Last Friday, someone was telling me that Detroit had an advantage over San Jose because they'd had more time off leading into their second round matchup (which didn't appear to be true). We always hear that a team with more rest prior to a series has an advantage and we'd expect the numbers to bear that out, no doubt. So here's the expected and actual winning percentage in the first game of 2nd round, 3rd round and Cup Finals since 1979-80 for varying amounts of rest:
| Days Off | Series | Expected | Actual | Delta |
| 2 | 69 | 0.485 | 0.449 | -0.036 |
| 3 | 42 | 0.473 | 0.500 | +0.027 |
| 4 | 90 | 0.494 | 0.467 | -0.027 |
| 5 | 81 | 0.514 | 0.556 | +0.041 |
| 6 | 53 | 0.500 | 0.509 | +0.009 |
| 7 | 37 | 0.516 | 0.541 | +0.024 |
| 8 | 29 | 0.520 | 0.448 | -0.072 |
| 9 | 17 | 0.529 | 0.588 | +0.059 |
Obviously, teams never have to play back-to-back games – 2 days off is just slightly less than 48 hours rest. If there's a trend in this data, it's not clear to me, aside from weaker teams getting less rest on average than stronger teams.
What if we look at the difference in the number of days of rest?
| Days Diff | Series | Expected | Actual | Delta |
| 0 | 66 | 0.500 | 0.500 | 0.000 |
| 1 | 37 | 0.490 | 0.568 | +0.078 |
| 2 | 52 | 0.525 | 0.538 | +0.013 |
| 3 | 36 | 0.515 | 0.583 | +0.068 |
| 4 | 31 | 0.524 | 0.581 | +0.057 |
| 5 | 9 | 0.545 | 0.333 | -0.211 |
| 6 | 10 | 0.525 | 0.500 | -0.025 |
| 7 | 2 | 0.500 | 0.500 | 0.000 |
This seems like a trend – teams that get more rest than their opponents prior to a series are not only better than their opponents but also outperform their expected winning percentage in the first game of the series. As for the claim that there's such a thing as too much rest, we don't have enough data to draw a conclusions for 5-, 6- and 7-game deltas.
