Bets are not created equal. The way your bankroll moves during a session can look different even when two games offer a similar house edge. One game might bleed you slowly and quietly. Another can swing wildly from one extreme to the other in just a few hands. People who understand the difference between volatility and edge and know how to adjust their bets for both can develop great betting skills.
Two Concepts That Get Confused All the Time
It helps to get these two ideas clearly separated in your mind.
- Edge. This is the mathematical advantage one side holds over the other. The house has the edge in most cases. If a game has a 2% house edge, you can expect to lose around $2 for every $100 wagered over enough samples.
- Volatility. This is about how bumpy the ride is on the way to that long-run result. A high-volatility game produces big swings in both directions. A low-volatility game moves more slowly and predictably.
You can have a game with a tiny house edge but enormous volatility. You can also have a game with a moderate house edge and almost no variance at all. Both situations call for very different betting approaches.
Why Both Numbers Matter for Every Bet You Place
Most bettors focus on the house edge. They look for the games with the lowest percentage and assume this is all they need to know. But ignoring volatility leads to some costly surprises. A high-volatility game can wipe out your bankroll before the long-run math gets a chance to even out. A low-volatility game keeps you in action longer, but might feel like slow-drip torture if you
How the Major Games Compare
Here’s a snapshot of how the most popular casino games sit on the edge-volatility spectrum:
| Blackjack (basic strategy) | 0.5% | Low-Medium | ±$50-$80 |
| Baccarat (banker bet) | 1.06% | Low | ±$40-$60 |
| European Roulette (even money) | 2.7% | Low-Medium | ±$60-$100 |
| European Roulette (single number) | 2.7% | Very High | ±$200-$500+ |
| Craps (pass line) | 1.41% | Low-Medium | ±$50-$90 |
| Video Poker (Jacks or Better, 9/6) | 0.46% | Medium-High | ±$80-$150 |
| Slots (low volatility) | 4-8% | Low | ±$30-$70 |
| Slots (high volatility) | 4-8% | Very High | ±$100-$800+ |
| Sports betting (sharp line) | 4-5% | High | ±$100-$300+ |
Roulette carries the same house edge whether you bet on a single number or on red/black. But the volatility is different. This difference alone should change how you size your bets.
The Standard Deviation: Your Best Tool for Measuring Volatility
Standard deviation measures how spread out the results are likely to be from the expected outcome. A higher standard deviation means wider swings.
| Baccarat (banker) | 0.93 | Very tight, predictable swings |
| Blackjack (basic strategy) | 1.15 | Moderate swings with strategy decisions |
| Craps (pass line) | 1.00 | Close to coin-flip feel |
| Roulette (even money) | 1.00 | Similar to craps short-term |
| Roulette (single number) | 5.84 | Extreme swings, rare big wins |
| Video Poker (Jacks or Better) | 2.70 | Significant swings from royal flush chasing |
| High-volatility slots | 8.00-15.00+ | Bankroll can vanish or triple very quickly |
You can use standard deviation to estimate what a session might look like. Over 100 bets, multiply the standard deviation per bet by roughly 10 (the square root of 100) to get a rough range of expected outcomes. For baccarat, this is about ±9.3 units. For a high-volatility slot, this same 100 spins could swing your balance by ±80-150 units.
Why Low Edge Doesn’t Always Mean Safe
Video poker is a perfect example. The 9/6 Jacks or Better variant has one of the lowest house edges of any casino game, just 0.46% with perfect strategy. But video poker has a standard deviation of around 2.7 per hand. This one rare outcome is responsible for a massive portion of the game’s total return. In a typical session, you probably won’t hit it. Thus, your actual experienced return rate is worse than 0.46%.
Research from gaming mathematics journals has shown that to have a reasonable probability of realizing the full theoretical return of Jacks or Better, a player needs to log upwards of 40,000 to 80,000 hands. In a single session of 500 hands, the variance from the royal flush alone creates swings wide enough to make the low house edge almost meaningless.
| 500 | ~2.5% | ~2.2% |
| 2,000 | ~9.5% | ~1.8% |
| 10,000 | ~40% | ~1.1% |
| 40,000 | ~86% | ~0.6% |
| 80,000 | ~98% | ~0.46% |
The lesson here is that low edge only protects you if you play enough volume for the math to average out. High volatility makes that volume requirement enormous.
The Right Way to Size Bets Based on Volatility
The higher the volatility, the smaller your individual bets should be relative to your bankroll. This keeps you alive long enough for the edge (or lack thereof) to actually express itself. A commonly used framework looks like this:
| Very Low (Baccarat) | 2-4% | $10-$20 per hand |
| Low-Medium (Blackjack, Craps) | 1.5-3% | $7.50-$15 per hand |
| Medium-High (Video Poker) | 1-2% | $5-$10 per hand |
| High (Sports Betting, Even-Money Roulette) | 1-2% | $5-$10 per bet |
| Very High (Single-Number Roulette, Hi-Vol Slots) | 0.25-0.75% | $1.25-$3.75 per bet |
These are starting points. You need more bets in your session to give variance time to even out when volatility is high. Smaller bets mean more bets. More bets mean a better chance of experiencing something close to the expected mathematical outcome.
How Edge Should Adjust Your Approach Too
Volatility sizing gets you the right bet scale. Edge sizing tells you how ambitious to be with that scale. You can afford to size up when you have identified a situation where you genuinely have an edge. You want to size down when the house edge is steep because you are paying more per bet for the entertainment.
| Under 0.5% | Near-zero cost to play | Slightly larger bets viable |
| 0.5% – 1.5% | Low cost, good value | Standard sizing |
| 1.5% – 3% | Moderate cost | Stay at or below standard |
| 3% – 5% | Meaningful cost | Size down by 25-40% |
| Over 5% | High cost | Minimum bets only, if playing |
Most slot machines, which carry house edges between 5% and 15%, should be played at the absolute minimum bet you are comfortable with if bankroll preservation matters to you. The entertainment value might be worth it, but you must go in knowing the math.
The Kelly Criterion: Edge and Volatility in One Formula
The Kelly Criterion is the gold standard for combining both edge and volatility into a single bet-sizing recommendation for bettors who have a genuine edge.
The formula is:
Kelly % = (Edge ÷ Odds)
Where edge is your estimated win probability minus the bookmaker’s implied probability. If you estimate a team has a 58% chance of winning, but the odds imply only 52%, your edge is 6%. At even-money odds, Kelly suggests betting 6% of your bankroll on this bet. But Kelly assumes perfect knowledge of your edge. This is why most serious bettors use a fractional Kelly approach, typically betting between a quarter and a half of what full Kelly recommends.
| Full Kelly (1x) | 6% of bankroll | Very High | Edge estimate is highly certain |
| Half Kelly (0.5x) | 3% of bankroll | High | Reasonable confidence in edge |
| Quarter Kelly (0.25x) | 1.5% of bankroll | Medium | Edge is estimated, not precise |
| Flat (ignore Kelly) | Fixed % regardless | Varies | No reliable edge identified |
The fractional approach sacrifices some theoretical maximum growth in exchange for dramatically reduced variance. This trade-off is worth making for most bettors.
Adjusting Mid-Session
Real sessions don’t always go to plan, even with a solid pre-session plan. Know when and how to adjust bets during a session. Adjust for bankroll changes. You can recalibrate your bet size upward if your session bankroll has grown because your absolute risk stays proportional. Your bet size should come down for the same reason if your session bankroll has shrunk. Never increase your bets because you are frustrated, chasing losses, or convinced a win is overdue.
Volatility makes this discipline harder to maintain. In a high-volatility game, losing streaks can feel catastrophic even when they are within the expected range. A single-number roulette player can go 50 spins without hitting their number and still be within the statistical norm.
Comparing Two Bettors
To see how this plays out in practice, consider two bettors who both sit down with $500 to play European roulette for two hours.
| Bet type | Single numbers ($35 payoff) | Even money (red/black) |
| Bet size | $10 per spin | $10 per spin |
| House edge | 2.7% | 2.7% |
| Volatility | Very High (std dev ~5.84) | Low-Medium (std dev ~1.00) |
| Expected loss (200 spins) | ~$54 | ~$54 |
| Realistic worst-case swing | -$350 to -$500 | -$120 to -$180 |
| Realistic best-case swing | +$250 to +$500+ | +$60 to +$120 |
| Probability of ruin ($500 bank) | ~28% | ~4% |
Both bettors face the same house edge. They bet the same dollar amount per spin. But Bettor A has a roughly 28% chance of losing their entire $500 in the session. Bettor B faces only a 4% chance of ruin. The only difference is where on the table they are placing their chips.
If Bettor A had adjusted their single-number bet down to $2 per spin to account for the higher volatility, their ruin probability would drop to under 5% and keep the big-win potential alive.
Sports Betting
Sports betting operates differently from casino games because outcomes are not purely random. The house edge comes from the vig built into the odds rather than from mathematical probability. It is possible to have a genuine edge.
But sports betting also introduces a different volatility problem, which is sample size. Winning 55% of bets at standard -110 odds is profitable long-term. But the standard deviation is large enough that even a skilled bettor with a real edge can go 40 wins and 60 losses by chance alone over a 100-bet sample.
| 50% | Losing (-4.5% vig) | 71% | 84% |
| 52% | Near breakeven | 55% | 62% |
| 55% | Profitable | 32% | 18% |
| 57% | Clearly profitable | 19% | 7% |
| 60% | Highly profitable | 9% | 1% |
Even a 57% win rate carries a 19% chance of showing a loss over 100 bets. This is why bet sizing in sports needs to stay conservative enough that a bad 100-bet run doesn’t threaten your overall bankroll. Quarter Kelly or flat betting at 1-2% of bankroll are the most commonly recommended approaches for recreational sports bettors.
Building Your Personal Bet-Sizing Framework
The most useful thing you can do is build a simple personal framework before each session. It only needs to answer four questions:
- What game am I playing, and what is the house edge? This sets your baseline cost per bet. Higher edge means smaller bets.
- What is the volatility of this game and bet type? This determines your minimum number of bets needed and how small each bet should be relative to your bankroll.
- What is my session bankroll, and what is my hard stop-loss? Decide this before you sit down. Your bet size should ensure you can absorb at least 50-100 bets before hitting your stop-loss.
- Do I have any edge in this situation? If yes, fractional Kelly gives you a mathematically sound way to size up. If no, aim for the lower end of the volatility-appropriate range.
| Blackjack, basic strategy | No | Low-Med | 1.5-2.5% of session bank |
| Blackjack, card counting | Yes | Low-Med | Quarter to half Kelly |
| Baccarat, banker bet | No | Low | 2-4% of session bank |
| High-vol slots | No | Very High | 0.25-0.5% of session bank |
| Sports, identified edge | Yes | High | Quarter Kelly (1-2% typical) |
| Sports, no identified edge | No | High | 1% flat or don’t bet |
| Video poker, 9/6 JoB | Marginal | Med-High | 1-1.5% of session bank |
Conclusion
Volatility and edge are two separate forces pulling on your bankroll at all times. Ignoring either one can lead to poor decisions. Edge tells you the long-run cost of playing a game. Volatility tells you how much punishment you need to be able to absorb before that long run arrives.
Size your bets to survive the variance of the game you are playing. Then, adjust within that range based on the quality of the edge you are facing. This means betting much smaller than feels natural in games with very high volatility. In games with a genuine positive edge, it means being disciplined enough to size up, but not so aggressively that a bad run can end your session early.
The bettors who manage both of these variables well can stay in the game longer and make better decisions under pressure. They can also give the math enough room to work in their favor.