These are estimates, and they don't change your odds of winning. Every draw is independent — no number is "due," and nothing here predicts the result. Split risk is about prize dilution: when more people play, a jackpot is more likely to be shared, which lowers what a winning ticket is actually worth. That's an expected-value point, not a way to beat the game.
Tonight's draws — projected participation
The figures below are the historical median for past draws in the same jackpot range — a descriptive benchmark for the upcoming draw, not a forecast of it.
Loading latest jackpots…
More people play as the jackpot grows
Each dot is one real past draw: its advertised jackpot (across) against our estimate of how many tickets were in play (up). Gold dots are draws where the jackpot was actually won. Participation is roughly flat for small jackpots, then climbs steeply once a jackpot makes national news.
Win & split odds by jackpot size
For each jackpot range, the typical (median) tickets in play across past draws, and the resulting odds that someone wins the jackpot and that it gets split if won.
Recent draws
Estimated tickets in play and win/split odds for the most recent draws, with whether the jackpot was actually won.
How we estimate the number of people playing
Lotteries don't publish national ticket sales before a draw, but Powerball and Lotto America do publish how many people won at every prize tier afterward. Those lower tiers have fixed, published odds — for example, matching just the Powerball pays $4 at odds of 1 in 38.3. So if a draw produced about 200,000 of those winners, roughly 200,000 × 38.3 ≈ 7.7 million tickets were in play. We compute that estimate from several stable tiers and take the median, which keeps the figure steady even when any single tier is noisy.
From the estimated tickets in play, the math is standard probability. With L tickets and jackpot odds of 1 in N, the expected number of jackpot winners is L / N. A Poisson model then gives the chance that at least one ticket hits the jackpot, and — given a winner — the chance that two or more hit it and split the prize. For the upcoming draw we can't know sales in advance, so we report the median tickets from past draws at a similar jackpot size. It is a benchmark, not a prediction.
Two honest caveats. First, these are estimates, not official sales figures, and they carry real uncertainty — especially at record jackpots, where only a handful of past draws exist to learn from. Second, and most important: none of this changes your odds of winning. The lottery is a negative-expected-value game by design, and each draw is independent of the last. The one thing participation genuinely affects is how likely you'd be to share a jackpot if you won — which is why choosing less-popular numbers (avoiding birthdays, dates and visual patterns) doesn't help you win, but can reduce the chance you'd split a prize.
Want the underlying ideas? See our guides on expected value, how the odds work, and quick pick vs self-pick, or the methodology page for how all our numbers are derived.
Participation is estimated by inverting each game's published fixed-tier odds against the reported number of lower-tier winners; win and split probabilities use a Poisson model on those estimates. Figures cover Powerball and Lotto America — the games that publish per-tier winner counts — and update after each draw. They are descriptive statistics with inherent uncertainty, not official sales data and not predictions of any future draw. For information and entertainment only; not financial or gambling advice.