THE RESULT
What the data shows
The largest computed metric is 0.958 for 1 / hour 7; the smallest is 0.6544 for 0 / hour 17. Metric: registered_share (Registered share).
Workingday 1 excludes weekends and source holidays. A registered rental is not necessarily a commute.
| workingday | hr | registered | casual | total | observed hours | registered share | group |
|---|---|---|---|---|---|---|---|
| 0 | 7 | 8,039 | 1,988 | 10,027 | 231 | 0.8017 | 0 / hour 7 |
| 0 | 8 | 19,688 | 4,718 | 24,406 | 231 | 0.8067 | 0 / hour 8 |
| 0 | 9 | 29,246 | 10,399 | 39,645 | 231 | 0.7377 | 0 / hour 9 |
| 0 | 12 | 55,443 | 29,163 | 84,606 | 231 | 0.6553 | 0 / hour 12 |
| 0 | 17 | 48,911 | 25,829 | 74,740 | 231 | 0.6544 | 0 / hour 17 |
| 0 | 18 | 45,272 | 19,371 | 64,643 | 230 | 0.7003 | 0 / hour 18 |
| 1 | 7 | 138,095 | 6,049 | 144,144 | 496 | 0.958 | 1 / hour 7 |
| 1 | 8 | 225,552 | 11,043 | 236,595 | 496 | 0.9533 | 1 / hour 8 |
| 1 | 9 | 107,734 | 12,059 | 119,793 | 496 | 0.8993 | 1 / hour 9 |
| 1 | 12 | 79,253 | 20,555 | 99,808 | 497 | 0.7941 | 1 / hour 12 |
| 1 | 17 | 233,729 | 28,391 | 262,120 | 499 | 0.8917 | 1 / hour 17 |
| 1 | 18 | 220,004 | 25,125 | 245,129 | 498 | 0.8975 | 1 / hour 18 |
THE METHOD
From source to answer
Aggregate rider counts before calculating shares at six declared hours.
The Python source exposes this study’s transformations. The complete project download includes shared preparation and evaluation routines.
THE NEXT DECISION
What follows from the finding
Use rider mix to frame service-planning questions without inferring individual traveler intent.
Where the conclusion stops
Historical system rentals measure realized use, not unmet demand or station inventory. Missing hours are unknown. Weather associations do not establish causal effects.
Related studies may reuse observations or holdouts. These are historical analyses; associations and backtests do not demonstrate commercial impact. Further model tuning needs new, untouched evaluation data.