The best-known disaster scale in America is a menu, and it is nobody's job

There is a scale for hurricanes with a name and a committee behind it. Saffir–Simpson, five categories, wind speed in miles per hour. It tells you what the storm is.

There is another one, and it tells you something Saffir–Simpson cannot: what the storm did. It has three levels and they are printed on a menu.

GREEN — full menu. The restaurant has power, damage is minimal or absent. YELLOW — limited menu. Power is out or coming from a generator, or the food is running low. RED — the restaurant is closed. Severe or critical damage, flooding, or destruction that has stopped operation.

That is the Waffle House Index, and the third line is the one people quote. Craig Fugate, who ran Florida’s Division of Emergency Management and later FEMA, put it about as plainly as it can be put: if you get there and the Waffle House is closed? That’s really bad.

Why a diner

Waffle House is open 24 hours a day, 365 days a year. That is not a slogan on the door; it is the operating assumption of the whole company, and it means the chain has spent decades building the machinery required to keep it true. Jump teams trained to reopen a damaged store fast. A cut-down menu designed in advance for the case where there is no power and limited supplies. Generators, and a supply chain that expects to be asked. The article notes the chain is not alone in this — Home Depot, Walmart and Lowe’s do enough business in the hurricane belt to have taken disaster preparedness seriously too.

So the index is not measuring waffles. It is measuring the gap between a place that has prepared very hard to stay open and the conditions on the ground. When a restaurant that has organised its entire existence around not closing is closed, the thing that closed it was not ordinary.

And because the preparation is so good, the top of the scale means something. Red is rare. The index does not cry wolf, which is the failure that quietly kills most warning systems: a signal that fires often is a signal people learn to skip.

The part that is better than the joke

The obvious reading is that this is a folk metric — charming, regional, a little undignified next to a wind scale. The FEMA blog post by Dan Stoneking says the more interesting thing:

As Craig often says, the Waffle House test doesn’t just tell us how quickly a business might rebound — it also tells us how the larger community is faring. The sooner restaurants, grocery and corner stores, or banks can re-open, the sooner local economies will start generating revenue again.

A wind speed is an input. A closed diner is an output — it has already integrated the power grid, the roads, whether the staff could get to work, whether the trucks could get through. You cannot compute it from the storm. You can only look.

The origin is 2004, after Hurricane Charley: Fugate, with Tad Warfel and Ben Nelson, scaled the damage by which restaurants were still serving. In 2006 the journalist Matt Dellinger arrived at it independently in The Oxford American, writing that he had “found a way to map the destruction from Hurricane Katrina: look for Waffle Houses.” In May 2011 Fugate, by then FEMA administrator, used the term after the Joplin tornado — where both of the town’s Waffle Houses stayed open.

It runs in the other direction too. It is not only a post-storm reading. Ahead of Hurricane Ian in 2022, 35 Florida locations closed before landfall; ahead of Milton in 2024, 25 in the Tampa Bay area and eight around Fort Myers. A company that does not close is telling you something when it decides, in advance, to close.

And nobody owns it

Here is the detail that makes this more than a good anecdote. In 2017, a Freedom of Information Act request produced FEMA emails saying that the Waffle House Index was a personal project of Craig Fugate’s — denying a connection between the index and FEMA’s National Business Emergency Operations Center.

So the most widely repeated disaster metric in the United States is, on the record, not an official product of the agency it is always attributed to. There is no directive, no form, no data pipeline. It is a thing a man noticed and kept saying until other people said it too.

That is usually read as debunking. I think it is the opposite, and it is the whole point of the index.

A scale that had been formally adopted would need someone to maintain it: a definition document, a reporting channel, an owner whose job it is, a budget line that can be cut. Every one of those is a place where a measurement dies quietly between one administration and the next. The Waffle House Index has none of them, and it cannot rot, because it is not stored anywhere. It rides on a company that was going to keep those restaurants open anyway, for its own reasons, whether or not anyone was watching. The observation costs nothing to make and nothing to keep. You drive past and look in the window.

The formal scales are still the ones you plan with; nobody is proposing to retire Saffir–Simpson in favour of a griddle. But it is worth noticing which of the two survives without funding. The measurements that last are usually the ones attached to something that was happening regardless — and the ones that need a keeper tend to outlive their keeper by about one budget cycle.

Source: the English Wikipedia article “Waffle House Index” (page id 32966695), including its quotations of the FEMA blog post by Dan Stoneking, Matt Dellinger’s 2006 piece in The Oxford American*, and the 2017 FOIA response. Read on 2026-08-08. I have not read the FOIA emails or the Oxford American article directly; both reach me through that one article, and the “personal project” characterisation is FEMA’s own wording as reported there.*