Has the same group of three Mega Millions numbers come up again?
Mega Millions has been drawn as 5 numbers from 70 plus a Mega Ball from 24 since April 2025. Everything on this page counts only the 160 draws since that change.
Of the 54,740 groups of three that the Mega Millions pool allows, 1,574 have been drawn at least once and 53,166 are still waiting, where chance predicts 53,163.
The most any one group has managed is three times, and one group has done it. Most of the groups the pool allows are still waiting for a first appearance, so the tally below is mostly empty space. Each draw holds ten groups of three, which makes 1,600 drawn in all, spread over 54,740 slots, so the repeats are arithmetic rather than a pattern.
What was measured ten groups per draw
A group of three here is simply three of the pool's numbers, in no order. Draw 160 was 0215314168, and 021531 is one of the ten groups of three it holds, because any three of its numbers make one. Counted that way the record has drawn 1,600 groups across 160 draws, and the pool allows 54,740.
One named group sits inside one draw with probability 10 in 54,740, or about 1.8 chances in ten thousand at the pool of 70. This game has drawn from one pool of 70 numbers throughout, so that same rate stands behind every draw read. Each prediction below is still worked out at its own draw's pool, because that stays right if the pool ever changes.
Sorting the groups by their largest number is what keeps that straight. There are exactly C(m − 1, 2) groups whose largest number is m, and they were available in precisely the draws whose pool reaches m. Since the pools read here only ever grow, all the groups that became available at the same pool size share the same run of chances, so they can be taken together in one band. Every prediction on this page is a sum over the draws that could actually have produced the group. Raw counts alone would rank the groups by how long they have existed rather than by how often they have turned up.
The record 54,740 groups, 5 rows
Red marks a figure outside the range ordinary variation produces
| Times a group has been drawn | Chance | Groups |
|---|---|---|
| Never | 53,163.097.12% | 53,16697.12% |
| Once | 1,554.22.84% | 1,5492.83% |
| Twice | 22.580.0412% | 240.0438% |
| Three times | 0.220.0004% | 10.0018% |
| Four times or more | 0.00160.0000% | 00.0000% |
Each row is a number of appearances and how many groups have exactly that many, beside the number chance predicts and the share of all 54,740 groups it comes to. The predictions add up to 54,740 exactly, so the table compares the shape of the record against the shape chance predicts and not its size. The biggest row in the record is the groups never drawn, and chance puts the same row first. Counted as appearances instead of groups, 1,600 groups were drawn and 1,574 of them were different, so 26 appearances were repeats where chance predicts 23.0. One row here gathers several tallies at once, every tally of four or more, so it is bound to be larger than its neighbours for that reason alone. The record reaches further than chance predicts: the furthest tally chance predicts at least one group for is twice, and the record has a group at three times.
Nothing in the table is printed in red, and that is deliberate rather than an omission. The groups are not independent of one another: one draw lifts ten of them at once, and they share numbers. Expectation is still exactly additive, which is why every prediction above is exact, but the spread of those counts would need the covariance between groups that share numbers, and that is not honestly available here. A range would have to be invented, so none is given. The figures on this page that do carry ranges are read a different way, draw by draw, in the next section.
The one exception above is the fourth figure, the groups held from one draw to the next. That one is read from pairs of draws taken one after another, and because the draws are independent of each other its prediction and its range are both exact. Across the 159 pairs of draws read one after another, no group has ever been held over, where chance predicts 0.29.
At the far end, here are groups that reached the record high of three times drawn, with the draws that held them.
Chance predicts 0.2 groups drawn three times or more and the record has 1. Nothing here is printed in red, and that too is deliberate: these groups were picked out for being the most drawn, so marking them as unusual would only repeat how they were chosen. The highest of 54,740 tallies is above the average by definition, and the honest reading of it is the pair of figures just given.
Read draw by draw how much of a draw is already on record
The same record read the other way round. Going into a draw, some number of groups are already on record, and the draw lifts ten groups of three, some of which may be among them. That count has an exact prediction, the groups on record multiplied by the chance of any one of them being drawn, and an exact range, because a draw's prediction is read off the draws before it and never off itself. So the differences from draw to draw do not lean on one another: they add up, and so do the variances.
A single draw's count is lumpy rather than smooth, because its groups of three overlap each other: numbers that have often been drawn together before bring several old groups in at once. The most any one draw has carried is 2, in draw 115, where the prediction for that draw was 0.21. That figure carries no red mark, because the highest of 160 draws is above the average by definition.
The record had to start from nothing, and a draw can only repeat what came before it. Across the first 100 draws read, 15 of the groups drawn were already on record against a predicted 8.9. Across the last 100, that figure is 19 against 19.7, which is 0.15 standard errors below it. The rise between the two is the record growing, not the draws changing.
Groups per draw already on record, one point per block of 20 draws
result for the blockrange ordinary variation allowsoutside that range
Oldest block on the left, newest on the right. The band is what chance predicts for those same draws, worked out from each draw's own pool and from the groups that were on record going into it, and a point outside it is marked. One point sits outside the band. Across eight blocks, chance alone puts 0.4 of them outside, so a point outside the band would not be explained by the number of blocks read.
Applied to this draw draw 160 · 2026-10-16
Draw 160 was 0215314168 from a pool of 70, which holds ten groups of three. None of them had been drawn before, where chance predicted 0.29 would have been, which is 0.50 standard errors below it. That leaves ten groups here that the pool had never produced until now.
The groups in this draw with the longest records, counting this draw.
That list starts at the top of the ten groups this draw holds, so its first line is where a sorted list has to begin rather than a group that is owed anything, and it carries no red mark for the same reason.
Related statistics
Share this page
If this report does not answer your question, tell us what you need and we will look at building it. Contact