literature on first base vs. third

they didn't post this to bjmath.com before publication. It looks like lots of work went into this. They state very early that they are neither probability specialists nor gamblers. It is kind of presumptuous to come along and make such claims without consulting experts in the field for an opinion.

I am speculating here, but I suspect that one of them may have read some of Bhr's work because of their emphisis on the long run. They said that they presume a longer long run to mean more bankroll volatility. We of course know this for a fact. N0 and DI are basically mirror images. If first base did have a smaller N0 then he would have a higher SCORE also.

One of the reasons this got so much attention last time is that Cacarulo, someone who is respected in BJ circles, ran some sims and calculated N0 for each table position. If I recall correctly, he initially got some results similar to what these two gentlemen were claiming. Much follow up work had to be done to find where his error might be. We had threads going on simultaneously on greenchip theory and bjmath.

I appreciate your theory but need to make a comment on your question:

"The $100 bill is gone before we even see under the last guy's cup. He has a zero % chance of winning now. Does this mean that he made a less accurate choice, that he had less than a 10% chance, that the first picker had more certainty of edge?

No, of course the first turn did not increase the certainty of an edge but this is not the issue. Fluctuations is what snapper is implying not EV.
Since I am not an expert theorist I will try to explain in simple terms of why I think this issue is
very controversial regardless if I am right or wrong.

Lets say I have first shot at winning a $100 and you
have last shot, fair enough? What I do absolutely know to be correct is I will ALWAYS be going off at a 10% chance of winning regardless of the outcome of exposed picks after I have picked, correct? However since you are last in turn there will be times you will have 0% of winning or %100 chance of winning depending on the previous choices being exposed,correct?
This would suggest to me that even though we both have a 10% shot at $100,your fluctuations are greater than mine because I will always be going off at a stable 10% chance of winning. I know this logic is confusing and misleading and is not true, but it is controversial and basically has nothing to do with cards because the cups do not have a memory and real cards do, therefore the $100 game is renderend useless in proving a point pertaining to fluctuations.

Mott

Mr. Don I am not trying to pester anyone or cause a disturbance. I am just saying I can see why this is controversial. I would not say someone is a fool or idiot for inquiring about the topic.
Just interested thats all.

Mott

among card counters to try and hide the evil truth...Mott...don't go away mad...just go away!

""Lets say I have first shot at winning a $100 and you
have last shot, fair enough?""
the high cards came out at the 3rd base position? You are not understanding DD's post. Like he said, if you have a +5 count(or whatever) and there are 12 cards remaining, who's to say first base, or third will receive the high cards?

It makes no difference where you are seated, play enough hands and you will see.

Regards,

Forensic

Your bankroll would fluctuate to the same degree. The second player has the exact same 10% chance, even if it is known that there is nothing under his cup. The variance is a function of the payout structure. Information that cannot be used can't possible have any affect on the variance.

Suppose you are playing heads up and have a 3% edge on the next hand of blackjack. Would your edge, or your standard deviation, change if someone else took a peek at the next few cards before they were dealt but did not give you any hint?

There are a few things that are throwing you off. I like adhoc's example of drawing cards from the middle of the table. Think about that some more if my the example I'm about to give doesn't allow you to see the light.

Let's revisit an example I gave before and we'll go very slowly:

We have five players. Before the deal, you have 52 cards remaining in the shoe. You have a true count of +4. So you have 4 more high cards than low ones in the pool of cards to be dealt. You and third base make your bets. They deal your card face down. Now the next three players get their cards face up. All three are tens or aces. Now third base gets his face down.

Now, you may think that you had better chances because you got dealt your first card while these three high cards were still in the deck and the true count was higher than it was when 3rd base got dealt his. You are thinking that you drew your card from a 52 card pool that was 4 high cards rich. Now third base is drawing from a 49 card pool that is one high card rich. But this is an illusion. The three cards that were dealt to the players between first and third were never really in the pool of cards available to you. You just thought they were. In truth, they were designated for these other players and you couldn't get them. So you were also drawing your card from a pool of 49 cards that was 1 high card rich, not a 52 card pool 4 high cards rich.
You can see that 1st base's faulty true count estimate, the best estimate you could make based on what he had seen so far, is off by exactly the same margin for both first and third base. So it is not only the edge that is the same. There margin of error, the variance of their true count estimate, is also exactly the same.

If, before betting, 1st base was allowed to see the three cards that would be dealt to the players between himself and third base then he would asses the true count as something close to +1 and bet accordingly.

The advantage and variance are the same for both players. If strategy variations are not being employed then the chances are entirely equal in every way.

The simlulation and the logical explanation must match or your work is not done. It is as simple as that. If they don't match you must keep looking for either the error in the program or the error in the logic. You don't present illogical simulation results until you have formulated a logical explanation.

In this case, the logic is air tight. Nobody was able to come up with a plausible explanation that could not be refuted. So the error had to be in the sims. If sims continue to consistently contradict logic then you just keep debating it until someone comes forth with an explanation that is not refutable.

This is the problem with claims made about the Pro's system as well. Unless he can give a logical explanation then he should assume that there is an error.

You realize this also translates into agreeing with him about the skew in the true count data too. The differences result in true count distributions that are not balanced also.

This is interestingly one area of such small effect even today's simulators don't add much. Your Power PC or Pentium will die of boredom!

If so did he concede because his comrades could not find a logical explanation of why his figures came up virtually the same as the ones conducted by the team?
If they assumed it was a mistake in Cacaluro and the teams sims, did they point out the mistake?

Thanks,
Mott

These are very good posts that you, and others, have made in this thread. I was away from my computer most of today, and I see that everything I would have said has already been said. And said very well.

It is a natural mistake for newbie's to make. They are at third base; running count is +3 with less than 1 deck left. They push out a big bet, only to see two Aces and 3 10s fall on the first five boxes. Then they get a 6 on their spot. They think "If only I had been at first base, my count would have been more accurate."

They forget all those times that three low cards came out on the first few spots.

But if this is a "natural" error, then so is progressionism. Many players see themselves many hands in a row on some day, and think how much they could have won if they had raised their bet, like everyone at the table told them �

And both these ideas have no mathematical justification.

That's why Cacarulo is respected as an honest researcher. He has no agenda and admits when he is wrong. I don't personally understand enough about computer programming to examine such things.

Information that is not available to you can not change the odds.

I appreciate the info and do understand how even the experts can make occasional errors when dealing with theory analysis. I do respect Cacarulo also and try to read all his postings. What thru me a curve was the results of a change in standard deviation. I knew that the EV would remain the same.

Maybe we'll get lucky and Cacarulo can enlighten us with a brief comment.

Mott

To clarify, I stated that I agreed with everything the three posters had written in their above posts . . . not everything they had ever stated in their lifetimes.

But, of course, everyone on the planet understood that except you.

Pathetic.

Don

But, it's all quite useless, David. They are going to believe what they want, no matter what you write, nor how eloquently you state it.

Don

my post above and realize that since 1st and 3rd base
place their bet at the same time they have, 1st base does not have a better bet (think of cards dealt upside douwn). now if 3rd base could see everyones
card then place his bet at the last minute before his card are given to him than it would be different.
There IS an error and you can arrive at it by using logic.

now we know that standard deviation of true count increases with n(as defined in article). and we also know that this only causes differences between playing accuracies for the different bases.

now does the true count changing for the dealer draw negatively affect the players? if it does, then it obviously affects all players the same? and also the effect should be larger with fuller table?

for those impatient folks who have seen and addressed these issues a million times please ignore.