Blackjack Tournaments with Elimination Hands

The Blackjack Page

Blackjack Tournaments with Elimination Hands - Part 5

Whenever I'm playing a "cash" game of Blackjack (as opposed to a tournament), I eagerly double hands like A-2 versus the dealer's up card of 6 because I know that play will produce a nice profit in the long run. Ah, there they are - those words - "the long run." But taking the long view is really the only way a professional Blackjack player can expect to succeed. Look, I know I can win at Blackjack on a consistent basis, yet I don't bet my entire bankroll on any one hand because I could easily lose that hand. But if I play several hundred hands at a decent game, I'm highly favored to show a profit for the effort.

In a multi-table (MTT) Blackjack tournament where one may play 75 or 80 hands at the most or in a single-table "sit & go" (SnG) tournament where one will play no more than 30 hands, those words - "the long run" - have much less meaning. The long run counts if I play a lot of tournaments, but in the case of just one specific tournament, playing my hand strictly according to Basic Strategy can oftentimes be a mistake. No, I'm not suggesting that you should start hitting all of your 17s against a dealer's up card of 10, but there will be times when that's the proper play. I discussed the role of Basic Strategy variations in a tournament setting in Lesson 3 of this series, so you might want to take a look at that when you get a chance. In there, I showed the difference between "expected value" in situations like standing with 12 against a dealer's 3, but that's not what I'm going to discuss here.

This time, I want to talk about doubling and splitting from more of what I call an "absolute" basis - which I define as winning or losing that particular hand. In other words, if you're playing Hand #1 of a tournament and you've bet the minimum, it really doesn't matter if you double, say, A-2 versus a dealer's 6 and go on to lose the hand; you can make up the loss in subsequent hands. But it might make a difference if you've placed a big bet on an elimination hand and now must decide whether or not you should double. A loss might cause you to be eliminated or a win might put you in first place and that cannot be measured by expected value alone. Just for the record, doubling A-2 vs. 6 in a game where the dealer hits soft 17 has an expected value of +.204, whereas the expected value for just hitting is +.166, so not doubling is a pretty serious error in most cases. But what if you really need to win this hand? Obviously, you have a better chance of beating the dealer by hitting, because when you double, your hand receives only one additional card and you're basically hoping the dealer will bust, rather than expecting to get an 8 as your double card, leaving you with a very satisfactory 21.

The way I view it, there are two types of doubling hands: "safe" doubles and "dangerous" doubles. This applies to splitting pairs as well and I'll discuss them in a bit. But back to doubling for the moment - specifically soft doubling where your hand contains an Ace - a hand like A-2 versus the dealer's 6. We know the dealer will bust this hand about 44% of the time - again, in "the long run" - but what if your tournament fate is tied to this one hand? There's a 56% chance the dealer will not bust, so hitting may or may not be the better play. If you double a hand of A-2, there are only four cards of the thirteen available to you that will improve your hand: the 5, 6, 7 and 8. Obviously, an 8 gives you 21; a 7 gives you 20; a 6 gives you 19 and a 5 gives you 18. Am I ignoring the 4, which gives you 17? Yes, because the dealer hits soft 17, she, he, (it?) will end with a 17 about 36% of the time and I'm assuming that a tie will do nothing for you - you're going to have to win the hand in order to proceed in the round. You really don't want to feel too comfortable when you're sitting with a 17 and the dealer shows a 6.

Anyway, this brings us to some pretty simple math. If you double A-2 versus a dealer's 6, there are 4 cards of 13 that will improve your hand to the point where you might beat the dealer - "might" being the operative word here; there are no guarantees, obviously. Four of 13 is a probability of 30.8%, which basically means you will likely lose (or push) the hand about 70% of the time if the dealer doesn't bust. To me, a 70-30 ratio makes doubling A-2 vs. 6 a dangerous double, so I do it only when necessary or when a loss doesn't mean a lot to my situation. Here are the numbers for other soft doubling hands, with "improves" meaning you end with 18 or more:
A-3 improves with 4, 5, 6, 7 or 4/13 = 30.8%
A-4 improves with 3, 4, 5, 6 or 4/13 = 30.8%
A-5 improves with 2, 3, 4, 5 or 4/13 = 30.8%
A-6 improves with A, 2, 3, 4 or 4/13 = 30.8%
A-7 improves with A, 2, 3, 10, J, Q, K or 7 /13 = 53.8%
A-8 improves with A, 2, 10, J, Q, K or 6/13 = 46.2%
A-9 improves with A, 10, J, Q, K or 5/13 = 38.5%
A-10 improves with 10, J, Q, K or 4/13 = 30.8%

You can see that the best hand to double is A-7 because most of the time (a slim majority, to be sure) you'll actually improve the hand or at least not make it worse. Doubling A-10 (a "natural" as we call it here) in order to improve the payout from 3 to 2 up to 2 to 1 is an "iffy" proposition at best because you'll actually make it a worse hand about 70% of the time, so you really have to hope that the dealer busts. Now don't get me wrong; sometimes doubling your hand in the hope of a dealer bust is your only viable choice, so that's what you should do. But at least by becoming familiar with the numbers I've presented here, you can make an informed decision about your play. Just so you know, most "hard" doubles like 11 vs. 10 or 10 vs. 9 often don't need the dealer to bust in order to succeed, so I'm not covering them here, but I might discuss it in a future lesson.

Now let's talk about splitting pairs, which I've also placed in the "safe" and "dangerous" categories. While a lot depends upon which up card the dealer is showing, some pairs just beg to be split because they're a lousy hand otherwise. The prime example is, of course, 8-8, which is a 16 if left alone. To my way of thinking, the 8s are followed closely by A-A, which is a 12 that's magically transformed into two 11s when split. In the previous lesson I showed that just hitting A-A versus a dealer's 10 is a huge mistake from an expected value point of view, so make darn sure you're in dire straits before you choose to just hit them. My advice is to split Aces every time you get them, period. So, both Aces and 8s are in my "safe" category, which means you'll be much better off most of the time you split them. I also put 9-9 in that category, so long as the dealer isn't showing a 7, 10 or Ace. Basic Strategy says to split 9s versus 2 through 9, except for 7; otherwise stand and that's what you should do most of the time.

Don't forget that, unlike doubling, all hands created by splitting a pair (except Aces, of course) can take additional hits so splitting/not splitting is less of a problem when the dealer doesn't bust. That said, there are still some pairs where you're likely to need a dealer bust, which makes them "dangerous" to me. Pairs like 2-2, 3-3, 6-6 and 7-7 are both poor starting hands and poor splitting candidates. Naturally, we don't get to choose our starting hands, but whether or not to split them is our choice. In a tournament setting I almost always follow Basic Strategy when playing those pairs, but if an additional loss will hurt me quite a bit, I'll just hit or stand with them, depending upon what up card the dealer has. I seldom split 4-4 (even though it's proper Basic Strategy versus 5 and 6 when double after split is allowed) and never split 5-5 or 10-10, unless I'm desperate. For the life of me, I cannot understand the affinity for splitting 10s in these tournaments. As you can see in Lesson 3, it's a costly move from an expected value point of view, yet players do it all the time. I suppose I should just shut up and be thankful. With that, I'll shut up.

The Blackjack Page

2007 Articles

2006 Articles

2005 Articles

2004 Articles

2003 Articles

2002 Articles

2001 Articles

2000 Articles

1999 Articles

1998 Articles

1997 Articles