I was BR2 acting last on the 8th hand out of 10. Min/max bet 100/1000 and surrender allowed. BR1 had 2704 and BR2 (me) had 2602 in chips. BR1 did a bad bet of 600, and I responded with a bet of 395. FIRST QUESTION: It was this the best bet I could put in or not ? (I was thinking at taking the direct low, and the same if both where about to use surrender, and also at taking the high with my double.) The dealt cards were a hard 13 for my opponent (and he did stand as normal), a hard 20 for me and a 6 up card for the dealer. SECOND QUESTION: Should I had split my hard 20 with the hope that I will get both hands winning or at least one push from these two hands while at the same time putting my opponent on a lose ? However, I did not splitted my hard 20 and the dealer eventually busted. And then I was starting the 9th (next to last) hand as BR2 with 2997 in chips while my opponent BR1 had 3304 chips. The bet I threw in was 615. THIRD QUESTION: It was a good bet (my 615 bet) or should I had bet something else (such as something between 115->158) knowing that on the next hand (last hand of the game) I will have the advantage to be the one who is acting last ?
Overall, I like your bet. As you said, it takes both the low and surrender-low and allows you to take the high with a double or split, if necessary. Even though you act last on the final hand, it is still important to try to be BR1 on that hand. Your bet is also large enough that you don't lose much ground should you both win a single bet. I also like that, should you both lose, you would have breathing room in the amount of 103, slightly more than the minimum bet. This can be used to avoid giving up the push/lose swing on the next hand, although some would bet 200 with the option to surrender back in this situation. I tried to reason this one out before turning to my simulator (but you know I did run a sim for this!). My first thoughts were these: By standing, you take the lead for all dealer outcomes except a dealer bust (42.3% with a 6 showing). You already have the low, so manoeuvring for a win/lose or push/lose or win/push while your opponent loses didn't make much sense to me. The only purpose of splitting would be in order to take the high by winning both hands. I didn't know the numbers yet, but I suspected that by splitting you would be giving up more(the low) than you gained (the high). However, the sim produced usable results while I was composing this and it turns out the the above is completely wrong. You have a 80.13% chance of becoming BR1 if you split and only 57.7% if you stand. Looking at the sim results and thinking about it some more I can now see why. By splitting, you take the high while giving up the low. The question is whether what you gain is more than what you give up. In this case, it turns out that answer is a resounding yes. Because your opponent has stood stiff, you can play a no-bust strategy after splitting and both of your hands are guaranteed to be as good as or better than your opponent's hand. You will become BR1 for all outcomes except if you lose both hands. For splitting 10's vs a 6 and then playing no-bust, it turns out that this only happens 19.87% of the time. This is a good reminder that when your opponent stands stiff before you act, it often opens up excellent opportunities to improve your situation and even sometimes to obtain a lock. Situations like this usually depend largely on the skill level of your opponent. Your bet of 615 (twice the lead plus a chip) is an optimal large bet for this situation, however, in this case, I favour a small bet on the second last hand regardless of your opponent. The numbers below are estimates but the difference in the percentages are large enough to make the estmates sufficient. If you bet minimum, he will either a) correlate to virtually lock in his lead (skilled player). In this case, you will most likely still be within a 1/2 max bet on the final hand and you will have a ~43% chance of winning if he bets low and you bet high or a ~49% chance if he overbets again and you bet low. b) overbet, giving you the low (unskilled player). In this case you have a ~49% chance of being BR1 on the final hand with a ~88% chance of preserving your lead on the final hand giving a probability of .49 x .88 = ~43% for winning the table. You also have a ~51% chance of being BR2 on the final hand and most likely in the same situation as a) above with at least a ~43% chance on the final hand for an additional probability of .51 x .43 = ~22%. This gives an overall probability of at least .43 + .22 = 65% of winning the table. These are both superior to your chances of becoming BR1 with a big bet on the second last hand (~43% if you can do it with a single bet win, ~30% if you need to double) combined with the probability of preserving your lead on the final hand (~88% since you can correlate). A large bet also introduces the possibility of you ending up more than a max bet behind, should you get swung.
Thank you very much ! Very valuable comments. Now I do realise how bad I played that game. (I have a feeling that I was the weakest player out there) And due to that huge difference in favour to split the 10s, I think I had to split them even if my opponent would have drawn a hard 17, or if I were to act first..
You weren't the weakest player there if your opponent overbet his lead on the 3rd last hand! A large part of the power of splitting for the high in the original situation comes from the fact that your opponent stood stiff before you acted. When an opponent does this, it allows to to achieve perfect correlation with his hand simply by not busting. If your opponent were to have stood on 17, suddenly the chance for a push/lose swing must be considered. There are more ways in which you can lose and therefore less value in going for the high by splitting. If we consider what you should do for other hands your opponent might have achieved, we will see that, as his hand improves, the importance of you taking the high increases, to the point where, should your opponent achieve 20 or better, you should split against every dealer up card. The optimal strategies for your opponents single hand outcomes are: Code: Opponent You 2 3 4 5 6 7 8 9 10 A ---------------------------------------------------------------- Stiff T,T | p p p p p p p p s s 17 T,T | p p p p p s s s s s 18 T,T | p p p p p p s s s s 19 T,T | p p p p p p p s p p 20 T,T | p p p p p p p p p p 21 T,T | p p p p p p p p p p BJ T,T | p p p p p p p p p p If your opponent were to act after you, your best play depends on his skill level. If you know he is a basic strategy player, then he will stand vs the dealer's 6 and the situation is the same as the original one, so you should split. If he plays optimal strategy after you, then you are better of standing (57.8%) vs splitting (56.8%), but not by much.
I would add some different scenarios, relatively close to the examples above: We are BR1 with 2201 chips first to act on the last hand and we did a bet of 400. Our opponent BR2 with 2000 chips responded with a bet of 802. 1. We have received TT, our opponent a hard 17 and dealer upcard shows 7. - I would split and stand on anything not busting. 2. We have received TT, our opponent a hard 18 and dealer upcard shows 8. - I would split and stand on anything not busting as long as I get at least one total of 18 or higher after split. 3. We have received TT, our opponent a hard 19 and dealer upcard shows 9. - I would stand. PS: But what if our opponent would have soft totals instead of hard ones ? - I would stand on all of the above. *** I am sure that I may be very wrong playing this kind of positions, but I would like to know about it so I can improve.
