Here is a question every player eventually asks: if the house has an edge, how come tonight went so well, and how long before that stops happening? The honest answer has a shape to it. The chance of being ahead after a blackjack session is a number you can actually compute, it falls slowly at first and then quickly, and the shape comes from just two numbers.
This article computes that chance for 100, 1,000, 10,000 and 100,000 hands, under assumptions stated up front: flat one unit bets on every hand, perfect basic strategy, and an ordinary 3:2 game. Everything is educational arithmetic, not a forecast of any real session, and nothing here is a reason to play or a way to bet. It is the math underneath the experience, which our pieces on session length and variance approach from other angles.
The two numbers that decide it
The first number is the house edge. With perfect basic strategy under liberal multi-deck rules it is roughly 0.5 percent, the figure our house edge guide works through. Over n hands at one unit each, the expected result is therefore minus 0.005 times n units: minus half a unit after 100 hands, minus 5 after 1,000, minus 50 after 10,000.
The second number is the standard deviation, the spread of results around that expectation. Per Wizard of Odds, as cited in our session length article, the standard deviation of one blackjack hand is about 1.15 betting units. It is above one unit because doubles and splits sometimes put more than one unit in play and a natural pays 3:2; those swings are already inside the figure. Standard deviations add up more slowly than expectations: over n hands, the standard deviation is 1.15 times the square root of n. After 100 hands that is 1.15 times 10, about 11.5 units; after 10,000 it is 1.15 times 100, about 115.
The chance of being ahead is then the probability that a normally distributed result with that mean and that spread lands above zero. The normal approximation is a simplification, roughest at 100 hands where results arrive in whole and half units, so treat the smallest row as a rounded guide rather than a precise odds quote.
The table
Putting the two numbers together for the illustrative 0.5 percent game:
| Hands | Expected result | Standard deviation | Chance of being ahead |
|---|---|---|---|
| 100 | -0.5 units | 11.5 units | 48.3 percent |
| 1,000 | -5 units | 36.4 units | 44.5 percent |
| 10,000 | -50 units | 115 units | 33.2 percent |
| 100,000 | -500 units | 363.7 units | 8.5 percent |
Read the first row twice. After 100 hands, a perfect basic strategy player finishes ahead 48.3 percent of the time, a whisker under a coin flip. That is the arithmetic behind the familiar experience of finishing a short session ahead: under these assumptions it happens almost half the time, even in a game the player cannot beat.
Why the chance falls slowly, then quickly
The pattern in the table has one cause. The expected loss grows in proportion to the number of hands, while the standard deviation grows only with the square root of the number of hands. Play 100 times as many hands and the expected loss is 100 times larger, but the spread is only 10 times larger.
Early on, the spread dwarfs the drift, so the drift barely dents your chances: between 100 and 1,000 hands the chance of being ahead slips only about 3.8 points, from 48.3 to 44.5. Keep going and the gap between the two keeps closing, because the drift is catching up. Between 10,000 and 100,000 hands the fall is 24.7 points, from 33.2 to 8.5. The crossing point where expected loss equals one standard deviation arrives around 53,000 hands in our session length piece's version of the same arithmetic; at that point the chance of being ahead is about 16 percent, and it keeps falling.
So the edge is not a wall that ends a winning run at some fixed point. It is a steady drift, faint over an evening and dominant over a year of play, and in this picture the number of hands is what decides how much of it shows. Our piece on streaks and what they prove looks at why a run of good sessions feels like more than noise.
21 TrainerFree on the App Store. Practice the strategy that keeps the edge near half a percent, with instant feedback on every hand.
Get the app
The same table in a 6:5 game
Now change one rule. A 6:5 payout on blackjacks raises the house edge by roughly 1.4 percentage points per the commonly published tables, our 3:2 versus 6:5 article works through the arithmetic, which puts the total edge near 1.9 percent (0.5 plus 1.4). Rerun the same computation:
| Hands | Expected result | Chance of being ahead |
|---|---|---|
| 100 | -1.9 units | 43.4 percent |
| 1,000 | -19 units | 30.1 percent |
| 10,000 | -190 units | 4.9 percent |
| 100,000 | -1,900 units | Effectively zero |
The standard deviation is held at 1.15 units per hand, since the payout change barely moves it; only the drift moved, and quadrupling the drift accelerates everything. After 1,000 hands the 6:5 player is ahead 30.1 percent of the time against 44.5 percent in the good game, and after 10,000 hands, 4.9 percent against 33.2. At 100,000 hands the chance rounds to nothing at all. The payout line, one small looking detail on the felt, does more to your long run prospects than any other rule on the table.
What the numbers assume, and how long the rows take
The assumptions are strict, and each one matters. Flat one unit bets: varying the stakes changes the size of the swings and the total wagered, though not the house edge on each unit, so the table does not cover it. Perfect basic strategy: commonly published estimates put a typical casual player's giveaway at one to two extra points, and every point moves every row. A 3:2 payout with liberal rules: the second table shows what the common alternative does. And a house game throughout: a skilled counter in a beatable game can at best flip the drift to a small positive one, a different subject with its own large swings.
For scale, the rows are long in wall clock time. Per the commonly published rates in our hands per hour piece, a full seven player table runs about fifty hands per hour and heads up play runs well over a hundred. So 100 hands is an hour or two; 1,000 hands is roughly 10 to 20 hours; 10,000 hands is roughly 100 to 200 hours; and 100,000 hands is in the thousand to two thousand hour range, the working hours of six months to a year of a full time job. The rows near the bottom of the table are not a weekend. They are a habit, sustained over seasons, which is exactly why the drift wins there.
What to take from it
Three lessons. Short sessions are mostly noise: winning one says almost nothing about how well you played, and losing one says almost nothing either, a point our expected value primer makes at the level of the single hand. Volume is where the edge lives: the casino's advantage becomes dependable only in aggregate, and the aggregate arrives faster than intuition expects. And rules accelerate or delay the slide: the same 10,000 hands hold a 33.2 percent chance of being ahead in a fair 3:2 game and 4.9 percent at 6:5, which is why the payout is the most important line on any rule card.
None of this is advice to play, and none of it is a way to bet. It is the shape of the game's arithmetic, worth knowing whether you ever sit at a table or only want to understand what the people at one are experiencing. For the number that measures whether a bankroll survives the swings rather than the chance of a good finish, see our piece on risk of ruin.
Frequently asked questions
What are the odds of being ahead after 100 hands of blackjack?
About 48.3 percent, in an illustrative game with a 0.5 percent house edge, flat one unit bets and perfect basic strategy. After 100 hands the expected result is only minus half a unit while the standard deviation is about 11.5 units, so the edge is buried under noise. The normal approximation is rough at this scale, since results arrive in whole and half units, so treat the figure as approximate.
How many blackjack hands does it take for the house edge to dominate?
There is no sharp threshold, but a useful marker: with a 0.5 percent edge and a 1.15 unit standard deviation per hand, the expected loss equals one standard deviation at roughly 53,000 hands. Well short of that, variance dominates the results; well beyond it, the edge does. By 100,000 hands the chance of being ahead has fallen to about 8.5 percent, and the trend continues from there.
Why can you still win sessions if the house has an edge?
Because in the short run the spread of results is far larger than the edge itself. After 100 hands the expected loss is half a unit while one standard deviation is about 11.5 units, more than twenty times larger, so finishing ahead is close to a coin flip. The edge is a small steady drift, and it only becomes visible once the number of hands is large enough for the drift to catch up with the spread.
How much does a 6:5 payout change your chance of being ahead?
A great deal, because it multiplies the edge roughly fourfold. The 6:5 payout adds about 1.4 percentage points on top of the 0.5 percent baseline, an edge near 1.9 percent. The chance of being ahead after 1,000 hands falls from 44.5 percent to 30.1 percent, and after 10,000 hands from 33.2 percent to about 5 percent, per the same normal approximation.
How long does it take to play 10,000 hands of blackjack?
At the commonly published rates, roughly 100 to 200 hours. A full seven player table runs around fifty hands per hour, putting 10,000 hands at about 200 hours, while heads up play with a fast dealer runs well over a hundred per hour, closer to 100 hours. Casinos often assume about seventy hands per hour when rating play, which lands 10,000 hands near 140 hours.