Open your mind - Page 10 - Backgammon Forum - Discuss Backgammon Strategy, Clubs and Books

5/2*
If black plays safe, white has a decent chance to get back in the game as black will have to break 22 and/or 20 point soon with terrible timing & no safe anchors to help him get home. The 5/2* move keeps white off balance gives black a chance at a lucky blitz, and gives black a very productive 1 to play on his next roll if he isn't hit on the 2 point. Even if black is hit on white's next roll that shouldn't be all that terrible.

Quote

05-24-2015 , 03:05 AM

#228

Alpha Fish

Pooh-Bah

Join Date: Aug 2008 Posts: 3,838

having no 5 can't be all that great I play 20-18 first then hit

Quote

05-24-2015 , 09:42 PM

#229

uberkuber

veteran

Join Date: Jul 2009 Posts: 3,043

Quote:

Originally Posted by Alpha Fish

having no 5 can't be all that great I play 20-18 first then hit

It ain't that bad either, because you would only have one legal 5 anyway, not like you would have to crush your board badly.

Quote

05-25-2015 , 07:48 AM

#230

yogiman

adept

Join Date: Jun 2012 Posts: 784

Code:

    1. 20/18 3/2*                   Eq.:  +0,628
       0,692 0,426 0,022 - 0,308 0,086 0,004 CL  +0,741 CF  +0,628
      
    2. 22/20 3/2*                   Eq.:  +0,537 ( -0,090)
       0,668 0,393 0,020 - 0,332 0,082 0,003 CL  +0,664 CF  +0,537

I thought that by playing 20/18 he easily gets in reach of the third white blot on a next turn, but this turns out not to be relevant. Moving 9w to 19:

Code:

    1. 20/18 3/2*                   Eq.:  +0,440
       0,661 0,333 0,006 - 0,339 0,090 0,004 CL  +0,565 CF  +0,440
      
    2. 22/20 3/2*                   Eq.:  +0,342 ( -0,098)
       0,632 0,299 0,006 - 0,368 0,085 0,003 CL  +0,481 CF  +0,342

5's do play a role. Moving 15w to 14 decreases the relative value of 20/18:

Code:

    1. 20/18 3/2*                   Eq.:  +0,728
       0,716 0,465 0,025 - 0,284 0,086 0,004 CL  +0,832 CF  +0,728
      
    2. 22/20 3/2*                   Eq.:  +0,675 ( -0,054)
       0,703 0,436 0,023 - 0,297 0,076 0,003 CL  +0,785 CF  +0,675

The other half of the story is likely that black is ahead and probably will get more ahead, but has a very bad connection in case white has entered his checkers, and by 20/18 creates the flexibility for making a point closer to home.

Quote

05-27-2015 , 06:17 AM

#231

yogiman

adept

Join Date: Jun 2012 Posts: 784

Position ID: mJ8ZhEC3XQCECQ Match ID: QykFAAAAAAAE

White - Pips 139

Black - Pips 108
Black to Play 3-1
Created with www.BGdiagram.com

Quote

05-27-2015 , 08:17 AM

#232

bleep69

grinder

Join Date: Sep 2008 Posts: 483

Our only options are 16/12 or 16/13 5/1. I prefer 16/13 5/1 simply so that we don't get hit. Wish I had more to say.

Quote

05-27-2015 , 09:53 AM

#233

Kamba

enthusiast

Join Date: Feb 2015 Posts: 93

We could also play our full move from the 5 point but i'm not a fan of that one. I don't think I would mind being hit, I want to make my 5 point, so I'll leave that guy there and hope for the best.

I move 16/12

I think this gives me a better chance to hit and a better chance to cover my 5 point. If he hits me I'll call it re-cycling and cross all my fingers.

Quote

05-27-2015 , 02:07 PM

#234

peachpie

centurion

Join Date: Aug 2014 Posts: 182

I will try 5/2 4/3. I am ahead in the race, so play safe. This also keeps the men on 16 and 22 connected. Plus there are more shots at the 10 point from 16 than from 13 or 12.

Quote

05-28-2015 , 05:04 AM

#235

yogiman

adept

Join Date: Jun 2012 Posts: 784

White - Pips 139

Black - Pips 108

Code:

    1. 16/13 5/4                    Eq.:  -0,178
       0,487 0,154 0,002 - 0,513 0,113 0,004 CL  +0,014 CF  -0,178
    
    2. 16/12                        Eq.:  -0,313 ( -0,134)
       0,444 0,164 0,003 - 0,556 0,147 0,006 CL  -0,098 CF  -0,313

Moving 4 and 1 to 20:

White - Pips 139

Black - Pips 143

Code:

    1. 16/13 5/4                    Eq.:  -0,527
       0,388 0,107 0,003 - 0,612 0,156 0,005 CL  -0,277 CF  -0,527
    
    2. 16/12                        Eq.:  -0,593 ( -0,066)
       0,375 0,127 0,004 - 0,625 0,224 0,009 CL  -0,351 CF  -0,593

Surprisingly, even with about even pipcounts killing a checker is preferrable.

Also with a safe defence:

White - Pips 139

Black - Pips 145

Code:

    1. 16/13 5/4                    Eq.:  -0,552
       0,382 0,104 0,002 - 0,618 0,151 0,005 CL  -0,287 CF  -0,552
     
    2. 16/12                        Eq.:  -0,579 ( -0,027)
       0,382 0,122 0,004 - 0,618 0,195 0,007 CL  -0,312 CF  -0,579

Moving 20-point to 24-point:

White - Pips 139

Black - Pips 116

Code:

    1. 16/13 5/4                    Eq.:  -0,468
       0,405 0,129 0,002 - 0,595 0,166 0,009 CL  -0,234 CF  -0,468
      
    2. 24/23 16/13                  Eq.:  -0,478 ( -0,009)
       0,403 0,146 0,003 - 0,597 0,203 0,015 CL  -0,263 CF  -0,478

The hitting of a second blot does play a role, but also without it 5/4 is still best. Moving 19w to 10:

White - Pips 148

Black - Pips 108

Code:

    1. 16/13 5/4                    Eq.:  -0,418
       0,414 0,130 0,001 - 0,586 0,146 0,006 CL  -0,193 CF  -0,418
      
    2. 16/12                        Eq.:  -0,504 ( -0,086)
       0,389 0,139 0,001 - 0,611 0,161 0,006 CL  -0,248 CF  -0,504

The answer is simple and obvious. It shows that the bot preferrably kills the checker in all circumstances, apparently because racing tactics outweigh strategical value with black's shifted homeboard.

PS Corrective remarks are welcome, also from Taper_Mike (in problem 28 I was mistaken myself)

Quote

06-04-2015 , 09:26 AM

#236

yogiman

adept

Join Date: Jun 2012 Posts: 784

Position ID: B7eDASjYbWCCDA Match ID: cIkOAAAAEAAE

White - Pips 132

Black - Pips 154
Black to Play 5-3
Created with www.BGdiagram.com

Quote

06-04-2015 , 10:19 AM

#237

peachpie

centurion

Join Date: Aug 2014 Posts: 182

I would need a pretty good alternative to not point on him while making a four point board. So what are the alternatives?

20/15 5/2* is the first one I consider. White's board is bad so the blot is not dangerous. The downside is that the 15 point doesn't look so great: it is disconnected from the 22 anchor, and the outfield coverage is redundant with the midpoint.

If I am going to hit loose, I like 15/10 5/2* better, bringing another checker into the attacking area.

15/7 is my next thought. With three dead checkers, white's position is pretty bad. This move preserves black's positional advantage and patiently waits for white's slow death. The idea here is that it may be better to let white keep moving, and eventually crashing.

15/10 13/10 has some appeal also, duplicating white's aces and inviting him to lose connectivity by breaking the midpoint. Similarly 20/15 13/10. But considering duplication leads me back to ...

Eh, I just spent a fair amount of time off my clock and I am still just playing 7/2* 5/2. The duplication of white's aces and treys clinches it for me.

Last edited by peachpie; 06-04-2015 at 10:25 AM.

Quote

06-04-2015 , 11:10 AM

#238

peachpie

centurion

Join Date: Aug 2014 Posts: 182

hmm, this problem seems harder the more I think about it. As it is, white's 4, 5, 6 are all pretty bad. But breaking the bar point gives him a couple good 6s, plus 34. Maybe a containment strategy is better after all.

Quote

06-04-2015 , 05:10 PM

#239

Robertie

old hand

Join Date: May 2004 Posts: 1,954

My first impression is to ask "Why is the cube in the middle here?" Black had a monster double last turn and it looks like a big pass with White's three dead checkers on his ace-point.

If we put the cube on White's side, I like 20/15 13/10 with a 5-3. White has a hard time playing rolls with 3s, 4s, and 5s and his position can quickly collapse if we leave him alone. Making the 2-point is reasonable but we don't have a lot of checkers to follow up a blitz, so I'll keep the prime for now and just let White move.

Quote

06-05-2015 , 02:19 AM

#240

yogiman

adept

Join Date: Jun 2012 Posts: 784

My apologizes, this problem was badly prepared. I was ill and had no energy to search for new problems. Likely I had been trying out some cube position, and forgot about the cube. The equity difference is a lot less impressive with the cube on 2:

Code:

    1. 20/15 13/10                  Eq.:  +0,536
       0,720 0,263 0,012 - 0,280 0,077 0,002 CL  +0,635 CF  +0,536
      
    2. 7/2* 5/2                     Eq.:  +0,496 ( -0,041)
       0,698 0,284 0,011 - 0,302 0,087 0,003 CL  +0,603 CF  +0,496

Quote

06-05-2015 , 02:25 AM

#241

yogiman

adept

Join Date: Jun 2012 Posts: 784

Position ID: sNtVABjMboMBUA Match ID: cIkNAAAAAAAE

White - Pips 142

Black - Pips 139
Black to Play 3-3
Created with www.BGdiagram.com

Last edited by yogiman; 06-05-2015 at 02:50 AM.

Quote

06-05-2015 , 09:41 AM

#242

peachpie

centurion

Join Date: Aug 2014 Posts: 182

After bar/22, I see three primary candidates:

24/21 7/4(2)
24/21 13/10(2)
13/7 13/10

The first two seem insane with all white's attackers. Meanwhile the last seems to be safer while aiming at making the 4 point, but this aim duplicates black's 2 and 6. Nothing seems really appealing, I guess I will try the latter, at least white's 2 are duplicated.

Quote

06-06-2015 , 01:04 PM

#243

yogiman

adept

Join Date: Jun 2012 Posts: 784

Code:

    1. bar/22 24/21 13/10(2)        Eq.:  +0,119
       0,530 0,163 0,011 - 0,470 0,142 0,006 CL  +0,086 CF  +0,119
    
    2. bar/22 24/21 13/7            Eq.:  +0,050 ( -0,068)
       0,513 0,164 0,011 - 0,487 0,162 0,008 CL  +0,032 CF  +0,050
    
    3. bar/22 24/21 7/4(2)          Eq.:  -0,026 ( -0,144)
       0,492 0,167 0,012 - 0,508 0,157 0,006 CL  -0,000 CF  -0,026
     
    4. bar/22 13/7 13/10            Eq.:  -0,148 ( -0,267)
       0,460 0,149 0,010 - 0,540 0,163 0,008 CL  -0,092 CF  -0,148

(I can remember that I presented this sort of problem not so long ago, but beware of the “six-factor”.)
The suggestion is that being on the bar while white has made a 5-prime is not bad at all, but is it?

White has played a 62 after black's bar/22 13/7 13/10:

Code:

Cube analysis
Rollout cubeless equity  -0,082

White has played a 62 after black's bar/22 24/21 13/10(2):

Code:

Cube analysis
Rollout cubeless equity  -0,234

It costs black 0,15 to be hit & 5-primed, which seems to be bearable relative to the benefits of being not so.

Benefits if white has played the following dice numbers after black's 24/21:
-1's will extend the prime from behind, and if black is not on the 21-point he will be fully blocked.
-63 and 32 hit loose, making black favorite to hit back
-43 leaves a lot of blots for white
-53 gives black an opportunity to spoil white's prime by making the 21-point
Above all, white's hits are good for black's timing.

The relevance of timing is illustrated by the following position. 14w to 20:

Code:

    1. bar/22 24/21 13/10(2)        Eq.:  +0,261
       0,567 0,179 0,012 - 0,433 0,140 0,006 CL  +0,177 CF  +0,261
     
    2. bar/22 24/21 7/4(2)          Eq.:  +0,036 ( -0,225)
       0,518 0,167 0,010 - 0,482 0,177 0,007 CL  +0,029 CF  +0,036
    
    3. bar/22 13/7 13/10            Eq.:  -0,031 ( -0,292)
       0,497 0,144 0,008 - 0,503 0,169 0,008 CL  -0,031 CF  -0,031

24/21 has even increased in importance, though black has 60% chance to extend the prime on white's head, instead of the original 35%. So big is the impact of 6 pips on white's timing.
7/4(2) is really bad, because it allows white to play a high number while white's timing has gotten worse.

But the timing factor can on it's turn be superseded by another consideration.
Shifting white's position 1 pip:

Black to Play 2-2

Code:

    1. bar/23 24/22 13/11(2)        Eq.:  +0,037
       0,524 0,149 0,009 - 0,476 0,194 0,006 CL  +0,006 CF  +0,037
 
    2. bar/23 24/22 13/9            Eq.:  -0,036 ( -0,073)
       0,519 0,149 0,010 - 0,481 0,233 0,019 CL  -0,055 CF  -0,036
    
    3. bar/23 13/7                  Eq.:  -0,044 ( -0,081)
       0,498 0,139 0,007 - 0,502 0,185 0,008 CL  -0,051 CF  -0,044

In spite of the fact that white's timing has terribly deteriorated, slotting is not best with that big a distance anymore. White has more chance now to deny black a defensive game by closing the board.

PS It safes me work when you participate with a good analysis, and I can address pitfalls with a bad analysis.

Quote

06-08-2015 , 02:04 AM

#244

yogiman

adept

Join Date: Jun 2012 Posts: 784

Position ID: udsGAAyxbRsAGA Match ID: cIkJAAAAAAAE

White - Pips 102

Black - Pips 125
Black to Play 3-2
Created with www.BGdiagram.com

Quote

06-08-2015 , 02:18 AM

#245

bleep69

grinder

Join Date: Sep 2008 Posts: 483

I'll move both guys off the 8 point. I want him to play his 5's and break his anchor.

Quote

06-08-2015 , 05:29 AM

#246

Kamba

enthusiast

Join Date: Feb 2015 Posts: 93

I'll move both of the 4 point. That way he's only getting out with 6-1 and is almost guaranteed to have to move off the 18 point next throw.

Quote

06-08-2015 , 06:09 AM

#247

Karol Szczerek

enthusiast

Join Date: Mar 2015 Posts: 92

I go with the odd-looking 4/2 4/1.

We don't overkill any checkers, and maintain a solid blockade for 1 more roll. Opp has to lift his bar point on most rolls, except 5s and 21. Then, our 6s and 5s escape, and lower numbers play OK by lifting one of our blocking points.

If he rolls a 5, he lifts 6pt. instead of bar point. Then, our 5s play OK too: 51, 52 and 53 cover the blot on 2nd point, 55 gives 5-pt. board, and 54 runs the back guy.

Quote

06-09-2015 , 01:27 PM

#248

yogiman

adept

Join Date: Jun 2012 Posts: 784

Code:

4/2 4/1                      Eq.:  +0,440
       0,618 0,134 0,003 - 0,382 0,082 0,002 CL  +0,289 CF  +0,440
    
9/7 9/6                      Eq.:  +0,263 ( -0,177)
       0,560 0,136 0,004 - 0,440 0,100 0,003 CL  +0,159 CF  +0,263

So if white gives up a point, it gets easier for black to skip the block. And once black has done that, it is like energy is released which will crumble white's homeboard.

The rule for this manoeuvre is that black can't escape a checker, and keeps the 5-prime intact. So only the lowest dicenumbers will do.

Black to Play 2-1

Code:

    4/1                          Eq.:  +0,440
       0,622 0,149 0,004 - 0,378 0,112 0,004 CL  +0,281 CF  +0,440
      
    9/8 9/7                      Eq.:  +0,270 ( -0,171)
       0,564 0,133 0,004 - 0,436 0,102 0,003 CL  +0,160 CF  +0,270

Code:

    4/2 4/1                      Eq.:  +0,861
       0,715 0,149 0,003 - 0,285 0,046 0,001 CL  +0,535 CF  +0,861
    
    9/7 9/6                      Eq.:  +0,736 ( -0,125)
       0,681 0,155 0,004 - 0,319 0,063 0,001 CL  +0,457 CF  +0,736

It is not necessary that white is at the end of his flexibility:

Code:

4/2 4/1                      Eq.:  +0,036
       0,511 0,118 0,004 - 0,489 0,125 0,004 CL  +0,015 CF  +0,036
     
9/7 9/6                      Eq.:  -0,027 ( -0,063)
       0,488 0,126 0,005 - 0,512 0,134 0,004 CL  -0,033 CF  -0,027

Quote

06-10-2015 , 06:51 AM

#249

bleep69

grinder

Join Date: Sep 2008 Posts: 483

A good indicator that you have made a mistake is when your chosen move doesn't make it in to the options!

I often see XG break a prime from the middle, rather than the back. Are there any rules of thumb for these scenarios?

Quote

06-10-2015 , 09:06 AM

#250

Robertie

old hand

Join Date: May 2004 Posts: 1,954

Generally speaking, you break a prime in the middle when you're either trying to create useful builders, or you want to fill in open points behind your opponent's anchor. Usually you're doing this when the race is somewhat close, and you think you'll need to attack your opponent when he runs the first man. You might also do it as a trap play, to create some extra gammon chances.

'How to break a prime' is actually a pretty complicated subject.

Quote

Page 10 of 15

First

5 6 7 8 9 10 11 12 13 14 15

Last

Post Reply Subscribe

...

Page 10 of 15

First

5 6 7 8 9 10 11 12 13 14 15

Last