Blackjack Tournaments with Elimination Hands

The Blackjack Page

Blackjack Tournaments with Elimination Hands - Part 3

When we're counting the cards, about 10% of the time we can vary the play of our hand according to the count, something we call "Basic Strategy Variations" (BSVs). Probably the best example is a hand of 16 versus a dealer's up card of 10. Basic Strategy says that's a hit, but if the count is above 0, the better play is to stand. Thus, the BSV for 16 v. 10 is to stand at any count above zero. Every hand in Blackjack has just one correct variation, like standing with 12 vs. 2 instead of hitting, doubling 11 v. A instead of hitting and, yes, splitting 10s v. 4, 5 or 6. But those plays should normally be made only when the count calls for it and the Blackjack tournaments played at Ultimatebet.com use 6 decks that are shuffled after every round of play, so counting the cards is of no use. (Darn it! The tournaments at Global Player Casino also use 6 decks, but they deal out 75% or more of the cards before a shuffle. I miss that a lot...sigh.

But just because counting the cards isn't possible, it doesn't have to mean that a Basic Strategy Variation is never the correct play in a tournament. There may come a time when you need to get more chips on the table or cannot risk busting your hand and making a Basic Strategy "mistake" may be your only choice. Yes, I called it a mistake on purpose - if you make a BSV without having the count to justify the play, you're making a mathematical mistake, plain and simple. Of course, your entire tournament may be riding on the hand, so the benefit of winning will likely far outweigh the "expected value" loss of the play, but don't kid yourself - there is a cost associated with failing to play perfect Basic Strategy - and, while it may not show up in this tournament or the next one, or even next week's tournament - it will eventually be collected from you. What I'm trying to say here is that you should vary from proper Basic Strategy only when absolutely necessary or be prepared to pay the price.

You won't have to play very many Blackjack tournaments to discover that players just love to split 10s. Most would never do it in a "cash" game, mainly because it draws such a heated reaction from the other players - "You're messing up the order of the cards, man!" - but in a tournament, it's apparently a different case. (It's BS, of course; how others play their hand will have no long-term impact on your results.) I have even seen players in a tournament make weird plays like hitting a hard 18 because they thought their actions may cause the table to lose, but in reality, they're only hurting themselves. I never comment on it, but if I did, it would be only to encourage them to continue with their "sophisticated" play. Sure, a player might take the dealer's bust card and cause you to lose the hand, but in the long run, they'll help you as much as they hurt you - they might also take the dealer's 5 that would have given him or her a 21. Anyway, let's get back to splitting 10s. You need to remember this: By splitting 10s, you're not only giving up the very strong hand of 20, but are also betting more chips for the privilege. If you need to get more chips on the table, that's a good way of doing it, but there's a cost. How much? Glad you asked.

Your hand of 10-10 versus a dealer's up card of 6 has an expectation of winning 67.7% of all the $$$ you bet in that situation when you stand. If you split the 10s, your expectation drops to 25.25% on two hands, which totals 50.5% of your original bet. (Don't forget that you must make a bet equal to your original bet in order to split a pair.) So by splitting, you're giving up 17.2 percentage points or nearly 30% of your original expectation. Yes, I understand splitting the 10s may be your only choice to, say, survive the elimination hand or win the match, but to do it as a matter of course throughout a tournament is very costly. The only saving grace is that 10-10 vs. 6 is a hand you'll get only about 723 times in every 100,000 hands of play, so you don't get that many chances to split them. The corollary to that is, of course, if you get it so seldom, why mess it up? But don't tell the "splitters" that....let them continue to have their fun and reduce their chances of beating you.

You'll also see a lot of players just hit an 11 versus a dealer's up card of 10 and you should applaud their timidity. As you might guess, it's a mistake to not double here; this is a hand you'll get about 1.66% of the time, which doesn't make any statistics associated with it earth shattering, but a mistake is a mistake and we all make enough without trying, so why add to them? Sure, it might be that betting more on this hand and losing could cause you to fall behind the lucky stiff who was LB-1 and is now sitting there with a 20, but not doubling 11 vs. 10 on a regular basis is costing you 5.9 percentage points, which is about a third of the expectation of doubling. The expectation for hitting 11 versus a dealer's 10 is to win 10.7% of all the $$$ you bet in that situation, whereas by doubling, the expectation is 17.8%. The numbers I'm showing here are situations where you double for the same amount as the original bet (you may double for less) and are based upon the return on the original bet. In other words, the percentages recognize that you're doubling the bet, but in this case specifically, the return isn't two times 10.7% or 21.4%, mainly because you get only one card on a double, but it is 8.9% on each bet, which totals to 17.8%. Clear as mud? I'll explain.

When I speak about the percentages here, I say "percent of all the $$$ you bet in this situation" because that's the most accurate way to describe it. As an example, if you were to always bet $10 per hand in a "cash" game of Blackjack and you were always dealt a hand of 10-10 against a dealer up card of 6, you'd have an advantage of 66.7% over the house. (Hell, the "house" would soon be yours.) Your expected value would be $6.67 per hand, but you would win or lose $10 at a time - losing when the dealer hits to 21 and basically winning every other hand - which basically means you'll win 667 hands out of 1000, pushes ignored, and it's better expressed (for me, anyway) on a per-hand basis. But if you wish, you can think of it this way: If you stand with 10-10 versus 6, you'll win 66.7% of the time, but if you split the 10s, you'll win only 50.5% of the time. Remember, when you split 10s, sometimes you'll win one hand and lose the other, resulting in an overall "push" or you'll win both or lose both, lose one and push one or win one and push one, etc. Take all of those possible outcomes and your expectation is to win 66.7% if you stand and win only 50.5% if you split. When you split, you're creating at least one additional hand and those "extra" hands kind of mess up the equation, but you have to count them, regardless. The best way is to calculate everything off the original bet, rather than start trying to figure out just how many times you might split the 10s (you may split to create 4 hands at Ultimatebet.com), whether one hand will win and the other may push, etc., etc. This is probably no clearer than before, but I think you get the idea.

Perhaps this is a good place to introduce my "Tournament Basic Strategy Variations" matrix, which is attached here as a separate page that you may copy for personal use. What I've done is list most of the BSVs a card counter will use when playing a cash (non-tournament) game and calculated the cost of making that variation when the count does not justify it. For example, Basic Strategy says to always hit a hand of 10,2 versus a dealer's up card of 2. However, if one is counting the cards using the Hi/Lo count we teach here, the Basic Strategy variation is to stand on 12 vs. 2 when the True Count is 3 or more, so you can see that standing with 12 vs. 2 when you're not counting the cards is a mistake. If you hit 12 vs. 2, the expectation is to lose 25.2%, but if you stand the expectation is to lose 29.0%. No matter what you do with a 12 versus a 2, you're going to lose in the long run. But at least by hitting, you lose less. Now, understand that I hate 12s, but it doesn't affect the percentages - hit, lose less; stand, lose more. But, there may be a point in a tournament where you simply cannot afford to bust, so standing is your only option. If you do, it's costing you 3.8%, which is not a big deal on one hand, but if you do it all the time, in the long run it'll add up. So that's what I've done here - give you the cost of making a Basic Strategy variation when the count doesn't justify it.

You can see that I've categorized the BSVs by Hit or Stand, Hard Double, Soft Double (hands with an Ace), Pairs and Surrender (late surrender, which is allowed in the tournaments at Ultimatebet.com/). As I mentioned earlier, there is one BSV for each hand, even though a hand like 8-8 may have several ways of being played. Obviously, you can hit 8-8, stand with 8-8, split 8-8 or surrender 8-8, but in a game that allows surrender, your choices are to split or surrender. In a cash-type of game, there is never a proper time to stand with 8-8 or just hit 8-8, although there may be a time in a tournament where that could be the case. Consequently, you'll find the hand of 8-8 in the Hit or Stand category, the Pairs category and the Surrender category. The proper way of playing 8-8 is to split it against all dealer up cards, except an Ace. If the dealer has an Ace up, you're better off surrendering, as my numbers will show.

Let's talk about those numbers for a moment. Mine came from an excellent program called "Blackjack Game and Basic Strategy Calculator (Version 5.0)", which is graciously made available for no cost by Eric Farmer at www.bjmath.com/. A tip of the GameMaster's hat to Eric for being a cool dude. This is a DOS-type of program (remember that?), but even if you don't know how DOS works, you can make this run. DOS programs do not recognize a mouse and its pointer, so you have to enter everything via your keyboard, but if you'll read the Readme.txt file that comes with it, you'll be up and running in no time. All you have to do is double-click the Strategy.exe file, answer some questions about the game you're playing and then you can analyze any hand for that game. You may also analyze three- or four-card hands, like I did in the Hit or Stand section of the matrix. Down near the bottom, you'll see a player hand of 10,2,4 versus a dealer's up card of 10. I put this hand in to show you what can be done if you have a starting hand of 10,2 versus a dealer's 10 and you hit it with a 4, giving you 16. Hit? Stand? Well, as the numbers show, it really doesn't matter because the expectation is to lose 54.1% either way. Of course you may not surrender because it's a three-card hand at this point, so just flip a coin.

In another column of the matrix I've made comments about various hands where I thought it might be helpful, plus I highlighted some plays with red (dumb play) and green (not a very big mistake) and one hand in blue (8-8 in the surrender section) so that you'll look at the footnote. Beyond that, the numbers should speak for themselves, but if you have any questions, doubt, concerns or disagreements, don't hesitate to email me.

I'll see you here next time.

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