Card Counting Values: Every Card from 2 to Ace, Explained
The Hi-Lo card counting values are three numbers across thirteen card ranks: cards 2 through 6 are worth +1, cards 7 through 9 are worth 0, and cards 10, jack, queen, king, and ace are worth -1. Add the value of every card the dealer turns. The total is your running count. When it climbs positive, the remaining shoe is rich in high cards and you have an edge. When it drops negative, the shoe is poor in high cards and the house has the edge it usually has. That is the entire system anyone uses to make money playing blackjack. The casino just really, really does not want you to know it.
Below is the long version: every card, why the values look the way they do, how they turn into a bet, and the order to drill them in before you ever sit down for real.
How the Hi-Lo card counting values actually work
The values track one thing: whether the rest of the shoe still has the cards a player wants. A new shoe is balanced. Twenty hands in, it is not. Some cards are gone and some are coming. The count is a running estimate of which way the remaining deck composition has tipped.
High cards favor the player for three concrete reasons. A natural 3:2 blackjack pays you 1.5 to 1 instead of 1 to 1, and you make more of them when the shoe is rich in tens. The dealer is forced to hit on 16 and stand on 17 or higher, which means a tens-rich shoe busts the dealer more often. Doubles on hard 10 and 11 win bigger when the next card is more likely to be a ten.
Low cards favor the dealer for the mirror reason. The dealer cannot bust a 12 against a low draw the way they can against a high one. A 6-rich shoe lets the dealer make hard hands the player paid to bet against. The 5 in particular is the single most house-friendly card in the deck. Wizards of the game have a saying: pull every 5 out of the shoe and the player edge climbs by close to a full percentage point. That is how much one card rank moves the math.
The Hi-Lo card counting values group those effects into three buckets. The values are not a guess. They are weighted by how strongly each rank shifts the player edge in or out of the shoe.
Card counting basics: every card from 2 to ace
Here is the full card counting chart for Hi-Lo, every rank, suits ignored.
| Card | Hi-Lo value |
|---|---|
| 2 | +1 |
| 3 | +1 |
| 4 | +1 |
| 5 | +1 |
| 6 | +1 |
| 7 | 0 |
| 8 | 0 |
| 9 | 0 |
| 10 | -1 |
| Jack | -1 |
| Queen | -1 |
| King | -1 |
| Ace | -1 |
A full deck contains five low ranks worth +1 (4 copies each, 20 cards), three neutral ranks (12 cards), and five high ranks worth -1 (20 cards). The +20 from the lows and the -20 from the highs cancel. Every clean 52-card deck adds up to exactly zero.
That is not a coincidence. Hi-Lo is a balanced count by design. The zero-sum property is what lets you check yourself: shuffle a single deck, flip every card, and you should land on zero. If you land on +1 or -2, you missed a card somewhere. Card counting basics start with that drill. Until a single deck reliably ends at zero, no other practice helps.
Suits do not matter in Hi-Lo. A 4 of clubs and a 4 of hearts are both +1. A jack of diamonds and a king of spades are both -1. The system ignores suit because suit has no effect on hand math at the table.
The values are not rounded. They are not approximate. They are exact. The math behind them was published by Harvey Dubner in 1963 and has not changed since.
The card counting high low system in plain English
The card counting high low system was the first balanced count anyone built that was simple enough to use at a real table. Edward Thorp's earlier Ten Count from Beat the Dealer proved card counting worked, but it tracked tens against non-tens as a ratio, which was clumsy in a casino. Dubner's contribution was the +1/0/-1 split. Same edge for the player, lower error rate at the table, much easier to teach.
The reason every modern trainer (including ours) builds around Hi-Lo is not nostalgia. It is the error math. A clean Hi-Lo player playing accurately captures around 97% of the theoretical edge available to a perfect Wong Halves player. The remaining 3% gets eaten three times over by the higher error rate at the table when you are tracking fractional values like +0.5 and -1.5 under casino conditions. Hi-Lo wins money. The fancy systems are a hobby for people who already mastered Hi-Lo and got bored.
Hi-Lo is also the system every major reference book teaches. Don Schlesinger's Blackjack Attack, Peter Griffin's Theory of Blackjack, and almost every modern guide on the Arnold Snyder archive use Hi-Lo as the working count. The CountEdge AP Analyzer is built on Schlesinger's formulas, which assume Hi-Lo. The canon is consistent for a reason. If you want a deeper read on what the canon is, the blackjack strategy book post lays out the order to read them in.
Running count, blackjack true count, and what the values become
The running count is the sum of every Hi-Lo card value you have seen since the last shuffle. The blackjack true count is the running count divided by the number of decks remaining in the shoe. Always floor the division. Never round.
The formula:
TC = floor(running count / decks remaining)
A running count of +9 with three decks left is a true count of +3. A running count of +9 with 4.5 decks left is still a true count of +2, because we floor. A running count of +5 with five decks left is a true count of +1.
Bets get made on true count. Deviations get made on true count. The running count alone tells you almost nothing in a multi-deck shoe. A running count of +6 with one deck left is a strong bet signal. The same +6 with six decks left is barely above neutral. Without the conversion, the values are tracking a number you cannot use.
Each true count point is worth about +0.5% to the player edge over the base house edge. That number comes from Peter Griffin's Theory of Blackjack and has held up across decades of simulation. At a 6D H17 DAS LS table with a -0.47% base edge, you cross into positive expectation around true count +1, and the edge grows linearly from there. At TC+4 with the standard Illustrious 18 deviations applied, a serious counter is playing at roughly +1.5% over the house. That is where the money lives.
Estimating decks remaining is the half of true count work that nobody emphasizes enough. The discard tray sits across the table. You eyeball the stack, estimate to the half-deck, and subtract from the total. A half-deck error in the denominator can swing your true count by a full unit, and a full unit on the bet ramp is the difference between $30 and $100 on the felt at TC+1 versus TC+2. Drill the estimation as hard as you drill the running count.
Card counting chart: where the values turn into plays
The values feed two things at the table. The bet and the deviations. Both are summarized on a card counting chart that should be in every counter's head before they sit down.
The bet ramp is the first half. It maps true count to bet size. A typical 1-8 spread at a $10 minimum 6D H17 DAS LS game looks something like this:
| True count | 1H bet | 2H bet per hand |
|---|---|---|
| 0 or below | $10 | n/a |
| +1 | $30 | n/a |
| +2 | n/a | $30 |
| +3 | n/a | $50 |
| +4 | n/a | $100 |
| +5 | n/a | $150 |
| +6 and above | n/a | $300 |
The bet ramp is the engine. Without the spread, the values do nothing for you. A counter who tracks every card perfectly but flat-bets every hand has the exact same expected loss as a recreational player who never looked at a single card. The count is the trigger. The spread is what turns the trigger into money.
The second half of the chart is the deviation set. The Illustrious 18 and Fab 4 are 22 specific plays that change at specific true counts. Insurance becomes correct at TC+3. 16 vs 10 becomes a stand at TC zero. 12 vs 3 becomes a stand at TC+2. 10,10 vs 5 becomes a split at TC+5. None of these match basic strategy. All of them are worth real EV when the count says yes. At TC+4 and above the index plays start to dominate the EV calculation, which is why every serious advanced blackjack strategy post leans on them heavily.
The chart is the bridge between the values and the money. Memorize it.
Card counting practice: drilling until the values disappear
The first goal of card counting practice is to stop translating in your head. A 4 of clubs is no longer a 4 of clubs. It is +1. A jack of hearts is no longer a face card. It is -1. Once the cards stop translating, the count gets fast. Until they translate, every card costs you cognitive load you do not have at a real table.
Start with one deck. Shuffle. Flip cards one at a time and run the count. End at zero. If you do not land on zero, do it again. Most published targets land around 25 to 30 seconds for a single deck. Under 30 seconds, five times in a row, is the basic checkpoint. Below 20 seconds is fluent.
Then move to a full 6-deck shoe at dealing pace. A real dealer at a 6D H17 DAS LS table runs about 80 hands per hour. Run a shoe at that speed, stop at random points, call the running count out loud, and finish at zero. If you bust the count once at minute 12, you bust the count for the rest of the shoe.
Then add true count conversion. Deal a shoe. Stop at random. Call running count. Estimate decks remaining. Floor the division. Drill until the conversion is automatic with no mental pause.
Then add deviations. Then add cover. The full ladder, with speed targets and the order to climb it, lives in the card counting practice post.
Anyone can memorize three numbers. Running the math 80 times an hour while a drunk tablemate keeps asking if you want to split tens is a different skill entirely. The trainer is the warmup. The casino is the test. Drill with distractions. Drill with a podcast playing at full volume. The translation has to survive noise or it will not survive a Saturday night in a real room.
The free CountEdge tier covers basic strategy training. The Pro and Elite trainer tiers add full shoe drills with running count, true count, deviation surfacing in shoe mode, weakness reports, and the bet ramp drills that build the muscle to actually push the chips out when the count says push.
Card counting strategy: when to bet what the values tell you
This is where most beginners leak money even after they have the values down. The Hi-Lo math gives you a clean signal. Card counting strategy is what you do with it.
The signal is true count, not running count. The trigger is TC+2 on a typical 1-8 spread. Below TC+2, you are flat-betting the minimum. At TC+2 and above, the bet ramps up. The exact shape of the ramp depends on the spread the table will let you get away with, but the trigger is the same.
The mistake almost every new counter makes is shading the bet down at the moment it matters most. You see TC+4 and the math says push $200 on two hands. The chair-internal monologue starts. The pit boss is watching. The cocktail waitress is watching. Maybe the guy two seats over knows what you are doing. So you push $100 instead. Next shoe, you tell yourself.
That fear costs you the entire reason you sat down. 90% of the time the pit is watching the dealer, not you. Their job is making sure the house does not make a payout error. You are not the priority you think you are. The count does not care about your ego. Neither does the variance. Put the money out.
The opposite mistake is worse: counting accurately and refusing to spread at all. A counter who tracks every card perfectly but flat-bets every hand has the same expected loss as a non-counter, plus a tax on attention. The entire edge in card counting comes from betting more when the count is positive. A 1-1 spread (no spread at all) at any count gives you a flat house edge regardless of how clean the count is. A 1-8 spread at a 6D H17 DAS LS game pulls you to roughly +1.0 to +1.5% edge depending on penetration and ramp shape. The count is the trigger. The spread is the engine. If you are scared to spread, do not count. Train, build a bankroll, and come back when you can.
Learn to count cards without lighting yourself on fire
This is the part nobody puts on a slick training page. Here is what NOT to do.
Do not switch counting systems before Hi-Lo is automatic. The marginal edge from a level-2 or level-3 system is real and small. The error rate at the table is huge if you are tracking fractional values before you can run a Hi-Lo shoe cleanly in a noisy room.
Do not skip basic strategy. A perfect Hi-Lo player at a 6D H17 DAS LS table is still only -0.47% behind the house off the top. Most casino players give up another 1 to 2% by misplaying soft hands, hitting hard 17s, and mishandling pairs. A new counter who is sloppy on basic strategy is donating a bigger edge than they will ever recover by counting.
Do not flat-bet through a positive count. If the bet ramp scares you, the bankroll is too small. Build it before you sit.
Do not chase a count into the night. When the count drops, the bet drops with it. Some sessions go cold for an hour. Variance does not care that you did everything right.
Do not telegraph the count with your eyes. A pit boss who watches your eyes track every card and watches your bet jump from $10 to $300 the moment the shoe goes positive does not need to know the math. They just need to know what the pattern looks like. They have known since 1963.
Do not take the count to a 6:5 table. The headline payout adds about 1.39% to the house edge and erases the deck advantage on every variant. Walk past them.
To learn to count cards cleanly, the order is: basic strategy, then the Hi-Lo card counting values, then running count under shoe conditions, then true count conversion, then the I18 and Fab 4 deviations, then the bet ramp, then cover. Each rung locks in before the next one. Twenty focused minutes a day beats two hours of distracted clicking. Most players who put in the work are ready to test live within 90 days.
The whole reason CountEdge exists is laid out on the About page. Nothing else on the market was good enough for a working counter to use at a real table. The free tier is the place to drill basic strategy. The 14-day Pro trial opens the running count, true count, I18, and Fab 4 drills the moment basic strategy is automatic. The math has been settled since Griffin published Theory of Blackjack. The values have not changed since 1963. The casino has been running the same game the whole time. Put in the practice and you take a small piece of theirs back.