Why Even a +EV Jackpot Usually Isn't Worth Chasing

Assume the hard part is done. You have run the arithmetic honestly — cash value, tax, Poisson sharing — and the number is genuinely above the ticket price. Should you load up?

Almost certainly not, for four reasons that have nothing to do with disapproving of lotteries.

1. The expectation is real; your experience of it is not

Expected value is a long-run average. "Long run" here means something specific and brutal.

At Powerball odds of 1 in 292,201,338, the jackpot term dominates any positive EV. To have a better-than-even chance of hitting it once you would need about 146 million distinct lines — see how many tickets to be more likely than not to win. Buying a thousand tickets at a +EV jackpot leaves you with a 99.9997% chance of not winning the jackpot at all.

A positive expectation you will realistically never sample is a mathematical property of the ticket, not a description of what will happen to you. Your actual distribution of outcomes is: lose everything, with overwhelming probability.

2. The edge evaporates as everyone notices

The sharing term is not a fixed penalty; it is a feedback loop.

Jackpot rises → coverage and excitement rise → ticket sales rise → λ = S × p rises → your expected share (1 − e^(−λ))/λ falls. The very condition that created the edge destroys it, and it destroys it fastest at exactly the jackpots that attract attention.

At a record US jackpot with 600 million tickets sold, λ ≈ 2.05 and you expect to keep about 42% of a jackpot you win. Push sales higher and it keeps falling. Any calculation using last month's sales figures at this month's record jackpot is overstating the edge, often by a lot — which is why the break-even jackpot is a floor rather than a target.

3. Even a real edge justifies almost no stake

This is the formal argument, and it is the strongest one.

The Kelly criterion gives the bet size that maximises long-run growth of a bankroll. For a bet paying odds b with win probability p, the optimal fraction is roughly (edge ÷ odds). With p ≈ 1/292,000,000, the denominator is astronomically large, and the optimal fraction of your wealth comes out at a number with a great many zeros after the decimal point — for any human bankroll, it rounds to less than the price of one ticket.

That is not hand-waving; it is what the formula returns. Growth-optimal staking says: even given a genuine edge, a bet with this variance should take a vanishing share of your capital. Full derivation in the Kelly criterion applied to a lottery ticket.

Betting more than Kelly does not merely add risk — beyond a threshold it makes long-run growth negative despite the positive expectation. Chasing a +EV jackpot with a serious fraction of your money is the textbook case of over-betting an edge.

4. The prize is not the number, and not when you want it

Three deductions sit between the headline and your bank account:

• Annuity vs cash. The advertised figure is usually annuitised. Taking the money now costs roughly half of it (annuity vs lump sum).
• Tax. In the US the top federal rate plus state tax removes a large fraction; elsewhere, nothing (lottery tax by country).
• Co-winners. Point 2, again.

The full stack is worked in what the advertised jackpot really means. An EV calculation using the advertised annuity figure is not slightly optimistic; it is wrong by a factor of three or more.

What the successful exploiters actually did

Every documented case of genuinely profitable lottery play shares a shape that is the opposite of jackpot-chasing:

• They ignored the jackpot. The Cash WinFall syndicates worked the lower divisions on roll-down draws, where odds were short enough to hit reliably and sharing was mild (Cash WinFall).
• They had enormous capital and deployed all of it repeatedly. About $40 million wagered over seven years for an estimated $48 million returned — a margin in the high single digits, earned by volume.
• They played only on qualifying days. Every other draw was as negative as any other lottery.
• They exploited a rule, not a jackpot size. Same for the 1729 Paris bond lottery and the 1992 Irish syndicate, which bought most of a small combination space rather than betting on a big number.

None of them were buying extra Powerball tickets because the jackpot got big.

The one honest reason to buy at a large jackpot

If you are going to buy a ticket anyway, buying at a high jackpot rather than a low one gives you more expected value per dollar. That is true and worth knowing.

It just is not a reason to buy more tickets, because "less bad" is not "good": at the billion-dollar Powerball worked through in positive expected value in a lottery, a $2 ticket returns about 78 cents. That is the best the game ever gets.

Buy the ticket for the daydream if the daydream is worth two dollars to you. Do not buy a hundred because a spreadsheet said the word positive.