Level two
You know what a price is. Now the part that separates a player who follows rules from one who thinks: ranges, sizing, and what the numbers are really telling you.
You cannot see their cards, so guessing them is a waste of effort. What you can do is narrow the set of hands they could hold, and ask how your hand does against that whole set.
Middle pair is a perfectly good hand against somebody who has checked twice. Against somebody who has bet three streets it is close to nothing, because the hands that bet three streets are almost all better than it. Your cards are identical in both cases. The read is doing all the work.
Every action narrows it. Someone who raised before the flop and bet three streets on a board of king, nine, four, two, seven does not have a random hand. Middle pair is a fine hand against somebody who checked twice and a bad one here, and the cards in your hand never changed.
When you bet, the hands that fold pay you nothing. Your bet is only ever paid by the hands that call. So the question before betting is not "is my hand good", it is "am I ahead of the hands that would put money in against me".
Top pair with the best kicker wants to be paid. Bet a quarter and the worse hands call but you win too little; bet the pot and the worse hands fold, which is the money you were betting for. Half the pot is where those two stop trading evenly. The sizing table on the table page works this out for every size.
The same hand bet six different ways. Bigger bets win more when called and fold out more of the hands that were paying you, and the peak is usually in the middle rather than at either end.
That distinction changes sizing. A very strong hand wants a big bet, because you are ahead of everything that calls and you want as much of their stack as possible. A decent but not great hand wants a small bet or a check, because the hands that call a big bet are exactly the ones that beat you.
A rough ladder, once you have decided you are ahead of the calling range: a third of the pot for thin value, half for a clear edge, three quarters or more when you are a big favourite and want to be paid.
Expected value is what a decision is worth on average if you faced it many times: your chance of winning multiplied by the pot you would win, minus what it costs you. It is the only yardstick that survives the short run, because it does not care whether this particular hand came in.
The objection people reach for is that it ignores how the other person is playing. It does not. It is worse than that, and more useful: EV is not a fact about your hand, it is a fact about your assumption. Every EV figure is built on a guess about what they could hold. Change the guess and the number changes with it. There is no EV without a read.
Which is why frequency is the thing to watch. A player who bets almost every hand has a betting range of almost every hand, so a bet from him narrows nothing and your assumption should barely move. A player who has folded for an hour and suddenly bets has a betting range of very few hands, and the same bet is enormous evidence. Same action, same size, entirely different information, because the information is in how often they do it, not in the doing.
Sizing carries information the same way. Someone who bets a third of the pot with everything tells you nothing by betting a third of the pot. Someone who has bet small all night and then bets the pot has told you something, whatever they hold.
So the honest order of operations is: watch what they do and how often, use it to decide what they could have, and only then work out what your hand is worth against that. People invert this. They compute a number against an imaginary average opponent, notice it does not match the person in front of them, and conclude the arithmetic is useless. The arithmetic was fine. It was answering a question about somebody else.
What EV genuinely cannot do. It cannot supply the read. It cannot tell you anything about this one hand, only about the average of many like it. And it says nothing about the hands you are not in, which is where a great deal of money is won and lost.
On this site the range is read from the betting, and weighted by who is doing it: a bet from a player who bets constantly moves the read far less than the same bet from one who rarely does. The table says which read it used, in words, underneath the percentage. If you disagree with the read, you should disagree with the number, and that disagreement is the skill.
Chance of winning is your equity against the players still in the hand, on the cards currently showing, reading them for a range based on how the betting has gone.
Versus average is that figure minus an even share of the pot. An even share is 100 divided by the players still in, so three handed it is 33.3%. Above zero you are ahead of the field, below it you are behind, and it means the same thing at any table size.
Bet signal is lesson six as a number: your equity against the hands that would call, minus the point where betting breaks even. Zero says check, and the further above zero, the more you can charge.
What to do turns those into a plain instruction. It is a rule of thumb from the same figures, not a solver, and it is right often enough to learn from and wrong often enough that you should understand why it says what it says.
Recent hands keeps the hands you have played so you can open one again and see what each decision was worth. A round goes past quickly, and this is where you find out what you actually did.
Play with the read-outs on until the numbers start to feel familiar, then turn them off with the stats switch and play a session honestly. The report says which way you played, because a rating earned with the answer on screen is not the same as one earned without it.
Watch the leak diagnosis at the end more closely than the rating. Being told you call too much, over a few hundred hands, is worth more than any single number, because it is the one thing you can go and fix.
Use the money and big blinds toggle. Money is easier to feel, big blinds are how strategy is actually discussed and how you compare a session at one stake with a session at another.
The three cards on the flop change what every hand at the table is worth, and they do not change them evenly.
The same pair is worth very different amounts on these two boards. On the left almost nothing can catch you; on the right half the deck can.
On a dry board there is little to draw to, so a good hand stays good and you can bet smaller and more often. On a wet board half the deck can change the answer, so a hand that is ahead now wants to charge for the privilege of catching up, and a big bet is easier to justify.
This is also what makes a bet believable. Somebody betting large on a board where nothing could have improved is telling a story that does not fit; the same bet on a board full of draws fits perfectly, whether or not it is true.
The same hand is played differently with 20 big blinds behind than with 200, and it is not a matter of taste. What matters is the effective stack: the smaller of the two, since nobody can win more than the shorter player has.
Deep, implied odds are real: a small pair is worth playing because the times it hits a set it can win somebody's whole stack. Short, they are not: hitting the set wins you what little is left, so the hand has to be good enough on its own. A drawing hand needs money behind it to be worth chasing, and a made hand wants the money in before anything can catch it.