ETA accuracy benchmark · 🇬🇧 Felixstowe

How Port Flow's predicted ETA compares to each vessel's own broadcast ETA at Felixstowe, measured on closed voyages over the last 30 days. We publish the number — good or bad. Lower error is better.

Port Flow predicted ETA (RMSE)

2.94 h

MAE 0.76 h · 166 voyages

Broadcast ETA (RMSE)

30.67 h

crew-declared, 140 voyages (same set)

Our advantage

90.4 %

lower error than the broadcast ETA

Broadcast ETAs more than 7 days from the actual arrival are excluded as AIS sentinel / placeholder values (24 excluded). Both figures are computed on the same remaining voyage set — apples to apples.

Closed voyages

error = predicted/broadcast minus actual arrival (hours)


230925000	container	10/12, 02:48 AM UTC	0.49 h	-45.30 h
563324800	container	10/12, 01:51 AM UTC	-1.06 h	-2.36 h
277548000	general-cargo	10/11, 09:00 PM UTC	-0.21 h	-14.85 h
245029000	container	10/11, 01:35 PM UTC	1.95 h	-62.59 h
636019364	container	10/11, 01:25 PM UTC	-0.37 h	-2.43 h
477745300	container	10/11, 11:56 AM UTC	0.08 h	-24.95 h
235054097	container	10/11, 11:08 AM UTC	0.44 h	0.86 h
235054097	container	10/11, 09:22 AM UTC	2.49 h	-496.37 h
277574000	general-cargo	10/10, 09:30 PM UTC	-0.14 h	-19.76 h
636023266	container	10/10, 10:33 AM UTC	-0.11 h	-2.06 h
266248000	container	10/10, 02:58 AM UTC	0.59 h	-11.97 h
277548000	general-cargo	10/10, 02:44 AM UTC	-0.05 h	-10.74 h
477308100	general-cargo	10/10, 12:25 AM UTC	-0.43 h	-2761.42 h
230925000	container	10/09, 05:16 PM UTC	0.80 h	-11.78 h
277574000	general-cargo	10/09, 02:16 PM UTC	-0.10 h	-12.52 h
266248000	container	10/09, 03:11 AM UTC	-0.10 h	-12.19 h
277548000	general-cargo	10/09, 02:55 AM UTC	-0.18 h	-10.92 h
220284000	container	10/08, 09:23 PM UTC	0.13 h	-11.82 h
245029000	container	10/08, 07:31 PM UTC	3.63 h	3.48 h
230925000	container	10/08, 06:16 PM UTC	1.27 h	-12.77 h
477967700	container	10/08, 04:38 PM UTC	0.12 h	-77.15 h
277574000	general-cargo	10/08, 02:10 PM UTC	-0.08 h	-12.43 h
245029000	container	10/08, 01:58 PM UTC	1.61 h	—
266248000	container	10/08, 03:01 AM UTC	-0.05 h	-12.02 h
277548000	general-cargo	10/08, 02:50 AM UTC	-0.05 h	-10.84 h
220284000	container	10/07, 08:58 PM UTC	-0.19 h	-11.39 h
230925000	container	10/07, 05:14 PM UTC	0.18 h	-11.75 h
563295600	general-cargo	10/07, 04:57 PM UTC	-0.03 h	-3082.95 h
277574000	general-cargo	10/07, 03:15 PM UTC	-0.10 h	-13.51 h
440489000	container	10/07, 10:07 AM UTC	-0.25 h	-3092.12 h
266248000	container	10/07, 03:00 AM UTC	0.20 h	-12.01 h
277548000	general-cargo	10/07, 02:47 AM UTC	-0.05 h	-10.80 h
563160200	container	10/06, 10:45 PM UTC	-0.56 h	-2.76 h
211839040	container	10/06, 09:58 PM UTC	0.98 h	-54.97 h
220284000	container	10/06, 09:05 PM UTC	-0.29 h	-11.50 h
230925000	container	10/06, 05:22 PM UTC	0.77 h	-11.87 h
277574000	general-cargo	10/06, 02:11 PM UTC	-0.12 h	-12.44 h
355216000	container	10/06, 10:49 AM UTC	-0.08 h	-3165.83 h
266248000	container	10/06, 03:14 AM UTC	0.32 h	-21.73 h
277548000	general-cargo	10/06, 02:50 AM UTC	-0.05 h	-21.67 h
220284000	container	10/05, 08:59 PM UTC	-0.07 h	-59.40 h
563308400	container	10/05, 05:28 PM UTC	-0.19 h	-1.98 h
230925000	container	10/05, 02:49 AM UTC	0.62 h	-45.33 h
277574000	general-cargo	10/04, 09:01 PM UTC	-0.04 h	-14.53 h
636025864	container	10/04, 04:36 PM UTC	-0.13 h	-2.11 h
477929200	container	10/04, 01:12 AM UTC	-0.25 h	-2363.21 h
277548000	general-cargo	10/03, 09:17 PM UTC	-0.21 h	-19.54 h
636023270	container	10/03, 03:19 PM UTC	-0.07 h	-1956.82 h
249767000	product	10/03, 10:18 AM UTC	0.52 h	-4.30 h
266248000	container	10/03, 03:01 AM UTC	0.37 h	-12.03 h

See it on a live desk

Predicted ETAs, congestion and sanctions screening across 51 ports — try it in one click, no signup.

Open dashboard See pricing →

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.