Webby's Backgammon Quiz 3
Analysis of Positions

Below you will find what the program Jellyfish V.3.00 (level 7, using the bear off database) regards as the top 4 plays of each position and the analysis of "The Backgammon Handbook" author and accomplished Backgammon player Martin Fischer, translated and edited by yours truly. All the positions have been rolled out 12,000 times using Snowie pro and some have been hand rolled to be 100% sure of what is happening. In addition, the equity differences between the top 2 moves are large enough for there to be little doubt as to the order of the moves. Thanks to Michael Depreli for the first 4 positions. The positions in Jellyfish .pos format can be downloaded from my download page, should you wish to roll them out yourself.

or scroll down for the first position

1) 3 point match. Match is even 0:0, Pip count: Blue 158 / White 155.

Blue to play 2-1?

Move	Rkg.	Equity	Eq. Dif.
(a) 24/22 23/22	2.	0.028	0.125
(b) 23/21 6/5*	3.	0.027	0.126
(c) 6/4* 6/5*	1.	0.150	0.000
(d) 6/5* 5/3	4.	0.017	0.133

The biggest problem with these kind of positions, is actually seeing the correct move. Here Blue starts 2 important points and places 2 white checkers on the bar with 6/4* 6/5*. White has 21 return shots, however, only 4 rolls hit 2 checkers. All other rolls give Blue an excellent chance to gain the initiative by building up his board. Any other moves gives white the chance to do good things on his side of the board and to make the 5 point with a 1. One of the most important factors early in a game is to fight for the initiative.

2) 5 point match. Match is even 0:0, Pipcount: Blue 88 / White 86.

Blue to play 5-2?

Move	Rkg.	Equity	Eq. Dif.
(a) 20/13	1.	-0.180	0.000
(b) 6/1 6/4	2.	-0.226	0.046
(c) 6/1 4/2	4.	-0.299	0.119
(d) 6/1 5/3	3.	-0.245	0.065

The race is close and could go either way here. Blue could play safe with 6/1 6/4, however, making this move would waste 7 pips, this could be crucial in the long run. If blue waits now then he will be in the same dilemma next roll, however White would probably have covered the blot in his home board and is likely to waste even more pips. Another factor is any 6 by Blue and he is forced to give up the anchor anyway. Blue just does not have the timing to play an effective holding game. Also of note is the duplication of 3s (3 to hit and 3 to cover) Half a point if you spotted that :-)

3) 5 point match. Blue leads 2:0, Pipcount: Blue 124 / White 128.

Blue to play 5-2?

Move	Rkg.	Equity	Eq. Dif.
(a) 6/1 4/2	2.	-0.119	0.051
(b) 18/13 18/16	3.	-0.120	0.052
(c) 13/8 13/11	1.	-0.068	0.000
(d) 6/1 3/1	4.	-0.133	0.065

This is what is known as a "pay now or pay later" position. Blue can play safe in his home board or leave a blot in the outfield. These kind of positions are quite difficult to master. A major factor for leaving an outfield blot here is the fact that White still has flexibility with spares on his 6 and 8 points, this will mean you are likely to be forced off an outfield anchor anyway and by that time Whites board would have no doubt improved. The most effective blocking position is that of a point 6 pips away. So to maintain the maximum outfield coverage Blue must move from the 13 point.

4) 5 point match. Match is even 0:0, Pipcount: Blue 152 / White 142.

Blue to play 2-1?

Move	Rkg.	Equity	Eq. Dif.
(a) 13/11 6/5	1.	0.041	0.000
(b) 13/11 24/23	3.	-0.033	0.074
(c) 8/5	2.	-0.023	0.064
(d) 13/10	4.	-0.045	0.086

White has got himself into a pretty inflexible position here with the ugly stacking on the 6 and 8 points. If white was a strong player he would hope that Blue splits here so he could attack. Getting hit would mean a checker recirculated to get some flexibility back. White also doesn't want to leave the midpoint so is unlikely to bring a builder down to make an inner board point. Therefore playing 6/5 and slotting a valuable point is not as risky as it may seem due to Whites inflexibility. The 2 should be played 13/11 covering the blot and adding an extra builder.

5) 5 point match. Match is even 1:1, Pipcount: Blue 97 / White 92.

Blue to play 6-1?

Move	Rkg.	Equity	Eq. Dif.
(a) 13/7 3/2	2.	0.213	0.052
(b) 13/7 13/12	4.	0.192	0.073
(c) 13/6	3.	0.193	0.072
(d) 8/2 3/2	1.	0.265	0.000

There is absolutely no need whatsoever to leave an indirect shot on the 13 point (Whites roll of 5/4 or 6/3). Blue is slightly behind here so Blue is not keen to get into a race. Playing 8/2 3/2 leaves Blue with 2 checkers to cover the outfield rather than just the one by playing 13/7 13/12. When behind in a race.. don't race.

6) 5 point match. Blue trails 1:0, Pipcount: Blue 154 / White 143.

Blue to play 1-1?

Move	Rkg.	Equity	Eq. Dif.
(a) 24/21 24/23	2.	-0.206	0.067
(b) 24/22 24/23 6/5	4.	-0.231	0.092
(c) 24/23(2) 4/2*	3.	-0.231	0.092
(d) 24/22(2)	1.	-0.139	0.000

Something has to be done about having 3 checkers on the 1 point. You will notice that White is stripped (only 2 checkers) on the 17 and 16 points. It is possible to "freeze" those checkers, this means make them inactive builders rather than active builders which they are at the moment. This makes life far more difficult for White to build up his/her board with ease. You will notice all the top 4 plays do something with the back checkers. The reason is because White is in an ideal position to make his bar, 4 or 3 point here, and with 3 men back Blue can't afford to let that happen with impunity.

7) 5 point match. Match is even 0:0, Pipcount: Blue 101 / White 86.

Blue to play 6-1?

Move	Rkg.	Equity	Eq. Dif.
(a) 11/5 6/5	1.	-0.408	0.000
(b) 8/2 3/2	3.	-0.703	0.295
(c) 13/6	2.	-0.528	0.120
(d) 8/2 8/7	4.	-0.729	0.321

There are many players who are worried about 7 and 8 pip away indirect shots. They have read or through experience are aware that 7 and 8 are the most common totals of a roll so therefore avoid leaving shots at that distance, sometimes however there are overriding factors. There are only 2 plays really here, 13/6 or 11/5 6/5. 13/6 is safe, but woefully unconstructive. 11/5 6/5 however makes the valuable 5 point making bearing in and off easier, the benefits of which far outweigh the indirect 7 shot. The plays are not close at all.

8) 5 point match. Blue leads 2:0, Pipcount: Blue 161 / White 150.

Blue to play 6-1?

Move	Rkg.	Equity	Eq. Dif.
(a) 13/7 8/7	2.	-0.215	0.138
(b) 13/7 6/5*	1.	-0.077	0.000
(c) 13/7 24/23	3.	-0.240	0.163
(d) 13/6	4.	-0.244	0.167

The favoured move in this position takes some nerve. Especially if you are a beginner, you will find it rather odd to leave such a loose position. It is only when you understand the tremendous value of your opponents and your own 5 point, that there can only be 1 move here. If you do not hit, then should White make the 5 point next move you become a serious underdog. The reason is White can take additional risks in building up his/her board in the knowledge that he is virtually unprimeable and has a solid anchor to come out on. 13/7 provides an extra builder and is pretty much forced.

9) 5 point match. Blue trails 2:0, Pipcount: Blue 86 / White 80.

Blue to play 3-1?

Move	Rkg.	Equity	Eq. Dif.
(a) 10/9* 9/6	1.	-0.009	0.000
(b) 8/5 7/6	3.	-0.159	0.150
(c) 8/5 6/5	4.	-0.190	0.181
(d) 10/7 10/9*	2.	-0.092	0.083

Blue is 6 points behind and on roll. Blue has the chance to complicate the game and should most definitely hit. Being behind in the race normally calls for a bold play. Blue is prepared to risk either of the 2 blots in his home as White also has 2 blots. There is some duplication with both 3s and 1s so White is not favorite to cover both his blots should he hit from the bar. Moving to the bar point adds an extra cover number for the 1 point. Blue might be prepared to give up the bar point but he won't the 6 point. Try and look for creating extra active builders wherever possible. 10/7 10/9* is therefore wrong.

10) 3 point match. Match is even 0:0, Pipcount: Blue 142 / White 129.

Blue to play 2-1 from the bar?

Move	Rkg.	Equity	Eq. Dif.
(a) 25/24 8/6	3.	-0.233	0.105
(b) 25/23 8/7	2.	-0.231	0.103
(c) 25/22	1.	-0.128	0.000
(d) 25/24 13/11	4.	-0.322	0.194

You MUST move up to the edge of the prime if you can. You never know when you might get another chance. At present White only has 1 active builder on the 19 point. If Blue waits, that will quickly change and it will then be too dangerous to split to the 22 point. Timing is also a factor here. Blue is likely to crunch earlier than White, as White has an extra checker on the 12 point. This makes it even more important for Blue to creep up to the edge of the prime so he can release a checker easier if need be. No other play comes close really.

Disclaimer: The answers to these positions are based on three separate evaluations. Jellyfish's standard evaluation at level 7, 12,000 Snowie rollouts and the study of the positions from a top drawer player. It is unlikely that any of these answers are wrong but cannot be ruled out as until someone or a program masters the game of backgammon then all positions in theory are debatable. If you are adamant that a position wrong then let me know and I'll check it out. You might also want to open it up for discussion, this can easily be done by using my Guestbook as a sounding board or perhaps download the position(s) and roll them out for yourself (Jellyfish Analyser required).

Please E-mail me or sign my Guestbook with any suggestions, comments, positions etc.

Should thee fail to sign my Guestbook, then thee shall be
cursed by rolling 5-1 forever more...

BG Home ~ Links ~ Quiz ~ Download ~ Webby's Homepage ~ Credits ~ E-Mail