actfap

Blackjack Side Bets

What 21+3, Perfect Pairs, Lucky Ladies and Royal Match cost, at your shoe size and pay table.

This prices one side bet at a full shoe. It says nothing about how to play the blackjack hand underneath it, and nothing about what the remaining cards are worth once a shoe is part way through.

Your two cards plus the dealer's upcard, read as a three-card poker hand. A pair pays nothing here, which is most of why it costs what it does.

Every figure below is counted over all 5,013,320 ways your two cards and the dealer's upcard can come off a full shoe of 6 decks.

Stepped: the flush pays little and the suited-trips line carries the advertising.

What the table advertises

100 to 1

Suited three of a kind

What it does not

1 in 4,821

hands at this shoe size, and $18.48 an hour to keep waiting for it at $5.00 a hand.

House edge on this bet

4.62%

Pays something

9.7%of hands

This side bet4.62%

The blackjack hand it sits beside0.41%

The blackjack hand this sits beside costs 0.41% at basic strategy on a six-deck shoe, so the side bet is about 11 times as expensive per dollar as the game it is printed on.

This pay table, at every shoe size

This bet gets 4.9 times cheaper for the player from 1 deck to 8 decks.

What it costs to keep making

Onto the felt / hour$400House edge4.62%Expected cost / hour$18.48

Over 3 hours

$55.45

This session3 hours

Average wait for the line on the sign60.3 hours

That is an average, not a forecast of your session. Most hands lose the side bet outright and a few pay small. The line on the sign arrives about once every 4,821 hands, which at 80 hands an hour is roughly 60.3 hours of play, and it is a long way from certain in that time.

Every figure here assumes a freshly shuffled shoe. A side bet placed deep into a shoe is priced by what is left in it, not by a full deck, so treat these as the cost of the bet over a session rather than the cost of the next hand. Side bets are gambling: play only with money you can afford to lose, and only where it is legal for you to do so. 21+.

Written and reviewed by Alexander Korsager · last reviewed September 26, 2026 · corrections

The idea

What this tool shows

A blackjack table sells its side bets on one number. The felt carries the top multiplier, the dealer announces it, and 21+3's suited three of a kind pays 100 to 1 in letters big enough to read from the next pit. What is nowhere on the felt is how often that line lands, or what the bet costs an hour while you wait for it.

  1. Those two facts are the whole story, and both move a long way with things a player can see.Perfect Pairs at the same 25-12-6 pay table costs 4.10% of every dollar on an eight-deck shoe and 22.33% on a two-deck one, because a perfect pair is two identical cards and a short shoe holds fewer of them. 21+3 with three of a kind paying 30 costs about 4.6% on six decks; cut that one row to 25 and it costs about 7.1%. The number a player half-remembers from another casino is usually the wrong number for the table in front of them.

  2. Royal Match takes that same lesson further than any bet on this page.One of its pay tables, 10 for the royal match and 3 for any other suited hand, prints the same two numbers on felt at every shoe size, and the shoe size decides which side of zero it sits on: it costs the player 3.77% on a single deck and pays the player 1.11% on six. Remembering the sign as well as the numbers still tells you nothing, which is exactly why that schedule is only ever dealt on a single deck; a multi-deck table paying it would be giving money away, not offering a rare edge to be found.

  3. The other comparison worth making is against the hand the bet is sold beside.At the default 21+3 pay table on a six-deck shoe, this bet runs about 4.6%, roughly 11 times the 0.41% a basic-strategy player is already paying on the blackjack hand underneath it. No felt states an edge that way, and the multiple moves with the shoe size and the pay table exactly as much as the edge itself does, which is why the panel computes it live from your own inputs rather than this page quoting one fixed ratio for every table.

  4. So this tool counts rather than looks up.Pick the bet, set the shoe size, and either take a pay table off the list or type in what your felt actually pays. Every frequency below is an exact count of the ways your cards can come off that shoe, and the panel puts the advertised prize, how often it arrives, and the hourly cost of waiting for it in the same view.

FAQ

Questions players ask

Which blackjack side bet has the lowest house edge?

It depends far more on the table than on the bet. 21+3 on a paytable paying 9 to 1 across the board runs about 3.24% on six decks, Perfect Pairs at 25-15-5 about 4.18%, Lucky Ladies between roughly 17% and 25% depending on the schedule, and Royal Match from about 6.46% down to a small player edge on one specific single-deck-only pay table, the widest spread of any bet this tool prices. Change the deck count or one row of the pay table and those figures move by several points, which is why this tool asks for both rather than printing a single number.

Why does the number of decks change the house edge so much?

Because the rarest lines need duplicate cards. A perfect pair is the same rank and the same suit, and a suited three of a kind is three identical cards, so neither can happen at all on a single deck and both stay scarce on a short shoe. Perfect Pairs at 25-12-6 costs 4.10% on eight decks and 22.33% on two for exactly that reason.

Does a soft 20 count for Lucky Ladies?

Yes, an ace with a nine totals 20 and pays the unsuited line, or the suited line if both cards share a suit. That is not an interpretation: leaving ace-nine out puts the six-deck figure for the standard schedule at about 25.4% against a published 17.64%, so the published figures only reproduce when soft 20s are included.

Do these figures hold once the shoe has been dealt from?

No. Every figure here assumes a full, freshly shuffled shoe. A side bet placed deep into a shoe is priced by the cards that are actually left in it, which is why side bets are countable in principle and why the number for one specific moment can differ from the average shown here.

How does a side bet's cost compare to the blackjack hand it's sold with?

As a multiple, which says more than the two edges on their own. This site's own six-deck, basic-strategy blackjack figure is 0.41%; 21+3 at the pay table this tool opens on runs about 4.6% on the same shoe, or roughly 11 times as much per dollar wagered. That ratio is not fixed: it moves with the shoe size and the pay table exactly as much as the side bet's own edge does, which is why the house-edge result computes it from whatever you have set rather than this page printing one ratio for every table.

Does Royal Match ever pay the player instead of costing them?

One pay table on this page prices that way: 10 for the royal match and 3 for any other suited hand runs a 1.11% edge in the player's favor on a six-deck shoe. That is not a bet worth going looking for. The same two numbers cost the player 3.77% on a single deck, which is the only shoe size a casino actually spreads them on, because a multi-deck table paying that schedule would be giving money away rather than running a game. Read a negative figure on this page as proof that the pay table and the shoe are one object, the same lesson the Perfect Pairs example above teaches, not as an edge sitting on a felt somewhere waiting to be found.

Methodology

How this is calculated

Every probability here is a count of card combinations, not a figure copied from a table.

For a given shoe size the tool counts how many of the ways your cards can come off a full shoe land on each winning line, divides by the total number of ways, and prices the pay table you entered against those frequencies. 21+3 is counted over every three-card combination of your two cards and the dealer's upcard; Perfect Pairs, Lucky Ladies and Royal Match over every two-card combination of your own.

Lucky Ladies' top line also depends on the dealer turning blackjack, which is worked as a conditional probability on the shoe with your two queens already removed.

Common misreadings6
  • A house edge is an average across thousands of hands, not what the next hand costs. Most hands lose the side bet outright.
  • A rare line landing early does not mean the bet is cheap, and a long gap does not mean the line is due. Each hand is dealt from a fresh shoe as far as this model is concerned.
  • The advertised multiplier is a payout, not a return. A line paying 100 to 1 that lands once in 4,800 hands returns about two cents on the dollar.
  • Two tables spreading the same bet can carry very different edges. The deck count and one row of the pay table both move the figure by several percentage points.
  • Switching the deck count reprices the side bet, not the blackjack hand it is compared against. Pick two decks and the times-more-expensive figure still holds against the fixed six-deck, S17 game above, not against a two-deck blackjack game, because this tool does not model how deck count moves the base game's own edge.
  • One Royal Match pay table, 10 for the royal match and 3 for any other suited hand, prices as a cost to the player on a single deck and a cost to the house on four decks or more: that describes which shoe a casino actually spreads it on, not an edge sitting at a real table waiting to be found, and the tool says so plainly rather than showing the negative figure as a win.
Assumptions6
  • A freshly shuffled shoe of the deck count you selected, with no cards removed.
  • The side bet is settled on the cards alone, with no dependence on how the blackjack hand is played out.
  • For Lucky Ladies, a soft 20 (an ace with a nine) counts as a 20, which is the reading that reproduces every published figure for the bet.
  • For Royal Match, every suited two-card hand is counted into one of three lines, the King-Queen royal match, a suited ace-and-ten-value blackjack, or any other suited pair, and the pay table assigns each of the three a payout; a schedule with one combined "easy match" line simply pays the same amount on the last two of those three.
  • Payouts are entered as a multiple of the stake, to 1, and the stake is lost on any hand that does not qualify.
  • The blackjack figure the panel and the FAQ above compare against, 0.41%, is one fixed game: this site's own house-edge figure for six-deck, dealer-stands-on-soft-17 basic strategy, read from the same engine the house edge and blackjack strategy pages use rather than retyped. It does not reprice itself when a different deck count is picked above for the side bet; only the side bet's own edge does. The hands-per-hour field opens on that same fixed game's pace, 80 an hour, for the same reason.
Limits4
  • A bet placed deep into a shoe is priced by the cards left in it, not by a full shoe, so a card counter's figure for a specific moment will differ from this one.
  • Lucky Ladies schedules that pay on a single queen, or on any pair of queens rather than two queens of hearts, split the hands differently and are not modeled here.
  • Progressive or envy-bonus variants, where part of the prize is a shared jackpot or paid on another player's hand, are outside this model.
  • The hands-per-hour default is a typical table pace, not a measurement of your table.
Evidence

The counting is derived in math.ts and checked two independent ways in math.test.ts. First, every count is compared against a brute-force walk of the shoe that classifies each deal by looking at the cards, which is a different method rather than the same arithmetic written twice.

Second, the resulting house edges are compared against 44 figures published independently by Wizard of Odds across four bets, fifteen pay tables and five shoe sizes (wizardofodds.com/games/blackjack/side-bets/, the first three bets retrieved 2026-09-08 and Royal Match retrieved 2026-09-16), to the two decimals those sources print.

Those are cross-checks, not sources: nothing the page renders reads them. A passing test means the code matches the model, not that a given table runs the rules assumed here.

Gambling can cause serious harm. The National Problem Gambling Helpline is free, confidential and open 24/7, call or text 1-800-MY-RESET (1-800-697-3738), or chat at ncpgambling.org/chat.

Related tools

Try next

We’d like to use privacy-friendly analytics (Google Analytics, Vercel Analytics and Speed Insights) to see which tools and pages are actually useful, so we can improve them. Nothing is sold, and nothing loads unless you say yes. See our privacy policy for details.