# How the Player's Edge Moves With the True Count

A simulation of about 414 million flat bet blackjack rounds shows the average result at every Hi-Lo true count, and how the player's edge climbs about half a percent per count.

Source: https://21trainer.app/articles/player-edge-by-true-count/
Updated: October 9, 2026
Publisher: 21 Trainer, makers of Blackjack Strategy: 21 Trainer, an educational blackjack trainer for iPhone and Android (App Store: https://apps.apple.com/app/blackjack-strategy-21-trainer/id6759527695, Google Play: https://play.google.com/store/apps/details?id=app.blackjacktrainer)

> **Educational content, not gambling advice.** 21 Trainer and this article teach blackjack strategy and card counting as skills, using virtual chips only. No strategy or counting system guarantees winnings, and nothing here encourages real-money play. If gambling is a problem for you or someone you know, call 1-800-522-4700 (National Council on Problem Gambling).

Ask an experienced counter what a true count is worth and you will usually get a rule of thumb: about half a percent of advantage per count. It is quoted widely, often without a source, and worth testing row by row.

So we ran the simulation ourselves. About 414 million blackjack rounds, one player against the dealer, a flat bet of one unit on every round, basic strategy for the rules with no count-based plays, and a Hi-Lo [true count](https://21trainer.app/articles/running-count-vs-true-count/) recorded at the start of each round. Then we grouped the rounds by that count and averaged the results.

The table below is the whole answer: how steeply the average result climbs, where it crosses zero, and how little any of it moves a single round.

## The simulation

The setup: six decks, dealer stands on soft 17 and peeks, blackjack pays 3:2, double on any two cards, double after split, split to four hands, one card to split aces with no resplitting, no surrender. The shoe is dealt to a cut card at 75 percent penetration. Before each round the true count is taken as the [Hi-Lo](https://21trainer.app/articles/hi-lo-card-counting/) running count divided by the exact decks remaining, then floored to a whole number, the convention our piece on [rounding conventions](https://21trainer.app/articles/true-count-rounding-conventions/) covers in detail. Flooring means the row labeled 0 holds every raw count from 0 up to just under 1, +1 holds 1 up to just under 2, and -1 holds -1 up to just under 0.

Three checks keep the run honest. The overall result across all 413,958,591 rounds was -0.43 percent, with a standard error of 0.006, in line with the [house edge](https://21trainer.app/articles/blackjack-house-edge-explained/) of roughly half a percent commonly published for liberal multi-deck rules. The standard deviation of a single round's result was 1.154 units, matching the 1.15 betting units per Wizard of Odds that our session length article quotes. And 4.74 percent of rounds dealt the player a natural, against the 4.75 percent in the commonly published probability tables that our piece on [how often you are dealt a blackjack](https://21trainer.app/articles/probability-of-being-dealt-blackjack/) uses.

Here is what came out, with standard errors of 0.01 to 0.05 percentage points on every row:

| True count at start of round | Share of rounds | Average result, percent of initial bet | Rounds with a player natural |
| --- | --- | --- | --- |
| -6 or lower | 2.68% | -4.23 | 3.29% |
| -5 | 2.12% | -2.81 | 3.72% |
| -4 | 3.90% | -2.20 | 3.94% |
| -3 | 6.74% | -1.65 | 4.17% |
| -2 | 12.42% | -1.12 | 4.40% |
| -1 | 18.95% | -0.63 | 4.62% |
| 0 | 26.62% | -0.18 | 4.83% |
| +1 | 11.78% | +0.34 | 5.08% |
| +2 | 6.47% | +0.86 | 5.33% |
| +3 | 3.67% | +1.32 | 5.59% |
| +4 | 2.09% | +1.77 | 5.83% |
| +5 | 1.18% | +2.14 | 6.10% |
| +6 or higher | 1.37% | +2.96 | 6.70% |
| All rounds |  | -0.43 | 4.74% |

## What the numbers show

The climb is real and close to linear. From -3 to +3, the average result rises from -1.65 to +1.32 percent, which is 2.97 points over six steps, or 2.97 divided by 6, about 0.5 per true count. The rule of thumb holds in this game. The steps are not perfectly even, though: the move from 0 to +1 is 0.52 points, while the move from +4 to +5 is only 0.37. The two end rows are not single steps at all: -6 or lower and +6 or higher each pool every count beyond them, which is why they sit so far from their neighbors.

The crossing point falls between the 0 row and the +1 row. A flat bettor's expectation turns positive in the +1 row, at +0.34 percent. Rounds starting at +1 or higher are 26.56 percent of all rounds and average +0.93 percent; rounds at -1 or lower are 46.82 percent and average -1.34 percent; the 0 row is 26.62 percent at -0.18. Weight those together, 0.2656 times 0.93 plus 0.4682 times -1.34 plus 0.2662 times -0.18, and the total is -0.43 percent, which is the whole lesson of the table. A player who bets the same amount every round and never changes a play gets that average over the long run. Knowing the count changes nothing unless a decision changes because of it.

One caveat about the shape of the shares: the negative rows look far bigger than the positive ones, and most of that is the flooring convention rather than the cards. The -1 row swallows every raw count from -1 up to just under 0, while the 0 row swallows everything from 0 up to just under 1, so the small negative fractions count as a negative row while the small positive fractions sit inside the 0 row. Most of the asymmetry is a property of the yardstick, not of the cards.

## Why the expectation rises

The one reason this table measures is naturals. The natural column climbs steadily with the count, from 4.40 percent of rounds at -2 to 4.83 at 0, 5.33 at +2 and 5.83 at +4. More tens and aces remaining means both sides make more blackjacks, and the payoff is asymmetric: the player's natural pays 3:2, while the dealer's natural only wins the original bet. That asymmetry is one part of the rise. It is also why a 6:5 payout, which removes most of the premium on the player's naturals, takes the most away at the high counts, where naturals are most frequent.

Other reasons are commonly cited alongside it: doubling hands get better when the next card is more likely to be a ten, and the dealer busts more often from a stiff hand when the remaining shoe is rich in high cards. Both are plausible and both appear throughout the counting literature. Our table does not measure either one, so we report them as commonly cited and leave the numbers out.

## What a counter does with this information

Mechanism only, because this is an educational article and not betting advice. Counters use the true count through two channels. The first is the bet: our piece on [bet spreads](https://21trainer.app/articles/bet-spread-explained/) explains that a spread is a ratio of largest bet to smallest, and that most of a shoe is spent at neutral or unfavorable counts, so a perfect count with unchanged bets returns essentially what basic strategy alone returns. The second is the play: our piece on [index plays](https://21trainer.app/articles/index-deviations-explained/) explains that an index is a true count threshold at which a specific basic strategy decision flips, with insurance at +3 the single most valuable one on Schlesinger's published list.

Both channels are described here as how the machinery works, not as encouragement to bet real money, and neither turns counting into income. It produces a small shift in long-run averages attached to very large short-run swings.

## Keep the scale honest

Look at the +2 row, a clearly positive count, and read the average: +0.86 percent of the initial bet. Then read the swing of a single round, about 1.15 units per Wizard of Odds, which our article on [session length](https://21trainer.app/articles/session-length-and-results/) unpacks. One round's noise is more than a hundred times the edge the count just handed you. Nothing in this table is a prediction for a session, an evening, or a month of evenings; it is an average over hundreds of millions of rounds.

The figures are also specific to this setup: six decks, these rules, 75 percent penetration and flat betting, with [penetration](https://21trainer.app/articles/deck-penetration-explained/) deciding how many of the high-count rounds a table ever deals at all. Other rule sets shift these figures.

And the standing cautions apply. Counting with your head is legal in the United States, but casinos are private businesses that may refuse service to suspected counters, and no technique removes risk from the game. In 21 Trainer, the Pro upgrade teaches counting as a skill with virtual chips and nothing at stake, which is the right scale on which to learn it, and our piece on [when counting is not worth the effort](https://21trainer.app/articles/when-counting-is-not-worth-it/) is the honest companion read.

## Frequently asked questions

### At what true count does the player have the advantage?

In our simulation of a six deck game with liberal rules, the average result for a flat bettor turned positive in the +1 row, at +0.34 percent of the initial bet. The 0 row and every row below it averaged negative results. The exact crossing point depends on the rules and the number of decks, so treat +1 as the answer for this rule set rather than a universal constant.

### How much does the player's edge change per true count?

The common rule of thumb is about half a percent per true count, and our simulation supports it for this game. From -3 to +3 the average result rose from -1.65 to +1.32 percent, which is 2.97 points over six steps, or 0.495 per count. The individual steps are not perfectly even: the move from 0 to +1 was 0.52 points while the move from +4 to +5 was only 0.37.

### Why does a high true count favor the player?

Partly because of naturals, the one cause our table measures. When more tens and aces remain in the shoe, both the player and the dealer get more blackjacks, but the player's natural pays 3:2 while the dealer's only wins the original bet. In our simulation naturals rose from 4.40 percent of rounds at a true count of -2 to 5.83 percent at +4. Better doubling hands and more dealer busts are also commonly cited reasons.

### Does knowing the count change your results if you bet the same every hand?

No. A player who bets one flat unit and never changes a playing decision gets the average of all rounds, which in our simulation was -0.43 percent. The count is information, and information only matters when something changes because of it: a different bet size or a different play. Without one of those channels the counting effort changes nothing about the outcome.

### Does a positive true count mean you will win?

Not even close. Even at a true count of +2 our simulated average was under one percent of the bet, while a single round swings by about 1.15 betting units per Wizard of Odds. Nothing in the table predicts a single session. Counting with your head is legal in the United States, but casinos are private businesses that may refuse service, and no technique removes risk from the game.

## Keep reading

- [How Card Counting Numbers Are Tested](https://21trainer.app/articles/how-counting-numbers-are-tested/)
- [Bet Spreads: How Counters Size Bets](https://21trainer.app/articles/bet-spread-explained/)
- [Running Count vs True Count](https://21trainer.app/articles/running-count-vs-true-count/)