♠︎

Counting Teacher

Advanced counting strategy guide

Play trainerv2026.06.29.003

Beyond basic strategy

Count, convert, bet, deviate.

Basic strategy is only the floor. Advanced counting adds four layers: choose a count system, convert to a true count, size bets when the deck is favorable, and deviate from basic strategy only when the math justifies it.

This guide is built for the trainer: blue for normal play, purple for high-leverage decisions and bet changes, green for positive edge/wins, red for negative edge/losses.

Choose the counting system by your goal

“Most advanced” does not always mean “best.” A harder system only helps if you can execute it perfectly while playing. Use Hi-Lo first, then upgrade only if your error rate is near zero.

System	Tags	Difficulty	Best use	Tradeoff
Hi-Lo	2–6 = +1 · 7–9 = 0 · 10/A = −1	Medium	Best all-around system; strong betting correlation; easiest to teach and drill.	Not the strongest playing efficiency for every hand.
KO / Red Seven	Unbalanced variants	Medium-low	Avoids true-count conversion; good for speed and lower mental load.	Less precise across deep shoes unless calibrated.
Hi-Opt II + Ace side count	2/3/6/7 = +1 · 4/5 = +2 · 10 = −2 · A = 0	Very high	Powerful play accuracy when you can separately track Aces for betting.	Hard; ace-neutral means betting needs side-count adjustment.
Omega II	2/3/7 = +1 · 4/5/6 = +2 · 9 = −1 · 10/A = −2	Very high	Advanced balanced count with stronger playing information than Hi-Lo.	Two-level tags slow many players down.
Wong Halves	2/7 = +0.5 · 3/4/6 = +1 · 5 = +1.5 · 9 = −0.5 · 10/A = −1	Elite	High theoretical performance for expert players comfortable with fractions.	Fractions under casino speed invite mistakes.

Recommendation for this trainer: master Hi-Lo first, because betting, true-count conversion, and index deviations matter more than using a fancy count badly. Add side counts after you can deal through a shoe with no errors.

Betting correlation

How well the count predicts your edge.

Hi-Lo is strong here because tens and Aces drive blackjacks, doubles, splits, and dealer bust patterns.

Playing efficiency

How well it changes hit/stand/double/split calls.

Multi-level counts can improve borderline play decisions, but only with accurate indices.

Insurance correlation

How well it predicts dealer blackjack.

Ten-density matters. This is why insurance becomes mathematically correct at high true counts.

Full composition mode: the theoretical ceiling

Balanced counts compress the shoe into one number. A theoretical composition-dependent system tracks the exact remaining count of every rank, then uses the actual next-card distribution instead of an approximate true count.

What it tracks	Why it matters	Trainer display
Every rank left	A 16 made from 10-6 is different from 8-8 or 5-5-6 because removal effects differ.	Card bubbles show how many of that rank remain in the shoe.
Ten-value density	Insurance, doubles, blackjacks, and dealer bust/made-hand rates are driven by tens.	Next-card panel shows 10-value %, Ace %, low-card %, and top remaining ranks.
Ace density	Aces change blackjack rate and betting value, especially in ace-neutral systems.	Full composition separates Aces from 10s instead of hiding both inside one tag.
Equivalent richness	You still need a compact signal for betting and UI coaching.	The trainer reports an approximate equivalent count from high-card richness.

Use this as theory/training, not casino execution: exact composition is the most informative model, but it is mentally expensive. It is excellent for understanding why index plays work and for simulator-style probability coaching.

True count conversion

The running count alone is incomplete. +6 is huge with one deck left and modest with five decks left.

True Count = Running Count ÷ Decks Remaining

RC +8

4 decks left → TC +2
Good, but not extreme. Small raise.

RC +8

1 deck left → TC +8
Very strong. Big bet and several deviations may activate.

Practical rounding: floor positive true counts for betting discipline. Example: +2.9 acts like +2. For negative counts, do not argue with the shoe — stay minimum or leave.

Deck estimation

The advanced skill is not adding tags — it is estimating decks remaining fast enough to bet before you hesitate.

Use half-deck precision in a 6-deck shoe once comfortable.
Practice with discard-tray photos: 0.5, 1, 1.5, 2 decks gone.
Convert both ways: cards remaining → decks remaining; decks gone → penetration.
Deep penetration matters: a TC +4 late in the shoe is more valuable than early noise.

Bet ramps: the count only pays if the money moves

Counting wins by betting more during positive edge and less during negative edge. The exact ramp depends on bankroll, table minimum, table maximum, rules, penetration, and your risk tolerance.

True count	Conservative ramp	Aggressive ramp	Trainer color	Reason
TC ≤ 0	1 unit	1 unit / sit out	Red edge	The house has the advantage. Do not donate volume.
TC +1	1 unit	2 units	Neutral	Often still near break-even after variance and rules.
TC +2	2 units	4 units	Raise	The player edge starts to become meaningful.
TC +3	4 units	6 units	Raise	More blackjacks and stronger doubles/splits.
TC +4	6 units	8–10 units	Green edge	High-card density is now clearly favorable.
TC +5 or more	8 units	10–12+ units	Max zone	Biggest edge, but biggest heat and variance.

Bankroll warning: big spreads create big swings. If you are practicing, the goal is not “always max bet at + count.” The goal is a repeatable ramp you can survive.

Unit

Your smallest planned bet. If the unit is $10, a 6-unit bet is $60.

Spread

Max bet divided by min bet. A 1–8 spread is common in shoe-game practice.

Wonging

Enter positive shoes and leave negative shoes when allowed. It is powerful because it avoids bad hands entirely.

Index deviations: when the count overrides basic strategy

Most hands are still basic strategy. Deviations are narrow, high-value exceptions. The trainer should make these obvious with blue/purple shading because they are the places where counting changes the move.

Priority	Play	Index trigger	Action when trigger hits	Why it matters
1	Insurance	TC +3 or higher	Take insurance	Only when enough 10-value cards remain to justify the side bet.
2	16 vs dealer 10	TC 0 or higher	Stand	High cards make hitting 16 worse; standing becomes less bad.
3	15 vs dealer 10	TC +4 or higher	Stand	At very high counts, the bust risk from hitting overtakes the gain.
4	10 vs dealer 10	TC +4 or higher	Double	Extra tens make the one-card double more profitable.
5	12 vs dealer 3	TC +2 or higher	Stand	Dealer bust risk rises enough to stop hitting.
6	12 vs dealer 2	TC +3 or higher	Stand	Similar logic, but needs a stronger count.
7	11 vs Ace	TC +1 or higher	Double	More tens make the double strong despite the Ace.
8	9 vs dealer 2	TC +1 or higher	Double	Positive count turns a marginal hit into a profitable double.
9	10 vs Ace	TC +4 or higher	Double	Advanced high-count leverage spot.
10	13 vs dealer 2	TC −1 or higher	Stand	If count is very negative, hit instead because low cards remain.

Use a prioritized index set: learn the top 6–10 perfectly before adding more. A rare advanced deviation done wrong is worse than no deviation.

Surrender deviations

If surrender is available, the “Fab 4” surrender indices become very valuable: 16 vs 9, 16 vs 10, 15 vs 10, and 15 vs Ace. This trainer currently labels rules as no surrender, so these belong in the advanced guide rather than the active buttons.

Composition effects

At the highest level, the same total can play differently depending on the exact cards. Example: 16 made of 10-6 is not identical to 8-8 or 5-5-6. Use composition only after count, true count, and primary indices are automatic.

Risk and bankroll

Advanced strategy is not just finding edge. It is surviving variance long enough for the edge to appear.

Kelly

Bet proportional to edge
Full Kelly is volatile; many players use half-Kelly or less.

RoR

Risk of ruin
The chance your bankroll dies before the math has time to work.

N₀

Long-run threshold
How many hands are needed before skill reliably overcomes noise.

Game selection

A weaker count in a great game beats a perfect count in a bad game.

Prefer 3:2 blackjack, not 6:5.
Deep penetration is critical; shallow shoes kill EV.
Good rules: double after split, resplit pairs, late surrender if available.
Low table minimum relative to bankroll gives room for a safe spread.
Fast dealers increase hands per hour, which increases both EV and variance.

Ace side count

Track Aces separately to sharpen betting and insurance decisions, especially with ace-neutral systems like Hi-Opt II.

Side counts

Advanced players may track 7s or specific ranks for certain deviations. This is only worth it when primary counting is effortless.

Shuffle tracking

Estimating clumps through a shuffle is an expert niche. Treat it as theory unless the shuffle is weak and repeatable.

Team play

Spotters keep counts at tables; a big player enters when edge is high. Powerful, complex, and execution-heavy.

Cover vs EV

Cover means hiding skill: slower bet ramps, occasional harmless-seeming choices, table talk, or shorter sessions. Every cover play has a price. Practice should show the clean EV move first; cover is a separate real-world constraint.

Heat signals

If bet jumps perfectly with the count, sessions run long, or decisions look robotic, the play becomes easier to identify. For learning, keep the trainer pure. For real-world discussion, separate “best mathematical move” from “best survivable move.”

Practice ladder: how to actually get advanced

Perfect basic strategy.
No count yet. If base play is shaky, every advanced layer compounds errors.

Count down one deck under 30 seconds.
Hi-Lo tags must be automatic before shoe play.

Add true count conversion.
Run drills at 1, 1.5, 2, 3, and 4 decks remaining.

Use a fixed bet ramp.
Say the next bet before the hand is dealt. Purple means the bet should change.

Learn top index plays.
Start with insurance, 16v10, 15v10, 12v3, 12v2, 11vA.

Add distraction and speed.
Talk, timer, dealer pace, multi-hand cards, bankroll tracking.

Only then test advanced systems.
Compare Hi-Lo vs Omega II/Hi-Opt II after measuring your actual mistakes.

What the trainer should teach next

The trainer now includes system selector, dark mode, bet ramps, index profiles, and a full-composition probability panel. Next useful upgrades are: index-drill mode, bet-ramp editor, surrender-rule toggle, side-count toggle, and post-round EV/explanation summaries.

Plain-English rule

Advanced counting is not “memorize more.” It is “make fewer mistakes while making larger bets only when the shoe deserves it.”