Which is why clubs have table limits. If you’re playing a $5 minimum bet game, be it roulette, craps, or 21, typically the maximum bet is $200. If you try to use a Martingale system, you’ve got only 6 bites at the apple to try and win $5 – 5, 10, 20, 40, 80, 160 – oops.
For a lot of sports bettors, I’m not sure making money is the point. It’s just fun to think you know a little more about the game than the next guy, and even more fun when you turn out to be right. If you’re a net loser – as most will be – you’ve paid a small price to have that fun.
Also – I know a guy who’s a rabid fan of his alma mater’s teams. He always bets against them, so at least if they lose he wins a little cash to salve his pain. 
That would be someone like me, who bets $25 or so per game and tries to get a drink or two in the sports book. I get to hang out watching just about any game in the world.
On the other hand, I definitely have known degenerate sports gamblers, I’ve ran across quite a few in the very male dominated world of financial market trade desks, and it’s sad to see them sweating the mortgage money on some game.
Well, the average person doesn’t bet all that much.
Let’s say you put $500 down on your team. The house vig is 10%. So it costs you $50 to make the bet. But you only make the one, and you get several hours of entertainment out of it. So maybe $15-25 per hour for the added entertainment of the bet. But of course, you could also bet $50, and then your entertainment is only $1.50-$2.50 per hour. Or you could bet $50,000.
The vig is certainly high, but it’s also of a different nature, because it requires a judgement call. Maybe not on the bookie’s part, who is usually adjusting the odds to balance his lines, but on the part of the bettor, who is simply betting against the wisdom of the crowd. The crowd is usually right, but the bookie doesn’t care so long as he’s getting equal action on either side of the bet. His risk is covered and the vig guaranteed, but from the player’s perspective it’s more like a house rake in a poker game. You aren’t playing against the house - you’re playing against the crowd. The house is just taking a cut.
Most people cannot beat a sports book, But it is possible with local knowledge to cut the odds. And there are people who can beat sports. These days it’s mostly done with college basketball, as I understand it, because there are so many games the bookies can’t keep great track of all of them, and someone who studies one particular team hard can sometimes find an advantage.
Also, the more thinly traded the sport is, the better the chance that the line is off. Also, more games means less variance and a better chance that the outcome of the year represents the actual expectation and not just luck. That’s why it’s possible to beat sports - the house actually doesn’t care.
But it’s really hard to beat the wisdom of the crowd. The odds you have to fight are emergent from the decisions of a whole lot of people. Like markets, it’s usually a pretty efficient way of discovering prices or value. Or in this case, the true odds of a matchup.
Sometimes the crowd is sentimental though. You might do well to bet against the home team in a big game with a local bookie. Their odds may be skewed against Vegas odds. Overall thought it’s a risky proposition.
Back to blackjack …
I found playing by the best statistical approach got very boring. Sure, I might get the least-bad return if I played all night, but it felt rote and joyless. (Plus, I hated that if I busted and the dealer subsequently busted I still lost.) It was a lot more fun when I started playing “wrong” and occasionally won a hand I “should have” lost. My results may have ultimately been worse but I wasn’t there to make money, I was there to have fun.
Bookies are generally really careful to keep their lines close to everyone else’s, for fear of being ‘middled’. If you can find two bookies whose lines are far enough apart to cover the vig, you can bet both sides of the game and guarantee a profit. And if the profit is guaranteed, you can bet the farm. The bookie whose line was wrong gets slaughtered. So most don’t take the risk.
Middling could be done profitably many years ago before the internet, where sports gamblers would subscribe to fax services and such that would alert them when a line on a game is way out of whack. Today with the internet, that’s extremely rare - the bookies just don’t let their lines move out of the vig window against others.
Whatever floats your boat. Gambling should be about fun. If you want to do that, however, I suggest you learn which ‘mistakes’ are least costly. For example, if you hate hitting 16, feel free to stand when you are against a 10, where the ‘wrong’ decision costs very little. If you stand on 16 against a 7, however, it’ll cost you 11% of your bet in the long run over hitting. The difference between facing a 7 and a 10 should be obvious.
Another aspect of playing against basic strategy: It’s less costly to do it in rare situations. For example, making some split errors, which don’t come ip that often. But you get faced with ‘potential bust’’ hands constantly, and if you play them incorrectly it will cost you a lot. For example standing on a 12 against a dealer’s 2 or 3 will cost you 5%.
On a related note, some people think that adopting a ‘never bust’ strategy, or playing by the same rules as the dealer will give them some advantage. Those are really, really bad ideas and would turn blackjack into one of the worst games in the house for player expectation
That reminds of an episode of “The Amazing Race” many years back. It was a family season, so 4 people per team. In one of the tasks, they went to a casino, and had to play rounds of blackjack until everyone won on the same hand.
Now 2 seconds of thinking shows that the easiest/fastest way to do that is never hit, and just wait until the dealer busts. But the dad of one family was convinced he was a blackjack expert, and was coaching his wife & kids what to do, with the unsurprising result that at least one of them would bust on every hand. I think it took them an hour to get out of there, when every other team figured out the “never hit” strategy within a few minutes.
Have you considered playing a video poker game that has a low house edge but is more volatile than blackjack? You can get more of a gambling rush with the ups and downs.
Me too. What I do is press my winnings twice and only pull it back and start over after 3 wins. I figure I’m playing with house money and getting a discount on splits and doubles but more over it’s thrilling to build up chips over a few hands and get a big win with low outlay than grind a quarter at a time.
No, I like the feel of a live game with a casino crowd. But not during a pandemic.
Me, too. I also like “wrong” splits like 9s or even 10s against a bust card.
Most casino blackjack tables now have some kind of side bet, like Lucky Ladies or whatever, where you can make additional bets on odd things happening - getting a perfect pair, things adding up to other things, stuff like that. Many are quite cheap, where even if the table minimum is $15 per hand you can play the side bet for a buck. You can play perfect strategy in blackjack but waste a little money on the side bet for extra fun.
If you want to bet a little more sometimes or change your strategy, you might consider a very simple count system. Nothing formal or difficult. But you could do something like this: On the first hand after a shuffle, if you see a lot of low cards come out and almost no tens or aces, just double your bet for the rest of the shoe, or until you see a big cluster of high cards go by. Then bet normally after. It won’t make you a winning player, but it will cut the house odds slightly.
An even simpler way is to just count Aces. The more aces left in the deck, the better it is for the player. On average, every 13th card should be an Ace. So let’s say you are at a table with four other people. That’s a minimum of 12 cards per round, including the dealer. So if you don’t see an Ace in the first couple of rounds, the deck is richer in Aces than average. Double your bet. If Aces catch up, lower your bet.
Again, that alone won’t make you a winning player, but it will help, and it will mean you will on average have more money in play when you get a blackjack, which is fun.
Or, you can do the card counting thing, which honestly isn’t hard. A simple hi-lo count is all you need for multi-deck games. Count 2-6 as +1, and tens and aces as minus one. The easiest way to,do that is to count in pairs. For example, wait until the dealer has dealt 2 cards to everyone, then look at each player’s hand. two low carss is +2, A ten or ace with a 2-6 is 0 and can be ignored, and a blackjack or two tens is -2. A neutral 7,8 or 9 is +1 when paired with 2-6 and -1 when paired with a ten. Keep the ‘running count’ in your head. When it is positive, your advantage grows. When it’s negative, the house advantage grows.
There are also playing changes you can make based on the count, but that requires more work. For example, you should always stand on 16 against a 10 on any positive count, and hit it on any zero or negative count.
There are 18 strategy changes to memorize if yoh want to be a good card counter.
The only other step is to divide the running count by the number of decks left to get the 'true count. So,if your running count is +6 and there are two decks left the true count is +3 - which just happens to be about the point where the player has an advantage over the house. So crank up those bets!
An evening of practice dealing out cards and counting is about all you should need to be able to do it in a casino. And even if you make mistakes, you are still better off than not counting at all.
If you want to take a bathroom break, wait until the count is negative. Just avoiding the negative count decks will improve your odds.
If you don’t want to count, then If the first hand of the shoe spits out a large number of face cards and/or aces, that’s a good time to take a break. The rest of the shoe will be worse for the players on average.
Here’s a good introduction to simple card counting:
My only nitpick with what you say is that I don’t think you would need a computer to calculate it. This isn’t chess or poker, there aren’t actually that many states.
For a of the 10 possible up cards by the dealer it is very easy to work out the probability of the dealer busting or stopping at a particular value between 17 and 21. Meanwhile there is only a 30 or so possible states for the player (allowing for aces and busts but ignoring doubling for now), and the probability transition between them based on whether you stay or hit or double is pretty easy, so all you have to do is work out the probability of winning if you stay hit or double in a given state, and choose the one that will give you the best expected value. Deciding whether to split 8s then becomes whether your expected value is better with a single hand with a value of 16 or two hands with a value of 8. The only hard part as I see it is dealing with the possibility of a 5 card charlie in which case the state of the player depends on the number of cards as well as their total value.
All of this would be pretty easy to work out in a day or so with pencil and paper and a pocket calculator. With just pencil and paper it would take maybe a week. But in any case, I’m sure that someone worked out the theoretical optimal strategy long before computers came on the scene.
Believe me when I say that I actually HAVE written a computer progam to calculate it, and there are millions of possibilities. I have also written a basic chess playing program. (It didn’t play very well). They are not entirely dissimilar. The blackjack tree is probably larger. There are known methods of pruning the chess tree, that is eliminating unproductive branches of the search early. I don’t know of any method of pruning the blackjack tree.
You are underestimating just how large the probability tree is.
Consider this, your hand is 6-6. Your choice, depending on casino rules, is stand, hit, double down, split, surrender. What if you hit? There is a chance that you might get an ace. So, now you have a choice, do you hit again or stand?
If you keep hitting, there is a small chance that you might end with:
6,6,A,A,A,A,A,A,A,A,A
It isn’t very likely to happen, but you must calculate it.
And then you must calculate the possibility of getting
6,6,A,A,A,A,A,A,A,2
6,6,A,A,A,A,A,A,A,3 (BUST)
6,6,A,A,A,A,A,A,2
6,6,A,A,A,A,A,A,3
6,6,A,A,A,A,A,A,4 (BUST)
6,6,A,A,A,A,A,2,A,A
And so on.
You can see that this makes a very large tree.
And for each of these you must consider the dealer’s possible response. That, too, is a very large tree. Rules are that he must draw until he has a soft or hard 18.
The dealer has , say, a 3 showing, and a hidden card.
Dealer might draw:
3,A,A,A,A,A - STAND
3,A,A,A,A,2 - STAND
3,A,A,A,A,3 - STAND
3,A,A,A,A,4 - STAND
3,A,A,A,A,5 - BUST
3,A,A,A,2 - STAND
and so on.
And remember, you have to calculate the dealer tree for every node in the player tree where the player doesn’t bust.
And that is just the calculation for stand. Splits add another level of complexity.
Double down or stand is comparatively simple, but you still have to calculate the dealer tree.
No sensible decision tree would ever reach the point of considering two sixes and nine aces, because well before then, you’d have reached the point where your chance on even one more hit would be suboptimal compared to the chance from staying, and therefore you wouldn’t consider anything that would happen after that one hit. Further, even to get to that point, you’re already duplicating a lot of effort, because the strategy when you have (say) two sixes and five aces is exactly the same as the strategy when you have a 9 and an 8, or a king and a 7, or a five and two sixes, or any other combination that adds up to 17.
Your state after your first card is completely described by one of 18 different possible total values, plus one more bit to indicate whether it’s hard or soft. The dealer’s state is completely described by one of 10 possible values showing and one of 10 possible hole card values. That’s a total of 3600 possible states.
Chronos, have you ever actually constructed a decision tree for Blackjack?
I have.
In my experience, there are occasions where the EV for hit, and the EV for stand are very close together, and what you should do really does depend on the exact composition of your hand.
We’re talking about a fraction of a percent, but its enough to make a difference to a professional gambler.
Or even someone like me who’s annoyed about the constant blackjack changes for the worst. Once I find a 3:2 game at a table minimum that’s in my budget, it’s almost assuredly a 6-8 deck game from a shoe. Probably no surrender option and no double down after splitting either.
Here’s a case in point, strategy cards for single deck blackjack.
Initial two cards, total 12. Strategy is different if you have 10+2 or 9 +3 or 8 +4.
Three cards, if you have 6,A,A vs a dealers A, you should hit. Any other soft 18, and you should stand.
And various other exceptions.