Final Adjustment and Cube Window Keith Count

[@lucapini6310]

thanks very very very much, with your videos I'm becoming a conscious player. When might we see lesson #6?

[@10bc-tenerifebackgammonclub]

Really enjoyed this series, im hanging off my seat for the 6th video.

[@evgenylee2515]

ty for great lessons nice doing!!!!!

[@BackgammonNoord-brabant]

Where is video 6 ?

[@GrantHoffman123456]

Another excellent video in an excellent series

[@backgammonskillset]

What are you doing here? Aren't you busy resting up and studying for the big tournamant next week? Go Kiwi's ! We expect a full report from the tournament right here in the comments . . . . . . .

[@bennemann]

Great stuff, thanks!

[@steveholman546]

Any idea when video No 6 is going to be available. Great videos what other topics do you have in mind

[@backgammonskillset]

Soon, and Opening play series to be released in a day or two. Stay tuned . . . . .

[@bennemann]

I want to understand why is there a 1/7 adjustment that INCREASES your Keith number (thus bringing you closer to a no double, for example). Given that you're on roll and that gives you an edge compared to the opponent, shouldn't the adjustment go in the opposite direction?

[@backgammonskillset]

Common question all backgammon players have had. Without the adjustment formula didn't work, with adjustment it works. You are correct, in that it IS counter-intuitive. Here is my answer copy/pasted from a previous question : It might be helpful to think of these adjustments as something different than the count itself. For example(s) : WHY would you penalize a gap on the 6 point? WHY would you penalize extra checkers on the 1 point? Why? Because when they tried that . . . it worked. At a glance it doesn't seem to make sense. Keep in mind these adjustments were all developed/discovered by trial and error. Let's try the following, we will run the following 10 experiments - decrease by 7% 9% 11% 13% 15% , increase by 7% 9% 11% 13% 15%. After running these 10 scenarios we found that increasing by 13% or 15% seems to give us , on average, the most accurate cube action. We then continue to experiment and find that increasing by something close to 14% is good enough for us mere mortals. To keep the math simple we settled on an increase of 1 for each 7 in adjusted pip count. Not perfect but a big big step from where we were when we first learned the game and we were just guessing. There is one thing about that increase that does make sense. It is only for the cuber's count. So it is a small 'penalty' when the pip count is low and a larger 'penalty' when the pip count is high. Which is in allignment with the following principle : The more play is left ( high pip count ) the weaker the cube and the bigger the take - the less play is left ( low pip count ) the stronger the cube and the smaller the take.

[@saeedabbasi-ym7wx]

Great. Tnx How ì can recognize which one is better ?(keith count or 8 9 12 rules)

[@backgammonskillset]

Yes sorry I know I'm behind 'schedule' to put out Video 6 for this. Once you have the pip count, which you need for both, you can do both as well as choosing just one. As you get faster this will be an option for you. If I have a position with no Keith Count penalties then often 8-9-12 will be good, as soon as I see crazy stacks and gaps I default to the Keith Count because that is exactly what it was designed for. If you think about it, if bad distribution didn't matter, the Keith Count wouldn't even exist as there would be no incentive or reason to develop an alternative to the % approach.

[@bennemann]

I found a position that seems to go against the formula. EDIT: No, it doesn't. Explanation at the end. Screenshot: prnt.sc/yeLgwn8q_Sp8 XGID=--AB-AD-------------abcbb-:2:-1:-1:00:0:0:3:0:6 White has 28 pips + 1 (from the vacant 6-point) + 1 (from the extra checker on 2-point) + 2 (from the extra checker on 1-point) = 32. Add 4 (from the 1/7th rule) = final Keith number of 36. Black has 37 pips + 1 (vacant 4-point). Final Keith number of 38. Since the difference is 2, White should redouble and Black should take, yet this would be a triple blunder to take according to XG! What gives? EDIT: I found the issue. White has 2 adjusted pips FEWER than Black, therefore this is a score of -2 for White. It would be a double/take if White had 2 adjusted pips MORE than Black. Leaving this comment regardless, in case it helps future people avoid this mistake.

[@backgammonskillset]

Yes indeed, when you are new to this it is easy to make mistakes. Solution? Do 100 random positions and practice practice practice! Keep in mind that when playing live you may want to take the time to do this but not too much time as to bog down the play, so you want to have it down as good as you can.

[@ericwazner6521]

👍

[@sjsphotog]

Why would you INCREASE the value for the player on roll since they have the advantage of having the dice so its at least a half a roll advantage and since the average roll is 8 pips (with 2 dice) it's about a 4 pip advantage to be on roll versus not on roll.? Confused here.

[@backgammonskillset]

It might be helpful to think of these adjustments as something different than the count itself. For example(s) : WHY would you penalize a gap on the 6 point? WHY would you penalize extra checkers on the 1 point? Why? Because when they tried that . . . it worked. At a glance it doesn't seem to make sense. Keep in mind these adjustments were all developed/discovered by trial and error. Let's try the following, we will run the following 10 experiments - decrease by 7% 9% 11% 13% 15% , increase by 7% 9% 11% 13% 15%. After running these 10 scenarios we found that increasing by 13% or 15% seems to give us , on average, the most accurate cube action. We then continue to experiment and find that increasing by something close to 14% is good enough for us mere mortals. To keep the math simple we settled on an increase of 1 for each 7 in adjusted pip count. Not perfect but a big big step from where we were when we first learned the game and we were just guessing. There is one thing about that increase that does make sense. It is only for the cuber's count. So it is a small 'penalty' when the pip count is low and a larger 'penalty' when the pip count is high. Which is in allignment with the following principle : The more play is left ( high pip count ) the weaker the cube and the bigger the take - the less play is left ( low pip count ) the stronger the cube and the smaller the take.

[@BackgammonFinesse]

very inspiring type of content and i really hoe so in one day i do the same on my native language with cartoon arts... but what about iSight and most important iSight (delta) for matches. this is marvelous method for matches. not only money games.