**9+3 landed closest.** Its revenue forecast missed the full year by $3K, or 0.02%, with a MAPE (average monthly miss) of 0.63%. It also had the lowest MAPE on operating expenses, at 0.30%.

**Revenue ran consistently high.** Every version forecast too high, and the bias shrinks as months close:

| Version | FY variance | Bias |
| --- | --- | --- |
| Budget | $362K, or 3.2% | 3.2% |
| 3+9 | $177K, or 1.6% | 2.1% |
| 6+6 | $158K, or 1.4% | 2.8% |
| 9+3 | $3K, or 0.02% | 0.10% |

- **Largest misses:** Budget missed July by $132K, or 15.0%. 6+6 missed September by $151K, or 15.0%. The data doesn't show what either forecast assumed for those months.
- **Why 9+3 looks so good:** it had only three months left to forecast, so part of its accuracy is the short horizon, not better forecasting.
- **Operating expenses:** no consistent direction. Budget's full-year variance was only ($649), but its MAPE was 4.1%, so large monthly misses cancelled out. Its worst month was May, at $32K over.

Both files are in `output/`: `accuracy.json` has the script figures and `accuracy-note.md` is the write-up with full tables.