Simple bet spread question
- Thread starter jopke
- Start date
With regards to ROR, I'm pretty sure that would amount to an effective spread of 1-to-27. The reason is that if you went with one hand of 270 as your max bet, it would be equal in overall ROR to two hands of 200 each. From the standpoint of EV though, your EV with 2 x 200 will be higher than with 1 x 270, but not as high as 1 x 400. That makes the benefit of your spread higher than 1-to-27. But from either viewpoint, your effective spread with 2 x 200 would be less than 1-to-40.
Last edited:
Could you explain this please? Same $$ on the table with the same advantage (iba) for the given count.
Playing 2 hands rather than 1, more cards are used up per round, thereby bringing out the cut card after fewer total dollars have been placed on the felt.
With only one other player at the table, two hands of 200 each equals 33% less money bet before the cut card pops out than 1 x 400. With five other players, it's 15% less money bet.
Last edited:
QFIT
Well-Known Member
I've never been comfortable with estimates. CVCX uses an estimation method to evaluate the switch to 2 hands from 1 hand that I borrowed from DD'. But, it is quite complex as a completely new set of optimal bets must be calculated and there are literally tens of thousands of calculations involved. And it still makes me uncomfortable, which is why CVData calcs it exactly.
There isn't always a large difference. But, I have seen a lot of quick calculations that just don't hold water.
There isn't always a large difference. But, I have seen a lot of quick calculations that just don't hold water.
thanks, renzey.
qfit, how much do you estimate the counts' frequencies change when playing multiple hands? i haven't run any experiments myself, but I was wondering if you had a "ballpark equation" for how much the frequencies should be modified.
qfit, how much do you estimate the counts' frequencies change when playing multiple hands? i haven't run any experiments myself, but I was wondering if you had a "ballpark equation" for how much the frequencies should be modified.
QFIT
Well-Known Member
Yes, you have identified the problem. Depends greatly on number of players. Switching from one to more hands at a count substantially affects TC frequencies -- and this is at the heart of counting. Like most of BJ, there is no formula. CVCX uses a bunch of tables that I created with several thousand sims. I had to run thousands of sims multiple times with different techniques until I got answers that matched real sims. In fact, I gave up a few times over multiple releases, but kept coming back and finally got results I trusted.
On the original question on the definition of spread with switching number of hands -- make up any answer you want. I tend to call it the min bet on one hand to the max bet on each hand, instead of the total bet, for convenience. Spread has two implications. First, the affect on results. This cannot be defined mathematically when switching numbers of hands. Second, the affect on attention. We all know that spread can bring heat. But, switching numbers of hands has a complex affect on heat, certainly not algorithmically calculable. Many gamblers switch to multiple hands based on gambling logic. Most, but not all, pits tend to look at the max bet on one hand instead of the total when spreading to two hands when calcing comps, which leads to their idea of heat. Clearly this isn't true if you spread to six hands and is highly dependent on your "act." In any case, we are talking about two concepts where pure algorithms don't apply.
This is interesting.
My understanding is that playing 2 hands reduces variance, so if your max bet is 300, that is roughly two hands of 400. So if there are other players at the table, it would make sense to do that, if you are heads up you should stick with the single hand. Am I way off?
If you know the cut card is coming out during a given round, it is clearly better to play multiple hands.
Isn't it the same as playing a 0-200 spread on one hand and 10-200 on the other? So you are basically wonging in one hand and playing the other normally. Couldn't you treat them as 2 totally separate situations, run the sims and add them up? In other words, if you are playing with 3 other people, you would run the sims with a 10-200 spread then run again, but with 4 other people (your other hand is the 4th person), wonging at positive counts. Does that make sense?
Also, where I play, one $400 bet seems to get a lot more attention than 2 $200 bets. I think it is the green-chip stack height that gets noticed.
If your max bet is to have $300 on the table, to have the approximately same variance (but more EV), that would be 2 hands of 72% * $300 = $216 each. To have approximately the same EV (but less variance) that would be 2 hands of $150 each. That is disregarding the change in count frequencies, which qfit and I were discussing.
The only problem with that analysis is covariance. If the dealer gets a 17 and you have 2 hands of 18, they both win. If the dealer gets a 21 and you have 2 hands of 18, they both lose. The probability of the 2 hands winning or losing together is what is the problem with that scenario that you proposed, as well as the 2 hands "eating up" good cards when you have a good count, and hence forcing the cut card to come out earlier, thus the entire table playing less rounds per shoe.
The covariance effect wouldn't have an effect on your EV, only your ROR, correct? (ignoring the "eating up good cards")
aslan
Well-Known Member
Personally, I like eating up good cards when I have a good count, because that is more chances for me to get the good cards, especially the naturals compared with the dealer. Am I wrong? Just give me three hands at max bet all the time! That's how I play it.
Hmm well, off the top of my head this is what I would think:
1. If both you and the dealer both have an equal chance of getting a natural, you're at an advantage already.
2. Regarding advantageous doubles and split situations, you usually start out with low cards, such as 5,6 when you double or split in good counts. Therefore, you don't need to start out with the high cards.
3. Variance: let's say there are enough cards left in the shoe to play about 6 hands. Your choice is to either play 3 rounds of 2 hands each, or 2 rounds of 3 hands each. The first choice yields a variance of (1.3 + 0.5 * 1) per hand, times 6 hands. The second choice yields a variance of (1.3 + 0.5 * 2) per hand, times 6 hands.
What is interesting about what you said is that if there's a very high probability of a natural coming out in a given round, and you play 3 spots, the fact that you are playing 3 of the 4 hands (your 3 plus the dealer's 1 hand), or 75% of the hands on the table, could mean you have a higher probability of receiving that natural vs. the dealer. I will think about this, and perhaps others can chime in.
aslan
Well-Known Member
Also, I want my hands while the count is positive. The next round it may have returned to neutral, or someone may drop in on the table to eat my good cards. Give me more hands when the count is positive. Multiple hands at low counts might work to eat up bad cards, but when max bet is out, it increases chances for getting the good cards. I hate seeing naturals and twenties all around me in a positive count while I have a six and a jack.
QFIT
Well-Known Member
Yes, covariance affects SCORE (i.e. risk-adjusted results). Another way of saying the same thing. But, if you adjust bets to keep the same desired RoR, then win rate would be changed by the existence of covariance.
As far as "eating up good cards," I can't agree with some of the statements in this thread. Ignoring cover considerations, you are better off always playing one hand or two hands for all rounds. How quickly you reach the cut card has no meaning to me, other than the overall speed of the game.
aslan
Well-Known Member
I can see that playing two or three hands when the count is positive gives the player a better chance of catching the tens and aces. The fact that Qfit doesn't see this as better than playing just one hand gives me pause. Is it because one also has the greater chance to lose both or all three hands when the dealer catches the natural or a twenty? I hope there is more discussion on this aspect of play in positive counts.
It wouldn't be the first time I followed a blackjack myth to my dismay.
I forget where I read it, perhaps in BJA, but isn't the basic idea that if there are 3+ other players, it's optimal to play 3 hands? I think it was "so you receive your fair share of good cards" or something like that.
Heads up, I'm not sure how disadvantageous it is to play 2 hands...