The Predictive Power of Waffles

wafflae

When breakfast food takes on hurricanes, who wins?

For another interesting take on the Waffle House Index, see this article the Fivethirtyeight blog, which they posted December 6, 2016.

Transcript:

Curtis: “I love waffles. I fill up each of the little squares with the precise amount of syrup so that each bite is a perfect distribution of syrupy goodness.”

Nathan: “I love owl-shaped waffles.”

Tiffany: “The kind you get at a hotel when they serve you those free breakfasts—they’re just perfect.”

Lily: “I love waffles with strawberries.”

Vince: “Liège waffles—Belgian waffles were pale in comparison. They’re sugar clumps in the shape of pearls, and they put this in the batter, and it doesn’t dissolve out, and they taste really good. I didn’t even need to add syrup.”

Ginette: “I’m Ginette, and I’m Curtis, and you are listening to Data Crunch, a podcast about how data and prediction shape our world. A Vault Analytics production.”

Curtis: “Today we’re talking about hurricanes, waffles, and predictions.”

Ginette: “It happened in 2004. Charley, Frances, Ivan, and Jeanne were four aggressors. With the group’s combined strength, they wrecked their victims. First, Charley attacked and was the most destructive. Frances followed quickly behind with a much weaker pummel, but, being so quick on the heels of Charley, the attack was effective. Then came Ivan with an unexpected one-two punch. And finally, Jeanne forcefully hit the same spot as Frances—but with much more intensity.

“To some, this wrecking ball of an attack is known as the Year of the Four Hurricanes. These four hurricanes ruthlessly shredded Florida’s east coast, west coast, panhandle, and interior in about six weeks, leaving $29 to $41 billion in damages. As a point of comparison, if Google had to cover these costs, it would take two to three years of the organization’s net income. Next to Hurricane Andrew, (the most destructive hurricane in US history at the time)—Charley claimed second-place that year.

“Charley obliterated mobile homes, savaged houses, knocked over water towers, caused the collapse of carports, obstructed roads by littering them with large trees and power poles, blew over semi-trucks, crushed large trailers, and rendered areas unrecognizable.

“We spoke with a couple that experienced a hurricane first hand, and their ordeal sounds harrowing.”

Melody Metts: “I don’t think we expected anything that we found when we came back. You couldn’t even recognize where you were.”

Ginette: “Christopher and Melody Metts lived within twenty miles of Homestead, Florida, where Hurricane Andrew hit with full fury.”

Christopher Metts: “There was nothing taller than the first floor. Any tree, any light pole, any anything that might have been higher than the first floor of a house was completely gone. Anything that would indicate where you were—a street sign, a light—it was all gone as far as you could see.”

Ginette: “Like most south Florida residents, they didn’t think much of the storm predictions.”

Christopher: “We saw it, and the predictions for it for many days.

“Because we were in south Florida and because every hurricane season that comes along has scares that could be very devastating but it’s a near miss or it turns at the last minute, you get into a pattern of they cry wolf too often and you’re lulled into a sense of ‘well not this time.’”

Ginette: “While this was their initial feeling, eventually the predictions became serious enough that the authorities issued an evacuation order, so the Metts prepped their house for wind damage and drove to Orlando with seven children in tow, ages one to eight, and it’s a good thing they did because their family would have been in extreme danger otherwise. This is where we start to see the power of prediction in people’s lives. Imagine if there had been little to no ability to predict the hurricane.”

Curtis: “Before modern hurricane prediction, people were generally caught off guard by hurricanes, and if you can imagine before we had all this technology what people might have used to predict that any kind of storm or hurricane was coming—maybe they watched the sky for clues or observed animal behavior—like birds when they fly low and stay silent or cows when they start clustering together—or there may have been some fisherman coming in to shore that had some stories about a storm coming or dark skies coming. Observations would have been really sparse, and it would have been really hard to prepare for a hurricane event.

“All of that started to change with the invention of the telegraph in the 1800s because now we have the ability to communicate over great distances very quickly. Since then technology’s continued to develop, so we’ve gotten better and better at predicting when hurricanes will hit and how hard that they will hit an area. So in the 1940s, you had airplane reconnaissance that could go and look at what was going; in the 50s, you had radar; in the 60s, you started to have satellites that could see weather patterns; and in the 70s, you had ocean surface measurement by buoy all of these expanded the days of warning that we had to three whole days. With modern computer modeling, that forecast expanded to a five-day forecast warning.

“There’s a lot of work that goes into forecasting where hurricanes will hit and how hard they’ll hit. One example of that is the Saffir-Simpson Hurricane Wind Scale which categorize hurricanes on a scale from one to five, and that’s to help us understand how fast the winds are, which is a predictor of how much damage that the hurricane could potentially do to know how serious the storm is.

“There are over thirty working models that meteorologists use in conjunction with other models to average the potential path. And some of these dynamic models can take hours to run on the world’s fastest supercomputers because they involve solving complex equations governing atmospheric behavior all over the globe. Other statistical models take only a few minutes to spit out a prediction. Obviously one is more powerful than the other, but there are a lot of ways to predict the weather nowadays.

“While the accuracy of these models becomes better and better, predictions can still be hundreds of miles off. But if we didn’t have that predictive power or the data that fuels it, we’d have no idea when storms would hit, how hard it would hit, and where it would hit.

“Conditions are favorable for the development of severe weather. Stay tuned to NOAA weather radio.”

Ginette: “Even with our ability to prepare, hurricanes are still devastating, and we have to deal with the aftermath. When the Metts returned home, they found a severely damaged structure.”

Christopher: “Nobody had windows or doors or roofs.”

Melody: “There was quite a bit of water damage to what was left in the house.”

Ginette: “They wouldn’t have electricity for five months, and since this was before the common cellphone era, everyone was disconnected.”

Christopher: “Nobody had phone service. Nobody had electricity. There was no way to dial 9-1-1. It became also then a matter of personal safety and security as well. They immediately went to a martial-law type of a situation. They didn’t let people come in without a reason for being there.”

Ginette: “People in the neighborhood couldn’t hide their stress.”

Christopher: “In the evenings, I think the stress would begin to find its way to families, and because you didn’t have the ability to close off the neighborhood, you’d hear raised voices and stress playing out around the neighborhood.”

Ginette: “People were basically stuck where they were until major repair efforts were underway.”

Christopher: “They were finally able to clear the roadways so people could drive out, but if they did have popped tires, they would have nowhere to go because the repair stations and tire stores were all destroyed as well.”

Ginette: “In a disaster like this, how quickly public and private organizations respond can be a question of life or death for victims who grapple with the aftermath.”

Christopher: “The army finally came in about two weeks later and set up food tents. They’d run a series of meals, and the neighborhood would come over and they would line up and go through and get grits and eggs from dried powder, and it was a welcome relief.”

Melody: “Yeah, really wonderful to have.”

Ginette: “FEMA, a government organization commissioned to help after large disasters, has to figure out how to prioritize assistance when a natural disaster has done vast amounts of damage. Since information is hard to come by in a disaster situation, this is a difficult problem to solve, and one that FEMA has to grapple with.

“In the year of the four hurricanes that we referenced earlier, Craig Fugate, then the director of the Florida Emergency Management Division and now the current FEMA administrator, noted a pattern—when the Waffle House was closed, he had a real emergency on his hands. Craig moved to FEMA in 2009 to be the Administrator, and he brought this concept with him, and this is the genesis of the famed Waffle House Index.”

Philip Strouse: “He said, ‘Hey, if there’s a Waffle House that’s closed, that’s where we know we need to go right to work. That’s ground zero’ if you will.”

Ginette: “This is Philip Strouse talking about Craig Fugate’s instructions to FEMA workers after the Joplin tornado. He works for FEMA’s region IV, which covers the southeast.”

Philip: “The Waffle House has a majority of its restaurants in the southeast, so when Director Fugate was going through his hurricane seasons and noticing the Waffle Houses, which are opened year round 24/7, if they were closed after a major event, then something was terribly wrong.”

Ginette: “The Waffle House Index is a creative lagging indicator of how hard hit an area is. It’s not highly technical, but it’s very actionable—the only thing that is required for a useful metric. When information about an area’s status prior to and after a storm is difficult to access, the Waffle House Index provides enough information to get know where to go to work.”

Philip: “He came up with this colloquialism, “hey, there’s a Waffle House Index: red, yellow, and green. If all the restaurants are open, and they have a full menu, then we consider that green—everything’s getting back to normal. If they’re yellow, it means restaurants are serving a limited menu, indicating there’s no power or only generator power or maybe food supplies are low. So that gave him a feel for what supply lines or other things might be doing in an impacted area. And then of course red would be the restaurants are closed, indicating either severe damage to that facility or right there in the surrounding area.”

Ginette: “While the Waffle House isn’t the only business on this index—it also includes Walmart, Home Depot, and Lowes—it’s the only one that bases its entire business strategy on being open 24 hours a day seven days a week, so it only closes in the most dire situations.”

Philip: “You put all these different pieces of the puzzle together and you get a better picture from the private sector and the public sector.”

Ginette: “There are some limitations to using the Waffle House as an indicator: the Waffle House isn’t all over the United States, so when storms hit New England or anywhere else that doesn’t have them, it’s not a useful tool. Also, this indicator supplements a much larger picture FEMA creates through other sources. But with all its limitations, what it does show us is that for certain situations, simple predictors can be the best predictors.”

Ginette: “If you know anything about Waffle House’s emergency procedures, you understand why Craig Fugate uses this quick and easy predictive tool.”

Pat Warner: “We started after Hugo putting together checklists and working on preparation mainly for hurricanes but we’ve evolved it for tornados and ice storms and whatever else may come up.”

Ginette: “This is Pat Warner, Waffle House’s Vice President of Culture.”

Pat: “It’s hard to plan for every little thing that’s going to happen in a storm. We get to the storm, and we decide what do we need to do to get the restaurants open. Since we’ve opened, we’ve been 24 hours a day. So we never close. We actually have to have a checklist on how do we shut the restaurants down because we don’t know how to do that. We, we, we don’t shut restaurants down! We’re open 24 hours a day. And really the philosophy that our founder Joe Rogers Senior has, if you’re going to be of service to the community, you need to be there for them 24 hours a day.”

Ginette: “What waffle house does to fulfill this mantra is impressive.”

Pat: “We start tracking hurricanes here in the office when they become a tropical depression.”

Ginette: “This is often when it’s off the coast of Africa.”

Pat: “We actually have all of our restaurants’ coordinates plugged into this software, so we can see not only the impact in the area, but down to our restaurant levels. We can look at a certain restaurant and see when the tropical storm winds will hit, when will the hurricane winds hit, what is the expected storm surge and flooding for that area, what is the expected damage in that area.”

Ginette: “This software is called HURRTRAK.”

Pat: “That helps us where we need to stage our resources before the storm.”

Ginette: “There are a lot of moving parts in this process.”

Pat: “We bring in generators, and if we bring in generators, you have to have fuel, and we’re going to have to have extra food shipped in because we’re going to be very busy. Whatever distribution center that is closest to where we think it’s going to hit, we start working with them. We get some restaurants some extra food before the storm. We start organizing jump teams from other markets that can come in to help the local operators, so now we’ve got to put them up some place.”

Ginette: ”For example, after Hurricane Matthew, they had 250 restaurant operators from around the nation come in and help.”

Pat: “We throw chaos at chaos. We get down in the middle of it, and then we figure out what we need to do to get the restaurants open.”

Ginette: “Because this is such a large undertaking, they bring in a team from the corporate office to direct the operations.”

Pat: “Our senior management’s in the field. Our CEO will be on-site. Our chairman will be there. Our executive vice presidents, our senior vice presidents, our area vice presidents, they don’t work out of Norcross [Georgia]. They work in whatever market we have the restaurants. We have a food safety expert. We bring a couple IT folks because if you shut the restaurants down you got to turn the computers off and getting them back on is always an ordeal, and getting back online. We bring in some other resources to help out with security and government relations.”

Ginette: “As an aside, one thing that impressed us was what they do for their employees before, during, and after storms. Besides making sure their employees and their employees’ families are safe during storms, helping them evacuate to safe areas, and providing housing for them when needed, they prep them before storms to know what they need in their home emergency kits, and after storms, they help them with their insurance claims and filling out FEMA grants if they qualify. If a storm like Hurricane Katrina is dire, they also conduct clothing drives and send clothing to help their employees get through the disaster.

“In addition to taking care of their employees, they go out of their way to help people during devastating times.”

Pat: “If somebody comes up, and they need to eat, we’re going to feed ‘em. I’ve been to some restaurants where they didn’t have any money, they had a check, so I was like ‘write me a check.’ Or just ‘you know what, come back in a few weeks and eat with us again.’ We’re going to do what’s right. We’re not going to turn anybody away.”

Ginette: “I asked him about the business sense of the decision to open so quickly after a major storm because of the large costs associated with it, and I appreciated the perspective Waffle House espouses.”

Pat: “To be quite honest, in the short term it’s not a very good business decision to do this. It makes more sense, and unfortunately a lot of companies do this, is to let the storm go through and wait weeks, months, or even after, like Katrina, years before you come back. From a business standpoint, that’s probably more in line with good business sense, but from the people’s sense, that stinks, and we’re going to err on side of the people’s side.

“People in Charleston still remember that we were the first place they had a hot meal. I can go back to the Gulf Coast, and Katrina’s over ten years ago, and people remember that Waffle House was the first place I got a hot meal. I was able to charge my cell phone. I was able to get air conditioning. The tying to the community after everybody rallies after a major event like that, that’s worth it. For whatever we put in the short-term loss of getting the restaurants open and getting people down there, and having all the resources there, whatever that costs in the short-term, the long-term way outweighs it.”

Ginette: “He explained that one part of this is the nature of private business.”

Pat: “We are very fortunate that we are a private company, so we don’t have to answer to Wall Street. Unfortunately a lot of businesses aren’t that way, because they have to meet their quarterly statements, and we can look more long-term than short-term.”

Ginette: “Kudos to the Waffle House for using their resources to help people who are trying to recover from catastrophic effects of natural disasters. Next time you hear about a hurricane in the weather forecast, don’t forget about the predictive power of waffles.

“If you love waffles and you also loved this episode—or one of our past episodes—help support the show by heading over to iTunes to give us five stars and a comment. Your reviews make a huge difference in the number of people we can reach so we’d appreciate it!”

Grandma: “I just like ‘em. I could eat them three times a day.”

Grandpa: “I like waffles, but I think I like hot cakes better. A little bit.”

Steve: “Remembering a time in Thailand eating waffles with mangos, chocolate drizzle, ice cream. I can’t think of a more perfect way to, to eat anything except to have it with a waffle.”

Rog: “They were all right. I just didn’t like ‘em a whole lot.”

Ryan: “I think really that there’s more, there’s more to that lovely comfort food than just a grid on a piece of bread. And I’m going to tell you why: it’s because waffles span the ages, from cultures to continents to epics. Waffles mean something to everybody. From royalty to poverty. From people like Marie Antoinette to people named Antoine. Waffles are always present. They’re always there, and they always make you feel happy. How do you feel about waffles? Well let the waffle talk to you. Put on some syrup, put on some jam, and then maybe wrap it in some chicken, and then put that whole sandwich in your face.”

Just Some Sources:

https://www.washingtonpost.com/news/morning-mix/wp/2016/10/07/hurricane-matthew-is-so-scary-even-waffle-house-is-evacuating/   

http://floridadisaster.org/hurricanes/2004/

http://www.nhc.noaa.gov/aboutsshws.php

http://www.nhc.noaa.gov/aboutsshws.php

http://clikhear.palmbeachpost.com/2014/world-nation/natural-disaster/the-meanest-season-floridas-hurricanes-of-2004/

http://www.miamiherald.com/news/weather/hurricane/article32006499.html

http://interactive.sun-sentinel.com/2004storms/  

http://usatoday30.usatoday.com/money/2004-09-07-florida-econ_x.htm

http://usatoday30.usatoday.com/weather/hurricane/2004-09-05-ivan_x.htm

http://www.accuweather.com/en/weather-blogs/hurricanefacts/what-is-the-saffirsimpson-wind/31021

http://www.nhc.noaa.gov/aboutsshws.php

http://www.nhc.noaa.gov/modelsummary.shtml

http://www.nhc.noaa.gov/climo/#cp100

http://www.nhc.noaa.gov/surge/

https://www.wunderground.com/hurricane/models.asp

https://www.wunderground.com/hurricane/models.asp

http://www.livescience.com/21850-hurricane-forecast-improvements.html

https://www.washingtonpost.com/news/capital-weather-gang/wp/2013/08/16/when-storms-were-a-surprise-a-history-of-hurricane-warnings/