When you start off playing Blackjack and go look for a basic strategy, you'll come across a bunch of basic Blackjack Strategy Charts, much like ours. These charts focus on the total value of your starting hand, and therefore are not a 100% accurate. In this article we will take a look at the standard multi-deck, dealer stands at soft 17 chart, and see how can we improve upon it.
The two key words in the sentence above are 'total value' and 'starting hand'. The total value includes several different starting hands, e.g. if you look at the row labelled 12, that means a starting hand of 2T and 75 at the same time, even though you have to play them differently in certain situations. The starting hand is also an issue, since on a lot of occasions we have more information: we either drew more cards and stood on that total, or we can see other players' cards, which can have a huge influence on our decision. It's not the same if you have 16 points via having a T6 starting hand, or say a 2345 hand, which we can get to legitimately. This article will showcase most of the situations in which you want to deviate from the Blackjack Strategy Cards available on casino-strategy.com.
About Expected Value
Now, before we get into it, we'd like to explain how we got to the conclusions in this article series. Every decision in gambling has an expected value. It's basically the end result of infinite runs of the same decision on the same situation. For example, if you have 65 versus a dealer 4, you have a very good doubling opportunity. In some situations, you'll spike a T and get 21 points instantly, in others you may get a 2, and the dealer might hit 19 and you lose. But the frequency of these events are not equal, one will happen way more often than the other. The Expected Value of doubling here is 0.587, which means you'll get 1.587 times your bet back here in the long run. The Expected Value of just hitting is 0.294, which is still positive, but way less than the Expected Value of doubling, therefore doubling is better.
In all of the situations listed below, the Expected Value of two decisions are very very close, and depend on minor details, like how many tens are out in the opponents' hands. In these situations the difference between EVs are small, but not negligable. Since there are a lot of situations where these borderline decisions occur, these add up slowly but steadily in the long run.
There are a lot of ways to determine the highest EV decision. In our case, we went with Threshold Analysis, which basically means we increase extreme (or in most cases, not that extreme) circumstances to the point, where we 'break' the rule. We take a simple situation, look for the easiest way to 'break' it, and take that as a baseline. We cannot list all the situations, because that would take ages and would not fit into a single article or charts, more like a lexicon. But these simple situations cover more than 95% of these infrequent situations, and the rest is even closer in terms of EV - in a way that probably will not affect our short or long term rewards.
The Blackjack Charts are surprisingly accurate in regards to most starting hands. There are basically two situations in a 6-deck, Dealer Stands at Soft 17 game (Double and Surrender allowed) in which case our starting hand will make the difference with the same total value.
The first is when we have a starting hand of 12 versus a dealer 4. In the Blackjack Strategy Chart you will see you should stand here. That is actually the case unless you have a 2T. In this case hitting is better by around 1%. It's not a favourable situation for us either way, but with hitting here we can get a little bit of an edge over those playing just by the chart.
In the second situation we hold 15 versus a dealer ten. This is one of the famous Surrender-situations, one of the very few cases you should do so... unless you have exactly 87. If you have 87 hitting is actually a better option than giving up half our bet for free (this is always a -0.5 EV play). Though the EV of hitting is still -0.49, it's still .01 better than surrendering. When we play in a game with such a low house edge, every little bit of advantage we can get is crucial.
Looking at our neighbours
Let's assume that we are still at our starting hands, but we've got others playing with us (or we play in a multi-hand table online). In this situation we have much more info than our starting hand. There are 5-10 additional cards (also called dead cards) which might very well influence our decision. In the following part we will list the situations where these 5-10 cards can have a relevant effect on our decision-making. Keep in mind that these are extreme situations, not calculating other cards into the equation. (E.g. if we say you need 2 or more tens to change your decision, that does not include that there might be 2s and 3s out there, that counteract the Ts.) As a general rule of thumb, subtract any number of low cards from the number of high cards or other way around to get the 'real' value of high cards out there (so if the board reads TTT2, it counts as two Ts).
When you have a total value of 8 or below, you want to hit regardless of the 'dead cards'.
When doubling with hands 9-11, the only situation where you don't want to double is if you have 9 vs a dealer 3, and there are no less than 7 cards worth ten out there. It is a very rare occurance, but not impossible.
Holding 12 versus a dealer 2 or 3, seeing 7+ middling cards will change our decision to stand rather than hitting. Middling cards are 6-9s generally.
Holding 12 versus a dealer 4 is a lot different. There are only 2 Ts have to be dead for hitting to be the good play. Versus a dealer 5 7 tens, a dealer 6 5 tens have to be dead. As a general rule of thumb, never stand versus a dealer 7 regardless of the dead cards.
If you have 13 points in hand, and the dealer has a 2, 5 tens have to be dead for us wanting to stand. In all other situations the threshold is so high, we won't even list them here. You can find them instead in the PDF below.
Surrendering by dead cards
Surrendering is always a big loss. Half of our bet for sure is a huge loss, while sometimes being correct. But as you will see from the following, even if we can find the smallest edge, surrendering becomes an option we don't really care about.
If you have a 9 and a 6 versus a dealer T, you only need 2 dead tens for hitting be better than surrendering. If you have a T and a 5, 3 other tens are required.
With any cards worth 16, you need 2 dead tens versus a dealer 9 for hitting. Versus a dealer A you need 6 or more tens to be dead.
As you can see, there are several situations in which he Basic Strategy Chart is not correct, and recognising these situations and acting differently can bring the house edge down quite a bit. But this is not all, if you're interested in more of these strange scenarios, take a look at our next episode of Advanced Blackjack Strategy. If you're interested in even more extreme scenarios (ones that are unlikely to pop up during a hand of blackjack), take a look at our PDF on extreme situations (coming soon).