A chart is a photograph. A range is a moving thing that happened to hold still long enough to be photographed. That distinction sounds like semantics until the moment the picture no longer matches the room you’re sitting in, which happens roughly every third hand you play.
The grids are genuinely good. A solver output compressed into 169 colored squares is one of the most efficient pieces of information transfer poker has ever produced, and players who drill them stop making the catastrophic preflop errors that used to define low stakes. The problem isn’t that the charts are wrong. It’s that they’re answers to questions with long, specific wording, and most people memorize the answer while skipping the question.
What follows is the part that tends to get skipped: what’s actually encoded in those squares, what isn’t, and how to tell which one you’re relying on at the table.
The fine print above the grid matters more than the colors inside it
Every chart carries a header, and the header is the whole thing. Table size, position, the action before you, effective stack, open size, ante, rake. Change any one of those and the grid underneath should change too. The people who build these tools say so plainly: a chart is only useful when its position, previous action, effective stack and game assumptions match the decision you are studying, and the habit worth building is to read the model label first, then compare the chart with the exact spot rather than with poker in general.
Two charts of the same spot can look meaningfully different because one assumes a 2.5x open and the other a 3.5x. Published solution libraries list both separately for a reason. A bigger open prices the blinds out more often, which lets you open a slightly wider set of hands and changes which ones you want.
Rake is the one that quietly breaks everything. A lot of free chart libraries are solved with no rake at all, and some don’t tell you either way. Rake is unspecified is a line you’ll find on perfectly reputable chart pages. Run the numbers on what that omission costs. In a $0.50/$1 online game with a 5 percent rake and a $3 cap, a $20 pot gives up $1 to the house. That’s a full big blind off the top of a pot you’re splitting equity in. Marginal calls that show a tiny profit in a no-rake solve are losers in that game. The hands at the bottom edge of every calling range are exactly the ones that flip.
Why counting colored squares misleads you about your own range
The grid squeezes 1,326 two-card combinations into 169 hand classes, and those classes aren’t the same size. A pocket pair has six combinations, a suited hand has four, and an offsuit hand has twelve.
Do the arithmetic and the chart stops looking like a map. Take the nine offsuit ace classes from ATo down to A2o. That’s nine squares, and 108 combinations. Now take every pocket pair from 22 through TT. Also nine squares, but only 54 combinations. The first block is twice the size of the second while occupying the same amount of visual space. A percentage based on colored cells is simply a different number from a combo-weighted percentage.
This is why eyeballing a chart gives you a distorted sense of what your range is made of. It’s also why blocker reasoning confuses people who think in squares. You hold AK, you’re deciding whether your opponent can really have aces here, and the relevant fact is that your ace removes three of the six combinations of AA. Half of it, gone, from one card. Nobody ever intuits that from a grid.
Mixed frequencies are the chart telling you it doesn’t care
The hands shaded two colors are where most of the misunderstanding lives. A hand marked 60 percent raise, 40 percent fold isn’t a hand the solver is uncertain about. It’s a hand worth almost exactly the same either way. The solver splits because at that precise point the two lines pay the same, and splitting keeps an opponent from building a counter-strategy around the hand.
Which means the mixed hands are, by definition, the hands where your decision matters least. That’s the opposite of how most players treat them. People agonize over whether to open KTo from under the gun at the exact frequency, then donate three big blinds on the turn with a hand they never had a plan for.
Good chart publishers already act on this. Red Chip’s grids round the mixes off, and their stated reasoning is that the simplification is much easier to remember and gives up very little profit compared to the raw, mixed ranges. They keep the overall frequencies intact while flattening individual hands to a single action. The money isn’t in the mix. It’s in not confusing a close hand with an important one.
Ranges are the thing that still exists on the river
Here’s the practical test of whether you’ve learned a chart or a range. Flop comes 7-5-4 with two hearts. You opened from early position, the big blind called. Who does this board favor?
If you memorized the chart, you know your opening range and you know it doesn’t contain 76s or 64s. If you understand it, you know what that implies on every street. The caller has roughly every two-card combination of low connected cards; you have almost none. You have the overpairs. You also have a pile of big offsuit hands that just became worthless.
That’s a board where your big bet is bluffing far more often than you think, and where the caller’s check-raise is credible in a way it wouldn’t be on king-high. Nothing in the chart said any of this. The chart only told you what to open. The range tells you what you’re holding three streets later, and the second thing is what gets paid.
The same gap opens up any time the hand stops being a duel. Two cold callers behind you reshape every equity assumption the solve was built on, and the hands that lose value in multiway pots aren’t the ones people expect. Dominated top-pair hands get much worse. Suited connectors and small pairs get better, because the payoff when they hit scales with the number of people paying it.
The deviations worth making, and the ones that aren’t
Charts describe an equilibrium against opponents playing the same equilibrium. Your $1/$2 table is not doing that. Knowing what the baseline is, then departing from it on purpose, is the whole point. The goal isn’t to play robotically, it’s to know what optimal looks like so your deviations are intentional rather than accidental.
Three deviations that hold up in most soft live games. First, open tighter from early position when the table has four players who call raises and never fold, because your thin opens realize far less equity against a crowd. Second, defend your big blind less than the chart says against small opens when the rake is high and the opener never barrels. Third, three-bet much wider against a player who folds to three-bets more than two-thirds of the time, including hands the chart never lists.
Three that mostly aren’t worth it. Calling off wider because an opponent “seems loose” over fifteen hands. Limping behind because the table does. And flat-calling with aces in a spot the chart raises, which turns a hand that plays itself into a hand that needs you to be a genius after the flop.
The honest self-test takes five minutes. Cover the chart. Name the worst hand you open from the cutoff, then say out loud why that hand and not the one below it. If the answer is playability and the ability to continue on more flops, you’ve learned the range. If the answer is because it’s blue, you’ve learned the picture. The chart will keep being right either way. It just won’t be there on the turn.