Quick Pipicount Assessment Jeremy Sherr

[@robbusby881]

Amazing something l found tricky during matches but now l am all over it thanks from the UK

[@jacques42]

Thank you for sharing this information!

[@pezn2077]

Instead of 4,3,2,1 I usually count quadrants as 3,2,1 and ignore home board (0). The difference comes out the same as this but with a bit less counting and smaller numbers.

[@bennemann]

Sounds like your method is a variation of the "crossover pipcount", but more prone to significant errors due to extreme situations (e.g. the checkers all the way to one side or the other in the quadrant). For a method that's way less susceptible to these extremes, I recommend the "HALF-crossover pipcount" which is by far the best pipcounting method I've tried. I suggest looking up an article that teaches it, but to summarize: points 24-22 are worth 6 for each checker there, points 21-19 are worth 5 etc., all the way to points 4-6 being worth 0 and 1-3 being worth -1 (subtract one). Do this count for both players, get the difference/delta and multiply it by 3. This is roughly by how many pips the player with the lowest of the two counts is leading by. The margin of error is typically 5 pips at most.

[@jedagelijksebraintraining]

I count all crossovers till homeboard. Difference times 5.5, exact same result, only faster.

[@Marcolino10DK]

This is a cool system, Jeremy is a great philosopher of the game. If an intermediate or advanced player prefer to use this system, that would be a big improvement. The only observation that I might add is: after timing myself on my Casio stop watch, on these positions, on these positions, I am still much faster at counting pip differences, than I am to calculate the "Sheer score". Like significantly faster. My counting differences is usually between 8-25 seconds per position, and the sheer count seems to be 40-60 seconds per position. Maybe I can improve my Sheer count time, with some practice, but still... counting differences is much much faster for me, and doesn't require much load on my system 1 (my deliberate system) to do calculations, and sheer count does. So I think it's still worth it learn how to count differences. That being said, this system is easier and quicker to learn (even though it's also quite easy to learn to count differences, if you set your mind to it, it doesn't take many hours).

[@sjsphotog]

Very cool method. For "ROUGH" counting to determine if Im ahead or behind my opp, I like an even quicker method that you dont need to multiply anything at all. Its seen on Page 30 (Concept # 3-20 and Figure 3-5) of the softcover version of "Winning Backgammon" by John Leet (Sterling Publishing 1998). It also uses the Quadrants system shown in this video but instead it assigns a value of how many rolls to get your checkers into your home board depending on where they are located. So: For *each pip* you have on your 24 and 23 points thats 2.5 rolls. On your 23 to 19 point thats 2 rolls. On your 18 to 15 point thats 1.5 rolls. On your 14 thru 11 point thats 1 roll and on your 10 thru 7 point thats 0.5 rolls. If any pips on the Bar thats 3 rolls. If you have more than 1 pip on each point then keep adding those up . Just add up how many rolls away from entering you are for each side and take the difference to know whos ahead or behind. I can do the math in less than 10-15 seconds easily. And if you know that each roll is 8 pips on average you can determine the rough pip count ahead or behind. for example if I need 4 rolls to get all my checkers home and my opp needs 5.5 rolls then its approximately 32 pips versus 44 pips so the difference would be about 12 pips which can be tine to double if its near the end of the game, Check it out. But this method shown here is great as well. Most times you dont need to know the exact pip count but only the difference in pips between you and your opponent.

[@bennemann]

Sounds like your method is just a more complicated version of the "half-crossover pipcount", which is by far the best pipcounting method I've tried. Instead of having to do math with "broken" numbers (1.5, 2.5, etc.), points 24-22 are worth 6 for each checker there, points 21-18 are worth 5 etc., all the way to points 4-6 being worth 0 and 1-3 being worth -1 (subtract one). Do this count for both players, get the difference/delta and multiply it by 3. This is roughly by how many pips the player with the lowest of the two counts is leading by. The margin of error is typically 5 pips at most. I would advise you switch to the half-crossover for easier arithmetic, you'll get it in no time since it's so similar to the one you already use.

[@jimsav1]

Thank you, this is very convenient and works well in the "normal" middle game positions as you mention. As I have been using it, I came upon some of the "weird distribution" positions with a lot of checkers near the edges (as mentioned at the 13:00 mark in the video) where this does not work as well. I will continue to experiment to try to get a sense of when to apply this approach, but wondered if you might already have some reference positions which highlight these outlier "weird distribution" cases? This will help to know when to go for a more accurate count method right away. Thanks again.

[@ryanrebelo133]

I play below a PR 5, and this is really awesome. Perfect for those middle game positions where you need just a rough estimate of the pipcount. Thanks Jeremy

[@jsherr1]

That is huge coming from you Ryan! So pleased you like it. I thought I would get roasted by the pros! An thanks for your 3500 series, love it

[@ryanrebelo133]

@@jsherr1 unfortunately I don't think the method works too well in backgame positions. It seems to underestimate the difference by a bit. But "normal" positions it works quite well on

[@ryanrebelo133]

@@jsherr1 thanks. I'm glad you like the videos

[@jsherr1]

@@ryanrebelo133 Agreed, it is good for mid games mainly. Have a look at my other counting video, that works for all positions kzbin.info/www/bejne/gKnUe5aNnK-merc

[@ryanrebelo133]

@@jsherr1 will do. Thanks

[@robertvledderman9764]

Thanks! Certainly worth trying!But how do you count a checker on the bar?

[@raymondmilcsik9840]

Hi Jeremy. I saw your first saw pip count video (from Phil) several years ago now. Do you have any other shortcuts related to that video now? Also, I've found that some of the positions I have come across and tried your QPA method results in pip count differences that are sometimes 5 pips off the actual difference. In your QPA video here, you mention about way to calculate the QPA more accurately. Could you share that?

[@ReignOfPraine]

Great stuff!

[@Sarchasm66]

Sherr Extraordinaire!

[@TrueLoner_]

How does this method count for on the bar checkers?

[@tgwnn]

I prefer to not multiply anything, just add up checkers and carry previous checkers. For example if you have (counting from the outside) 2, 5, 3, 5, I count it as 2 + 7 + 10 = 19, and I don't count my home board as that is always 15. PS you should mention that you're straight-up counting crossovers, that would help viewers put the video in context.

[@WebColorsGr]

Brilliant!! Thanks J