Advanced Blackjack Strategy - Part 3

In our previous articles, we took a look at the changes you should make from the Blackjack Strategy Charts when the dealer stands at soft 17. However, many casinos live and online offer games in which the dealer hits soft 17, making the mathematics behind the charts and therefore the deviations a little bit different. In this article we will take a look at scenarios and the way to adapt to them in a 6-deck, dealer hits soft 17 game.

Small differences make for big changes

At first sight the fact that the dealer has to act differently in such a rare circumstance may sound like a miniscule thing that most players do overlook, but in fact just this tiny change in the rules will make our strategy vastly different in certain cases.

Just to illustrate the situation, let's take a look at the following graph:

I won't bore you with the details of the simulation, let's just get into analysing this graph. As you can see, the lower the upcard is, the higher the chances are that the dealer has an extra chance to bust in a 'hits soft 17' game. The main exception from this is 6, since with a 6 the dealer will have A6 a significant amount of the time. This would be an instant stand in in a 'stands at soft 17' game, but the dealer has to hit it in our case. The other situation where the chance of busting spikes is if the dealer has an A. In this case, the dealer will surely have a soft hand, and even if it's not soft 17, it can easily become one (e.g. if they draw a 4 to a soft 13). In the case of other low cards the difference is quite small (but not mathematically negligable). At high cards there is no difference between the chances: there is no way the dealer can make a soft 17 with a 7+ upcard.

This means our plays will be quite different when the dealer has and A or a 6 as an upcard, and the thresholds might be a bit higher (or lower, if you wish) when the dealer has a low upcard. The overall chance of the dealer busting is increased by ~0.5% which leads to quite a big edge for us. Let's see how we can get an even bigger edge by straying away from the Blackjack Strategy Chart.

Doubling and surrendering

In this game we want to double more than using the 'standard' rules. The increase in the chance of the dealer busting will leave us more doubling opportunities and loser thresholds. Let's take a look at the most common cases.

If the dealer has a 2 up and we have 9 total, we should only hit, but not double. However, if we see 2 middling cards (7-9) dead, we should go ahead and double. If 3 small cards are dead, we should consider doubling again.

If we have 10 points total, and the dealer has an A, we shouldn't double generally, but if we see 4-5 middling cards dead, it becomes a marginal double. This is so borderline and depends so much on the exact cards that you might want to overshoot here and go for 6+ middling cards (not including ours).

Generally soft 13 is a double against a dealer 5, but even a single dead high card (8-T) makes it a hit, while if we have a lot of those dead (~3-4+) it becomes a double again.

Having a soft 17 ourselves (how ironic) we should generally hit against a dealer 2, but if there are two tens or one middle to high card (5-9) dead, we should go ahead and double. This quite nicely shows how easily we can improve upon the General Blackjack Charts, since unless the board is full of low cards, we need to make a deviation almost all the time.

Finally, soft 19 is a double against a dealer 6, but if we have 4-5 small cards dead, it becomes a marginal stand.

These are the situations your game of doubling changes, now let's take a look at surrenders. I promise, there will be jaw-dropping results here as well.

With a total of 15 against a dealer T we should surrender, unless there is a single ten or two high cards (7-9) are dead, in which case we should hit. If the dealer has an A instead, 2 tens or 3 high cards have to be dead. It's amazing how a couple of cards can change our decision so drastically out of a 312 card pool.

With 16 it's a lot simpler. If the dealer has a 9 and there are 2 or more high cards (7-T) are dead, we should hit instead of surrendering. In the case of us having a pair 8s, 1 high card (9-T) or 2 middling high cards (7-8) are the threshold, but we should split in these cases.

If we have soft 17 against a dealer A, and two or more tens are dead, we should stand instead of the usual option.

Other cases

Weirdly, cases outside of doubling and surrendering are mostly clear-cut, and no deviations have to be made unless the circumstances are quite extreme.

You can see how just a tiny change in rules can bring a great difference to the optimal plays especially surrounding decisions that hinge on the dealer busting rather than us having or making a made hand. This study is a great introductory on the decision-making process we should follow when playing without a chart or accounting for information other than our starting hand and the dealer upcard.