Problem of the Week #82: November 1


Cash game, White owns the cube. Black on roll.





Black to play 5-3.


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.

since we are trying to run home as quickly as possible here i would play 21-13. At this stage I think the midpoint is superior to white's 4 point.

21 - 13 cheers

Black has a very strong position here. His most important objective right now is to make the bar point as that would substantially increase his gammon chances. If white is lucky enough to roll 6-3 or 6-4 next roll, black would like to have an opportunity to hit from the 13 point. For these reasons, I think the best play is 21/13. It gives black a better chance to make the bar point on a later roll, rebuilds the midpoint which will make bringing home the back checker easier and gives black a chance to hit on the 10 or 11 point.

Other possibilities: 24/21, 13/8 doesn't accomplish anything important. I'm not even sure if this is safer than 21/13

9/1* - This play appears to have some merit because it makes it more difficult for white to escape on the next roll. However, it also doesn't accomplish much long-term. If white hits it may give him a chance to get back in the game. Even if he doesn't, if he comes in on the 3 point and the 9 point is now broken, he also has better chances to escape. Without a good threat to close out, I don't like the hitting play.

Answer - 21/13

Grunch.

Race is 135-134. After moving, we are leading 7 pips with a better blocking structure against the straggler. Unfortunatly, we have to care for the rear position with three blots. Because the race is in our favour, the main issue is to connect the three blots to the main troops.

a) 24/21 13/8. This will clean up all blots. But the anchor is a lone fort, far from home. If we don't throw doubles, going home can be difficult. With 13/8 we have no outfield control and little chances, to make the bar. So escaping the orphan is a work, white now can do.

b) 21/13. This will connect the position and establish outfield control. Since white has no blocking structure, escaping the last man will be easier for black, compared to white. By now, the left man on the ace is even an asset, then a liability, because he is more a sniper in the unbuild white position then an orphan. The rebuild midpoint will also give more opportunities to make eventually the barpoint. This is the play which supports more possibilities and game plans, so this must be the winner.

21/13 will be my choice.

Make the midpoint (13 point).

I'm always scared when they look too easy, but the job is to escape your guys while hopefully keeping white's checker blocked.. and hitting another one would be nice as well. I see no way to play 53 usefully at the front of the board, so it's 21/13 or 24/21 13/8. Making a 21 anchor doesn't really help escape those guys. You're going to have to roll a big number not to leave a double shot, and some lucky doubles, fast, not to leave single shots. And he has a whopping 2-point board, 5 guys on the wrong side of the board, and stacks, so you don't really need an anchor for defense anyway. 21/13 actually does escape one of the back men, and in addition, the blot on 24 covers his infield, so white can't just dump a guy to the 3 or 1 safely, and the 13 anchor covers the outfield so 63/64/66 aren't clean escapes. 21/13 has all the upside here IMO.

21/13.

My problem with 21/13 is that the race is as close as it is. Black's advantage is not in racing but in the strength of its board, and the weakness of White's. Black *wants* to create contact. Perhaps 21/13 is still right since I'm not seeing that many other options. 21/16 13/10 popped into my head, though it seems crazy it brings a builder down for the prime, maintains some continuity, and plays to contact. Just a thought.

The two obvious choices are 21/13 and 24/21 13/8, and 21/13 seems much the better move. So much better, in fact, that I have to wonder if

1. 24/21 13/8 really is the better move, but for some reason none of us have figured out yet;

2. there is some other non-obvious move that's better than the two obvious moves;

or

3. 21/13 is easily best and this problem is really just a test to see how prone we are to overthinking things.

I can't see 1. Sure you make an anchor and safety all blots, but with white's scraggly position I'm not scared of having lots of blots around, and you give up too much -- removing a builder for the bar, stacking on the 8, isolating the two back men.

If 2 then maybe you're supposed to do something like 9/4 21/18, creating another builder for the inner board (and maybe looking to blitz) and connecting the three stragglers. Your 9 pt isn't directly blocking white's 1pt blot, you still have 3 builders for your bar, and getting hit's not that bad because with the rear connectivity there's a good chance you can come in and make white's bar or your midpoint. (Actually this would be my 2nd choice over 24/21 13/8)

But 21/13 seems so simply better that I figure it's a test to see how much a position can be overthought. A test I failed, of course, given the length of this post.

The hit doesn't make sense at this point and anchoring would slow us up and let white catch up and possible start his own prime.

21/13

I think the best play by far here is 21/13 re-making the midpoint and leaving just a single checker to scramble home from the 24 point against whites weak structure. Anchoring with 24/21 and playing 13/8 (the only reasonable 5 after 24/21) is the wrong idea. While it looks like buttoning up and playing safe, its just the opposite. An anchor at this point without a midpoint lacks two critical Robetie-isms for checker play. Connectivity and robustness.

21/13

I'm tired, just checked in here before going to bed, and I don't like to stare at these positions for any longer than I would in a real game. I considered 3 moves and settled for

21/13.

The race is practically even. Black has a better structure, but how can he take advantage of it?

By trying a blitz with 9/1*? If he's hit back, all of a sudden White will have a nice lead and Black will be hard pressed to hit again. At some point, if White escapes his back man, he might have a redouble along the way. On the other side, if White fans, Black could go for the kill since gammons are activated.

By "equalizing" with 21/13? That play, although it seems passive, has some merit. It reestablish the midpoint which could be really useful and secure a back man at the same time. It prevents 6-6 for White, so unless White rolls a 6-5, his back man will still be under a lot of pressure next roll.

What else? 24/21 13/8? Is making an anchor really a priority at that point? With White having a rather weak board, I don't think so.

What about 9/6 9/4? It would prepare for an eventual blitz with additional builders if White tries to move the back man forward. But it could be too late. And Black would be dismantling a nice broken 6-prime.

In the end, I hesitate between 9/1* and 21/13. Try the blitz and be greedy or play quietly and wait for a better roll to either blitz or prime? If Black goes with 9/1* and White replies with 1-1, 2-2, 2-1, 4-1, things could change quickly. Is it worth it? I say no.

My pick: 21/13

Not really, since from the 13-pt, that checker is already a builder for the prime (for the bar point). The 4-pt is already made and there's no value in making the 10-pt, so 13/10 isn't really useful here.

Good point, I hadn't factored this in my analysis. Makes 9/1* even less appealing.

9/1*. Like a blitz attempt with our structure. Lots of bad rolls for white. Downside not huge.

Funny. I didn't even think about an attack because of my 3 blots. Hitting with 9/1* would add additional two blots. So 5 blots in return to shoot on, is a no no.
But suppose the spare is on the six point. Button up with 21/24 and hit with 6/1*. Only aces left. And if white is unlucky, he has to give a second blot to shoot on.
Funny. 1 checker 1 point different, and a hit can be a move to be considered.

if the answer is not 21/13 I am gonna bang my head against the wall

come on, other moves would leave the position pretty disconnected and 21/13 is also the only way to regain central control.

The only problem with hitting is white has the cube

Even with a centered cube, hitting is a triple whopper. If you consider here leaving five blots, it is a different universe. Too loose, l'autrec.
But yes: cube leverage and volatility is on the side of cube ownership. But i'm carrying coals to newcastle.

There are options enough in Problem 82 to make it worth listing them before trying to choose the best. The race is a dead heat, with Black trailing by just 1 pip, 135 to 134.

• Play A: 21/13 -- Remake the midpoint.
• Play B: 9/6, 9/4 -- Prepare to attack next turn.
• Play C: 9/1* -- Launch an attack now.
• Play D: 24/21, 13/8 -- Make an advanced anchor.
• Play E: 24/21, 9/3 -- Make an advanced anchor (alternate).

The plays that give White complete freedom of movement to hit a blot on the midpoint seem wrong. Unless White is on the roof, Black should pick up his blot on the midpoint, or else cover it. That eliminates plays B and E.

Were White stronger, remaking the midpoint would be a priority. Here, however, Black has great control of all boards, while White is all stacked up. Black's blot on the twenty-one point threatens many of the moves White might make into his outer board. But if Black remakes his midpoint, he'll have to give up that coverage. If I were White, I would be happy to see Black give me some space to play in this tied race. So, Black should reject play A.

Making an advanced anchor, play D, is better. It's a move that let's White play behind Black with certain rolls, but retains outfield coverage. Because the game is a near tie, this "safe" play will be favored by many. It does have the rather large downside of isolating Black's back men.

The alternative, play C, is not as safe, but gives Black a connected position. Of course, it has its own downside: if Black is hit, he'll fall back twenty-four pips. The race would then be suddenly one-sided, and with three checkers back for Black, he'd be forced to discard any racing options. Still, this is the play I prefer. When White misses, Black will have significantly improved his position. Even when White hits, there are many awkward rolls that will force him to expose a blot for Black to hit on the return.

My solution: 9/1*.

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. My record at this writing is 53%.

Hi Bill,

again a very nice problem.

I like 24/21 since it gives black a very good 4-anchor, especially in this situation with the two killed checkers on the 2-point, the hole on the 5-point,
and the stacked white position, this anchor alone will give black already 30% defense equity.
All white moves except 63,64,65 and D6 give black also forward changes, the anchor now makes it also possible for black to blitz relatively safe with all 4's and 7's.

I guess you meant 24/21 8/3.

I read your analysis carefully because so far, you are the only one that has put forth any sort of argument for 9/1*
Even assuming that white misses, I still don't see how this improves black's position (let alone significant improvement). Can you please elaborate?

White is not significantly outboarded. We have a 3 point board against a 2 point board. Only one blot to shoot, and when we shoot, leaving 5 blots in return? Haaalloooo? This is not tailoring the risk according to the demands of the position. This is throwing away a position. With the cube on the other side it is russian roulette with 3 bullets.