A measurement shot-scale three-row scan holds the prepared state fixed at amplitudes three-fifths and four-fifths while only the trial total changes, so the expected counts scale exactly with the trial count rather than with any change in the underlying probabilities. The three rows use 25 trials (9 zero, 16 one), 100 trials (36 zero, 64 one), and 225 trials (81 zero, 144 one) for the stated toy qubit ledger; real measurement batches have finite-shot statistical noise around these expected counts, not the exact counts themselves.
A single fixed state is measured at three trial totals, scaling both expected counts by the same factor as the trials. Exact arithmetic here means exact results for the stated model inputs; measured inputs still carry uncertainty and significant-figure limits.
highlighted = computed this step
A fixed state gives exact counts at the smallest total
The state stays fixed at three-fifths and four-fifths throughout this scan. At 25 trials, the expected counts are 9 zero and 16 one.
Nzero=9,None=16
The trial total is the only source that changes
Chapter six scanned three different states at roughly matched trial counts. Here the state is held fixed and only the trial total moves: quadrupling the trials quadruples both expected counts.
N25100225Nzero93681None1664144
Nine times the trials gives nine times each count
The largest row uses 225 trials, nine times the baseline. Both expected counts scale by the same factor nine, since the probabilities never changed.