The Cone of Uncertainty, Explained

The "cone of uncertainty" is the most recognized, and the most misread, graphic in all of weather. Every hurricane season it spreads across news broadcasts and phone screens, a tapering white funnel sweeping toward the coast. People look at it and ask the wrong question: Am I in the cone? The cone was never built to answer that. It shows the probable path of one thing (the storm's center), and it says nothing about how big the storm is or where its dangers will reach.¹ Misunderstanding that distinction has put people on the wrong side of preparation decisions for years. This article explains exactly what the cone represents, what it deliberately leaves out, how the National Hurricane Center builds it, and how to read it without being fooled.

What the Cone Actually Represents

The official definition is precise and worth quoting: "The cone represents the probable track of the center of a tropical cyclone, and is formed by enclosing the area swept out by a set of circles (not shown) along the forecast track."¹ Those circles sit at the forecast positions for 12, 24, 36, 48, 60, 72, 96, and 120 hours out.¹ Wrap a smooth boundary around all of them and you get the familiar funnel that widens with time, wider toward the end because forecasts five days out are less certain than forecasts twelve hours out.

The key word is center. The cone is a track-of-the-center forecast. It is a statement about where the eye, the geometric heart of the storm, is likely to go. It is not a map of the storm's reach, its wind field, or its hazards.

The Two-Thirds Rule

Here is the part most people never learn. Each circle along the track is not arbitrary; it is sized from history. "The size of each circle is set so that two-thirds of historical official forecast errors over a 5-year sample fall within the circle."¹ In other words, the NHC looked at how far off its own track forecasts have actually been over the past five years and drew the circle big enough to contain about two-thirds (roughly 67%) of those misses.

The immediate, sobering consequence: if the center stays inside the cone about two-thirds of the time, then about one-third of the time the center tracks outside the cone entirely. The cone is a likelihood, not a fence. A storm that drifts outside its own cone is not behaving strangely; it is doing exactly what one storm in three does.

How Wide Is the Cone? The Actual Numbers

The cone's width comes from a published table of circle radii, one radius per forecast hour, recomputed every year. The 2026 Atlantic cone is built from these forecast-error radii:¹

• 12 hours: 25 nautical miles (29 mi; 46 km)
• 24 hours: 39 nautical miles (45 mi; 72 km)
• 36 hours: 49 nautical miles (56 mi; 91 km)
• 48 hours: 62 nautical miles (71 mi; 115 km)
• 60 hours: 77 nautical miles (89 mi; 143 km)
• 72 hours: 95 nautical miles (109 mi; 176 km)
• 96 hours: 134 nautical miles (154 mi; 248 km)
• 120 hours: 200 nautical miles (230 mi; 370 km)

By day five, the circle has a radius of roughly 230 mi (370 km), meaning the center could plausibly be anywhere across a span of more than 450 mi (740 km) of coastline, and still be a "good" forecast. The Eastern and Central North Pacific cones use their own, somewhat smaller table; for 2026 the day-five radius there is 138 nautical miles (159 mi; 256 km), reflecting that basin's distinct forecast-error history.¹ These are the reasons a cone five days out is so wide that it feels almost useless for pinpointing a landfall, because, honestly, no one can pinpoint a landfall five days out. In our own work that's exactly how we treat the day-five cone: as a signal for who should start paying attention, never as a read on who will be hit. The narrow end near the present is the part you can lean on; the wide end is a planning aid, not a landfall prediction.

The Misconceptions That Get People Hurt

It Is Not the Size of the Storm

The cone's width is a measure of forecast uncertainty, not storm size. A tiny, compact hurricane and a sprawling giant produce cones of identical width, because the cone is drawn from the NHC's track-error statistics, not from the storm in front of you. Two storms with the same forecast track get the same cone regardless of how far their winds extend. Reading the cone's edge as the edge of the storm is a category error.

Impacts Routinely Land Outside the Cone

This is the misconception that matters most. The NHC states plainly that the cone "does not provide any information on potential impacts from storm surge, wind, flooding, rainfall, or tornadoes."² Users, it notes, have often misinterpreted the cone as the area of potential impacts, and it is not.²

Hazards do not respect the cone's boundary:

• Wind — hurricane-force and tropical-storm-force winds can extend well beyond the cone. The cone traces only the center; the damaging wind field can be hundreds of miles across.
• Storm surge — surge is driven by the storm's wind field pushing water onto the coast, and it can inundate areas far from where the center comes ashore. Surge is consistently the deadliest hurricane hazard, and it is precisely the hazard the cone says nothing about. (See our full explainer on storm surge mechanics.)
• Rainfall and inland flooding — a storm's heaviest rain can fall hundreds of miles from the center and persist long after the center has passed, flooding communities the cone never crossed.
• Tornadoes — tropical systems spin up tornadoes in their outer rainbands, frequently to the right of the track and outside the cone.

Living just outside the cone has never meant being safe. The cone simply does not carry that information.

The Edge Is Not Safer Than the Center Line

Because the cone is a swept area of equal-probability circles, a point at the cone's outer edge is not meaningfully less likely to see the center than a point near the middle of the funnel. The cone has no "bullseye." Treating the center line as the forecast track and the edges as long shots misreads the geometry; the whole enclosed area represents the probable path of the center.

How the NHC Builds the Cone Each Year

The cone is rebuilt before every season from the most recent five years of verification data. The 2026 cones "are based on error statistics from 2021-2025."¹ Each year the oldest year drops off and the newest is added, so the cone is a rolling five-year memory of how accurate the forecasts have actually been.

This is why the cone shrinks over time. As the forecasting system improves (better satellites, more aircraft reconnaissance, sharper numerical models), the average track error falls, the circles get smaller, and the cone narrows. A five-day cone today is tighter than the five-day cones of a decade ago. The funnel you see is a statistical fingerprint of recent forecast skill, not a measurement of the particular storm on the map.

What's Changing in 2026

The cone graphic itself is evolving. For the 2026 season the NHC's standard cone now incorporates inland tropical-storm and hurricane watches and warnings for the continental U.S., Hawaii, Puerto Rico, and the U.S. Virgin Islands directly on the cone graphic, a deliberate push to pull readers' eyes away from the center line and toward the areas under threat.² The NHC is also running an experimental cone that uses ellipses anchored at each forecast point and is drawn to capture 90% of forecast track possibilities, rather than the traditional roughly two-thirds.² Both changes share one goal: getting people to stop reading the cone as a damage map.

How to Use the Cone Safely

A few rules turn the cone from a misleading picture into a useful one:

1. Read it as the path of the center, not the storm. The funnel shows where the eye might go. Your hazards come from the storm's reach, which the cone does not draw.
2. Never conclude you're safe because you're outside the cone. About one storm in three sends its center outside its own cone, and impacts spill outside it routinely even when the center stays in.
3. Go to the hazard-specific products. For your actual risk, read the storm-surge watches and warnings, wind-arrival and wind-speed forecasts, rainfall and flash-flood outlooks, and tornado outlooks — not the cone.
4. Don't fixate on the center line. There is no bullseye. The whole cone is the probable track of the center.
5. Listen to local officials. Evacuation and shelter decisions are made by emergency managers using the full hazard picture, not by which side of a white line your house falls on.

The cone is a genuinely elegant summary of one hard problem: where will the center go? It answers that honestly, uncertainty and all. The danger comes only when we ask it to answer a different question, where will I be hurt?, that it was never designed to address. Read it for what it is, then look past it for everything it leaves out.

Frequently Asked Questions