Complete two threats together

If you finish your turn with two valid lines sharing no tile, the opponent’s next removal cannot destroy both. Any tile belongs to at most one of those lines. Removing from one leaves the other intact; removing elsewhere preserves both. This is an exact conclusion that does not depend on the opponent’s skill.

The easiest pattern to recognize is two lines in separate areas, such as A1–D1 and A8–D8. They need not be horizontal or far apart: their sets of cells simply must be disjoint. A diagonal and a vertical can also create an unavoidable double threat when each is valid and they do not touch.

Suppose that after your removal you have blue 1s at A1, B1 and C1, and blue 8s at A8, B8 and C8. D1 and D8 are empty and you retain one copy of 1 and one of 8. Placing D1=1 and D8=8 completes two separate lines. You used the fourth copy of each number without exceeding your inventory.

Check this pattern after your removal, when you know the board on which you will place. In a real game, the opponent may previously have removed a tile from one of the trios. If that has happened, count the gaps again: completing both projects will no longer be possible with the same pair you had prepared.

Two trios are not already two winning lines. If you finish your turn leaving both incomplete, the victory check does not force your opponent to remove from them. They can choose their preferred defense and then place. “I threaten to complete” and “I have completed two resilient lines” are different claims.

Also distinguish two possible places for one tile from two compatible placements. If two projects require different values at D4, one tile cannot satisfy both. If both require the same last copy of 6 in different cells, the reserve prevents the plan. Spatial and numerical compatibility are part of the tactic.

After resolving removal, identify windows needing one or two tiles. For each, note the required cells and numbers. Combine projects you can finish with a total of two legal placements. Then list every line on the final board again and ask whether any removal can break them all.

The best pair may look modest when you examine only its first tile. One placement finishes a vulnerable line and the other eliminates its saving defenses by creating an independent line. Compare the result of both tiles together. Choosing an attractive placement first and then fitting in the remaining tile may cause you to overlook a better combination.

To prove your pair works, try to defeat it yourself: remove a tile from the first line and locate the survivor; remove one from the second and repeat. If the lines are disjoint, this argument covers every defense. If they cross, it no longer suffices and you must study the global intersection in the next chapter.

As strategic preparation, developing projects with different requirements can make defense harder. That recommendation is a heuristic: it does not prove the projects can be completed against perfect play. In exercises, demand a concrete proof of the immediate outcome. In games, use projects to generate candidates and check again after every removal.

One threat draws attention to one place; two fronts force a choice. If your opponent removes a tile from your first staircase, you can spend one placement rebuilding it and the other preparing a second. Before repeating, check that the second front really improves: rebuilding without progress can lead to a loop.

Work backwards from the desired finish. For example, a 3–4–5–6 staircase on B3–E3 and a 4–5–6–7 staircase on B6–E6 share no tiles. If both are complete when you hand over the turn, no single removal can break both. This is a teaching pattern, not a forced sequence from the opening.

Two unfinished projects are not two resilient lines. Your opponent may remove a supporting tile before you complete them. In this chapter’s exercises, use both placements and then try the defense: check what survives rather than merely counting threats.