A turn is a claim about what matters next.

Begin with the binding constraint

Inventory the position before selecting actions. Record available resources, board space, hand limits, health pressure, known effects, relevant breakpoints, and the opponent responses that deserve real weight. Then identify the binding constraint: the condition your turn must satisfy before any extra benefit matters.

A constraint might be survival against a common damage range, preventing an opposing engine from remaining active, developing enough pressure before the opponent's stronger phase, or keeping a specific answer available. It is narrower than play for tempo and more useful than make the best board. Two positions that look similar can have different constraints because time, role, and remaining resources differ.

Separate required work from optional improvement. If survival requires removing one threat, that action belongs in every serious line. The remaining resources can then be compared honestly. Without this separation, attractive optional actions compete with necessities and make a line look more flexible than it is.

Build at least two complete lines

A complete line states every meaningful action in order. It accounts for resources, targets, board space, information gained during the sequence, and what remains afterward. Do not compare clear plus develop with maybe use the other card. Give the alternative an exact order and a plausible resulting position.

Include a baseline line that meets the constraint with low complexity. Then build a line that pursues a different advantage, such as preserving material, increasing pressure, gaining information, or protecting flexibility. A third line is useful only when it represents a genuinely different tradeoff. Listing minor permutations can create the feeling of analysis without exposing a new decision.

For each line, sketch the next turn. What resources remain? Which draws or opponent actions can be answered? Which card or effect becomes stranded? Does the line create a clear follow-up, or does it borrow strength from the future without a repayment plan? The next-position sketch is often where a superficially efficient line reveals its cost.

1. State the required end condition.
2. Write the action sequence in order.
3. Name information that could change later actions in the same turn.
4. Record resources and options left after the sequence.
5. Describe one credible opponent response and your next-turn answer.

Sequence for information and flexibility

Two lines can contain the same actions and differ because of order. Ask whether an early action reveals information, creates board space, changes a random outcome, commits a scarce resource, or closes off a branch. When possible, gain useful information before making the decision that depends on it. When the information has no practical chance to change your line, do not spend resources merely to feel informed.

Distinguish reversible and irreversible actions. A reversible action preserves more branches if new information appears. An irreversible action may spend the only answer, fill the final space, select a target, or expose a key resource. Delaying commitment has value, but it is not free. Sometimes the strongest sequence must commit early to satisfy the turn's constraint.

Check execution after strategy. Count resources, targets, slots, and ordering effects. A strategic idea can still fail through a mechanical omission. Keeping a short execution checkpoint separate from the line comparison prevents counting accuracy from masquerading as strategic depth, and prevents strategic language from excusing a simple sequencing error.

Review the reason at three levels

First review execution. Did the actions occur in the intended order, and were resources, targets, and board limits counted correctly? Second review the line. Given the opponent range you held then, did the resulting position meet the constraint and retain acceptable options? Third review the process. Did you identify the right constraint and compare a real alternative before committing?

Later information belongs mainly in the line review. It can reveal that an opponent response was more common or more damaging than you estimated. It should not be inserted into the original observation. Preserve both records: what you believed and what the replay taught. That difference is where learning lives.

End with a conditional test. When a turn has a mandatory defensive action and several optional gains, I will define the required end condition before ordering the rest. Or: when two lines use the same actions, I will check whether one order gains information before commitment. A precise test can be repeated without demanding the same answer in every position.