ETA accuracy benchmark · 🇪🇸 Algeciras
How Port Flow's predicted ETA compares to each vessel's own broadcast ETA at Algeciras, measured on closed voyages over the last 7 days. We publish the number — good or bad. Lower error is better.
Port Flow predicted ETA (RMSE)
226.87 h
MAE 85.82 h · 31 voyages
Broadcast ETA (RMSE)
Building…
only 8 comparable voyages so far (need 20)
Gap
—
benchmark builds as more voyages close
Closed voyages
error = predicted/broadcast minus actual arrival (hours)| 210707000 | container | 10/11, 11:51 PM UTC | 0.05 h | -72.86 h |
| 211886810 | general-cargo | 10/11, 04:29 PM UTC | -56.80 h | -2508.49 h |
| 224879000 | container | 10/11, 04:18 PM UTC | -22.23 h | -73.30 h |
| 256370000 | product | 10/11, 04:15 PM UTC | -32.58 h | -201.25 h |
| 413055000 | general-cargo | 10/11, 12:50 PM UTC | -2.92 h | -259.85 h |
| 563289400 | general-cargo | 10/11, 02:00 AM UTC | -443.65 h | -412.01 h |
| 255915724 | product | 10/11, 01:47 AM UTC | -55.15 h | — |
| 219734000 | container | 10/10, 01:55 PM UTC | -10.22 h | — |
| 277562000 | container | 10/10, 01:28 PM UTC | -42.69 h | -269.22 h |
| 209613000 | container | 10/10, 11:16 AM UTC | -8.36 h | -342.78 h |
| 305176000 | container | 10/09, 11:37 AM UTC | -326.56 h | -327.38 h |
| 229699000 | container | 10/09, 04:18 AM UTC | -175.75 h | -2427.31 h |
| 538010773 | container | 10/09, 02:00 AM UTC | -12.41 h | -37.01 h |
| 563068400 | product | 10/09, 12:24 AM UTC | -4.68 h | -585.41 h |
| 566158000 | container | 10/08, 04:34 PM UTC | -7.22 h | -1827.08 h |
| 563141600 | container | 10/08, 04:12 PM UTC | -22.73 h | -455.21 h |
| 229706000 | container | 10/08, 03:33 PM UTC | -1.97 h | -649.55 h |
| 232043372 | container | 10/08, 02:08 PM UTC | -2.92 h | -3.65 h |
| 563168300 | container | 10/08, 12:48 PM UTC | -261.89 h | -275.81 h |
| 210707000 | general-cargo | 10/08, 12:34 PM UTC | -3.11 h | -141.57 h |
| 256744000 | container | 10/08, 07:44 AM UTC | -11.23 h | -783.74 h |
| 224879000 | container | 10/08, 12:26 AM UTC | -2.63 h | -153.45 h |
| 563377000 | general-cargo | 10/07, 11:11 PM UTC | -11.29 h | -438.19 h |
| 563193300 | container | 10/07, 02:07 PM UTC | -1086.75 h | -2323.13 h |
| 477665200 | container | 10/07, 07:53 AM UTC | 5.91 h | -394.89 h |
| 255806016 | container | 10/06, 12:11 AM UTC | -10.31 h | -148.20 h |
| 219508000 | container | 10/05, 11:56 PM UTC | -7.21 h | -1932.95 h |
| 563058900 | container | 10/05, 01:35 PM UTC | -23.37 h | -27.59 h |
| 636023006 | product | 10/05, 01:07 PM UTC | 3.74 h | — |
| 215335000 | product | 10/05, 09:44 AM UTC | -1.92 h | — |
| 636093143 | container | 10/05, 08:10 AM UTC | -2.30 h | -581.18 h |
See it on a live desk
Predicted ETAs, congestion and sanctions screening across 51 ports — try it in one click, no signup.
How it's measured
- Predicted ETA: distance-to-berth over speed-made-good, refreshed on every AIS position, with a seasonal correction from rolling medians.
- Broadcast ETA: the crew-entered ETA in the vessel's AIS message — the de-facto industry default.
- Both are compared to the actual arrival timestamp on the same closed voyages. We report RMSE (penalises large misses) and MAE (typical error).
- Broadcast ETAs more than 7 days off are dropped as AIS sentinel / placeholder values, and our model is scored on that same set — so the comparison is honest, not cherry-picked.
- Coverage is public AIS only; ports with weak coverage show fewer closed voyages. We'd rather show a small honest sample than a big misleading one.