You finally opened a solver, ran your first sim, and stared at a grid full of colored squares, tiny percentages, and numbers that felt like they belonged in a spreadsheet from another planet. Most players have that exact moment. They expect the solver to hand them clean answers, and instead it hands them a wall of data that seems to raise more questions than it settles.
Here is the truth that unlocks everything. A solver does not give you rules. It gives you frequencies, and learning how to read solver output is really about learning to translate those frequencies into actual understanding you can carry to the table. Once you crack that translation, the wall of colored squares turns into one of the most powerful learning tools in poker.
This guide walks you through exactly how to interpret what a solver shows you. We will break down frequencies, expected value, and the mixed strategies that confuse almost everyone at first. By the end, you will look at solver output and see a strategy rather than a headache.
What a Poker Solver Actually Does
Before we read the output, let me explain what is actually happening under the hood, because it shapes how you should interpret everything.
A poker solver calculates a Game Theory Optimal strategy for a specific situation you define. You set the parameters, the stacks, the positions, the ranges, the bet sizes, and the solver runs millions of calculations to find the strategy that cannot be exploited. It converges toward what is called a Nash equilibrium, a point where neither player can improve by changing their approach.
So the solver is not telling you what your opponent will do. It is telling you the unexploitable baseline against a perfect opponent. That distinction matters enormously, because it explains why solver output rarely says “always do this.” Instead, it says “do this a certain percentage of the time,” and that percentage is the first thing you need to learn to read.
If you want a deeper foundation on the theory itself, our guide on GTO poker explains why unexploitable strategies work and where they help versus where exploitative play beats them. Understanding that background makes solver output far easier to interpret.
Reading Frequencies: The Heart of Solver Output
The single most important concept in solver output is frequency. So let me make this crystal clear, because everything else builds on it.
When a solver shows you a hand, it usually assigns percentages to each possible action. For example, it might tell you that with A♠Q♠ on a particular flop, you should bet 65 percent of the time and check 35 percent of the time. That does not mean the solver is confused. It means the optimal strategy involves taking both actions, split across the many times that exact situation arises.
Think of it this way. If you played that identical spot one hundred times, the solver wants you to bet in roughly 65 of them and check in the other 35. This is the concept of a mixed strategy, and it is the thing that trips up nearly every player who opens a solver for the first time.
Here is a simple way to read frequency output:
- 100 percent one action: This is a pure strategy. The hand always takes this action. These are the easiest to learn and the most important to get right.
- Split frequencies (like 60/40): This is a mixed strategy. The hand takes different actions at different times to stay balanced and unexploitable.
- Tiny frequencies (like 95/5): The dominant action is almost always correct, and the small percentage is a minor balancing adjustment you can often ignore in practice.
Whenever you study frequencies, focus first on the pure strategies and the heavily weighted actions. Those carry the most value and are the easiest to apply. The tiny mixed frequencies matter far less in real games, especially at lower stakes.
Understanding Expected Value in Solver Output
Alongside frequencies, solvers display expected value, usually abbreviated as EV. This number tells you how profitable each action is, and reading it correctly changes how you interpret everything else.
Expected value represents the average outcome of an action across all possible runouts. So when a solver shows that betting has an EV of 4.2 and checking has an EV of 4.0, it is telling you that both actions make money, but betting makes slightly more on average. Understanding EV in poker is essential here, because the entire solver framework is built on maximizing expected value across millions of simulated hands.
Now here is a subtlety that confuses many players. When a solver mixes between two actions, it usually means those actions have nearly identical EV. The solver splits its strategy precisely because both options are close to equally profitable. So if betting and checking both show an EV of 4.1, the solver mixes them, and in practice you can often simplify by choosing whichever action is easier for you to execute correctly.
This leads to one of the most practical insights in all of solver study. When two actions have close EV, the exact frequency barely matters for your bottom line. When one action has significantly higher EV, that is where you must pay attention, because deviating from it genuinely costs you money.
Mixed Strategies: Why the Solver Won’t Give You One Answer
Let me address the concept that frustrates players more than any other, the mixed strategy. Because once this clicks, solver output stops feeling contradictory.
A mixed strategy exists for one reason: balance. If you always bet your strong hands and always checked your weak ones, a good opponent would read you instantly and exploit you. By mixing, you make your ranges impossible to read. The solver builds these mixes to keep your entire strategy unexploitable, which means individual hands sometimes take actions that feel counterintuitive in isolation.
For example, the solver might check some of your strong hands and bet some of your weak ones. In isolation, that looks insane. But across your entire range, it creates a balanced structure where your opponent cannot tell strength from weakness based on your action. That is the whole point.
Here is how to handle mixed strategies in practice:
- Identify the dominant tendency. If a hand bets 70 percent, its primary identity is a betting hand. Treat it that way most of the time.
- Don’t obsess over exact ratios. No human executes perfect mixes in real time. Approximating gets you the vast majority of the value.
- Understand why the mix exists. Ask what balance the solver is protecting. This teaches you the underlying logic, which transfers to spots you have not studied.
- Simplify for real games. Against exploitable opponents, you often abandon mixing entirely and exploit their specific leaks instead.
The goal of studying mixed strategies is never to replicate them perfectly. It is to understand the logic behind them so your instincts improve in every spot, including the thousands you will never run through a solver.
Common Mistakes When Reading Solver Output
Even players who understand the basics fall into predictable traps. So let me cover the errors that waste the most study time.
First, treating frequencies as commandments. New solver users see a 55/45 split and agonize over hitting it exactly. In reality, that split means the two actions are nearly equal, so the stress is misplaced. Focus your energy on the pure strategies and the large EV gaps instead.
Second, ignoring the range and studying single hands in isolation. A solver’s recommendation for one hand only makes sense in the context of your entire range. Whenever you study a spot, look at how the whole range is constructed, not just the one combo you held. This is where real understanding lives.
Third, memorizing outputs instead of understanding them. Some players try to memorize solver charts hand by hand, which is both impossible and useless. The value comes from understanding the principles that generate the output, such as range advantage, board texture, and blocker effects. Learn the why, and the specific numbers become far less important.
Fourth, forgetting that solvers assume a perfect opponent. Your actual opponents are not perfect. Solver output is your baseline, but the real money often comes from deviating to exploit specific mistakes, which is a skill built on top of solver understanding, not instead of it.
Turning Solver Output Into Real Improvement
Reading solver output is only the first step. The real value comes from connecting it to the hands you actually play, and this is where most players fall short.
Whenever you finish a session, take the spots that confused you and run them through a solver. Compare what the solver recommends against what you actually did. The gap between the two is your specific leak in that situation, and closing that gap is how solver study translates into a higher win rate. Pairing this with disciplined poker hand review is the fastest way to turn abstract solver knowledge into instinct you can use at the table.
Over time, you stop needing the solver for common spots. You have seen the patterns so many times that you recognize them instantly. That is the entire goal of solver study. Not to memorize outputs, but to internalize the logic so deeply that your real-time decisions start to mirror the equilibrium without any conscious calculation.
Frequently Asked Questions
What does a poker solver output actually show?
A poker solver output shows the Game Theory Optimal strategy for a specific situation, expressed as frequencies for each possible action along with the expected value of each. It tells you how often to take each action to remain unexploitable against a perfect opponent, not what your specific opponent will do.
What does frequency mean in solver output?
Frequency is the percentage of the time the solver recommends taking a particular action with a given hand. A frequency of 70 percent bet means that across all the times you hold that hand in that exact spot, you should bet 70 percent of the time and take the alternative action the remaining 30 percent, in order to stay balanced.
Why does the solver mix between two actions?
The solver mixes between actions when they have nearly identical expected value and when mixing keeps your range balanced and unexploitable. If you always took the same action with the same hand, an observant opponent could read and exploit you. Mixing disguises your range so strength and weakness look the same.
Do I need to follow solver frequencies exactly?
No. Perfect frequency execution is impossible for humans and unnecessary in practice. Focus on the dominant tendencies and the large EV differences. When two actions have close EV, the exact split barely affects your results, so you can simplify by choosing whichever action is easier to execute well.
How do I use solver output against real opponents?
Solver output gives you an unexploitable baseline. Against real opponents who make mistakes, you often deviate from that baseline to exploit their specific leaks. The solver teaches you the equilibrium, and exploitative adjustments layered on top of that understanding are where much of your profit comes from at lower and middle stakes.
Final Thoughts
Learning how to read solver output feels overwhelming at first, but it comes down to three ideas. Frequencies tell you how often to take each action. Expected value tells you how profitable each action is. Mixed strategies exist to keep your ranges balanced and unexploitable. Master those three concepts and the wall of colored squares transforms into a clear, readable strategy.
The mistake most players make is trying to memorize the output. The players who actually improve focus on understanding the logic that generates it. So the next time you open a solver, do not ask “what should I do here?” Ask “why does the solver want this?” That single shift turns solver study from a frustrating chore into the most powerful learning engine in your poker toolkit.
Check Replay is a software used by coaches, winning players, and champions to review hands with speed, clarity, and flow.
