How To Win At Craps

The tables don't care about your feelings

Craps is a miserable game for people who refuse to accept basic math. I've spent more hours than I want to admit sitting at those green felt tables, watching people bet $25 on the field because they heard it was "easy," while the house edge on that particular wager sits at roughly 5.5%. Meanwhile the pass line with full odds is sitting right there at less than 1% if you're taking advantage of 5X or 10X odds depending on the casino. It makes my teeth ache. The fundamental problem is that craps forces you to make decisions fast and most people just parrot what the guy next to them is doing. Nobody explains why. You learn nothing. You lose money consistently and then blame the dice.

How To Win At Craps Is Actually Simple and Also Kind of Annoying

Here's the thing nobody tells beginners: winning at craps isn't about predicting rolls. It's about minimizing the juice you give the house on every dollar you risk. The pass line bet carries a 1.41% house edge on its own. That's already one of the best bets on the floor. But then you add odds behind the pass line, and those odds bets carry literally zero house edge. The casino pays true odds on them. A $10 pass line bet with $50 in odds behind it drops your overall house edge to something closer to 0.4% or lower depending on the table's odds multiplier. That's not a typo. That is legitimately one of the best numbers you will find in any casino. The don't pass bar is the same deal in reverse. You're betting the shooter will seven out before making the point. The house edge is 1.36%. Slightly better than pass line. The catch is that you lose on a 2 or 3 and push on 12, which makes it feel worse emotionally even though the math is fine. People hate losing on a 12 because they think it should be a push regardless. It's not. That's just how the game is coded. Now here's where most people mess up. They see the low house edge on pass line with odds and think they've found the holy grail and start betting absurd amounts on all the other bets around the table. "I'm covering everything!" they say. The moment you add the place bets, the field bets, the hardways, and the proposition bets, you are setting your own edge back to something closer to 4% or worse depending on what you stack on. It doesn't matter that your pass line bet is playing beautifully. The side bets are dragging your expected value down into the dirt.

I ran into this problem with a regular who'd come in every Saturday. He'd bet $25 pass line with max odds, which was great, but then he'd also throw $50 on the field, $25 on any craps, and $10 on the six and eight. His overall house edge on his total action was sitting somewhere around 2.8% instead of the sub-1% he was getting on his main bet. I told him three times over three different Saturdays. He came in the fourth Saturday and told me he'd cut the side bets and finally understood why. Took him four visits. Some people need it to hurt before they listen.

The bets that actually matter

Pass line and don't pass with full odds. Place the six and eight if you want additional action, since those pay 7 to 6 and carry a 1.52% house edge, which is tolerable if you're going to bet them. Avoid everything else unless you have a specific reason. The horn bet, the Any Seven, the Hardways, the Big 6 and Big 8 - these are all terrible bets dressed up as exciting ones. The house edge on a horn bet ranges from about 11% to 17% depending on the specific number. The Any Seven is 16.67%. I once watched a woman lose $400 in twenty minutes on horn bets alone while her pass line bet sat untouched in the corner. There's also the question of how much odds you take. Most tables offer 3X 4X 5X. Some go 10X or 15X. The higher the multiple, the better your overall expected value. If you have a choice between two identical tables and one offers 5X odds while the other offers 10X, take the 10X. The difference in house edge on your total action between those two setups is measurable and real. Over a session of maybe 100 to 200 decisions, it can amount to dozens of dollars in expected loss saved.

What doesn't work and why people keep believing it

Roll control is the biggest myth in craps. Players will tell you they can set the dice and influence the outcome by gripping them a certain way and releasing them on a specific axis. The physics don't support this at all. Dice are designed to tumble unpredictably. Even if you could somehow control the initial rotation, the cushions, the rail bounces, and the sheer randomness of the landing surface make any attempt at directional control statistically meaningless. I tested this myself once by rolling dice with a specific grip for about 200 throws and tracking the results. My win rate was exactly what random chance predicted. I stopped bothering with it after that. Progression systems like the Martingale, the Labouchere, the Fibonacci - none of these change the house edge. They just rearrange how you lose money. A Martingale where you double after every loss will produce a long string of small wins followed by one catastrophic loss that wipes out everything. The math is brutal and simple. You're still playing negative expectation bets. No betting pattern changes that. Some players swear by betting the come bet after the point is established. This is actually a reasonable strategy and mathematically equivalent to the pass line. The house edge on a come bet is identical at 1.41% before odds, and adding odds brings it down the same way. It just feels different because you're placing a new bet after the round has started, which gives you the illusion of having more information than you actually do.

The part nobody talks about: bankroll management and knowing when to leave

You can do everything right and still lose because variance is a real force. A shooter can go on an eleven-roll hot streak and take your entire buy-in before the law of large numbers has a chance to smooth things out. The only thing you control is your bankroll size relative to your bet size and how long you plan to stay at the table. A good rule of thumb is to bring at least 50 to 100 times your base bet. If you're betting $10 pass line, you should have $500 to $1,000 in front of you. That gives you enough room to survive a cold streak without tilting into bad decisions. I had a guy once who came in with $200, was betting $25 pass line with odds, and lost his third roll. He stood up and said the table was "cursed." It wasn't cursed. He had about eight standard deviations of variance in his favor and then eight in its disadvantage, and $200 wasn't enough to absorb the swing. He left before he could even get close to the expected value of his bets playing out over a meaningful sample size. The hardest part of this game is walking away while you're up. The casino wants you to sit down and give it back. Every bet you place after you've reached your win target is a bet with negative expectation that slowly drains your profits back. Set a stop-win before you sit down. $100? $200? $500? Decide it in advance and stick to it. The people who walk away winners aren't lucky. They're disciplined. There's a difference and it shows up in their bank accounts.

When the strategy completely falls apart

Even the best craps strategy has hard limits. If you're playing at a table that only offers 1X odds, your edge reduction is minimal and the game becomes much less attractive. The pass line with 1X odds drops your overall house edge to roughly 0.8%, which is decent but nowhere near the 0.2% or better you can get at a 10X or 15X table. If your local casino only has low odds tables, the math shifts significantly and you should consider whether you're better off playing blackjack or video poker instead, where the house edges can be similarly low with proper strategy. Another scenario where craps becomes unplayable is when the table is overcrowded. I've been at tables where six people are all shouting different bets, the dealer is calling out numbers faster than you can process them, and you accidentally place a $25 hardway bet because someone knocked your chip into the wrong zone. That happened to me at a busy holiday weekend. I lost $150 in an hour partly because the chaos forced me into mistakes I wouldn't have made on a quiet Tuesday night. Don't play crowded tables if you're trying to execute a real strategy. The environment itself becomes a negative factor. The ultimate limitation is that craps, like every casino game, is designed to take your money over time. No amount of discipline, bankroll management, or smart bet selection changes the fundamental fact that the house has the edge. What these strategies do is shrink that edge from 5% or more down to below 1% on your main action. That's the best you can do. Anything promised beyond that is a scam.