ETA accuracy benchmark · 🇺🇸 New York / Newark

How Port Flow's predicted ETA compares to each vessel's own broadcast ETA at New York / Newark, 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)

640.58 h

MAE 203.33 h · 50 voyages

Broadcast ETA (RMSE)

Building…

only 17 comparable voyages so far (need 20)

Gap

—

benchmark builds as more voyages close

Closed voyages

error = predicted/broadcast minus actual arrival (hours)


477886800	container	08/26, 11:54 PM UTC	2.10 h	-11.91 h
354972000	container	08/26, 10:12 PM UTC	-0.16 h	-4.22 h
366999412	container	08/26, 06:17 PM UTC	0.02 h	—
636025463	container	08/26, 04:00 PM UTC	0.27 h	-2.00 h
636019061	container	08/26, 03:15 PM UTC	3.07 h	-3.26 h
636024506	container	08/26, 12:08 PM UTC	3.25 h	-4.15 h
563293200	container	08/26, 10:56 AM UTC	0.34 h	-2.94 h
636092783	container	08/26, 10:08 AM UTC	1.12 h	-2114.64 h
563277100	container	08/26, 09:21 AM UTC	0.80 h	-1297.37 h
636093178	general-cargo	08/26, 07:35 AM UTC	2.27 h	102.41 h
538009817	container	08/26, 02:04 AM UTC	0.26 h	-1540.07 h
563006800	container	08/25, 09:12 PM UTC	1.05 h	-1387.70 h
566956000	container	08/25, 07:10 PM UTC	-0.47 h	-1105.18 h
366999412	container	08/25, 06:54 PM UTC	-0.10 h	—
366999412	container	08/25, 06:03 PM UTC	-0.95 h	—
636026193	product	08/25, 04:42 PM UTC	-0.16 h	-2.70 h
338403000	container	08/25, 10:12 AM UTC	-1.54 h	-1779.20 h
538005625	container	08/25, 09:13 AM UTC	4.62 h	-1138.22 h
367686920	product	08/25, 08:48 AM UTC	1.24 h	—
636023225	container	08/25, 12:30 AM UTC	-0.27 h	-2.51 h
636015771	container	08/24, 11:08 PM UTC	-0.22 h	-3.15 h
636093178	general-cargo	08/24, 06:46 PM UTC	2.08 h	-2041.77 h
366999412	container	08/24, 05:44 PM UTC	0.60 h	—
366999412	container	08/24, 05:18 PM UTC	0.14 h	—
366999412	container	08/24, 04:12 PM UTC	1.02 h	—
366999412	container	08/24, 11:20 AM UTC	0.28 h	—
636015445	container	08/24, 11:14 AM UTC	-0.04 h	-102.24 h
367686920	product	08/24, 11:06 AM UTC	-2.04 h	—
563292600	container	08/24, 10:56 AM UTC	0.16 h	-2162.93 h
636020494	container	08/24, 09:05 AM UTC	0.42 h	-3.60 h
218566000	container	08/23, 08:33 PM UTC	0.85 h	-2126.06 h
366999412	container	08/23, 06:58 PM UTC	-0.26 h	—
538006871	container	08/23, 06:27 PM UTC	2.26 h	-110.45 h
366999412	container	08/23, 06:14 PM UTC	-0.36 h	—
636021529	container	08/23, 04:53 PM UTC	0.62 h	-1700.89 h
338418000	container	08/23, 03:56 PM UTC	0.98 h	-3.94 h
229557000	general-cargo	08/23, 02:54 PM UTC	-2.73 h	-1159.90 h
366999412	container	08/23, 11:03 AM UTC	-15.51 h	—
636015819	container	08/23, 10:00 AM UTC	3.22 h	-3.00 h
636025251	container	08/23, 09:35 AM UTC	0.29 h	-3.60 h
563051200	container	08/23, 08:40 AM UTC	1.61 h	-1040.66 h
338078000	container	08/23, 03:29 AM UTC	-4.51 h	-1628.49 h
229949000	general-cargo	08/22, 07:59 PM UTC	0.32 h	-779.50 h
366999412	container	08/22, 05:48 PM UTC	0.48 h	—
563219100	container	08/22, 05:24 PM UTC	-2193.27 h	-126.16 h
255806358	container	08/22, 04:28 PM UTC	-664.65 h	-428.47 h
565382000	container	08/22, 02:54 PM UTC	-1062.39 h	-1040.91 h
636021013	general-cargo	08/22, 02:53 PM UTC	-1110.64 h	-1089.89 h
431654000	container	08/22, 02:40 PM UTC	-2691.11 h	-871.18 h
563217600	container	08/22, 02:36 PM UTC	-2379.20 h	-1045.61 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.