The "Hot" Shooter Fallacy: Why Past Rolls Don't Predict the Future
One of the most persistent beliefs at craps tables is that a shooter can be "hot" or "cold"—that a sequence of past rolls somehow increases or decreases the chances of specific outcomes on subsequent rolls. This is a classic example of the gambler's fallacy and misunderstanding of independent trials. Each roll of two fair dice is an independent event with fixed probabilities: 2 and 12 each have 1/36 chance, 3 and 11 each 2/36, 4 and 10 each 3/36, 5 and 9 each 4/36, 6 and 8 each 5/36, and 7 has 6/36. Those probabilities do not change because of what happened on previous rolls. So a run of sevens or a streak of passes doesn't make the next roll more or less likely to be any particular number.
Another related misunderstanding is seeing patterns or streaks and assigning meaning. Humans are pattern-seeking: we notice runs and assign causality. But statistical randomness produces streaks frequently enough that they are expected. A "hot" shooter who seems to roll many winning hands is just experiencing variance; over a very large number of independent trials, outcomes will converge on expected frequencies, but in any short session, deviations are normal.
This doesn't mean streaks are useless for entertainment—players often enjoy momentum and camaraderie—but it does mean you should not change your betting strategy based on the false belief that the dice have memory. Sound strategy treats each decision in light of fixed probabilities and your bankroll, not in response to imagined trends.
The Martingale Miracle: Why Doubling Down Isn't a Winning Strategy
The Martingale system—doubling your bet after each loss until you win—appeals because it seems to guarantee eventual recovery of losses plus a small profit. In practice, two facts destroy that illusion: limited bankroll and table limits. Suppose you start with a $10 bet and double after each loss. After N losses in a row you must bet 10 * 2^N. A sequence of 7 or 8 consecutive losses is not uncommon in many casino games; after just 7 losses the required bet becomes $1,280, and after 10 losses it’s $10,240. Most players can't sustain that, and most tables cap bets far below the astronomical levels needed to keep doubling indefinitely.
Mathematically, the Martingale does not change expected value. Each bet still has the same negative expectation (house edge). Doubling shifts the distribution of outcomes—fewer small wins, rare but catastrophic losses—while leaving the long-run expectation unchanged. It magnifies variance and increases the probability of ruin. For example, even if the chance of losing a sequence long enough to bust seems small, the payout structure and bankroll limits mean that over time those rare busts dominate and make the system unprofitable.
A practical alternative is disciplined bankroll management: decide on session loss limits, use bets with lower house edge (pass line with odds), and avoid schemes that require exponential wager increases. Fixed-percentage betting and flat betting keep variance manageable and avoid catastrophic sequences of losses.
House Edge Misconceptions: What Bets Really Offer Value
Craps offers a wide range of bets with dramatically different house edges, and many myths arise from confusing payout sizes with true value. The pass line (and don’t pass) bets are popular for good reason: with proper use of full odds behind them, they become the best overall expected-value play on the table. The pass line alone carries a house edge of about 1.41%; backing it with a true odds bet (which pays true odds and carries no house edge) reduces your overall expected house edge on the combined wager. For example, a single-odds behind a pass line keeps the average house edge lower than many other simple bets; taking the maximum odds allowed by the casino is mathematically optimal if you can afford the variance.
Contrast that with so-called "sucker" bets like Any 7 or hardways. Any 7 pays 4:1 but the true odds are 5:1 (6/36 chance), producing a house edge around 16.67%—a very poor value. Hardways and many one-roll propositions (like the field or specific one-roll numbers) often have significantly higher house edges despite offering tempting payouts. The Big 6 and Big 8 are another common trap: they pay even money but the true odds are better if you place 6 or 8 as a place bet; Big 6/8 carry about 9% house edge versus roughly 1.52% if you place the 6/8.
Understanding true odds is essential. True odds are based on the precise probability of an event occurring; casino payouts often subtract a margin from true odds. When you see a large payout, compare it to the mathematical probability. If the payout is worse than the true odds, the bet favors the house. Practical advice: focus on bets with low house edge (pass/don’t pass with odds, come/don’t come with odds, place bets on 6/8) and avoid proposition bets and any 7 unless you’re playing for entertainment and expect higher variance.
Dice Control and Precision Throwing: Myth vs. Evidence
The idea that skilled shooters can "control" the dice with precise grip and release to reduce sevens and increase favorable numbers is alluring, and it has motivated books, videos, and seminars. In reality, the evidence supporting dice control is tenuous. Casinos design the game environment to thwart consistent manipulation: table surfaces, back walls, and dealer procedures (such as forcing a "fair" toss or requiring that dice hit the back wall) all increase randomness. Laboratory studies and statistical analyses of large samples have not produced robust, reproducible evidence that a player can alter long-term probabilities enough to overcome the house edge.
Even if a shooter could modestly bias outcomes in a short sample, several practical obstacles limit usefulness: the need for sustained consistency across thousands of rolls, the reality that casinos may restrict or ban suspicious players, and the fact that small biases require huge sample sizes to convert into a meaningful edge. For advantage play based on dice control, the theoretical edge would have to be significant enough to offset variance and operational risk—something not demonstrated reliably in real casino conditions.
A pragmatic approach recognizes that while you should try to make a legal and comfortable throw for personal enjoyment, you should not expect to beat the game by mechanical skill alone. Instead, rely on legitimate advantages like exploiting mispaid odds (where permissible), using correct bet selection, and managing bankroll. Treat dice control as folklore rather than a dependable strategy, and allocate time and money accordingly.
