Speed of play

I finally clocked my casino play head-on for a dozen shoes and the results are quite interesting.I played all negative rounds with one hand at minimum.In positive situation,I spread to two hands with one or two blacks each according to the advantage.
The shoes which stayed negative all allong and were thus played exclusevely with one hand lasted on average 5:40 for an average penetration of 4.5 decks. This means that it took me and the dealer only 76 seconds on average to deplete one deck or 8.5 seconds per round.
To my own surprise,the shoes with positive rounds where I played two hands per round lasted slightly longer.This might be due to the fact that dealers are more careful with payouts at higher stakes or may be I play slightly slower myself.
So for a final verdict, I will make a second test in another casino where two hands minimum have to be played.

Francis Salmon

We know by now that the profitability of playing multiple hands in negative situations is based on the premise that this form of play can deplete the deck considerably faster than one hand in head-on play. After my first test with one hand the results of which I posted above I went to another casino with equally skilled dealers where I had to play two hands always.
The results are rather disappointing for the adepts of multihand play.On average the dealer and I depleted one deck in 75.5 seconds compared to 76 seconds in my one hand test.This means that one round with two hands lasted on average 11.8 seconds.
If we take into consideration the relative frequencies of the two forms of play (1 always and 2neg/1pos) derived from sims we must come to the conclusion that the multiple hand approach doesn't yield more positive situations per hour (even slightly less).
That's bad news for DD' but I'm sure he hasn't spoken his last words.

Francis Salmon

I explained this notion to DD' years ago. He chose to simply continue extolling the virtue of the "considerable" gain in speed for two-hand play, which I always claimed was pure illusion.

Experiments like yours may not be the final word, but they help to add a little balance to the discussion, as does the fact that, in DD''s scenario, it appears that we're also required to raise the top bet somewhat, when playing two negative-EV hands, and so, immediately, we're comparing apples to oranges. With the top bets no longer the same, why even bother to discuss card eating? Why not just claim that, if I raise my top bet (with one hand all the time), I win more? Who would argue with that notion? But then, it has nothing to do with card eating, now, does it?

Don

but even your half second savings per deck would illustrate the point theoretically. Any time savings at all results in a value on the top end that balances the values on the bottom end. Players need only find where their values meet.

Your commentary seems to miss the point. The rate at which rounds are played are not what is of importance. It is the rate at which cards are depleated. Your test showed a time savings for multi-hands, but not a substantial savings. Going through negative segments faster means more frequent high counts, not less. In your test we would get about one extra round dealt per six shoes if they are cutting 1.5 decks by playing multi-hands. The 4.5 decks sees a savings of 2.25 seconds per shoe. If the average round takes 12 seconds it takes a little more than five shoes to gain a round. Given the ev per count, the frequencies of the true counts, and the bet sizes, it would be a simple matter to calculate how big the bets would have to be for multi-hand play in negative counts to be superior given your speed scenario. The larger the top bets, the stronger the case would be for multi-hands.

For a given top bet it may already be optimal to play more hands on the bottom end. Then again, it may not be. Finding out how big the bets need to be on the top to profit from two hands in negatives is not much different from finding out how big the top bets need to be to afford one hand. The backcounter who plays no hands at all in negatives does not need any spread at all. But as soon as he decides to play some negative ev hands we must calculate to make sure his big bets can support it.

By the time we reach the point where the multi-hand model is playing his extra round that is afforded him by virtue of depleating cards faster, both models have seen the same number of positive ev rounds, although these have been a higher percentage of the rounds dealt for the multi-hand model. The two models have seen the same number of negative ev cards, but not the same number of rounds. The multi-hand model has put twice as much in action per negative ev round, but has used up the negative segments in 33.33% fewer rounds by virtue of using up 50% more cards per round. So the multi-hand model has put a total of .66667 * 2 as much in action at disadvantage, or 1.333 times as much, or 1/3 more negative ev. Now he gets his extra round. It is a simple matter of seeing whether the average win rate per round is greater than what he gave up. Whether it will be practical will vary among individuals. But it is not clear that you have yet even grasped the theory.

Again, stay with me. Some people really don't understand the theory behind all this and it is not all that complicated.

We are going to say that 4.5 decks of 52 cards are dealt per shoe. And we are going to say that it takes five shoes for the multi-hand strategy to gain enough extra cards to deal one extra round.

4.5*52=234
234*5=1170

1170 cards are dealt in 5 shoes. Let's say 2/3 are negative ev, averaging -1.5%, and 1/3 are positive ev, averaging +1.5%.

Again, these are just examples to illustrate the point. The actual frequencies and EVs are not important.

1170*.6667=780 negative ev cards.

For the one handed model that is 780/5.4=144.45 rounds. Let's call it a $10 bet on one hand. That's $1444.50 in action at -1.5% for negative ev of $21.67

For the two handed model that is 780/8.1=96.296 rounds. We'll call this 2 hands of $10, or $20 in action per round. That's $1925.93 in action for a negative ev of $28.89.

So the two handed model has invested $7.22 to get the extra round. The cross over point in this model would be (7.22*3)/.015=$1444.

We multiplied the negative ev by three since we are saying only 1/3 of the hands are positive ev, thus 15 shoes needed for an extra positive ev round.

In this scenario, with the minimal time savings you clocked, a $1500 top bet (total action) would make more money for playing multi-hands in negatives. I've certainly spread 3@$10 to 3@$500 before, but the point of this was not to illustrate practicality but just to show how this works in theory. Obviously advocates such as myself believe that more than a half second time savings per 52 cards is typical.

You are mistaken if you think that a greater rate of card consumption means automatically more positive rounds per hour.The disadvantage of using up more cards per round is that you will miss some opportunities namely when the count turns positive but then the high cards come out before the round is completed and also you might get a high count but unfortunately the cut card has come out.
That's why we have to derive the frequency of advantage situations from sims for the different models and believe me: The half second per deck time savings for two hand play is not enough to overcome the above mentioned disadvantages or in other words: You will not get more positive rounds per hour but less.

Francis Salmon

going faster means getting to the postive segments faster. It is just as likely to result in a round being more positive than it would have been had only one hand been played. But in either case we are talking about hands very close to the zero ev threshold. But go ahead and use actual sim frequencies. Everyone has the information needed to make the calculations. There is no one correct answer. In some close call scenarios we can afford the extra hand only at the TC of zero, which is always the biggest contributor to the overall gain.

In your particular case, not the one you did for this study, I think you could gain considerably from the multi-hand strategy. You are an experienced player who should be able to gain much more by efficiency than a half second per deck. Even beginners should do better than that but experienced professional level card counters, like yourself, should be able to do much better.

In the hundreds of posts you have written on the subject you never stated any hard facts. The vocabulary used was always vague like "considerable","enormous" and so on.This is not serious for a professional player like you.You ought to do your own speed test and I bet you won't do any better than the half second time savings I got.
Actually,the possibility of time savings for two hand play depends mainly on the dealer.The player should be able to remove the payout and place new bets while the dealer is removing the cards so that no time is lost between rounds.The dealers I had on my two hand test were very efficient with payouts: When one hand won and the other lost they wouldn't bother moving any chips.They would simply say "one pays the other" and deal for the next round. But many casinos wouldn't allow that.
Beginners have absolutely no chance to get any time gain with multihand play and head-on play is anyway not advisable for them (too much pressure from the dealer).Time used by the player is roughly proportional to the number of decisions he has to take.So a slow player is better off playing just one hand because here he will have less decisions to take per number of cards used.
You certainly know Wong's study about speed of play where he found that one hand play is fastest.This might well be true for the average player and average dealer which totally contradicts your teachings.
My final conclusion is: Time savings with two hand play are only possible with a very skilled dealer and a very skilled player in a casino with very liberal procedures and even then the gains are so marginal that they don't even bring you more positive rounds let alone cover your extra costs.

Francis Salmon

I came to your conclusion ages ago, told DD' so, but was mainly ignored. He kept hammering home his "enormous time savings" routine, and, after a while, I just gave up trying to enlighten him. Such is life. (And, all this, of course, was separate from the provision of possibly having to raise one's top bet, which, in my mind, completely invalidates the whole discussion.)

Finally, playing head-on in a shoe game, for most people, is not their normal mode of play. I think the argument becomes weaker and weaker as other players are added to the table.

Don

users can supply their own variables. There is no reason for anyone to take your word or mine. A sufficient amount has been posted for the motivated student to find the answer. The work that I had on this is three computers ago and it is unlikely I can find it. I remember finding time savings at face up shoe games and little to none on handhelds. I still use the strategy on handhelds because if one decides upon a fixed spread size then multi-hands in negatives provides a superior win rate per unit of spread... same is true at shoes but I believe actual time savings at shoes occurs as well.

Your time study has encouraged me to look at the time issue again as I believe you are telling the truth and made an honest attempt to play quickly. My belief was that we could go through cards at a speedier pace such that we would gain a half round to a round per shoe, thus a few extra rounds per hour and a few extra big bets over the course of a few hours. I'm sure that in some situations this is true and some not.

For the first time I get the impression that you're starting to doubt. The way you put it shows me that you never really made a speed test because why would you have the results on a computer rather than on an easily retrievable piece of paper.
Your equations are all very well but they are worthless if the premise (more positive rounds per hour) is not met.Up to now I haven't seen any of your motivated students confirm your claim in an actual speed test.They probably despair trying.
I don't even discuss face-down games.It seems obvious that multihands must be slower here because the player's actions take much longer.
One more remark: I tested only two hand play but I saw you advocating playing as many hands as possible, even seven.I don't need to make a test for 7 hands. I can see immediately that this must be slower because time will be lost when the player has to pick up his winnings and/or place new bets.

Francis Salmon

I have not studied this particular thread that closely. This subject has been discussed over and over again so often, that I don�t pay that much attention to the discussion.

However, you did says something that struck me:

Beginners have absolutely no chance to get any time gain with multi-hand play and head-on play is anyway not advisable for them (too much pressure from the dealer).

My final conclusion is: Time savings with two hand play are only possible with a very skilled dealer and a very skilled player in a casino with very liberal procedures �

I don�t think DD�s point is directed to beginning players. Lower stakes players would be better off not playing negative rounds at all, by wonging out to another table. What DD� has pointed out is that these tactics cannot always be used at his level. He may get the pit to open one high stakes table, which he has to play at. If he gets a negative shoe, he has to play through it.

So his approach is oriented to experienced players. Typically, the high-stakes table will also have the most experienced dealers.

As you gained in speed, even if only slightly, with the second hand you continue to gain speed with additional hands. Best is to put the frequencies on a spreadsheet, tinker with bets at different levels, and see what is best. If you provide the complete rule set and penetration for the game you time tested then someone could provide sim data and various models could be tested. Also, their is less cost per additional hand. The first extra hand increases the negative ev by 100% per round. The third hand adds only 50% to this, and the next adds only 33%. I fail to see why you think that a one size fits all model is best and will always be best. You saved a half second a deck with the second hand. You may save a total of 2 seconds per deck, or more, with three or four hands. Each efficiency adds to the time savings.

My movement is simply recognition of you as a real player who is concerned with maximizing profit as opposed to a writer who is concerned that something he may have written in the past will become outdated. I respect that you are not going to cost yourself money for the sake of winning an argument. But I do not think, at least at the start of the discussion, that you even had an appreciation for the dynamics involved in what is going on. You probably do understand now that if you are in a game where cover is of no concern and your spread is absolutely huge that you can probably profit from eating negative counts more quickly. I happen to think that there are lots of reasons for the strategy I employ. My advice in the future will be what I usually tell people when this comes up several times per year. I tell them to sim it and test the individual game for themselves and I mention to them that when they consider the time factor that the gain will not be all that the sims suggest. Your time trials are probably a much better value to put on a spread sheet than are Wong's when one is doing this exercise.