I’ve been logging every fuel fill for my Toyota Yaris Cross Icon HEV since I bought it in April 2023. Three years of data, reconciled against till receipts, with every anomalous figure traced to a cause rather than smoothed away. The headline: 58.8 MPG lifetime (UK imperial, brim-to-brim method). The interesting part is what it took to get to a number I trust.
The full dataset is published as an interactive page with charts. This post is about the method and what the corrections revealed.
Method: brim-to-brim
MPG is calculated brim-to-brim: miles between two full tanks divided by litres refilled, converted to imperial gallons (4.54609 L). Partial fills are rolled into the following full tank — they affect the per-fill MPG of the next segment, not the current one. This is robust to occasional missed logs because the lifetime figure only divides total distance by total fuel between the first and last full tank.
The formula sounds simple. The hard part is keeping the log clean enough that the formula means something.
The six corrections
Raw logs from a fuel app have errors — typos, misread pumps, partial fills logged as full. I traced every figure above ~75 MPG or below ~45 MPG back to a receipt or the odometer. Six corrections were needed:
| Date | What changed | Effect |
|---|---|---|
| 22/12/2023 | Date typo — was 22/12/2024 | Chronology only |
| 24/04/2024 | Date typo — was 24/02/2024 | Chronology only |
| 31/05/2024 | 6.95 L / £10 corrected to 30.42 L / £43.77 | Removed a false 97 MPG |
| 21/10/2024 | Date corrected (was 28/10/2024) | Same fill, 16,624 mi |
| 11/10/2024 | Missing 31.48 L / £41.52 fill added | Removed a false 82 MPG |
| 27/04/2024 | Partial reclassified as full (29.25 L confirmed) | Revised adjacent fill from 58.5 → 59.5 MPG |
The most instructive was 31/05/2024. The app showed 6.95 L — a genuine receipt for a £10 top-up I’d added as a partial, but it had been promoted to full in the log. That one misclassification produced a 97 MPG figure that looked like a recording error but was actually a data entry error. The receipt confirmed the correct volume; the correction dropped that segment to a plausible 53 MPG.
What the data shows
Seasonal variation is real and consistent: winter economy runs 8–12 MPG below the summer peak. The hybrid system relies heavily on regenerative braking and electric-only mode at low speeds — both are less effective when the battery is cold and when short journeys don’t allow the system to warm up. The effect is visible every year in the same months.
The Toyota official figure is 64.2 MPG (WLTP combined). At 58.8 MPG lifetime across real mixed use — including motorway runs, cold winters, and the occasional short school-run errand — that’s a reasonable real-world outcome. WLTP hybrid figures tend to overstate real economy more than petrol figures because the test cycle favours the electric-only portion disproportionately.
The spreadsheet
The underlying model is an .xlsx file with a per-fill log tab, a brim-to-brim calculation tab, a courtesy-car sheet (separating out the period the car was in for warranty work), and a summary chart. The formula work is straightforward — the value is in keeping the log clean enough that the formulas mean something.
I’ll publish the spreadsheet template alongside the interactive page once I’ve stripped the remaining identifying fields.