Edge with certain counts

[Ulysses]

OK, here is the scenario.

Basic hi-lo count. No side count.

Standard rules (w/ dealer hits soft-17) at the Venetian, multi-deck (6 decks or whatever it is). Looks like she put the cut card a little bit too far back, so penetration will be a little deeper than normal, but pretty much normal.

There is about one deck's worth of cards left to be dealt. Three players. If the count is +20, +25, +30, is there any way to quantify the edge? If so, what are the percentages for those counts. Or is this impossible without knowing how Ace-heavy the deck is?

[cardcounter0]

You aren't going to see a count of +10,+20,+30.....
A count of +6 or more occurs about 3 hands out of every 100.

A rough rule of thumb is every +1 count in HiLo equates to a 1/2% increase in advantage. So at +1 you are at break even or slightly positive. By the rare +6, you have about a 2 1/2% advantage. It doesn't get much better than that.

Also advantage increases the deeper you are in a shoe. A +3 at the top of the shoe is worth slightly less than a +3 with one deck to go.

HiLo incorporates the Aces into the count, so it is not necessary to know Ace density to arrive at a good advantage figure (of course, knowing exactly how many aces are left is good, but with 6 deck shoes it would take a long time for the effects of an ace side count to show its effects on your bottom line).

[Ulysses]

Thanks. The numbers I was referring to were "running count" not "true count." I think those are the right terms. I was playing at a table this weekend where the running count got higher than I have ever seen at a table before. While I could of course have been a little off with my count, I don't think I was. At the height of the count, of the next 20 cards that were dealt (two rounds of three people), the cards dealt were:

AT AT AT 9T 8T TT
TT 7T 289 9A

[Homer]

FYI, with one deck left, the true count is equal to the running count, since true count is the running count density or running count per deck [TC = RC / (# decks left)]. Anyway, assuming the game was 6 deck with H17 and DAS, and 5.5 deck penetration (1/2 deck cut off), the frequency of each true count is as follows (from Blackjack Attack, p. 208):

<font class="small">Code:</font><hr /><pre>TC Frequency
&lt;=-1 30.58%
0 39.64%
1 10.47%
2 6.98%
3 3.29%
4 3.32%
5 1.65%
6 1.35%
7 0.53%
&gt;=8 2.19%
</pre><hr />

I know this doesn't answer your question regarding player advantage, but you can see that having a +20 TC is rare, to say the least.

-- Homer

[Homer]

I forgot about www.bjstats.com. I yanked the table below from that site, for a game with 5/6 penetration. If you look at the "Advantage Percent" column, you can see that at a +20 count, the player advantage is around 10-11% (ignoring the rows with a low sample size), which coincides with what the other poster said about the HA going up about 0.5% for each increase of 1 in the TC. Also, note that a +20 count or higher came up 93 times in 514 million total hands, so it's rare to say the least (probably a little more common if the actual penetration was 5.5/6, but still very rare).

[Ulysses]

Thanks much for the info, Homer.

[PokerCat69]

You might like KO. It starts out with a negative count and just uses the running count to determine bets. No need to divide out the number of decks. This saves the hassel of looking at the discard tray and estimating all the time.

With hi/lo +1 TC is break-even, +2 and higher you want your bets increasing.