In this blog post, we’ll explore the concept of “alternative data” and look at real-world examples to see how small pieces of information we casually overlook in our daily lives—such as pizza orders or mobility data—can come together to provide clues for predicting future events.
There is a theory that before international conflicts like the Gulf War break out, pizza orders at restaurants near the Pentagon surge dramatically. Information used to predict the future based on such subtle changes in everyday life is called “alternative data.”
“This is the Pentagon, CIA. Send us 21 pizzas right now!”
On August 1, 1990, this order placed at a Domino’s Pizza store in Washington, D.C., appeared at first glance to be a simple request for a late-night delivery. However, it took on special significance when it became known the next day that Iraq had invaded Kuwait, sparking the Gulf War. This phenomenon was seen as more than just a coincidence, as it was observed that when urgent situations related to national security arise, overtime work increases at major institutions like the Pentagon and the White House, naturally leading to a surge in orders for convenience foods like pizza. Consequently, people began referring to this as the “Pentagon Pizza Index” or “Pizza Meter” and using it as an intriguing predictive tool.
This concept gained widespread attention in 1990 when Frank Mix, who operated a Domino’s Pizza store in the Washington, D.C., area at the time, reported the unusual early-morning order patterns to the media. While analyzing past data, he discovered a common pattern: on the nights before the U.S. engaged in military action in 1983 and 1989, pizza orders near the Pentagon more than doubled compared to usual levels, leading him to suggest the possibility of a consistent pattern. Over time, this intriguing hypothesis seemed to fade into obscurity, but it regained attention when Google introduced the “Popular Times” feature to its Maps service. This is because it is now possible to check in real time how busy restaurants in a specific area are at any given moment, allowing for direct observation of changes in pizza orders around the Pentagon.
In fact, this data has shown a surprisingly high degree of correlation with various events. In April 2024, when Iran launched drones toward Israel, pizza orders in the Washington area surged; and on June 12, 2025, an Israeli military attack began approximately one hour after a sudden spike in orders at pizza shops near the Pentagon. Furthermore, on the night of January 2, 2026, pizza orders around the Pentagon increased by approximately 770% compared to usual levels. As U.S. military actions and political events unfolded in the early hours of the following day, this “pizza index” once again drew significant attention. Of course, these cases do not always lead to accurate predictions, but the recurring patterns make it difficult to dismiss them as mere coincidence.
Pizza orders are not the only indicator used in this way. For example, when people work late into the night, the number of bar patrons actually decreases, and this change is sometimes interpreted as a precursor to a tense situation. Furthermore, data from ordinary daily life—such as the number of Uber rides near the Pentagon or orders at fast-food restaurants near the White House—can also serve as important clues. In this way, the traces of daily life that we casually overlook function as signals reflecting the movements of society as a whole, going beyond mere consumption records.
The scope of alternative data is expanding. It is being utilized in various ways, such as analyzing the number of vehicles in supermarket parking lots via satellite to predict changes in consumer activity, or tracking aircraft flight paths to assess military tensions in specific regions. Recently, “prediction markets” that forecast the future based on this data have been growing rapidly, and their scope is expanding to include predictions of election results and international political trends. Major platforms like Robinhood and Coinbase are also paying close attention to this trend and actively participating in related markets.
Ultimately, traditional economic indicators like gross domestic product (GDP) or unemployment rates are not the only tools for predicting the future. Sometimes, even the smallest traces of daily life—such as a pizza ordered late at night from an office—can serve as the first signals of events unfolding around the world. The small changes we casually overlook in our daily lives may already hold clues that allow us to read the future.
