Statistics
Most and least frequently drawn Powerball, Mega Millions & Millionaire for Life numbers, last 90 days
Based on 38 draws, 2026-06-15 to 2026-09-09.
Main Numbers
Powerball Number
Based on 25 draws, 2026-06-16 to 2026-09-08.
Main Numbers
Mega Ball Number
No recent Millionaire for Life draw data available right now β this refreshes automatically every few hours.
Pairs of numbers drawn together most often, based on 38 draws, 2026-06-15 to 2026-09-09.
Pairs of numbers drawn together most often, based on 25 draws, 2026-06-16 to 2026-09-08.
Hot numbers are those drawn most frequently in recent Powerball, Mega Millions, and Millionaire for Life draws β specifically, the last 90 days of official results. Cold numbers are drawn least often in that same window. Hot combos take this a step further, showing which pairs of numbers have landed in the same draw most often. While past draws don't influence future results β each draw is completely random β many players use frequency data as part of their number selection strategy.
Powerball uses 5 main balls drawn from 1β69 and one Powerball from 1β26. Mega Millions uses 5 main balls from 1β70 and one Mega Ball from 1β24. Millionaire for Life uses 5 main balls from 1β58 and one Millionaire Ball from 1β5.
It's worth being direct about the math here: every ball in a fair lottery machine has exactly the same chance of being drawn on any given night, regardless of how often it came up over the last 90 days. Lottery machines don't have memory β a ball that hasn't come up in months isn't "due," and a ball that's come up five times recently isn't more likely to keep coming up. This is a basic property of independent random events, the same reason a coin that's landed on heads five times in a row is still exactly 50% likely to land on heads on the sixth flip. Statisticians call the opposite belief β that a "cold" number is now overdue β the gambler's fallacy, and it's one of the most common misconceptions in games of pure chance.
So why publish hot and cold numbers at all? Mostly because people enjoy tracking patterns, and a genuinely random process will always produce some numbers that happen to cluster more than others over any finite window β that's expected variance, not a signal. Some players use hot/cold data (or its opposite β deliberately picking "cold" numbers on the theory they're due) as a way to pick numbers when they don't have a preference, which is a perfectly reasonable way to choose, just like using a Quick Pick. What this data cannot do is improve your odds of winning, because the underlying machine doesn't know or care what it drew last week.
The 90-day window used here is a rolling sample, not a lifetime total, so don't be surprised if a number that looks "cold" today looked "hot" a few months ago β that's normal statistical noise working itself out over a longer sample, not a meaningful shift in the game's underlying odds.