"Lab day," Comet says, fanning out 12 index cards on the picnic table. "One lap time on each."
"Shuffle, deal six and six, find the difference," Wren says. "Fifteen times. Nova keeps the log."
The first shuffle gives -2.7. Then -1.3. Then 0.3. Comet plots each dot as it lands.
"What do you notice?" Wren asks after fifteen. "Not one shuffle reached 4."
"Zero of fifteen," Comet says. "Chance never made our gap, not once."
Nova glows. "Would you like a hint? Fifteen shuffles is a small picture. Twenty-five was better. A hundred would be best."
"So we trust the shape, not the exact zero," Wren says. "The gap is still far out on the edge."
"Then our runners really did get faster after the jog," Comet says, and writes it in the Run Log.
The crew's 15 shuffles are made up for the Run Log. Yours will differ, and the comparison is the point.
| Shuffle | 1 | 2 | 3 | 4 | 5 | 6 | 7 | 8 |
|---|---|---|---|---|---|---|---|---|
| Difference | -2.7 | -1.3 | 0.3 | 2 | -0.3 | 1 | -1.7 | 0.7 |
| Shuffle | 9 | 10 | 11 | 12 | 13 | 14 | 15 |
|---|---|---|---|---|---|---|---|
| Difference | 3 | -2.3 | 1.3 | -0.7 | 0.3 | 2.3 | -1 |
In the crew's lab, 0 of 15 shuffles reached 4. With Tuesday's 25 shuffles the count was 1. Both are under 5%.
A small number of shuffles gives a rough picture. The judgment is the same, but the crew trusts it more with more shuffles.
| Statement | True or false? |
|---|---|
| None of the crew's 15 lab shuffles reached 4. | ? |
| 15 + 25 = 40 | ? |
| Fifteen shuffles give a sharper picture of chance than a hundred. | ? |
| Shuffled differences pile up near zero because chance alone splits the groups. | ? |
Good lab work. Tomorrow a second experiment gives a gap that chance can explain, and you learn to say so.