Solver Study & Heuristics

How to turn a solver from a thing you stare at into a thing you learn from. The goal is never to memorize outputs — it's to extract a few transferable rules you can actually use at the table.

Read range-first, not hand-first

When you open a solution, resist looking up "what does my hand do?" Instead, zoom out:

1. Aggregate first: what % of the range bets vs checks? At what sizes?
2. Then the pure strategies: which hands always do one thing? (e.g., "nutted hands always bet"). These are your anchors.
3. Treat mixed strategies as indifference, not instructions. A hand that bets 50% means both lines are about equal EV — do not try to memorize the percentage. Understanding why it's indifferent is the lesson.

Aggregation reports = the heuristic machine

The single most valuable solver feature for learning. An aggregation report shows the strategy across all 1,755 distinct flops at once, sortable and filterable. Instead of studying one board, you see the pattern across textures — which is exactly the transferable principle you want, not a memorized board.

Use it to answer questions like: "How does my turn-barrel frequency change by turn card?" — then write down the rule.

Node-locking for exploits

To study how to beat a specific tendency (your pool over-folds turn, say), node-lock the opponent's strategy to that mistake and re-solve. The solver shows you the maximally exploitative response. This is how you turn population reads into concrete adjustments. → see Hand Reading & Reads

The weekly solver loop

1. Target a high-frequency, recurring spot (not the random big pot you lost).
2. If it's new to you, study the solution first (it's complex — worked examples beat struggling when you're new).
3. Run the aggregation report; find the cross-board pattern.
4. Distill 3–5 heuristics — write the rule + the why.
5. Card them in Anki, then drill them closed-book (Drilling), then test in play.

The output is rules, not memories

A good solver session ends with 3–5 written heuristics, each with a reason. If you finish a session having "looked at a lot of sims" but can't state the rules you extracted, you studied the wrong way.

Key takeaway

Study ranges, extract rules, ignore the noise. Read aggregate-first, treat mixed strategies as indifference, and leave every session with a handful of why-backed heuristics you can drill and test — not a pile of memorized frequencies.

---

Sources: GTO Wizard — aggregation reports · PioSOLVER — node locking · Using a solver effectively