Is TrueCount Blackjack fair? We audited it.

Game engine audit · TrueCountEngine v1.0 · Audit date: 18 March 2026 · Updated: 7 October 2026 · Result: all 76 tests pass

TrueCount's game engine passed a 76-test audit across 8 categories, every test passing. The tests check that the shuffle is random, the payouts are exact and the odds match a real casino. Here is every check, and what it does not prove.

First, the honest part

This is a self-audit, not a third-party certification. We built a test suite modelled on the kinds of checks gaming labs run (GLI-11 / eCOGRA-style), but we ran it ourselves, and we're not accredited by GLI, BMM, or any gaming laboratory. We don't need to be: TrueCount is an entertainment app with no real money and no gambling. This audit exists for one reason: transparency. Every test uses a fixed random seed, so the audit is fully deterministic: every run reproduces identical results.

What we checked, in plain English

The 76 tests fall into 8 categories. Here's what each one is really asking:

What it proves	Tests
The shuffle is genuinely random No patterns, no memory, no bias, across 100,000+ dealt cards per test.	10	PASS
Every shuffle is a complete, correct deck No missing cards, no duplicates, the cut card lands where it should.	6	PASS
Every rule plays by the book Dealer hits/stands correctly (S17 & H17), doubles, splits, insurance and surrender (an engine rule; not offered in the app) all follow real casino rules.	20	PASS
You're paid exactly right Blackjack 3:2 and 6:5, wins, doubles and insurance pay to the exact chip, even on odd bet sizes.	12	PASS
The odds match a real casino Hundreds of thousands of simulated hands per table, checked against Wizard of Odds reference values.	12	PASS
The strategy advice is correct 280 basic-strategy cells verified against published Wizard of Odds charts.	6	PASS
The card-counting math is right Hi-Lo running count, true count and player edge all compute correctly.	7	PASS
No chips ever appear or vanish Across 10,000+ full rounds per table, every chip wagered and returned balances exactly.	6	PASS

The numbers that matter to you

Real, published house edges

Every table was simulated over 500,000 hands using perfect basic strategy, and the house edge converged to its expected value, verified against the Wizard of Odds calculator. Lower is better for you; these are honest casino numbers, not slot-machine odds:

Table	Decks	Reference value
Jean-Luc LaChance	2	0.31%
Czarita Fortuna	4	0.39%
"Dice Man" Jack	6	0.57%
Lady Lux	6	0.72%
Aunty "Lucky Lei"	8	2.02%

Reference values from the Wizard of Odds house edge calculator. Our simulation of 500,000 hands per table, with basic strategy and the dealer standing on soft 17, matched each one within 0.40 points. The house edges we publish for each table (0.30% to 2.20%) come from the table rules.

Everything else lines up too

• Blackjack shows up about 4.7 to 4.8% of hands, exactly as often as it should, on every table.
• Dealer bust rates by upcard match published values within 3% (e.g. a dealer 6 busts ~42% of the time; an Ace ~11.5%).
• Payouts are exact to the chip: 3:2 and 6:5 blackjacks, doubles, and insurance, including correct rounding on odd bet sizes.
• Player win / loss / push rates (≈42.5% / 49.0% / 8.5%) land right where the math predicts.

How big was the test?

Big enough that a hidden bug would have nowhere to hide. The fairness simulations ran 500,000 hands per table for house edge, 200,000 hands for dealer bust rates, and 100,000+ dealt cards per shuffle test, millions of hands in total, checked against a 99.9% confidence threshold (chi-square, serial-correlation and runs tests). And because every test uses a fixed seed, the audit is fully repeatable: it isn't one lucky run, the same checks produce the same results every time.

The full caveats

• This is a self-audit, not a third-party certification. We are not accredited by GLI, BMM, or any gaming laboratory.
• TrueCount is an entertainment app: no real money, no gambling, no cash-out. This audit exists for transparency and quality assurance.
• Reference house-edge values are approximate (from published calculators), and tolerance bands account for finite-penetration effects.
• Statistical tests are probabilistic: a passing result means no statistically significant deviation was detected, not that deviation is impossible.

Questions?

Email admin@monarchgames.io, or head back to TrueCount support. See also our Privacy Policy.