My Theory on CSMs - critiques welcome

OK, so we all know CSMs suck for card counters, but here's my theory on them.

The reason why we count cards to begin with is because we know that tracking the "ratio" of high-cards to low cards can tell us when we have an advantage. One of the reasons why that works is because each shoe, once dealt, is essentially "static" - the cards are already in order and they will come out, no matter what.

In this scenario, a person sitting with a Basic Strategy card can find that they are quickly at a disadvantge. EXAMPLE: Let's suppose that in a six deck shoe, the early count is ridculously high with many small cards coming out. Half-way through the shoe, the count stands at around +3. Dealer shows a 7 and the player has a stiff 15. BS dictates that they must take a hit until they are at 17 or above - assuming they are unaware of the account, they will hit and most likely bust-out. However, a counter, knowing that the shoe is "rich" in face cards may or may not take a hit - either way, the extra knowledge of the shoe ratio can help the counter while the BS player will simply play out the hand as if it was the first one dealt.

It's obvious that a CSM machine is worthless for counters, there's no need to debate that. However, BS is perfectly suited for CSM since every hand dealt will have the same odds as every other hand - which makes having a thorough understanding of BS a necessity. Depending on the skill of the player with BS, they can cut down the casino advantage to less than 1% (if I remember correctly) which makes a short-run win possible. Even though we should all think of BJ as one long game, the fact that the deck is essentially "reset" after every hand of BJ, combined with the tiny casino advantage, can make playing with CSMs at least somewhat worthwhile.

Common sense tells us that casinos are getting rid of these machines because profits are much worse than expected.

People will bet crazy against a shoe, but not so with the csms.
Trying to convince a ploppy that csms are the better game is like trying to convince someone that water is not wet...it wont work.

Golf Cart,

If the off the top edge on a blackjack game is 0.5%, and I play the game dealt on a shoe, and you play the game dealt on a CSM, and we each flat bet the same thing per hand, both you and I should lose at the same rate in the long run.

There will be more VARIANCE in my game - maybe I'll catch many more high or low counts than I should - but in terms of expected value, the two are the same.

One good thing about CSMs - if I want to hit on a fine young female dealer, and she's dealing a game at MUCH higher limits than my $ can handle, I'd much rather she be dealing from a CSM. I couldn't justify continuing to bet when the count tanked, or not cranking my bet when the count goes up. Plus, easier to concentrate on the important things in life :)

M.

Moose, I have to respectfully disagree with you. Recall that my thoughts are based on the fact that your shoe game and the decks within it already have a "fixed" ratio of high to low cards while my CSM game has the same ratio during every single hand.

Imagine this - let's suppose that by some miracle, your first twenty hands are all cards with a value less than 8. Assuming heads up play, that means that the first 80 cards (give or take, depending on hits you take) dealt were low cards - you now know that the deck is rich in ten-value cards. You would almost _certainly_ abandon some parts of BS and modify your play, and your decisions based on this knowledge. My CSM game, on the other hand, would require me to play nothing but BS for the entire duration since I know that at the beginning of each hand, the ratio of high-to-low cards remain the same. Your shoe game could just as easily cause you large losses while my CSM game would allow me to keep a steady loss rate while having enough variance to allow for a short-term gain.

BS player wins slightly more at CSMs for one reason only: the cut-card effect doesn't manifest itself. The difference is minuscule.

As for variance, Moose, there is no difference. CSMs don't cut down variance for the BS player.

Don

BS player wins slightly more at CSMs for one reason only: the cut-card effect doesn't manifest itself.

When did you actually study a CSM? Are you talking about all CSM's or an individual brand?
How do you have any idea what they may or may not do?

For your information,Golf Cart, the index for standing on 15 against 7 is +9 which means that you practically always hit.
Of course you don't need to know that when you only play CSM's!

Francis Salmon

My question is, what shall I do in this
case, hit or stand?
Maybe if more 2-6 cards on the table stand?
More 10's or A's hit ?
Rainer

why don't you ask a specific question regarding my comments? Do you think I stated something that was incorrect? If so, just say what you think the error is.

Don

OK, so maybe I did pick a bad example, but I think that it still made my point. You could further argure that in a shoe game a basic strategy player ONLY LOSES their advantage as the cards are dealt. The more the ratio changes (the direction in which is changes is irrelevant), the more that they should (but don't) deviate from BS.

Of course you don't need to know that when you only play CSM's!


Yes! That is exactly my point! A CSM player can play EVERY hand in the same way (obviously, using BS) since they know that the number of 10s and cards less than 6 are ALWAYS the same at the beginning of each hand. Even if they are playing at third base at a full table, the ACTUAL count is still so completely irrelevant that they can play BS over & over again and have a better expectation against a shoe game.

With a CSM, the concept of "count" becomes meaningless. However, this does not mean that the high and low cards will be evenly distributed through the pack.

Just as with a shoe, there will be times when a preponderance of high cards are popping out, and other times when the players see nothing but low cards.

In theory at least, short term variance could be even worse than a shoe. When nothing but low cards come out early in a shoe, we can reasonably expect to see some high cards later in the shoe, assuming decent penetration. After all, this is what counting is all about.

However, with a CSM, a bunch of low cards coming out means nothing. It could be followed by high cards, or more low cards.

It is a misconception to think that, with a CSM, one will not see rounds that resemble a high count, or rounds that resemble a low count.

It is merely that there is absolutely no way to predict when this will happen.

Golf Cart has made his point very clear. While your concept of "anything can happen" is possible, lets take a close look at the simple mathematics of a csm. It is plain to see we have a continuous even flow of cards regardless of this card clumping theory you perpetrate.

between CSM & shoe play is that with a shoe, you get a re-shuffle--a fresh start, once every 14 to 16 hands or so (assuming normal penetration). You can then "rely" on the count into the deck somewhat, whereas with the csm, you can simply never rely on the "count" because it is constantly changing--behind the scenes--inside the little machine :-)

Obviously, everyone already knows this--that's why counters don't get along with csm's.

That being said, I do not agree that the ratio is static with csm's--for the very reason outlined above--the deck is constantly being shuffled inside the little machine :-). As others have stated in this thread, you can certainly get "clumps" with a csm just as with a shoe. Why? Shuffling, that's all. It's not that the little machine :-) shuffles any more or less effeciently than the well-endowed blonde--it's just that the little machine :-) does it much more frequently--and without letting you know exactly when it's being done. It's kinda like playing from the start of a new shuffle every hand.

Assumption.

For sake of discussion, we assume (maybe it could have been explicitly stated by Don) that the deck is shuffled thoroughly after every round. This may or may not be a perfect model of the CSM, but I think it's valid for our discussion of the differences between shoes and CSM's. Based on the discussions I've read, that seems to be the conventional wisdom, so I think it can be assumed without explicitly calling that out on every post theorizing about the differences of CSM's and shoes.

If one starts discussing more detailed aspects of the inner workings of a CSM - how quickly can the cards get back into the shuffle, is it truly random between every round, can cards from a bust and picked up hand get into the same round? Those are questions we're sort of overlooking for the sake of a simple discussion, and would require a serious study that you elude to. I don't think Don postulated anything that would require such study.

I was going to mention the same thing Don said that the cut card effect will definitely be a difference between the shoe game and CSM game. Golf Cart, I think your idea that CSM will give you nice even wins/losses is not necessarily correct. There should be an equally random chance of having big plus and minus chunks of cards coming out in succession, even if they aren't "fated" the way they are in a shoe. The big difference is that we just can't predict by process of elimination the same way we do in counting a non-replenished pack. I'm not ruling out the possibility of tracking a CSM, I just don't know about any way to do it.

I think your statement:

"You could further argure that in a shoe game a basic strategy player ONLY LOSES their advantage as the cards are dealt. The more the ratio changes (the direction in which is changes is irrelevant), the more that they should (but don't) deviate from BS. "

underestimates the value of bet correlation. It is possible to have an edge without any deviation from basic strategy, simply by knowing when to bet big in a good count. Play variation is gravy helping us win a little more when we have the big bets out, or lose a little less with the small bets.

A perfectly random sample can have many large or small cards come in a row. It's not clumping, it's randomness. Random does not mean high and low cards are evenly distributed. That's why the normal distribution curve is a bell not a spike.

Based on the discussions I've read, that seems to be the conventional wisdom, so I think it can be assumed without explicitly calling that out on every post theorizing about the differences of CSM's and shoes.

It may be the conventional wisdom but it is incorrect. No CSM actually operates that way. Because of this, any actual effects in the real-world will be different. That does not seem like a minor point to me. Nor does it seem "smart-assed".

Those are questions we're sort of overlooking for the sake of a simple discussion, and would require a serious study that you elude to.

It is a simplification too far, to my mind. These discussions have no applicability to any game spread in the real world.
Pure theory is fine, but this discussion is just flat out misleading.

...that the opposite scenario could happen too? Sure, you could face a low count for most of the shoe and lose more than you would with a CSM, but you could just as easily face a HIGH count for most of the whole shoe and win MORE than you would with a CSM.

Guess what the average count is with a CSM.
Guess what the average count is with a shoe.
I rest my case.