By the bjshoe team · Reviewed against Wikipedia, Stanford CS visualizations, and Blackjack Apprenticeship · Updated May 2026
Blackjack is one of the only widely-offered casino games where the deck composition changes meaningfully as cards are played. After a Jack and a Queen come out, the shoe has fewer ten-value cards left — your odds of being dealt a natural blackjack on the next round go down. After three 2s and a 5 come out, the opposite happens: the shoe is now slightly richer in high cards, and the player gains a small edge.
That edge is real but small. With perfect basic strategy alone, the house edge in a typical six-deck game is around 0.5%. A skilled card counter using the Hi-Lo system can flip that edge to roughly 0.5–1.5% in the player's favor, depending on rules, deck penetration, and bet spread.
Hi-Lo is the simplest accurate system and the one almost every professional learns first. Every card you see gets assigned one of three values:
| Cards | Hi-Lo value |
|---|---|
| 2, 3, 4, 5, 6 | +1 |
| 7, 8, 9 | 0 (neutral) |
| 10, J, Q, K, A | −1 |
The system is balanced: every full deck of 52 cards sums to zero. If you count every card in a single deck without error, you'll end at exactly 0.
From the moment the shoe is shuffled, add and subtract as each card is dealt. You count every card you see — yours, the dealer's, every other player's. The total at any moment is your running count.
Example. Dealer turns up 6 (+1), 10 (−1), 5 (+1), King (−1), 4 (+1), 7 (0), 2 (+1). Running count = +2. The remaining shoe is slightly richer in tens and aces than a fresh shoe would be.
Start with a single deck. Flip cards one at a time and call out the running count. When you finish the deck, the count must equal zero. If it doesn't, you missed a card. Time yourself.
Jim Makos recommends counting a full deck in under 20 seconds with zero errors before you ever sit at a real table. His personal account is also a warning worth taking seriously: laboratory speed doesn't translate to casino speed — the sensory overload of a real pit (noise, lights, rapid dealing, social pressure) derails counters who only practiced quietly at home. Watch from behind players for several sessions before risking money.
The running count alone isn't enough — a +6 with one deck left is far more powerful than a +6 with six decks left, because the high cards are concentrated. The fix is the true count:
You estimate decks remaining by looking at the discard tray (or the shoe itself). Round to the nearest half-deck.
Example. Running count is +9, and you see about three decks left in the shoe. True count = 9 ÷ 3 = +3. That's when you bet bigger.
A Stanford CS visualization of Hi-Lo simulations found a near-linear relationship between true count and player edge:
| True count | Approx. player advantage |
|---|---|
| 0 (basic strategy only) | −0.35% |
| +3 | ~+2% |
| +6 | ~+5% |
| +9 | ~+8.25% |
The Stanford analysis also showed lower counts occur much more frequently than high ones — most of the time you'll be sitting at neutral or slightly negative counts, betting the minimum. That's why bet spread is so important: the rare high-count rounds are where almost all the profit lives.
The whole point of counting is to bet more money when the odds favor you and less when they don't. Two common ladders:
| True count | Aggressive (1–12 spread) | Conservative (Makos) |
|---|---|---|
| ≤ +1 | 1 unit | 1 unit |
| +2 | 2 units | 2 units |
| +3 | 4 units | 3 units |
| +4 | 8 units | — |
| +5 or more | 12 units | — |
The "unit" is your minimum bet. Aggressive spreads (1–12 or 1–15) draw casino attention quickly. Wikipedia notes that with a 1–15 spread you can expect roughly +1.2% advantage with a standard deviation of about 3.5 — meaningful variance, even when you're playing perfectly.
Card counting has a small edge with very high variance. Wikipedia summarizes the standard guidance: around 100× your maximum bet as a bankroll. So if you're spreading 1–12 units and your max bet is $300, you want about $30,000 set aside for the long haul. Otherwise you'll go bust during a normal losing streak even with perfect play.
Realistic earnings expectations from Wikipedia: a counter spreading appropriately at a six-deck game with about a 1% edge, averaging $100 per hand at 50 hands per hour, can expect roughly $50/hour — long-term, smoothed over hundreds of hours.
At high true counts, a handful of basic-strategy plays become wrong. The most valuable deviations are known as the Illustrious 18. The top two account for a large share of the gain:
You don't need all 18 deviations to capture most of the edge — the top 4 or 5 do the heavy lifting.
Hi-Lo is the most common, but it's not the only one. There's a real tradeoff between accuracy and how fast you can keep up:
| System | Level | Notes |
|---|---|---|
| Hi-Lo | 1 | The standard — simple, accurate enough for almost everyone. |
| KO (Knock-Out) | 1 | Unbalanced — skips the true-count division entirely. Easier mentally. |
| Hi-Opt I | 1 | Aces neutral; aces tracked as a separate side count. |
| Hi-Opt II | 2 | Uses ±2 values for finer accuracy. Harder to do at speed. |
| Omega II | 2 | +2 for 4/5/6, −1 for 9. More precise but slower. |
| Zen Count | 2 | Comprehensive level-2 system. |
| Wong Halves | 3 | Uses ±0.5 fractional values. Maximum precision, maximum effort. |
Wikipedia summarizes the tradeoff bluntly: "Some counters earn more money playing a simple count quickly than by playing a complex count slowly." Master Hi-Lo first.
Card counting wasn't invented by one person, but it has a clear timeline of milestones:
Wikipedia is unambiguous: "Card counting is not illegal under British law, nor is it under federal, state, or local laws in the United States provided that no external card counting device or person assists the player."
That said, most casinos are private property and can refuse service to anyone. In practice:
Using a device (phone, hidden computer, earpiece signal) crosses into cheating in most jurisdictions and can be charged criminally.
Modern pits are well-equipped to spot and stop counters. The full toolbox:
Blackjack Apprenticeship — founded by Colin Jones, who ran a multi-million-dollar counting team featured in the documentary Holy Rollers: The True Story of Card Counting Christians — puts it bluntly: "Card counting isn't for everyone." Beyond the weeks of pure counting drills, a serious counter has to learn:
If you find that exciting, there are professional training resources (Blackjack Apprenticeship's mini-course is the best-known starting point). If you find it tedious, basic strategy + small recreational bets is still a near-even game and far less hassle.
In bjshoe's solo mode the shoe is reshuffled at the start of each round, so deck penetration is effectively zero — counting can't help you here, by design. The game is meant to feel like a casino table for fun, not be a card-counting trainer. If you want to practice counting drills, use a physical deck and a timer.
▶ Play bjshoe — free blackjack, no signup