The True Count, Explained

Why the running count is not enough

A running count of +6 means six extra high cards are waiting — somewhere. Spread across five remaining decks, that surplus is thin soup. Concentrated in a deck and a half, it is dinner. The same running count means different things at different depths, so the number you bet from has to account for how much shoe remains.

The divider

The conversion is one division: true count = running count ÷ decks remaining. The number of decks you started with minus the decks in the discard tray equals the decks remaining — that subtraction is the whole operation. At a real table you cannot see inside the shoe, so you read the discard tray by eye and subtract. That estimate is a skill of its own, and it is trainable: real trays, real stacks, at the precision the game actually allows.

Truncate toward zero

This system drops the fraction — rounding toward zero, for negative counts as well as positive. A computed +2.7 plays as +2; a −2.7 plays as −2. One convention, applied everywhere, so your bets and your practice grade against the same arithmetic. Conventions differ across books; what matters is picking one and being consistent, and this is ours, stated plainly.

Training the whole chain

Count → tray read → divide → truncate is a chain, and chains break at their weakest link. The free true-count trainer on this site shows you a real tray photograph, takes your read, does the subtraction with you, and then asks for the true count — grading each link separately so a tray miss doesn't hide an arithmetic miss.

Why the running count alone will mislead you

A running count of plus six means something completely different depending on how much of the shoe is left. Six extra high cards spread through five remaining decks is a faint tilt. The same six concentrated into one remaining deck is a strong one. The running count cannot tell those apart, and betting off it directly means overbetting early in a shoe and underbetting late — which is the wrong error in both directions at once.

Converting fixes this by asking a different question: not how many extra high cards are left, but how concentrated they are in what remains. That is the number that actually corresponds to your advantage, and it is the number bet sizes and playing decisions are built from.

Estimating what remains, and how accurate that has to be

The conversion needs an estimate of the decks still to be dealt, and that estimate comes from looking at the discard tray rather than from arithmetic. This alarms people more than it should. You are not trying to be exact; you are trying to land in the right neighbourhood, because the decision the number feeds into is coarse.

Bet sizes move in steps, not continuously. Playing decisions flip at thresholds. An estimate that is off by a quarter of a deck will usually produce exactly the same action as a perfect one, and on the occasions it does not, the two actions are close in value anyway. What actually costs money is not imprecision — it is being wrong by a whole deck or more, which is a different kind of error and one that practice fixes quickly.

Where the number is used, and where it is not

The converted count drives two things: how much you bet, and a small number of playing decisions that change when the deck is rich enough to justify it. It drives nothing else. It does not tell you to sit down, stand up, or change tables, and treating it as a general-purpose signal is a way of finding patterns that are not there.

Most of the value is in the bet. The playing changes are real and worth learning, but they are a refinement on top of correct bet sizing — not a substitute for it, and not where a beginner should be spending attention.

Counting down, and why the conversion is where errors hide

Most people who lose accuracy at a live table do not lose it on the tags. They lose it on the conversion, because it is the only step that asks for a judgement rather than a rule. A tag is right or wrong; a deck estimate is a reading, and readings drift when attention is elsewhere.

The drift is also silent, which makes it worse than an error you notice. A reading that is off does not announce itself; the count still feels correct, the bets still feel reasoned, and nothing in the session flags that the number feeding both has quietly moved. That is why the estimate is worth drilling separately rather than trusting to the moment.

The fix is unglamorous: practise the estimate on its own, away from the count. Look at a tray, say a number, then check it. Do that until the reading is instant and unbothered, because at a table it has to happen while a hand is being decided and a bet is being sized, and any step that still requires thought at that moment is the step that will bend.

Questions

How accurately do I need to estimate the discard tray?

Half-deck accuracy is the working standard the eye can actually hold at a table, and half-deck steps are what you train against in a six-deck shoe. A double-deck game is shallow enough that quarter steps are readable, so that mode trains against quarters. Perfect is not on the menu; consistent is.