Blackjack betting system as proxy for card-counting

[NickMPK]

So I know that in general, a "betting system" that relies on varying your bets based on whether you won or lost a previous hand cannot overcome the house edge.

However, it seems like if you combine such a system with the principles of card counting (without actually counting cards), you should at least be able to reduce the house edge.

To explain, if the player has an advantage when the deck has a lot of high cards left and a disadvantage when there are a lot of low cards left, it would follow that on hands where the player wins, a greater than average # of high cards are dealt, while on hands where the player loses, a greater than average number of low cards are dealt.

Therefore, it would seem that if the player notices that he has won a bunch of hands in a row, it is likely that the deck has become less favorable to the players over the course of that sequence of hands (and vice versa if the player has lost a bunch of hands in a row).

Therefore, wouldn't it make sense to increase your bet after a losing streak, and reduce your bet (or leave the table) after a winning streak?

I'm not sure if you could ever turn this into a winning strategy, but it seems like a simple substitute for counting that could reduce the house edge, and you would always have a plausible non-counting explanation for varying your bet.

[SheetWise]

[ QUOTE ]
"...it would seem that if the player notices that he has won a bunch of hands in a row, it is likely that the deck has become less favorable to the players..."

[/ QUOTE ]

This isn't counting, it's observing. There's no foundation to believe, unless you can eliminate confirmation bias. You can only effectively eliminate bias by employing a count, to set a foundation.

This is the sort of game I used to play from the inside with big players. Enteratin them, serve them drinks, confirm their beliefs, and raise their limits. Almost all of them came along ...

Except those few who knew better.

[BSQJamie]

I think I would be wary of any system which involved pumping up your bets when you are on a losing streak.

Chasing cas be pretty disastrous.

[NickMPK]

So am I wrong that, on average, the count will become more favorable to the player after he loses a hand, and less favorable to him after he wins?

It seems that this should be true for the same reason that card counting works in the first place, right? If the player wins more often when the count is high, then winning should generally reduce the count.

I'm not talking about chasing. I'm talking about winning and losing as a rough short-hand for counting.

[ChipWrecked]

I think if you want a counting shorthand you should just count aces and fives or something.

Here's a common loser: you have sixteen (6-T), hit, bust with a ten, dealer had a ten showing and one under.

That's a -3 count.

Maybe if the dealer drew to a winner two or three hands in a row with low cards or something.

In the end, I don't see this working.

[knicknut]

A main reason why card counting is profitable is because you get more 3-2 blackjacks when the count is positive, not just that you win a higher % of hands. Therefore, I don't think there is a high enough correlation between high winning % and count to make this work at all.

And what others have said--it's a good way to increase the magnitude of your downswings (won't help/hurt your bottom line, but if you play infrequently the variance can kill).

You're thinking the right way, though.

[bpb]

Your expected return on a hand following a loss is marginally higher than that of a hand following a win (assuming no shuffle in between). But tailoring a betting system to this observation isn't going to produce a noticeable effect in a lifetime of blackjack play, let alone to erase the house edge. If you search online you should be able to find simulation results showing that there isn't any practical use to this knowledge.

If you're interested in reducing the house edge of blackjack without diving into counting, I'd recommend reading Fred Renzey's "Blackjack Bluebook II".

[Wongboy]

What he said, except I recomment Blackjack Attack by Don Schlesinger instead. KO blackjack is also a good book for any beginner who wants to learn a simple winning system.

Seriously, if you are going to the trouble of learning perfect basic strategy, then learning to count is an insignificant increase in your time/effort investment. For a complete newbie of above-average intelligence, I would estimate 15-20 hours to learn basic strategy, 5 minutes to learn the hi-lo counting method and 10-20 hours of practice to be semi-effective. Another 15-20 hours to memorize a decent amount of play indexes (I would say this is voluntary).

[knicknut]

15-20 hours to learn basic strategy? I think an hour or two, 5-10 minutes at a time, is enough to memorize a simple chart. I mean you can boil 90% of it down to about 5 rules.

[Wongboy]

[ QUOTE ]
15-20 hours to learn basic strategy? I think an hour or two, 5-10 minutes at a time, is enough to memorize a simple chart. I mean you can boil 90% of it down to about 5 rules.

[/ QUOTE ]

I said above average intelligence. Have you had a discussion with an average person lately? They are not very smart.

Further evidence is that 99% of all people playing blackjack do not know perfect basic strategy. Surely a rational person would spend an hour or two to learn correct strategy before plopping their hard earned cash on the table, would they not?