ETA accuracy benchmark · 🇺🇸 Sabine Pass
How Port Flow's predicted ETA compares to each vessel's own broadcast ETA at Sabine Pass, 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)
391.00 h
MAE 107.62 h · 30 voyages
Broadcast ETA (RMSE)
Building…
only 12 comparable voyages so far (need 20)
Gap
—
benchmark builds as more voyages close
Closed voyages
error = predicted/broadcast minus actual arrival (hours)| 366273000 | product | 08/26, 11:51 PM UTC | 3.60 h | -118.86 h |
| 367561320 | product | 08/26, 10:28 PM UTC | 3.67 h | — |
| 538011663 | product | 08/26, 06:36 PM UTC | 2.77 h | -104.61 h |
| 367561320 | product | 08/26, 05:08 PM UTC | 6.43 h | — |
| 367561320 | product | 08/26, 03:30 PM UTC | 6.79 h | — |
| 538010566 | product | 08/26, 01:39 PM UTC | 0.23 h | -3.65 h |
| 247486100 | chemical | 08/26, 08:38 AM UTC | 6.49 h | -2141.64 h |
| 219490000 | product | 08/26, 01:42 AM UTC | 3.64 h | -12.70 h |
| 310652000 | product | 08/26, 01:34 AM UTC | 0.88 h | -1.55 h |
| 367561320 | product | 08/25, 07:00 PM UTC | 6.02 h | 6696.99 h |
| 538008271 | product | 08/25, 05:39 PM UTC | 7.56 h | — |
| 367561320 | product | 08/25, 01:37 AM UTC | 4.25 h | — |
| 538008271 | product | 08/25, 12:24 AM UTC | -2.40 h | — |
| 374171000 | container | 08/24, 11:25 PM UTC | 3.89 h | -24.42 h |
| 367561320 | product | 08/24, 11:01 PM UTC | 4.10 h | — |
| 367561320 | product | 08/24, 09:51 PM UTC | -1.03 h | 6718.14 h |
| 367561320 | product | 08/24, 09:34 PM UTC | 4.05 h | — |
| 538009545 | container | 08/24, 07:38 PM UTC | 0.05 h | -2.64 h |
| 368042000 | container | 08/24, 05:55 PM UTC | -0.08 h | -0.43 h |
| 367561320 | product | 08/24, 02:01 PM UTC | 2.81 h | 6725.98 h |
| 563251200 | product | 08/24, 01:24 PM UTC | -1.42 h | -2.23 h |
| 538008271 | product | 08/24, 11:06 AM UTC | 7.27 h | -44.11 h |
| 367390890 | container | 08/23, 09:06 PM UTC | -0.11 h | 5590.89 h |
| 248720000 | product | 08/23, 09:02 PM UTC | -1.65 h | -329.05 h |
| 538007137 | product | 08/23, 03:19 AM UTC | -1.19 h | -120.33 h |
| 538006761 | product | 08/22, 08:48 PM UTC | 5.08 h | -87.81 h |
| 256721000 | product | 08/22, 03:53 PM UTC | 4.25 h | -432.89 h |
| 256512000 | product | 08/22, 02:44 PM UTC | -737.49 h | -459.74 h |
| 215578000 | lng | 08/22, 02:37 PM UTC | -1962.39 h | -1995.63 h |
| 636020504 | product | 08/22, 02:36 PM UTC | -437.05 h | -395.61 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.