*/1
32 -- 6/3*/1
51 -- 23/22*, 6/1

Hit Once, Make Inside, but Leave Inside Blot
33 -- 6/3*(2), 23/20(2)
53 -- 8/3*, 6/3

Hit Twice Outside
61 -- 23/22*/16*

Switch Points To Hit Twice, but Leave Inside Blot
11 -- 23/22*(2), 4/3*(2)

Strong Plays
Switch Points To Hit Once, but Leave Inside Blot
22 -- 5/3*(2), 13/11(2)
42 -- 5/1*, 5/3

Hit Inside and Out, Leaving Two Inside Blots
31 -- 23/22*, 6/3*

Escape Back Men
66 -- 23/11(2)

Risky Hits -- Retaining the Initiative
Hit Once Outside
21 -- 23/22*, 6/4
43 -- 23/16*
41 -- 23/22*, 8/4

Hit Inside, Leaving Two Inside Blots
64 -- 13/3*
54 -- 8/3*, 8/4

Based on all these chances, let's give Black a double. Although White's anchor won't protect him against a closeout and gammon, together with his own three-point board, it will give him some counter-play.

My solution: Double, take.

For the Record
I am so often wrong that I like to post my record in these messages. It's kind of a truth-in-advertising thing. Grunch: I have been answering these problems without the use of a bot, and before checking the excellent solutions of others, since Problem 28. Including the tossups 39a and 62, my record at this writing is 46% correct.
23 Correct: 28a, 29, 30, 32, 35, 36, 38, 39a, 39b, 42b, 43, 44, 45, 47, 48a, 48b, 50, 52b, 53, 57, 62, 66, 67. 27 Incorrect: 28b, 31, 33a, 33b, 34, 37, 40a, 40b, 41, 42a, 46, 48c, 49, 51, 52a, 54, 55, 56, 58, 59, 60, 61, 63, 64, 65, 68, 69.

My gut reaction: No double/beaver

Double drop.
Regarding 52, I prefer 23-16* , and 1-1 : 23-22* 4-3*(2), 6-5
Rolls left out: 44: 13-9 (2) 6-2 (2) puts even more pressure on whites back man. 65 and 63 are ugly 42: am I really thinking 5-1, 5-3*?

Even though black's 6's aren't great, and white has a 6 to cover his 3 point, white has two dead checkers, and 3 blots right now, and not too many rolls to not leave direct shot next roll.

8/3*/1 looks much stronger to me. It puts a ton of pressure on white while he has two exposed blots. Should lead to more wins and quite a few gammons (assuming a double/take somewhere)

You can't be serious about double/drop, it is barely even a double!

New to the backgammon section but just thought I'd drop my 2 cents in. Long time poker player and amateur gammon player but been playing for 15 years. Played a little online back when US was allowed on partygammon and did very well at $5 and $10 best out of 3 games.

Anyways, I would DOUBLE/TAKE

Black is actually behind 3 pips (131 to 128) but with it being our roll and the average roll being 7, we are technically ahead by 4 pips after taking turn into account.

Look a little deeper into the math.

If we roll any combination with a 1 in order to 23/22.

1/6, 1/5, 1/4, 1/3, 1/2, 1/1, 2/1, 3/1, 4/1, 5/1, 6/1. This is 13 combinations. Coming in on the next roll white will have the same 13 combinations to hit us unless we hit 1/5. This is a 30.56% chance to roll a 1.

Of all these ways to roll a 1, the 30% chance that we will do so, is helped by the fact that if we roll a 1/5 or a 1/6, we will either hit our opponent twice, or we will hit and cover our only open man at home. There is an 11% chance this will happen. We could also roll one of the other 4 combinations for a 7, brining our total chance to hit our opponent on the next roll up to 41.67%. Add in the 4/2 combinations to hit and cover at home and the picture becomes clear.

Total chance to hit opponent once on next roll (not covering) = 30.56%
Chance to hit twice = 5.56%
Chance to hit and cover = 5.56%
Roll 4/2, hit and cover at home = 5.56%

47.24% chance to hit opponent on next roll in some fashion.
16.68% chance to own him in the face on the next roll.

I would opt to DOUBLE/TAKE with these odds. Still a close call, but I like my chances, if I don't win a gammon, I still stand a very good chance to win the doubled bet depending on the next roll or 3.

For those analyzing rolls, the issue is not whether Black can hit, but what his winning chances are after the hit.

I'm not going to try to analyze every scenario, but in most scenarios either a loose blot on the Ace-point and/or a White runner on the 3-point (or a Black blot on that point), plus potential loose blots in the outfield, means that White will have useful entering rolls, and if he counter-punches we're gonna be real glad that we didn't double.

If White enters but not constructively, now we can offer a very strong "efficient" double, making the best use of the cube at that point.

Even if White dances, I'm not going to be hugely disappointed as Black to just cash the game, since there will still be a fair amount of work to do to bring home the win (or gammon).

The only "big" market losers, where I really wish I had doubled, involve double hits followed by a dance, and there aren't many of those.

Meanwhile if Black whiffs to begin with, he's very glad he didn't double.

Anyway, lecture over.

Cliffs: Is Black losing his market by a big margin when things go his way? If not, it's probably better to wait to turn the cube.

Agree with this.

Sorry, about all the counting. I guess I started it. But take a look at the thirteen rolls I listed under the heading "Very Strong Rolls." Ten of them make a new inner board point while putting one or two checkers on the roof. The other three put two men on the bar against a three-point board. Against a four-point board, White will have only 5-to-4 odds of entering one man. When he fans, Black gets to munch on White's other blot(s). In the variations where two White checkers are in the air, White has at most 9 rolls in 36 to enter both. Even when he does, Black gets to shoot at the third.

As you say, White certainly has lots of counter play. Black could easily find himself dropping a cube -- or losing a gammon -- in just a few rolls.

But thirteen rolls is more than a few jokers. In this highly volatile position, it's some serious market-losing potential.

Doesn't seem like time to double yet to me. There's some good rolls that might lead to a drop and on those occasions you may regret not doubling. But maybe you miss and white is offering you the cube before his 2nd roll. My guess:

Double/Take

Hi,

what a beautiful problem, I had to put it aside for awhile to have it sinked in.
The position is a sort of blitz, it's not a pure blitz.
In a pure blitz the benchmark position for a double/take is a 3 vs. 1 board position, 1 on the bar, and 2 attacks on the second blot.

Here white is doing better in 3 ways:

1. It's an anchor-blitz, with a good anchor on the bar-point.
2. There's no checker on the bar.
3. The innerboards are 3 vs. 3.

So black needs compensation for these 3 points will it again be a Double/Take.

A direct shot and 3 attackers instead of two is generally a good compensation for the missing checker on the bar, this compensation is there.

The only extra compensation black has is one more attacker (the 4th) and the better board (654 instead of 631), this is by far not enough compensation for two lacking inner board points, and the bar point anchor.

Conclusion it's No Double.

Something more: after 14,13,12 for example 12: 23x22 6/4, a nice question is if white has a re-action double?

Black doesn't have an anchor then, and he has a direct shot and 3 attackers instead of two to compensate for the missing bar-blot, not enough for a re-action double, but after this moves (12,13,14) as much as 23 out 36 numbers will start a counterblitz.

Another thing about the discussion about volatility and market-losers, aren't those two words that describe the same situation: when the position is volatile it means the chance is big that you will lose your market.

So to put it in the perspective of this problem: Black might consider a double because there are a lot of market-losers, but he shouldn't because there are to many market-switchers.

greetings k.

I said double/drop in my above post, and probably got it wrong. I'm hoping it is a instance of losing the battle, but not the war. Over the board, I rather have an analysis plan that gets most of decisions correct, and a few wrong, than an over my skill level plan that I could really screw up. In this position, I see white race as even. I see white has worse- prime, board, pureness, getting gammoned, play options. These are basic merits of a position, and I will be interested in Robertie's solution.

Lol. I meant "No Double/Take".