"Draw a token, read it, put it back, shake," Comet says. "Twenty times. Wren keeps the tally."
She draws. "One." Again. "One." Again. "Three." The tally fills three rows.
After twenty draws Wren reads it. "One pennant 11 times, two pennants 5 times, three pennants 4 times."
"What do you notice?" he asks. "The bag says 1/2, 3/10, 1/5. The tally says 11/20, 1/4, 1/5."
Nova projects two sets of bars side by side. "Would you like a hint? Find the mean of the tally."
"1.65 pennants," Comet says. "The bag's expected value is 1.7. Close, not equal."
"Twenty draws wobble," Wren says. "An empirical distribution from a small tally drifts toward the theoretical one as the draws pile up."
Here is the crew's own tally from twenty draws. Yours will differ, and that is the point of a lab.
| Pennants | Tally (of 20) | Empirical probability | Theoretical probability |
|---|---|---|---|
| 1 | 11 | 11/20 | 1/2 |
| 2 | 5 | 1/4 | 3/10 |
| 3 | 4 | 1/5 | 1/5 |
The crew's empirical expected value is 1 × 11/20 + 2 × 1/4 + 3 × 1/5 = 1.65. The bag's is 1.7.
Both are means of a distribution. One comes from twenty draws, the other from the ten tokens themselves.
| What the lab shows | True or false? |
|---|---|
| 11 + 5 + 4 = 20 | ? |
| The empirical distribution comes from the tally; the theoretical one comes from the tokens in the bag. | ? |
| A tally of twenty draws must match the bag's distribution exactly. | ? |
| 1 × 11/20 + 2 × 1/4 + 3 × 1/5 = 1.65 | ? |
Careful lab work. Tomorrow expected value weighs two decisions, and a two-way table judges the worn-tire check.