Backgammon Cube Handling in Races

Cube Handling in Races

Forum Archive : Cube Handling in Races

EPC examples: stack and straggler

From:   Carlo Melzi
Address:   camelzi@tiscali.it
Date:   15 December 2008
Subject:   Effective Pip Count with stragglers
Forum:   BGonline.org Forums

I seem to have problems in handling cubes in bearoff positions with stragglers. Here are 2 examples that recently gave me some troubles. - - - - - - - - Example 1. 13 14 15 16 17 18 19 20 21 22 23 24 +---+---+---+---+---+---+---+---+---+---+---+---+---+ | O | | O | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | X | | | | X X X | +---+ | O | | X X X | | 2 | +---+---+---+---+---+---+---+---+---+---+---+---+---+ +---+ 12 11 10 9 8 7 6 5 4 3 2 1 X doubles to 4. Should O take? X has an EPC of about 29 (4*7+1). O has a pip count of 25. How can you calculate O's wastage? GnuBG estimates O's wastage as 6.675, for an EPC of 31,675. O is then trailing by 2.675 in the EPC. How do you use this number? Is this a pass or a take? - - - - - - - - Example 2. 13 14 15 16 17 18 19 20 21 22 23 24 +---+---+---+---+---+---+---+---+---+---+---+---+---+ | | | O | | | | O | | | | O | | | | O | | | | O | | | | O | | | | | | | | X | | | | X X | | | | X X X | +---+ | O | | X X X | | 2 | +---+---+---+---+---+---+---+---+---+---+---+---+---+ +---+ 12 11 10 9 8 7 6 5 4 3 2 1 X doubles to 4. Should O take? X has an EPC of about 36 (5*7+1). O has a pip count of 19. How can you calculate O's wastage? GnuBG estimates O's wastage as 18.885, for an EPC of 37,885. O is then trailing by 1.885 in the EPC. How do you use this number? Is this a pass or a take? - - - - - - - - I remember reading in Backgammon Boot Camp by Walter Trice a formula to use to understand EPCs and related doubling decisions when stragglers are present. However, I don't have access to the book now. Can anybody post the formula or a link to where I can find it? Many thanks for your help, Carlo Melzi

Gregg Cattanach writes:

The formula to estimate the EPC of a stack + straggler (stack should be super-speed board with basically no misses) is: (#checkers * 3.5) + straggler pip count. If there are two stragglers: (#checkers * 3.5) + both stragglers total pip count - 4.

Walter Trice writes:

In Example 1, I would treat O's position simply as 3 checkers with a pip count, not stack-and-straggler. (Where's the stack??) Boot Camp chapter 23 suggests using wastage = 4.7 for a one checker far outfield position and 5.2 for two checkers (like what you have after a coup classique), and of course it's something more than 7 for 15 checkers. Mindless linear extrapolation from 4.7 and 5.2 would yield 5.7 for three checkers, which is probably not so bad here. For sure GBUBG's 6.675 is too high. As for Zare's method, if you use it here you get EPC = 3*3.5 + 10 + 13 - 4 = 29.5. 29.5 means wastage = 4.5, which has got to be too low. With the 5.7 estimate we get 30.7 for White's epc, which puts him down 1.7. The pips-vs.-rolls guideline says that in a 4-roll position you can take down 1, so it looks like I'm predicting a close pass.

Keene writes:

OK, according to my best guesstimates, I have Example 1 as a 4-roll position (or transpose to 4 rolls for each side), and Example 2 as a 5-roll position. Each roll average is 8 pips, so by playing 8 pips per figurative roll to get in, then whatever its going to take to bear them off. My reference point here is the 4-roll positions for each side are redouble/take. So Example 1 is a redouble/pass for me, and Example 2 is a redouble/take for me. Just because there are fewer sets that will work as "bearoff sets" in the short term in Example 1 is where you are losing enough equity to make it a pass. The way I worked it out was to calculate how many rolls to get the stragglers into the homeboard, then how many rolls to bear them off, then the same for X. After some brief checking, I can see that my answers are at least half right ... but, nonetheless, this is how I looked at it, and how I would assess it OTB. Although much less precise than Gregg's numbers, I feel its a reasonable way to go about it, and will give you at least a good approximation.

Did you find the information in this article useful?

Do you have any comments you'd like to add?