How Big Is the Lottery Scam Problem? The FTC's Own Numbers

Most claims about the scale of lottery fraud are unsourced. There is, however, one substantial public dataset: the Federal Trade Commission's Consumer Sentinel Network, which aggregates consumer reports from the FTC and around two dozen other federal, state and international contributors and publishes them annually.

The reporting year matters, so it is stated up front. The figures below are from the Consumer Sentinel Network Data Book 2024, covering calendar year 2024 and published in March 2025. As at the last check of this page, no data book covering 2025 had been published, so 2024 is the current edition. If you are reading this well after that date, check the FTC's reports index for a newer one.

The headline figures

For context, the same data book records 2,600,678 fraud reports in total for 2024, $12,537,194,708 in total reported fraud losses, and a $497 median loss across all fraud.

Reading the numbers properly

Prize fraud is a small share of the money and a visible share of the reports.

• Reports: 97,350 ÷ 2,600,678 = 3.7% of all fraud reports.
• Losses: $351M ÷ $12,537M = 2.8% of all reported fraud losses.

So it is neither the largest category nor a rounding error. Investment fraud alone accounted for $5,697 million in 2024 — sixteen times the prize-fraud total — on fewer reports. Prize scams are high-contact and comparatively low-yield per attempt.

The loss rate is unusually low, and that is the interesting part.

Across all fraud in 2024, 987,520 of 2,600,678 reports involved a dollar loss — about 38%. In the prize, sweepstakes and lottery category it was 22%. Substantially more people report the approach without losing money.

That is what you would expect from a fraud whose opening move is implausible on its face. Most recipients recognise a prize they did not enter, and report it rather than pay. The category is heavy on attempts and comparatively light on conversions.

But the losses that do occur are large. Work it through:

• Reports with a loss: 22% × 97,350 ≈ 21,417
• Mean loss among those: $351,000,000 ÷ 21,417 ≈ $16,390
• Median loss: $1,000

A mean roughly sixteen times the median is the signature of a severe right tail. The typical loss is around a thousand dollars; the average is dragged upward by a comparatively small number of people who lost tens or hundreds of thousands. That is the escalation stage showing up in national statistics — victims who pay once mostly lose a modest amount, and victims who keep paying lose everything.

Note also that the prize-fraud median loss of $1,000 is roughly double the $497 median across all fraud. When this scam lands, it costs more than fraud generally does.

Older adults and the tail

The FTC's own advisory language matches the tail in the data. Mega Millions, in its operator scam advisory, states these scams "often target older people and have been known to wipe out victims' retirement savings."

The data book also reports the category separately for military consumers (active duty, veterans and their families), where in 2024 there were 3,469 prize, sweepstakes and lottery fraud reports, 33% with a loss, $29.2 million total and a $1,665 median — a higher loss rate and a higher median than the general population. Targeted subgroups fare worse, which is exactly the pattern you would expect from a fraud that works by building trust and applying pressure.

Where the approaches come from

The 2024 data book breaks all fraud down by contact method. Of the 1,509,002 fraud reports where a contact method was identified:

This is all fraud, not prize fraud specifically, so it does not tell you how lottery scams arrive. It does tell you something worth knowing: social media produced the largest total reported losses of any contact method in 2024 — $1,858 million — despite ranking fourth by report volume. That is the backdrop to social-media prize-draw scams and impersonated operator accounts.

What these numbers are not

This is the part most articles skip, and it matters more than the headline.

They are reports, not incidents. Sentinel counts what consumers told the FTC and its data contributors. There is no denominator and no survey weighting. A rise in reports can mean more fraud, better awareness, or a new reporting channel.

Under-reporting is severe and unmeasured. Prize fraud carries embarrassment, which suppresses reporting more than it does for, say, a mis-delivered parcel. The published loss figures should be read as a floor, not an estimate of the true total.

Losses are self-reported and unverified. Nobody audits the dollar figure a consumer enters.

It is a United States dataset. Sentinel includes some international contributors, but it is not a world total. Other jurisdictions publish their own: Australia's National Anti-Scam Centre through Scamwatch, the Canadian Anti-Fraud Centre, and the UK's fraud reporting service. Adding figures across these systems is invalid — the categories, thresholds and collection methods differ.

Category boundaries are imperfect. A fake-prize approach that ends in identity theft may be coded as identity theft. A lottery-themed message that harvests card details may land in a different bucket. The 97,350 is not a clean census of every lottery-branded fraud.

The defensible summary

In calendar year 2024, US consumers filed 97,350 reports of prize, sweepstakes and lottery fraud with the FTC's Consumer Sentinel Network — the sixth most-reported fraud category — with 22% reporting a loss, $351 million in total reported losses and a median loss of $1,000. These are minimum figures: reporting is voluntary and incomplete.

That sentence has a source, a year and a caveat. Any figure quoted about lottery scams without all three is worth checking before you repeat it — the same standard applied to the 70% broke myth, which turned out to have none of them.

For what the individual approaches look like, see ten red flags in a win notification. To add your own report to the dataset, the reporting routes by country are collected in one place.