Read Stella locks and target probabilities

Compare the same unlocked shard count, arrangement, locks and target area. This virtual experiment uses the site placement and area model; it does not run game-server draws or spend real currency.

What does a lock preserve?

Locked shards retain their positions. Other shards occupy distinct positions among the 12 slots excluding locked positions. More locks reduce choices but may also preserve unfavorable positions, so they do not always increase the target probability.

Displayed odds are a sample estimate

The display is outcomes meeting the target ÷ sample count × 100. For example, 25 qualifying outcomes out of 1,000 give 2.5%. More samples generally stabilize estimates, but zero observed hits do not prove impossibility. A deterministic random seed is used for the same conditions, so estimates may repeat.

A stopped run is not the true probability

A first success on roll 100 gives a converted run value of 1÷100×100=1%. This describes one waiting-time record, not evidence that the true success probability is 1%. Do not compare different locks, targets or shard counts as equivalent conditions.

Cost per reroll by lock count

Example: with 2 locks and a 100 SP cap, floor(100÷30)=3 rolls are allowed, spending at most 90 SP. The run stops earlier if it reaches the target.

Even when attempt and SP limits are zero, execution is not infinite. The current site safety ceiling is 100,000 rolls. Stopping at a cap is not counted as success. Reset totals clears the displayed spending and run statistics; it does not create a better arrangement.