Blackjack has an unusual problem for a casino game: played correctly, it is too cheap. Basic strategy under liberal multi-deck rules brings the house edge to roughly half a percent, which is among the lowest numbers on a casino floor. Side bets are the industry's answer. They sit in a small circle next to your main wager, they pay in flashy multiples, and they carry house edges an order of magnitude larger than the hand they are attached to.
That is not a scandal. It is a design. But it is worth understanding what you are buying, because the numbers are not close.
What a side bet is
A side bet is a separate wager, resolved separately, on a pattern in the cards rather than on the outcome of your hand. Your two cards being a pair. Your two cards plus the dealer's upcard forming a poker hand. The dealer busting. Your total hitting some particular value. The main hand plays out as normal and the side bet is settled on its own terms.
Because it is separate, none of your basic strategy skill applies to it. There is no decision to make after placing it and no way to play it well. You put the chips down and the cards decide. That alone makes it a different kind of thing from the game beside it.
The published numbers
Here is the comparison that matters, using figures from the commonly published analyses. Note the ranges: they are wide because the same-named bet is not the same bet everywhere.
| Wager | Published house edge |
|---|---|
| Blackjack hand, correct basic strategy, liberal rules | around 0.5% |
| Perfect Pairs, common version, 8 decks | roughly 2% to 8% by paytable |
| 21+3, by paytable and deck count | roughly 3% to 9% |
| Insurance, as a standalone bet | roughly 6% to 8% |
| Lucky Ladies, 6 decks, common paytables | roughly 18% to 25% |
Read the first row against the last. The blackjack hand is priced at about half a percent of what you bet. A Lucky Ladies bet on some published paytables is priced at more than forty times that. Both are on the same felt, settled by the same dealer, in the same two seconds.
Why the same bet has so many different edges
Two variables move these numbers, and casinos control both.
The first is the paytable. Wizard of Odds analyzes four common paytables for one version of Perfect Pairs at eight decks and reports house edges spread from roughly 2.2 percent on the most generous to about 8 percent on the least. Same name on the felt, same rules for what wins, nearly four times the cost depending on which multiples the operator printed. A second version analyzed separately sits around 8 percent at eight decks and climbs steeply as decks come out. There is no way to tell these apart except by reading the paytable printed at the table.
The second is the deck count, and it behaves backwards from what blackjack players expect. In the main game, fewer decks favor the player: moving from eight decks to one is worth roughly half a percent, which is why single deck games are prized. For pair-based side bets the direction reverses. Once you hold a card, the chance that the next one matches its rank depends on how many matching cards remain relative to the shoe, and that ratio is worse in a small deck than a large one. Published tables for Perfect Pairs show the edge deteriorating sharply as decks are removed, in one version from around 8 percent at eight decks to figures many times larger at two.
So a player who has learned that fewer decks are better carries that instinct to the side bet circle and gets it exactly wrong.
21 TrainerSee how each rule moves the house edge with the built-in calculator. Free, works offline.
Get the app
Insurance is the side bet nobody calls a side bet
When the dealer shows an ace and offers insurance, that is a side bet on whether the hole card is ten-valued, paying 2 to 1. It is not a hedge, despite the name and despite the way it is offered, and treating it as protection for a good hand is the most common misconception in the game.
The arithmetic is straightforward. Four of thirteen ranks are ten-valued, so the hole card is ten-valued somewhat less than a third of the time, while the bet pays as though it should happen a third of the time. The gap is the house edge, and published analyses put it in the same range as the flashier side bets. Basic strategy declines insurance on every hand, including a blackjack, without exception.
The exception belongs to counters, and it is precise rather than general. When the remaining deck is rich in ten-value cards the bet crosses into positive territory, and the Hi-Lo insurance index is commonly published as a true count of about plus three. That is not intuition, it is a threshold, and it only means anything if you are actually counting. We covered the whole question in insurance explained.
What a side bet costs over a session
The edge is a rate, so its cost scales with how much action you put through it. Blackjack is a fast game, commonly dealt at dozens of hands an hour, and a side bet placed every hand is exposed on every one of them.
That is the part that makes side bets expensive in practice rather than merely in theory. A small side wager, placed every hand for an evening, can put a comparable amount of money at risk as the main bets do, at ten or twenty times the rate of loss. The main hand's low edge is doing its work while the circle beside it quietly undoes it. Our piece on what half a percent really means works through the same arithmetic on the main game.
Can they be counted?
Sometimes, in the sense that the published literature contains specialized side counts for a handful of bets. In practice the answer for nearly everyone is no. A standard Hi-Lo count tracks high versus low cards, which is the right information for the main game and mostly the wrong information for a bet that resolves on suits or ranks matching. Doing better means tracking something else entirely, alongside the count you are already keeping, under conditions the casino can change by swapping a paytable.
And where the composition is kept near neutral, by a continuous shuffling machine for instance, the question does not arise at all.
The honest position
Side bets are entertainment with a price on them, and the price is high relative to the game they sit next to. If a two dollar swing on whether your cards pair up is worth two dollars of enjoyment to you, that is a legitimate reason to place it, the same way a lottery ticket is. What does not hold up is the idea that a side bet is a smart play, a hedge, or a way to make a session more profitable. Every published number points the other way.
The useful habit is simply to know which number you are accepting before the cards come out. Read the paytable, count the decks, and remember that the cheapest bet on the table is the one you already know how to play well.
Spend your attention on the main handFree basic strategy training with an explanation for every decision. No account, no ads.
Download
Frequently asked questions
Are blackjack side bets worth it?
Not on the mathematics. Every common side bet carries a house edge many times larger than the blackjack hand beside it, which sits near half a percent with correct basic strategy under liberal rules. Published analyses put typical side bets somewhere between two percent and well over twenty, depending on the paytable and the deck count. They are priced entertainment, not value.
What is the house edge on Perfect Pairs?
It depends entirely on which paytable and how many decks. Wizard of Odds publishes eight-deck figures for one common version ranging from roughly 2.2 percent on its most generous paytable to about 8 percent on its least, with the edge worsening as decks are removed. A different version analyzed on the same site reaches about 8 percent at eight decks and far higher at fewer. Two tables with the same felt name are not the same bet.
Why do side bet odds change with the number of decks?
Because most side bets pay on matching cards, and matches get rarer as decks are added. After you are dealt a card, fewer of the remaining cards share its rank in an eight-deck shoe than in a two-deck game, proportionally speaking. Pair bets therefore get worse for the player with fewer decks and better with more, which is the opposite of the main game's behavior.
Is insurance a side bet?
Yes. Insurance is a bet that the dealer's hole card is ten-valued, paying 2 to 1, offered whenever the dealer shows an ace. Basic strategy always declines it because the ten-value cards are not frequent enough to justify the price. Card counters are the exception: the Hi-Lo insurance index is commonly published as a true count of about plus three, at which point the remaining deck is rich enough to make the bet worthwhile.
Can card counting beat side bets?
Standard counts like Hi-Lo are built for the main game and do not track what most side bets depend on. Specialized side counts exist in the published literature for a few bets, but they demand different tracking, deep penetration and unusual conditions to matter, and casinos change paytables freely. For nearly every player the practical answer is that the side bet stays a losing proposition.