The Mega Millions Jackpot Hit $1.4 Billion — The Probability Math Says You're Paying $4,300 in Expected Loss

The Mega Millions jackpot rolled over to $1.4 billion in mid-May 2026, the fifth time the jackpot has crossed a billion dollars in the game's history. Ticket sales surged, as they always do above the billion-dollar threshold. And as always, the public conversation framed the question the same way: "$2 ticket, billion-dollar prize, even the tiny chance of winning makes it worth it." It's an intuitive framing. It's also wrong by every measure of expected value, which is the math the lottery itself uses to set the odds.

This isn't a moralizing piece about whether lotteries are bad. Plenty of people buy a ticket as entertainment and have a clear-eyed view of what they're spending. The argument here is narrower: the expected value math on a high-jackpot lottery ticket is the worst it has ever been, and the headline jackpots that make tickets feel more valuable actually make them less.

What "Expected Value" Means in This Context

Expected value (EV) is the weighted average outcome of a bet across every possible result, weighted by the probability of each. For a $2 lottery ticket where the jackpot is the only prize, EV = (probability of winning × jackpot) − (probability of losing × $2). When EV is positive, the bet is in your favor on average. When EV is negative, you lose money on average.

The probability of winning the Mega Millions jackpot is 1 in 290,472,336. Five numbers from a pool of 70, plus the Mega Ball from a pool of 24. The combinatorial math is straightforward: C(70,5) × 24 = 12,103,014 × 24 = 290,472,336.

At a $1.4 billion advertised jackpot: EV from the jackpot alone = (1 / 290,472,336) × $1,400,000,000 = $4.82. Subtract the $2 ticket cost and you'd expect to gain $2.82 per ticket. By that crude calculation, the ticket looks like a positive-EV bet.

This is the math most coverage stops at. The next three adjustments invert the conclusion.

The Three Corrections That Change Everything

The advertised jackpot is not what a winner takes home. Three deductions apply, in sequence, and each one is large.

Correction 1: The advertised jackpot is the 30-year annuity, not the lump sum. Almost every winner takes the lump sum, which is the present value of the annuity stream. For Mega Millions, the lump sum is typically about 51%–54% of the advertised jackpot, depending on prevailing interest rates. At today's rates, a $1.4 billion advertised jackpot pays approximately $735 million as a lump sum.

Correction 2: Federal and state taxes apply immediately. The IRS withholds 24% federally; the actual top-bracket rate for a sudden $735 million windfall is 37%. State income tax varies widely — 0% in Florida, Texas, and Washington; 10.9% in New York. For an average state rate of ~5%, total tax on the lump sum is around 42%. Net after tax: roughly $426 million.

Correction 3: Split-jackpot risk grows with jackpot size. When jackpots cross $500 million, ticket sales spike dramatically — sometimes tripling. More tickets in play means more winning combinations purchased and a meaningful probability of a split jackpot. At the $1.4 billion level, historical data suggests roughly a 30% chance of a split between two or more winners, with smaller probabilities of three-way or four-way splits. Adjusting for expected split: take-home value drops to approximately $310 million on average.

The revised EV calculation: Probability × adjusted prize = (1 / 290,472,336) × $310,000,000 = $1.07. Minus the $2 ticket cost = −$0.93 per ticket. The expected loss per ticket at a $1.4 billion advertised jackpot is just under a dollar — meaning the higher the headline jackpot, the worse the math becomes once split risk kicks in.

For comparison, the probability calculator handles single-event probability math directly; the lottery scenario adds the tax and split adjustments on top.

How a $4,300 Annual Spend Becomes a Confident Prediction

The negative expected value compounds with frequency.

Someone who buys five tickets per drawing, twice per week, year-round. Annual spend: $1,040 in ticket cost. Expected return at average jackpot levels: roughly $440 in winnings (including small prizes, which improve the EV from the −47% above to about −58% in actual long-run lottery returns). Expected annual loss: $600.

Someone who buys 10 tickets per drawing, twice per week. Annual spend: $2,080. Expected annual loss: $1,200.

The "big jackpot only" buyer who buys 20 tickets for every drawing above $500 million — typically 8–12 drawings per year. At 20 × $2 × 10 drawings: $400 annually. But because the EV is worse at high jackpots due to split risk, expected return is closer to 35% — expected annual loss: $260.

The serious lottery player. Some lottery players spend $20 per drawing or more. At $20 × 2 × 52 = $2,080/year baseline plus jackpot escalations, annual spend can easily reach $4,000–$5,000. Expected annual loss: $2,300–$3,000. Over a 30-year adult lifetime: $70,000–$90,000 in expected loss, before considering opportunity cost.

That same $4,000/year invested at a 7% real return for 30 years compounds to approximately $377,000. The expected-value cost of lottery participation, properly accounted, is the retirement account that didn't accumulate.

What the Probability Numbers Mean in Recognizable Terms

The lottery odds are abstract. Translating them into everyday-life probabilities makes the magnitude concrete.

Compared to other rare events. The probability of winning a Mega Millions jackpot (1 in 290 million) is roughly:

• 70× less likely than being struck by lightning twice in your life
• 200× less likely than dying in a commercial plane crash on any given flight
• 10,000× less likely than being canonized as a Catholic saint (estimated 1 in 20 million for any given individual)

Compared to other bets. A negative EV of about 50% on a high-jackpot lottery ticket is worse than:

• Slot machines (typically −5% to −15% EV)
• American roulette on red/black (−5.26% EV)
• Most sports betting at standard juice (−4.5% EV)

The lottery is, by any conventional gambling math, the worst-value bet available to consumers. Casinos cannot legally operate games this unfavorable to the player; state-run lotteries can.

The cognitive bias driving the appeal. Probability neglect — humans systematically over-weight extremely low-probability, high-magnitude events. A 1-in-290-million chance is functionally indistinguishable from zero, but the brain processes "1 in 290 million" and "1 in 100,000" as similar categories ("very unlikely"). The actual gap between those probabilities is roughly 3,000× — the difference between buying a ticket every week for a lifetime and never having a realistic chance, versus having a marginally plausible chance.

When Buying a Ticket Is Still Rational

The above is the math. The math isn't the only consideration.

Entertainment value. Many ticket buyers describe the few-day window between buying a ticket and the drawing as having genuine value — the daydream of winning, the casual office-pool conversation, the cheap source of optimism. If that's the framing, $2 is a low-cost form of entertainment, and the EV is irrelevant the same way the EV of a movie ticket is irrelevant.

Pooled-purchase syndicates. Office lottery pools pool the loss too. If 20 coworkers each contribute $5 per week, the expected annual loss per person is $130 — small enough to fit a comfortable budget — while the social benefit of the shared participation is real.

The "first ticket has the most value" principle. The vast majority of the entertainment / hope value of a lottery ticket comes from the first ticket purchased per drawing. The second ticket doubles the probability of winning (still essentially zero) but adds almost nothing to the dream. The fiftieth ticket is purely a loss with no marginal psychological return.

The mathematical case against the lottery is clean. The behavioral case for limited participation — small spend, entertainment framing, pooled with friends — is defensible. The case that fails completely is the one most lottery marketing rests on: "you can't win if you don't play, and someone has to win." Both halves are true. The unstated arithmetic that follows — that the expected ticket buyer loses about half their wager — is what the probability calculator makes visible the moment you put the numbers in.