Unofficial Indicator
Live
Pentagon Pizza Index
An unofficial indicator that measures US military activity and tension through pizza shop orders around the Pentagon
1
DEFCON
Imminent War
2
DEFCON
Dangerous State
3
DEFCON
Increased Alert
4
DEFCON
Enhanced Alert
5
DEFCON
Normal State
DEFCON history
Plotted by observation time. The newest sample was observed 17 minutes before we retrieved it.Window
Aug 1 โ Aug 22
Observations
277
Level changes
103
Observations with a spike
128
Peak alert
DEFCON 1
Cross-check this tension elsewhere
Pizza orders are circumstantial. These are the prices that move when the risk is real.
Recent Cases
Recent cases frequently cited alongside the Pentagon Pizza Index.
In the cases below the military operations themselves are widely reported facts, but claims about how far orders or busy-ness rose are a different matter. No verifiable order records have ever been published, and days when orders surged and nothing followed were never counted, so no hit rate can be calculated.
2026 - Iran Strikes and War (Operation Epic Fury)
On February 28, 2026, the United States and Israel launched joint military strikes against Iran, sparking the largest US military buildup in the Middle East since the 2003 Iraq invasion. Approximately 7,000 additional troops were deployed, including the 82nd Airborne Division. The tracker's records show busy-ness surging at several pizza venues near the Pentagon shortly before the strikes.
Sources: the military operation is widely reported ยท the busy-ness surge is pizzint.watch's own record
2026 - US Invasion of Venezuela
On January 3, 2026, the United States executed an operation to raid Caracas, Venezuela's capital, and capture President Nicolas Maduro. In the days leading up to the operation, late-night orders at pizza shops around the Pentagon surged dramatically, with the sharpest reported surge on the night before the operation, similar to patterns observed during the Panama invasion and Gulf War.
Sources: the military operation is widely reported ยท the order-surge account has no identifiable original source
The subject of the index announced it would interfere
On 23 February 2026, Defense Secretary Pete Hegseth told Fox News he had considered ordering large amounts of pizza on quiet nights to throw off the people tracking this index. Whether any such order was ever placed is unknown. But once the observed party has publicly said it may intervene, neither a surge nor the absence of one reads the way it did before.
Source: Secretary Hegseth on Fox News (23 Feb 2026), as reported by The Hill.
Historical Background
The evidence behind this indicator begins with what pizza delivery operators in 1980s Washington, D.C. told reporters. The five cases below can be traced to a named speaker and a named outlet; after them come the episodes that circulate widely but whose original source cannot be found.
All five rest on participant accounts and contemporaneous reporting. No order ledger has ever been published in verifiable form, and days when orders rose and nothing followed were never counted, so no hit rate can be calculated.
October 1983 โ Grenada invasion (Operation Urgent Fury)
The earliest episode anyone cites. That this case and every one through 1998 traces back to the same person is the index's central weakness. Frank Meeks, who ran 43 Domino's stores in the Washington area, told reporters his drivers were out at 2 a.m. and so knew something was coming before the press did.
Source: press interviews with Frank Meeks
December 1989 โ Panama invasion (Operation Just Cause)
Late-night orders around the Pentagon were said to rise ahead of the December 1989 invasion of Panama. The account carries no multiple and no time window; it is a delivery operator relaying to reporters what he noticed.
Source: press interviews with Frank Meeks
1 August 1990 โ 21 pizzas to CIA headquarters
The one episode that comes with both a date and a number. On the night of 1 August 1990 a record 21 pizzas were delivered to CIA headquarters, and Iraq invaded Kuwait the next day. As the story spread, CNN's then-Pentagon correspondent Wolf Blitzer wrote that the lesson for reporters was to always watch the pizzas, and most accounts of this indicator still start here.
Sources: press interviews with Frank Meeks, a Domino's franchisee in the Washington area ยท Los Angeles Times (Jan 1991) ยท contemporaneous CNN reporting
January 1991 โ Gulf War (Operation Desert Storm)
Orders around the Pentagon were again described as rising around the opening of the Gulf War that followed Iraq's invasion of Kuwait. This is also when the phenomenon picked up the name 'Pizza Index' among reporters and analysts. What survives is again the delivery operator's account and the articles that carried it, not the orders themselves.
Source: Frank Meeks, Los Angeles Times interview (Jan 1991)
1998 โ White House order surges
After the Clinton scandal broke in January 1998, the White House is cited as having ordered $2,600 worth over three days, and during the December impeachment period and Operation Desert Fox, extra-cheese orders were counted at 32% above normal. Here the same pattern appears around a political calendar rather than a military one, which shows a surge does not by itself point at military activity.
Source: counts relayed to the press by the franchise. No order ledger was published
Widely cited, source unverified
Accounts of pizza orders surging around the Pentagon before the 2001 attacks, the 2003 Iraq war, the 2011 bin Laden raid, the 2018 Syria strikes and Russia's 2022 invasion of Ukraine circulate widely online. This page carried the same accounts for a while; on review, no original report or testimony could be identified for any of them. The versions that specify a multiple and a number of hours are, if anything, the most widely repeated, so they are recorded here separately as unverified.
Analysis Methodology
This page does not compute its own figure. Collecting busy-ness at the six pizza venues around the Pentagon and summarising it into five steps is pizzint.watch's work; econix displays the result and records one sample per hour. The three signals below carry the most weight in that call.
The step logic and thresholds follow the values pizzint.watch publishes. The underlying data is public per-venue live busy-ness from map services.
Late-night weighting
The same busy-ness reading counts for more at night. Daytime crowding is explained by lunch demand, while activity well outside the usual level late on a weekday is treated as a sign of extended hours or standby.
Size of the deviation
Each venue's current busy-ness is compared with its own typical level for the same weekday and hour. The count of venues above 150% and above 200% of usual is taken, and the largest single deviation among them carries the most weight in the level.
Breadth and persistence of simultaneous surges
One busy venue can be explained by that venue's own circumstances, so it counts for little. What matters is whether several venues rise in the same hour and whether that state holds into the next observation. The number of venues currently open is counted too, because a closed venue is an absence of signal rather than a quiet one.
TACTICAL MAP - PENTAGON SECTOR
Venues observed
The six venues pizzint.watch tracks. Distance and bearing are straight-line from the Pentagon, and the map plots them from the same figures. The state says only whether a live reading is arriving right now and whether the source has flagged it as spiking.
Extreme Pizza
1.3km ยท S
We, The Pizza
1.9km ยท S
District Pizza Palace
2.1km ยท S
Domino's Pizza
2.7km ยท W
Pizzato Pizza
3.1km ยท W
Papa Johns Pizza
3.4km ยท W
Pentagon Pizza Index explained
What is the Pentagon Pizza Index?
The Pentagon Pizza Index is an unofficial indicator built on the observation that pizza restaurants near the U.S. Department of Defense sometimes get unusually busy shortly before a major military or security announcement. It is not an official statistic produced by any government or statistical agency; it is usually described as an example of open-source intelligence, where publicly visible information is used to infer how busy an organization is. The reasoning is simple: when staff are held late or put on standby, they order dinner in, and that demand shows up in nearby restaurants. The idea entered public awareness in the late 1980s when pizza delivery operators in Washington, D.C. mentioned order surges around certain events, and it has circulated in the press and online ever since.
How do pizza orders become an index?
This page uses the public feed from pizzint.watch, which tracks live busy-ness at the main pizza locations around the Pentagon. Each venue's current activity is compared against its own typical level for the same weekday and hour, and the index looks at how far it deviates from that baseline. Two patterns carry the most weight: unusual late-night activity, and simultaneous surges across several venues. A single busy restaurant can be explained by that restaurant's own circumstances, but several nearby locations spiking at the same hour is harder to attribute to any one of them.
How should the DEFCON level be read?
The DEFCON level shown here is this page's own five-step summary of the order activity, and it has no connection to the readiness condition the U.S. military actually operates. A lower number means the observed activity departs further from normal, with 5 representing an ordinary day and 1 the strongest anomaly. Rather than treating the level as advance notice of an event, it is better used as a single glance at how unusual the current activity is. If the level drops, the practical use is as a prompt to check the news and other geopolitical-risk indicators.
What are its limitations?
The main limitation is that the evidence is anecdotal and assembled after the fact. Surges that preceded a notable event get retold widely, while the far more numerous days when activity spiked and nothing followed leave no record, so there is no way to compute a real hit rate. Busy-ness readings are also easily moved by things unrelated to national security, such as sporting events, holidays, weather, nearby gatherings, or a change in a restaurant's opening hours. Structural shifts in demand, including remote work and the use of dining facilities inside the building, mean the multiples quoted from historical episodes do not transfer cleanly to today. Treat this as an indicator worth watching out of interest, not as a basis for investment or risk decisions.
How often is the index updated?
The level on screen is re-read from pizzint.watch's public data every ten minutes. Because the source sits behind an edge cache, the value that arrives is typically 25 to 45 minutes older than the observation it describes, which is why this page shows the observation time next to the retrieval time. The history chart below accumulates separately, one sample per hour, so movements shorter than an hour never appear in it.
Is the DEFCON on this page the real US military readiness condition?
No. The military's defense readiness condition is an actual alert system run by the Joint Chiefs of Staff, and its current step is not published. The number here is this page's five-step summary of pizza venue busy-ness wearing the same name, and there is no link between the two. This index showing 1 says nothing about the military's alert posture.
Has the Pentagon Pizza Index ever been right?
There is no way to compute a hit rate. Nearly forty years of accounts describe orders rising just before a major event, but nobody counted the days when orders rose and nothing followed. The frequently cited episodes from 1983 through 1998 also come from one person, a Domino's franchisee in Washington who relayed them to the press, and no order ledger was ever published. The modern tracking method uses live busy-ness from map services, and that data cannot be queried for past moments, so a record can be built going forward but past episodes cannot be checked retrospectively.
Sources and disclaimer
The DEFCON level and each venue's state are observed and decided by pizzint.watch; this page cites that result and records one sample per hour. The underlying data is public per-venue live busy-ness from map services.
The historical cases rest on contemporaneous reporting and participant accounts. No order ledger has ever been published in verifiable form, and days when orders rose and nothing followed were never counted, so no hit rate can be calculated.
The Pentagon Pizza Index is not an official statistic produced by any government or statistical agency, and the DEFCON shown here has no connection to the readiness condition the US military operates. Nothing on this page is a suitable basis for an investment or risk decision.
Related indicators
Advertisement
Advertisement