Picking Your Departure Day From the Canaries: Read the First 72 Hours

For most boats crossing from the Canaries to the Caribbean this winter, the season is already settled. The harder call is the day. Our short answer: pick the day for the first three days of the passage, not the whole crossing. Look for a window that lets you clear the islands and work south toward the trades without a low, a front or a stretch of motoring in your way. Check that several forecast models broadly agree on those first days, and be ready to wait when they don't. Beyond about five days, treat any forecast as an outlook, not a promise.

Season first, day second

Our Canary Islands to the Caribbean guide covers the season: late November into December is the classic window, when the trades are usually established and the hurricane season is winding down. That choice is made months ahead, from climate data.

The day is a different question. Inside a good season you can still leave into a blow, a week of calm, or a low that stalls north of the islands and sends a cross swell down the route. Only a forecast can help with that, and only a few days at a time.

What a good first 72 hours looks like

The first few days set the tone. The crew is not yet in a watch rhythm, the boat is at its heaviest, and seasickness is most likely. The first job is getting clear of the islands and south toward steadier trade winds.

A day worth taking usually looks like this:

• Wind from somewhere behind the beam, at a strength a tired crew handles comfortably.
• No low or front close to the route in the first days, and nothing in the models suggesting one will develop across your track.
• A sea state you would be happy to sail in for a full night, not just an afternoon.
• Enough wind to sail. A calm start means motoring and burning fuel you may want later.

Wind around the islands is often stronger and gustier than the open-sea forecast suggests, so plan the first evening with a reef in mind.

Read more than one model

A single forecast tells you what one model thinks. Several forecasts tell you how sure anyone can be. When the main global models agree on the first three days, that is a reasonable sign the picture is stable. When they disagree about a low, nobody knows yet, and the usual remedy is to wait a day and look again.

A few habits help:

• Start watching the pattern a week or more before your target date.
• Compare the same time step across models, not just the summary.
• Give the first few days the most weight, and a pretty forecast for day eight very little.
• Look at the barometer and the sky on the morning you leave. If they don't match the forecast, the forecast may be wrong.

Where PredictWind fits

You can do all of this with free models and a GRIB viewer, and every skipper should learn to. Paid tools save time by running the comparison for your route and boat.

PredictWind's Departure Planning is built for this decision. According to PredictWind, it runs your route for four departure dates and summarises what each would meet: average, maximum and minimum wind, waves, currents and rain, plus how much of the route would be upwind, reaching or downwind. The Professional plan adds boat-motion estimates for each date, covering roll, vertical acceleration and slamming. PredictWind's new AI Weather Router can then read the result and give a written view of which day looks best, including saying that none is fully recommended.

On PredictWind's pricing page on 30 October 2026, Departure Planning was included in Standard ($249 USD a year) and Professional ($499 USD a year), each with a seven-day free trial. The free plan includes a GRIB viewer and several forecast models. Prices change, so check before you subscribe.

We have not tested Departure Planning at sea, and everything above about what it does is PredictWind's own description. The output is only as good as the boat data you enter, so be pessimistic about a loaded cruising boat's speed. For where a routing briefing can mislead, see our piece on PredictWind's AI Weather Router or your own GRIBs.

Disclosure: the PredictWind link above is an affiliate link. If you subscribe through it, we may earn a commission, at no extra cost to you, and it does not change what we recommend. See our affiliate disclosure.

Write your no-go list before you look

The most useful step costs nothing. Before you open any forecast, write down the conditions you will not leave in. Include a wind strength, a sea state, how close a low may come to the route, and how many days of model agreement you want. Agree it with the crew.

That list is for the moment when the marina bill is mounting, other boats are leaving and the forecast is borderline. A limit written in calm weather is easier to keep than one argued at the dock. If conditions turn worse than forecast, our heavy weather tactics guide covers heaving to and running off.

The boat and crew are part of the window

A perfect forecast is wasted on a boat that isn't ready. Count these as weather too:

• Jobs still open on the rig, steering or autopilot.
• A crew that is exhausted from the final week of preparation.
• Weather downloads and satellite links not yet tested from the boat.

Leaving a day later, rested, into a slightly less perfect window is often the better passage.

Who should skip the paid tools

• Skippers who already compare models confidently. A subscription won't add much.
• Crews sailing with a rally that provides weather briefings. Use what is included first.
• Anyone tempted to follow a score instead of reading the weather.
• Boats without honest performance data, until they have some.

Bottom line

Choose the season from the climate and the day from the first three days of forecast. Look for model agreement, a clean start and wind you can sail in. Write your limits first, and treat waiting as part of the plan. A routing tool can speed up the comparison, but the decision to cast off stays with the skipper.

Charts, Checklists & Sea Stories — practical gear and passage notes for bluewater sailors. Free email when a new piece goes up.