THE RESULT
What the data shows
The largest computed metric is 33.7 for 1980.0; the smallest is 17.1 for 1973.0. Metric: mean_mpg (Miles per US gallon).
These are sampled historical models, not sales-weighted fleet statistics.
| model year | vehicles | mean mpg | median mpg | calendar year |
|---|---|---|---|---|
| 70 | 29 | 17.69 | 16 | 1,970 |
| 71 | 28 | 21.25 | 19 | 1,971 |
| 72 | 28 | 18.71 | 18.5 | 1,972 |
| 73 | 40 | 17.1 | 16 | 1,973 |
| 74 | 27 | 22.7 | 24 | 1,974 |
| 75 | 30 | 20.27 | 19.5 | 1,975 |
| 76 | 34 | 21.57 | 21 | 1,976 |
| 77 | 28 | 23.38 | 21.75 | 1,977 |
| 78 | 36 | 24.06 | 20.7 | 1,978 |
| 79 | 29 | 25.09 | 23.9 | 1,979 |
| 80 | 29 | 33.7 | 32.7 | 1,980 |
| 81 | 29 | 30.33 | 31.6 | 1,981 |
THE METHOD
From source to answer
Model-year sample counts, means and medians.
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
Separate vehicle-composition changes from within-design improvements before explaining a trend.
Where the conclusion stops
Historical city-cycle vehicle observations are not current fleet performance. Vehicle mix and model year are confounded. A '?' horsepower value means missing, not zero.
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.