Comparing Halves and Zen counts
Halves 22 positive tags
Zen 36 positive tags
I imagine it is not quite add or subtract approx 50% between counts?
Thanks for your time
Comparing Halves and Zen counts
Halves 22 positive tags
Zen 36 positive tags
I imagine it is not quite add or subtract approx 50% between counts?
Thanks for your time
> Comparing Halves and Zen counts
> Halves 22 positive tags
> Zen 36 positive tags
> I imagine it is not quite add or subtract approx 50%
> between counts?
When different counts reckon different card ranks, no perfect comparison between the two of them is really possible. The best you can do, to convert one running count to the other -- realizing that it is just an approximation -- is to take the square root of the ratio of the sums of the two counts' tags.
So, in this case, (36/22)^.5 = 1.28. So, if you had a Halves RC of, say, +7, the corresponding Zen count would be about +9.
Don
I read your post a bit too hastily and David Spence was kind enough to point out that my answer didn't appear right to him.
I thought you had already squared the values of each of the tags, because that's what needs to be done before one forms the ratio of the two and then takes the square root.
So, once you square all the tags of Zen, you get 32. For Halves, it's 11. (I'm not sure where you got your values from above). (32/11)^.5 = 1.71. Whatever the Halves RC is, Zen would be 1.71 times as great.
Don
> I read your post a bit too hastily and David Spence
> was kind enough to point out that my answer didn't
> appear right to him.
> I thought you had already squared the values of each
> of the tags, because that's what needs to be done
> before one forms the ratio of the two and then takes
> the square root.
> So, once you square all the tags of Zen, you get 32.
> For Halves, it's 11. (I'm not sure where you got your
> values from above). (32/11)^.5 = 1.71. Whatever the
> Halves RC is, Zen would be 1.71 times as great.
> Don
Don, I am interested in this as well and have a few questions about your answer (which I thank you for as well). I understand there is never a perfect conversion between two counts using different tag values for a given card.
That said, can you explain:
1. Where did the concept of taking the square root of the ratio of (tag values)^2 come from when trying to estimate a conversion between two counts?
2. Do you sum the positive tags of a count first and then square the sum, or do you square each tag and then sum the squares? That's a tongue twister, ey?
3. The reason I ask #2 above is because I am having trouble seeing where your values of 32 and 11 come from for "squared" values of Zen and Halves. Are you doing this just for a quarter deck (one suit) or for a whole deck of tag values?
Let's take Zen as an example (Let's just say I'm partial to it).
Here are the positive tag values --
Card: Tag value
2: 1
3: 1
4: 2
5: 2
6: 2
7: 1
Sum of the individual squares for one suit: 15 (1+1+4+4+4+1)
Square of the sum of the tag values for one suit: 81 (9^2)
For a whole deck, these values would be even bigger. So I guess you can say I'm confused at where your 32 came from! Maybe I am missing something or misunderstanding?
Thanks, and apologies for the message length,
Rukus (without the C)
I'm sure Don will give you the answers you need without delay, but, well, I'm up early and I have nothing else to do :-)
> 1. Where did the concept of taking the square root of
> the ratio of (tag values)^2 come from when trying to
> estimate a conversion between two counts?
The purpose of summing the squares of the tag values, then taking the square root of this sum, is to capture the fact that a few very high or very low tags tend to produce higher and lower counts than do many moderately high or low tags, even if the sums of the absolute values are the same.
For example, a system that counts ace=-5 and five=+5 will tend to produce greater count fluctuations that does the Hi Lo, even though the sums of the absolute values of the tags of the two systems are both 10.
> 2. Do you sum the positive tags of a count first and
> then square the sum, or do you square each tag and
> then sum the squares? That's a tongue twister, ey?
You square each tag, then sum the squares. One important point: you consider ALL the tags, not just the positive ones.
> 3. The reason I ask #2 above is because I am having
> trouble seeing where your values of 32 and 11 come
> from for "squared" values of Zen and Halves.
> Are you doing this just for a quarter deck (one suit)
> or for a whole deck of tag values?
You consider one tag per card rank, or a quarter deck as you phrased it above. For this calculation, don't forget to consider 10, J, Q, and K individually. And, again, make sure you include both the positive and NEGATIVE tags.
> I'm confused at where your 32
> came from!
Here are the tags for the Zen count (one tag per card rank):
-1 1 1 2 2 2 1 0 0 -2 -2 -2 -2
Their squares:
1 1 1 4 4 4 1 0 0 4 4 4 4
The sum of squares:
32
I hope that helps.
David
I can't improve on David's answer except to tell you that I think I first read of the conversion concept in one of Griffin's writings, and I once discussed it with Ken Uston, in a letter he wrote to me.
Don
> I can't improve on David's answer except to tell you
> that I think I first read of the conversion concept in
> one of Griffin's writings, and I once discussed it
> with Ken Uston, in a letter he wrote to me.
> Don
You are right, David explained it perfectly. I guess I was just looking at the positive tags and ignoring the negative ones. Woops! So thank you David, that was excellent, and thank you Don for your time as well.
Rukus
> Comparing Halves and Zen counts
> Halves 22 positive tags
> Zen 36 positive tags
> I imagine it is not quite add or subtract approx 50%
> between counts?
> Thanks for your time
Does the same method work for comparing unbalanced to balanced counts?
Thanks again for your time!
> Does the same method work for comparing unbalanced to
> balanced counts?
Never gave it any thought. You'd have to TC the unbalanced count and then use the tags to compare. Or, you'd have to specify at what level of penetration you are, before making the comparison.
May I ask why any of this matters?
Don
Posting Permissions
Bookmarks