How bad was this play ?

How bad was this bet ?

Code:

Hand 9 out of 10, min/max bet 100/1000 in increments of 1. Surrender allowed. ReSplits allowed making up to 4 hands.

BR1 (me) 1611 chips - bet 105, and betting after me BR2 1600 with - bet 722 !

*I was expecting from my opponent a bet of 222, maybe 327, but 722 .. wow !

- Now the question is, how much did he lose/(gain?) in terms of % (assuming perfect play thereafter) with this bet ?

BR1 first to play and got a pair of K for a 20, BR2 got 10 and dealer a 9 up card.

- Should BR1 have split ? [in this case would have mattered the count towards splitting decision ? (-when yes/no ?)]

 

I Like That Bet

It looks like your opponent wanted to gain a 1/2 max bet lead over your possible double/split by winning his hand. That way, betting first on the final hand, he would be able to cover you high and low. If he loses, he will be in very bad shape, but still not completely out of. Once he has bet enough to accomplish his goal, I like that he kept the remaining chips for the slim chance on the final hand. A slim chance is better than no chance.

Overall, he has a much better chance of hold his lead (should he win) by being more than a 1/2 max bet ahead of you. You would need a full swing or you would have to double/split to beat him. With less than a 1/2 max bet lead, he would take the low, giving you the high and you could beat him by winning your hand.

Given that he has your double/split covered for his goal, you don't gain anything that I can see by splitting. I would keep the 20 which would give you more than a max bet lead on the final hand should you win or push your 20 and he loses.

 

Yes, I have remarked that he made this bet in order to gain a half+1 of a max bet lead. And I think he risked about 13% (more than normal) in order to get this lead. (If he was betting 222 and we both lost, he would keep about 25% chances to win the game, but now if we both lose, he would need a full swing to win the game which happens about 12%) - But my question is, how much he gained in % with this bet ? - So, overall which bet is better: 172, 222, 327, 722 or.. ?

And in regards to splitting my 20, I think I could gain something if I could split twice (at least) so if we both won our hands (mine with a 3 way bet - after split) then his lead would be less than half of a max bet, and some improvements could be made ? Generally thinking about it.. (but in this case he is likely to double)

PS: Re-thinking about it, since my opponent was more than likely to double his 10 vs dealer 9 in this situation, if he win his double then, he will end up with 2944, and me (by not splitting) with 1716 needing a full swing (and maybe made with a double - in case he bets something between 273 and 454 - which I think is better than 490? and then surrender if he see that I am not likely to swing him with a double - say I get a 20 or so..)

- What I want to say is that, I am likely almost locked out if my opponent wins his double down now, and if so, how worse could it be if I split as many times as I can my 20 lowering as much as I can the difference, in case he win his double ? And in regards to making this split decision according to the count, if there is a "+" count, then 10s are more likely to come, should not I try to take the next card from my opponent ? - But on the other hand, if the count is "-" then I should not split and draw any other cards since my opponent is not so likely to get a 10.. - Does all this theory have any real clue, or I am mixing it all up so badly ?

 

There is nothing special about the next card. All unseen cards are equally likely to be a 10.

 

I still like the 722 bet the best. He is betting/playing first on the final hand, so he really should go for the lead on this hand. Covering your double/split + 1/2 max gives him an excellent chance of holding that lead. Better than if he simply covers your blackjack (172) or your double (222) or your 3 bet win (327). In order to get exact percentages one would have to consider every possible outcome of both players on the second last hand, for each bet, combined with the chances of success from each of those outcomes on the final hand. Sometimes, exact numbers are not needed. One need only know that X will be better than Y.

It is likely that your opponent would double this hand. I probably would myself with the thought that those extra chips provide more value by increasing my lead to more than a max bet than they do on a "Hail Mary" on the final hand. In this case, I suppose trying to keep up in order to prevent that by splitting and/or doubling might be reasonable.

As far as using the count to decide goes, I don't think it's relevant here. The count might tell you that there are extra 10's and aces are coming, but it doesn't tell you when or in what order. We're discussing splitting/doubling in order to guard against falling more than a max bet behind because there is really no risk, assuming your opponent will double. Assuming that you will or will not take a given card from your opponent or that you will take a given card for yourself based on the count is a step in the direction of voodoo-land in my opinion. It would be equivalent to doing the same with regard to the dealer when playing 3rd base at a regular table.

 

If we are only discussing the split decision in regards to not fall behind a max bet lead, then I have no chance for that unless I bring on the table a 5 way bet.

You are correct when you say that the count will not tell me when the 10s will come or in what order, it may very well be a count of +10 (which is very high and indicates a lot of 10s left in the shoe) and the next 8 cards to be all in the range of 2 to 6, in which case all I do is to only increase even more the chances that the next card will be a 10. Correct, everything can happen, I even heard of one person getting 4 blackjacks in a row all on a "-" count, where that was very not likely to happen. But in the same order of ideas, when playing regular blackjack where you can count cards, sometimes is correct to take insurance, sometimes is correct to stand on 12 vs dealer 2 or 3, sometimes is correct to stand on 15 and 16 vs dealer 10, and sometimes is even correct to not double 10 vs dealer 2 to 9, and all that turns the edge from the house to the player ! So.. yes, I am thinking that someone might see some reasons in taking the next cards out of the shoe with the real hope that he will be more likely to take out that 10s which should be come soon, if the count says so.. (not saying I am right, just reasoning)

 

I've already tried to point out the flaw in this reasoning -

There is no more reason to expect that the next card is a 10 (which you can take away from your opponent) than there is to expect that any other card is a 10 (including the one which your actions will now be giving to your opponent).

To repeat - all unseen cards are equally likely to be a 10.

Yes, when we make strategy changes or raise our bets in regular blackjack it is because of the change in probability of the card(s) we will receive on our hand. But this changed probability affects every card in the remaining deck, not just the next one. The dealer could shuffle the remaining cards, burn the next card(s), swap the order of the next two cards, or whatever - nothing about the situation will have been changed.

So there is no reason for you to ever alter your strategy in order to take the next card away from someone else.

 

Certainly I have missed that bolded part big time part til now, thank you for explaining ! (but no, I did not alter my strategy based on that - I was just thinking)

 

Very clear, thank you very much .. Well in this case, certainly I have learned something which will improve my winning percentage pretty much, because until now I did not thought to take this extra risk when I am behind on ante last hand in exchange to gain a +1/2 max bet when I will be first to bet on the last hand!

 

Just out of curiosity, what happened? What was your play on the second last hand, what was the result and what happened on the final hand?

 

On hand 9, I stand with my 20. He doubled and got a 10 for 20, dealer had a 10 under. On the last hand he made a bet of 500 and got a hard 13, I got 11 and dealer a 9 again. He did hit and got a 4 for a 17 then stand, I doubled, got a 10 for 21, but dealer had a 7 under and drawn a 10 after busting out with 26.

- Now do you think that a bet of anything between 273 to 454 instead of his bet of 490 (or more) could further improve things for him, or he did the best bet ?

- And hitting his hard 13 - I think it was a bad move, because if he was busting it were improving my chances if dealer was going to bust too (which has happened) - I think he should double down his hard 13 in case he was betting 273 to 454, but otherway stand in case he placed a bet of 490 or more.. ?

 

So you won your hand and he won his double leaving the situation as follows:

BR1: bankroll: 3044, Bet: 500
You: bankroll: 1716, Bet ??

He has bet enough to cover your best possible result (double/split win) and he can still surrender to beat your single bet win. He could have accomplished the same thing with a bet of 389, but sometimes it's best not go be too fine with calculations in the heat of the battle. This is especially true for live tournaments where your chip count might be a little off (I suspect that this was an online tournament, however).

Covering your double/split is his top priority, so he needs to bet at least 389. Anything between 389 and 654 would still allow him to surrender to beat your single bet win. There is no reason to bet more than 654.

I'm going to assume that you bet 858 with the intention of doubling or splitting anything. With your 11, he knows that you will double, so his top priority is now to win his hand. Vs the dealer's 9, this is best accomplished by hitting to 17 or better.

 

No.. on the last hand he was BR1 with 2944 chips, and me BR2 with 1716 chips. He did a bet of 500 and I did max bet of 1000. - Should I bet 858 instead ?

So, in order for him to cover my double bet win he should bet at least 490. But with a bet of 490 he could not surrender and still keep the lead if I was going to win only with my max bet of 1000, so this is why I think a bet in the range of 273-454 could be better for him ? (he could still surrender or double if the case)

And if I think even further, I think he should place a bet between 326 to 454 because with a bet of this size, if he gets a blackjack he is 100% winner, and also can cover my blackjack single bet max win, and also he can surrender if it is looking that I am likely to swing him with a single bet, but not with a double.

As regarding for him to hit his hard 13 .. Ken`s "Avoid a full swing acting first when opponent have a hard total" chart says that the proper action for the situation at hand (BR1 hard 13 vs BR2 hard 11 when dealer shows a 9 up card) is to hit to hard 16 or better. But if I have read that chart correctly, the chart only says that it is a strategy to protect against a full swing, and not against a full swing with double down - so I am not sure if that exactly apply here ?

EDIT: Sorry my calculations went wrong !!!

Yes, you are right he ended up on the last hand with 3044, and not with 2944 as I said previously, my bad sorry readers !

And you are right Gronbog with 389 being the best bet in this case, sorry for the confusion I have created !

But just for the teaser, lets say that BR1 had 2944 chips, and BR2 1716 chips on the last hand.

- What is the best best in this case for BR1 326 to 454 or 490 ? And what BR2 bet would be the best, a max one or split bank ?

*But the part with hit his hard 13 I think it still should remain valid, Ken`s chart is meant to protect against a full swing made with a single bet, no ?

 

BR1's first priority is still to cover BR2's double/split, so he should bet at least 489, even though this gives up the opportunity to surrender against BR2's single bet win and also exposes him to a full single bet swing. SInce BR1 has more than a max bet lead, I believe that any bet 489 or greater would be equivalent. In a live tournament situation, I would probably bet 500 as a cushion.

Once BR1 has covered BR2's double/split, BR2 needs some kind of swing. Either lose/win (BR1/BR2) or push/win2 (where win2 indicates winning a double/split). If he bets 858, then he has a chance at both scenarios.

I was wrong when I said before that BR1's priority is to win his hand. You are correct that it is about guarding against a fatal swing. What BR1 wants here is to avoid the push/win2 and lose/win swings (BR1/BR2), assuming that BR2 will double. He wants to engineer a win/any, any/lose or push/push outcome (BR1/BR2). My simulator says that standing on his 13 is the best play for a 70.88% chance of winning vs 67.69% for both hitting and doubling. If he were to hit, he would take only one card.

 

Thanks a lot, but still one question about BR2 best bet: - If BR2 split his bank by betting 858, then he allows BR1 to surrender against a swing without double by BR2. On the other hand if BR2 bet 1000, then BR1 with his bet of 489+ will no longer be able to surrender against a single bet win by BR2. - So, still 858 is best ?

PS: And I am also thinking that the all-in win frequency is much greater with a double than it would be by a split. (so somehow .. I feel the split is negligible ?)

 

Ahhhh, good catch! The reason for betting 1/2 bankroll was to preserve the ability to split, since one can't split for less. The question is whether preventing the surrender is more valuable than being able to split. I had no idea, so I ran two more simulations. It turns out to be very close, however, with both players playing optimally, BR2 has a 14.95% chance by betting 858, before any cards have been dealt vs a 14.11% chance by betting 1000.

This is almost certainly correct.