Simple Question for 'True Count Distribution' by Tag Values

[greg16394]

True Count Distribution Analysis.PNG

This is just to find TC distribution by tag values of each count. This is from CV/Data Sim Results with 3 strategies attached.
If you play just TC+2 or over, how many percentage of total hands you would play ?

When I attach Felt-F strategy, will play 26.47% of total hands.

When I attach 'Complete HiLo', it drops to 16.28%. Does this make sense ? EV or other factors (Score/N0) do not matter.

When I attach NLLC (new count), it drops to 8.45%.

These three strategies have different tag values.
NLLC is count with new tag values for Lucky Lucky Side Bet.
I just want to know how many percentage of total hands I would play this side bet if I want to play at advantage.
Is my question and simulation worth to effort ?

[Three]

I just want to know how many percentage of total hands I would play this side bet if I want to play at advantage.
Is my question and simulation worth to effort ?

It isn't really clear what you are asking about. I assume you want to use two counts? You need to know the overlap of the sets listed (NLLC and the other count). Some information about the overlap would be available if you listed NLLC count tags. I assume LL is either Lucky Ladies or Lucky Lucky. Since most counts are strong on the former I assume using a different count would mean you are talking Lucky Lucky side bet count. Since the trigger seems to be TC +2 I assume you are using a level 2 count with the 7 and 8 weighted as the heavy "high" cards and the A and 6 as the weak "high" cards. All the other ranks are weak low cards? The BJ count and the Lucky Lucky count would be quite independent without much of a correlation.

[ZenMaster_Flash]

These 3 constricted data points can easily be misconstrued,
as any two True Counts will have different e.v.'s. at T.C. +X.

A novice must understand that you cannot compare different
counts in this fashion.

Ergo, you may want to be Wonging In at +1 or +3
with certain counts and certain rule sets, etc.

[Dog Hand]

greg16394,

What penetration did you use in your sim? I posted the NLLC over at bj21 some years ago, and in that sim (6D, 75% pen, heads-up) the NLLC was +2 or higher on over 20% of the rounds.

Dog Hand

[greg16394]

What I don't understand is what makes tc distribution in 0 or +1 -1 high or low.
For tc 0, first strategy 15%, second one is 25%, third one is 33%.
Dog Hand, I wanted to find it. If I play over 20% of total hands with advantage, it's worth to effort.
But, just under 10%, I doubt it.
What makes tc distribution varies ?
Should it be very similar regardless of tag values ?

[ZenMaster_Flash]

"Should it be very similar regardless of tag values ?"

There is a great difference when the T.C. computation is "floored"

[the best method] compared to others.

When done in this fashion the most common [0] "bin" expands

to include all results between [+0.9] and - [0.9]

Had the T.C. been done by "truncation" the [0] "bin"

includes all results from [0] to - [0.9]

A significant metric is the percentage of T.C. that lie between -1 and +1.

To answer greg16394's question -- NO -- Level One Counts e.g. Hi-Lo do not

reach the much larger T.C. (absolute values) as do Level 2 and Level 3 Counts.

[Three]

When done in this fashion the most common [0] "bin" expands

to include all results between [+0.9] and - [0.9]

We may just have a pronoun reference error here but if "this" means flooring, as is implied by the context, the bin is +0.9 to 0 not to -0.9. The latter is rounding and truncating while the former is flooring.