Unbalanced Counts and Efficiency

London Colin

Well-Known Member
#1
Does the formula for calculating the correlation of a point count system hold equally well for unbalanced counts as it does for balanced? (And provide the same measure of efficiency of the count?)

This would imply that both KO and Red-7 have marginally higher betting efficiencies than HiLo, which is something I had not realised.
 

London Colin

Well-Known Member
#4
Thanks. I had put together a little spreadsheet to do the BC calculation, as defined in The Theory of Blackjack, and was surprised to see that KO and Red-7 score a little higher than HiLo. But then I had to stop and think what BC actually means for an unbalanced system.

Presumably direct comparison of balanced and unbalanced BCs only makes sense if you are doing TC conversion? (So that you get an equally accurate measure of the actual advantage at any count, not just at the pivot.)
 

daddybo

Well-Known Member
#5
London Colin said:
Thanks. I had put together a little spreadsheet to do the BC calculation, as defined in The Theory of Blackjack, and was surprised to see that KO and Red-7 score a little higher than HiLo. But then I had to stop and think what BC actually means for an unbalanced system.

Presumably direct comparison of balanced and unbalanced BCs only makes sense if you are doing TC conversion? (So that you get an equally accurate measure of the actual advantage at any count, not just at the pivot.)
Yes, That's my thinking. Although, I usually use the running count as a rough rule of thumb for bet sizing using a balanced count. (with an eye on the remaining decks)
 

iCountNTrack

Well-Known Member
#6
London Colin said:
Thanks. I had put together a little spreadsheet to do the BC calculation, as defined in The Theory of Blackjack, and was surprised to see that KO and Red-7 score a little higher than HiLo. But then I had to stop and think what BC actually means for an unbalanced system.

Presumably direct comparison of balanced and unbalanced BCs only makes sense if you are doing TC conversion? (So that you get an equally accurate measure of the actual advantage at any count, not just at the pivot.)
I am stuck at the airport in Paris :), but if i am not mistaken only Playing efficiency is only defined/meaningful for balanced, i think betting correlation works for both.
 

London Colin

Well-Known Member
#7
iCountNTrack said:
I am stuck at the airport in Paris :), but if i am not mistaken only Playing efficiency is only defined/meaningful for balanced, i think betting correlation works for both.
I think that's right, in terms of what information is intrinsically available from the count. But there is a distinction to be drawn between 'an unbalanced counting system' and 'a betting strategy that only uses the running count'.

I'm a bit unclear as to whether Betting Correlation is a measure of how accurately the magnitude of any advantage is determined, or a measure of what proportion of the time a player advantage (of whatever size) is correctly identified. I'd have thought it must be the former, but the wording of some descriptions I've seen is a bit ambiguous.

Only using the running count, you accurately detect when the advantage first swings to the player (if the pivot is at a suitable point), but you need to do a true-count conversion to get the same accuracy at other running counts.


P.S. Hope you escape the airport without too much delay. (Or failing that, have a great time stuck in Paris! :))
 
Top