Chips & Truths No spin. Just the math.
Home/The Game Library/Craps/CRA 110: Craps Odds Chart — Probabilities, Payouts and House Edge

CRA 110: Craps Odds Chart — Probabilities, Payouts and House Edge

A practical craps reference that separates dice probability, true odds, casino payout, and house edge so similar-looking bets are not mistaken for equal value.

CRA 110: Craps Odds Chart — Probabilities, Payouts and House Edge
Point Value
House Edge 0% on true-odds portions; much higher on some proposition bets
Difficulty Medium
Skill Ceiling High

A useful craps odds chart should answer four different questions without mixing them together: How often can this result happen? What does the bet pay? What would fair odds pay? What percentage advantage does the casino keep? Those numbers are related, but they are not interchangeable.

That distinction matters in craps because several wagers can involve the same number while having completely different prices. A 6 can appear as a Pass Line point, a Come point, a Place 6, an odds bet on 6, a Hard 6, or a one-roll proposition. The dice total is the same; the winning conditions and payout are not.

Use this page as a lookup table. For the probability logic behind the numbers, see craps odds. For bet-by-bet recommendations rather than raw reference data, use best craps bets.

Start with the 36 possible dice combinations

Two fair six-sided dice produce 36 equally likely ordered outcomes. Totals near 7 have more combinations than totals near 2 or 12.

TotalCombinationsProbability on one roll
212.78%
325.56%
438.33%
5411.11%
6513.89%
7616.67%
8513.89%
9411.11%
1038.33%
1125.56%
1212.78%

The shape of this table explains much of craps. Seven is the single most common total. That helps the Pass Line on the come-out roll, but once a point is established the same 7 becomes the principal losing result for many bets.

The basic probability formula is:

P(total) = combinations for that total / 36

For a 6:

P(6) = 5 / 36 = 13.89%

That is the probability of rolling a 6 on the next roll. It is not the probability that a Place 6 wins before it loses, because that wager can survive many neutral rolls. Multi-roll bets require conditional probability instead.

Core craps bet chart

The house-edge figures below assume common traditional rules. Field bonuses, buy/lay commissions, proposition payouts, odds limits, and some variant rules differ by casino, so check the posted layout and house rules before treating any number as universal.

BetMain win conditionCommon payoutTypical house edge
Pass Line7/11 on come-out, or point before 71:11.41%
Don’t Pass2/3 on come-out, or 7 before point; 12 usually pushes1:11.36%
ComePass Line sequence beginning after come-out1:11.41%
Don’t ComeDon’t Pass sequence beginning after come-out1:11.36%
Pass/Come oddsPoint before 72:1, 3:2, or 6:50% on odds portion
Don’t odds7 before point1:2, 2:3, or 5:60% on odds portion
Place 6 or 8Number before 77:61.52%
Place 5 or 9Number before 77:54.00%
Place 4 or 10Number before 79:56.67%
Field2, 3, 4, 9, 10, 11, 12 on next rollRule-dependentOften 2.78% or 5.56%
Hard 6 or 8Hard pair before easy total or 7Commonly 9:19.09%
Hard 4 or 10Hard pair before easy total or 7Commonly 7:111.11%
Any Craps2, 3, or 12 on next rollCommonly 7:111.11%
Any Seven7 on next rollCommonly 4:116.67%

A payout ratio alone does not tell you whether a bet is inexpensive or expensive. Any Seven pays 4:1, which may look attractive beside an even-money line bet, but there are only 6 winning combinations and 30 losing combinations. Fair odds for a one-roll 7 would be 5:1. Paying only 4:1 creates the large 16.67% house edge.

True odds after a point is established

Once a point is on, only the point and 7 matter to the eventual resolution of a Pass/Come odds wager. Other totals simply delay the decision.

PointWays to roll pointWays to roll 7Probability point wins before 7Fair Pass/Come odds payout
43633.33%2:1
54640.00%3:2
65645.45%6:5
85645.45%6:5
94640.00%3:2
103633.33%2:1

For point 5, for example, there are 4 combinations that roll 5 and 6 combinations that roll 7. Ignoring neutral rolls, 10 combinations can resolve the race:

P(5 before 7) = 4 / (4 + 6) = 40%

A fair win must compensate for losing 60% of those resolved races, so the correct true-odds payout is 3:2. That is why the craps odds bet can carry 0% house edge on the odds portion: the payout matches the underlying conditional probability.

Why Place 6 is not the same as odds on 6

This is one of the easiest places to misuse a chart.

Both wagers can win when 6 appears before 7, so they have the same 5-versus-6 combination race. But they pay differently:

  • Odds on 6 pay 6:5, the fair mathematical price.
  • Place 6 commonly pays 7:6, a little less than fair value.

Suppose $30 is at risk.

At true 6:5 odds, a win pays $36 profit. A $30 Place 6 cannot be paid neatly at 7:6 because place bets are normally made in multiples of $6; a $30 Place 6 pays $35 profit. That $1 difference on each winning resolution is where the casino advantage comes from.

For payout-unit handling and returned-stake examples, see craps payouts.

Field-bet numbers depend on the bonus rule

The field wins on 2, 3, 4, 9, 10, 11, and 12. Those totals account for 16 of 36 combinations. The other 20 combinations lose.

If every field winner paid only 1:1, the wager would be poor. Casinos therefore usually pay a bonus on 2 and/or 12. The exact bonus changes the edge.

A common structure pays 2:1 on both 2 and 12. Under that rule:

Expected value per $1 = [(14 × $1) + (2 × $2) - (20 × $1)] / 36
                      = -$2 / 36
                      = -$0.0556

So the house edge is 5.56%.

If one of the two single-combination totals pays 3:1 while the other pays 2:1, the edge drops to 2.78%. The word Field on the felt is therefore not enough information; the bonus printing beside 2 and 12 matters.

One-roll proposition bets: read probability before payout

High payouts are visually prominent in the center of many craps layouts, but they are often attached to rare events with a large pricing gap.

Take Any Seven at a common 4:1 payout. Six combinations win and 30 lose:

EV = (6/36 × $4) - (30/36 × $1)
   = -$0.1667 per $1 wagered

That is a 16.67% house edge.

Compare that with Any Craps, which wins on 2, 3, or 12: four combinations in total. At a common 7:1 payout:

EV = (4/36 × $7) - (32/36 × $1)
   = -$0.1111 per $1 wagered

The payout is larger, yet the wager still costs 11.11% in expectation. This is why a craps chart should never sort bets by payout size and call the largest number “best.”

House edge is a price, not a prediction

The simplest long-run cost estimate is:

Expected loss = total resolved amount wagered × house edge

If $500 of resolved action goes through Place 6 at about 1.52%:

$500 × 0.0152 ≈ $7.60 expected loss

If the same $500 goes through Any Seven at 16.67%:

$500 × 0.1667 ≈ $83.35 expected loss

Neither result predicts what will happen in a particular session. A player can win on a high-edge wager or lose repeatedly on a low-edge one. The figures describe the average pricing built into repeated action, not the next roll.

For a more complete treatment of that distinction, use craps house edge or the expected-loss calculator.

A table-limit example: same bankroll, different exposure

Imagine a $15 table and a player deciding how to put $60 into action.

One choice is $18 each on Place 6 and Place 8, leaving $24 uncommitted. Each hit pays $21, and both place bets are exposed to a 7 while working.

Another choice is $15 on the Pass Line and, once a point is established, adding legal odds behind it. The total amount at risk may eventually be similar, but only one point is being backed and the odds portion is paid at true odds.

A third choice is to scatter $5 chips across several one-roll propositions. The visible payouts may be much larger, but the percentage price of that action is generally far higher.

The chart cannot choose the player’s preferred experience. It can show that these three uses of the same bankroll are not mathematically equivalent.

What changes from one casino to another

Published rules can alter the values in a chart. The current Massachusetts Gaming Commission craps rules, for example, formalize recognized wagers and payout procedures, but an individual casino can still offer table-specific odds limits and approved payout variations within its regulatory framework.

Before relying on a chart, confirm:

  • the maximum odds multiple;
  • whether field 2 and 12 pay double or triple;
  • whether buy/lay commission is charged up front or only on wins;
  • the posted hardway and proposition paytable;
  • whether a variant such as crapless craps changes which totals act as come-out winners or points.

A generic chart is useful only after those rule differences are separated from the universal two-dice probabilities.

How to read this at the table

You do not need to memorize every percentage before playing. A compact mental hierarchy is enough:

  1. Know the 36-combination dice table. Seven has 6 ways; 6/8 have 5; 5/9 have 4; 4/10 have 3.
  2. Separate fair odds from casino-priced bets. True odds mirror the point-versus-7 combination ratio.
  3. Check the payout rule when the bet name is not enough. Field, buy, lay, and some proposition bets can vary.
  4. Judge total action, not just percentage edge. A low percentage applied to much more money can still create a large dollar swing.
  5. Use the chart before changing bets after a loss. Previous rolls do not alter the next-roll probabilities of fair dice.

The craps odds calculator is useful when you want to convert the table into dollar payouts for a specific point and stake.

The central lesson is simple: probability describes the dice; payout describes what the casino returns; house edge describes the gap between the two. A craps odds chart becomes useful only when those three numbers stay separate.

Play smart. Gambling involves real financial risk. If the game stops being entertainment, it's time to stop playing.