The Most Common Powerball Numbers (and Why That Does Not Help You)

Across the 1,389 Powerball draws under the current 5/69 matrix, through August 3, 2026, the most frequently drawn white balls are 61 and 21, tied at 123 appearances each, followed by 64 (121) and 28 (119). The least common is 13, at 75. Those gaps are real, they are also precisely what random draws are supposed to produce, and none of them tells you anything useful about the next drawing.

Key takeaways

• 61 and 21 are the most drawn Powerball numbers, at 123 appearances each across 1,389 draws — 22.2% above the expected 100.7 per number.
• 13 is the least drawn white ball at 75 appearances, 25.5% below expected. The full top-to-bottom spread is 48 appearances.
• That spread is not evidence of anything. With 1,389 draws the expected standard deviation per number is about 9.7 appearances, which predicts a top-to-bottom spread of roughly 46 across 69 balls. The observed 48 is normal.
• A chi-square test on all 69 counts returns 79.8 on 68 degrees of freedom (p ≈ 0.16) — no statistical evidence that any ball is favoured.
• All-time tables covering all 1,981 Powerball draws since February 3, 2010 are broken: 592 of those draws used a different ball pool, which is the only reason 60–69 look cold.
• Whichever numbers you play, a Powerball ticket is 1 in 292,201,338 to hit the jackpot.

The most common Powerball numbers

Here are the ten most frequently drawn white balls across the 1,389 draws under the current 5/69 matrix, from October 2015 through August 3, 2026. The expected count for any single number over that window is 100.7.

Rank	Number	Times drawn	vs. expected	Last drawn
1	61	123	+22.2%	2026-06-17
2	21	123	+22.2%	2026-06-24
3	64	121	+20.2%	2026-06-06
4	28	119	+18.2%	2026-06-27
5	27	117	+16.2%	2026-08-01
6	36	117	+16.2%	2026-07-29
7	32	116	+15.2%	2026-06-06
8	23	116	+15.2%	2026-03-09
9	63	115	+14.3%	2026-07-06
10	33	115	+14.3%	2026-04-27

Where does 100.7 come from? Each draw pulls 5 white balls from 69, so any given number has a 5/69 chance of appearing — about 7.25% per draw. Over 1,389 draws that's 1,389 × 5 ÷ 69 = 100.7 expected appearances. Multiply 100.7 by 69 and you get 6,945, which is exactly 5 balls × 1,389 draws. The books balance.

The least common Powerball numbers

The other end of the same table. These are the ten white balls that have shown up least often in the same 1,389 draws.

Rank	Number	Times drawn	vs. expected	Last drawn
1	13	75	−25.5%	2026-06-24
2	26	81	−19.5%	2026-07-27
3	49	81	−19.5%	2026-06-17
4	46	83	−17.5%	2026-07-27
5	34	84	−16.5%	2026-06-08
6	25	88	−12.6%	2026-07-13
7	65	89	−11.6%	2026-07-27
8	51	90	−10.6%	2026-05-27
9	5	91	−9.6%	2026-07-22
10	22	91	−9.6%	2026-07-22

Notice that 13 — the number everyone expects to be cursed — is genuinely last. Notice also that it was drawn on June 24, 2026, and that 65, one of the "coldest" numbers in the game, came out on July 27, 2026. Being on the bottom of a frequency table has no bearing on whether a ball comes out tonight.

What the counts should look like if nothing is wrong

This is the part most frequency pages skip, and it's the part that actually answers the question.

The expected wobble is bigger than people think

Each number's appearance count is a binomial process: 1,389 independent trials, each with probability 5/69. The standard deviation of that count is the square root of 1,389 × (5/69) × (64/69), which works out to about 9.7 appearances.

So even with a perfectly fair machine, a typical number should land roughly 9.7 appearances away from 100.7. One standard deviation covers 91 to 110. Two standard deviations covers 81 to 120. In the real data, 46 of the 69 numbers fall inside one standard deviation and 63 fall inside two — almost exactly the 68% and 95% you'd expect from a normal distribution.

The spread is the wrong statistic to be impressed by

People look at "123 versus 75" and read it as a 48-appearance chasm. But you are not looking at one number, you are looking at the widest gap among 69 of them, and the maximum of 69 draws from a distribution sits far out in the tail by construction.

For 69 samples, the expected highest value sits about 2.36 standard deviations above the mean and the expected lowest about 2.36 below. That's a predicted range of roughly 4.73 × 9.7 ≈ 46 appearances. The actual observed spread is 48. Powerball's frequency table is not just consistent with randomness — it's boringly close to the textbook figure.

The formal test

If you want the single-number version: a chi-square goodness-of-fit test across all 69 counts gives a statistic of 79.8 on 68 degrees of freedom, for a p-value of about 0.16. Anything below 0.05 would be a flag worth investigating. 0.16 means "this looks like a fair drum." The same result holds for other games — we ran it across the full data set in 15 lottery myths tested against real draw data, and frequency bias never survives contact with a significance test.

The 64% gap that means nothing

Here's the headline that frequency pages love: 21 has come up 123 times and 13 only 75 times, so 21 has been drawn 64% more often than 13. That number is arithmetically true (123 ÷ 75 = 1.64) and practically worthless.

The reason is that the balls have no memory and no mechanism. A lottery drum has no way of storing the fact that 21 is ahead, and no way of acting on it. The next draw is 5 balls from 69 with every combination equally likely, exactly as it was on the first draw under this matrix and exactly as it will be on the ten-thousandth.

Put concretely: if you play 61, 21, 64, 28 and 27 — the five hottest white balls in the game — your jackpot odds are 1 in 292,201,338. If you play the five coldest, your odds are 1 in 292,201,338. If you play 1, 2, 3, 4, 5, your odds are 1 in 292,201,338. We break down where that figure comes from in Powerball odds explained, but the short version is that there is only one winning combination and every combination is equally likely.

If you want a set of numbers without spending an hour staring at these tables, the Number Generator will produce one in a second — and it will be exactly as good as any set you agonise over.

The most common red Powerball numbers

The red ball is drawn separately from a pool of 26, so the expected count over 1,389 draws is 1,389 ÷ 26 = 53.4.

Most drawn	Times		Least drawn	Times
4	66		16	40
14	65		17	44
21	63		7	45
24	62		8	46
18	61		15	46

The same check applies. With 1,389 trials at probability 1/26, the standard deviation is about 7.2 appearances. The leader, 4 at 66, is 1.75 standard deviations above expected. The laggard, 16 at 40, is 1.87 below. The observed spread of 26 appearances is actually narrower than the roughly 28 you'd predict for the range across 26 numbers.

In other words, the red ball data is, if anything, slightly less varied than chance would predict. Nobody writes "Powerball red balls suspiciously uniform" headlines, because that's not a story.

The matrix trap: why all-time frequency tables are broken

This is the single biggest error in published Powerball frequency data, and it produces confidently wrong answers.

Powerball has been drawn 1,981 times since February 3, 2010, but only 1,389 of those draws — since October 2015 — used the current 5/69 + 1/26 format. The other 592 draws used a different white-ball pool. Numbers in the high 60s did not exist for those draws. They could not be drawn. Their count for that period is structurally zero.

So when a site publishes an all-time table, every number from the old pool gets a 1,981-draw head start and the numbers added in October 2015 get counted over 1,389 draws. The result is a table where the highest numbers look reliably "cold." That coldness is an artifact of arithmetic, not of the drum.

You can see the effect vanish the moment you scope correctly. In the properly scoped 1,389-draw window, three of the ten hottest white balls are 61, 64 and 63 — all of them numbers that an all-time table would file under "avoid." The same trap applies to the other big game, where the white-ball pool moved to 1–70 in October 2017 and the special-ball pool shrank in April 2025; we handle both windows separately in the most common Mega Millions numbers under the current game.

Rule of thumb: if a frequency table doesn't tell you how many draws it covers and which matrix those draws used, the numbers on it are not usable.

The most common Powerball pairs

Pairs get asked about a lot, so here they are. These are the white-ball pairs that have appeared together most often in the 1,389-draw window.

Pair	Times together
52 & 64	15
21 & 32	15
37 & 44	15
61 & 69	15
51 & 61	14
3 & 16	14
44 & 50	14
19 & 21	14

Run the same sanity check. Each draw contains 10 distinct pairs (5 balls choose 2), and there are 2,346 possible pairs (69 choose 2). So the expected count for any specific pair is 1,389 × 10 ÷ 2,346 = 5.9 appearances.

A top pair at 15 is therefore about 2.5 times its expected count, which sounds enormous until you remember you're looking at the maximum across 2,346 pairs. The standard deviation for a pair count is about 2.4, and the expected maximum across that many pairs lands near 14. Observed: 15. Once again, dead-on for random data.

"Overdue" numbers are the same error, wearing a different hat

The mirror image of hot numbers is the overdue number — a ball that hasn't shown up in a while and is therefore "due."

Number	Draws since last appearance	Last drawn
23	63	2026-03-09
11	53	2026-04-01
33	42	2026-04-27
15	37	2026-05-09
51	29	2026-05-27
1	28	2026-05-30

Look closely at the top of that list and the contradiction becomes obvious: 23 and 33 are simultaneously two of the ten most drawn numbers in the game and two of the most "overdue." A hot-numbers system says play them. An overdue system says play them. A cold-numbers system says avoid them. The three systems disagree because none of them is measuring anything real. We take that apart in detail in why hot, cold and overdue numbers don't work.

Gaps of 40 or 60 draws aren't anomalies either. Each number appears in about 7.25% of draws, so long droughts are routine — with 69 numbers in the pool, somebody is always in the middle of one.

What frequency data is actually good for

Nothing on this page changes your odds of winning. Nothing can. But there's one genuine, non-mystical use for thinking about numbers, and it has nothing to do with the drum and everything to do with the other people holding tickets.

Your odds of winning are fixed. Your odds of winning alone are not. Jackpots are split among all winning tickets, so the numbers other players choose determine what your share would be if your combination came up.

And player choices are wildly non-random. Most people pick dates, which confines them to 1 through 31 — that's 31 of 69 numbers, or 44.9% of the pool, taking a hugely disproportionate share of the tickets sold. Real draws don't cooperate: only 21 of the 1,389 draws in this window, or 1.51%, produced five white balls all at 31 or below. The full cost of that habit is laid out in the birthday number trap.

There's a sting in the tail for this article specifically. Published "most common" lists get read by millions of people, and some of them play exactly those numbers. Playing the top of a hot-numbers table doesn't lower your odds of winning — but it plausibly raises the chance you'd share the prize if you did. The most common Powerball numbers may be the single worst set to play, for a reason that has nothing to do with statistics and everything to do with crowds.

How to actually use any of this

Be honest about what you're buying. A $2 Powerball ticket has overall odds of 1 in 24.87 of winning something, and the non-jackpot prize structure returns about $0.32 per ticket — roughly 16% of the price back in small prizes, with the rest riding on a 1-in-292-million jackpot. That is the deal, and no number selection method alters it.

What you can control is how much you spend and how many tickets you buy. Play what you'd cheerfully lose, treat it as entertainment, and never spend more to make back a loss.

If you'd rather not pick at all, quick-pick style selection has one small advantage: machine-generated numbers spread across the full 1–69 range instead of clustering in the birthday zone, which cuts your exposure to splitting. Our Number Generator does the same thing and lets you exclude the popular clusters if you want to. It won't help you win. It might help you keep more if you do.

Frequently asked questions

What are the most common Powerball numbers?

Across the 1,389 draws under the current 5/69 matrix, through August 3, 2026, the most drawn white balls are 61 and 21 with 123 appearances each, then 64 (121), 28 (119), and 27 and 36 (117 each). The most drawn red Powerball is 4, with 66 appearances. Expected counts over that window are 100.7 for a white ball and 53.4 for a red one, so all of these sit within normal random variation.

What is the least common Powerball number?

The least drawn white ball is 13, with 75 appearances across 1,389 draws — 25.5% below the expected 100.7. It's followed by 26 and 49 (81 each), 46 (83) and 34 (84). The least drawn red Powerball is 16, at 40 appearances against an expected 53.4. None of these numbers is defective, and none is more or less likely to appear in the next drawing.

Does playing the most drawn Powerball numbers improve my chances?

No. Every combination of five white balls plus a Powerball has identical odds of 1 in 292,201,338, whether you pick the hottest numbers, the coldest, or five consecutive digits. The drum has no memory of past results and no mechanism to favour any ball. Frequency data describes what already happened; it carries no information about what will happen next.

Why do high Powerball numbers like 65 and 66 look cold?

Usually because the table is scoped wrong. Powerball has been drawn 1,981 times since February 3, 2010, but only 1,389 of those used the current 5/69 pool. In the other 592 draws the highest numbers did not exist and could not be drawn, so any all-time count gives them an artificially low total. Scoped to the current matrix, 61, 63 and 64 all rank in the top ten.

Is a 48-appearance gap between the hottest and coldest numbers unusual?

No — it's almost exactly what randomness predicts. Each number's count has a standard deviation of about 9.7 appearances over 1,389 draws. Taking the widest gap among 69 numbers, the expected top-to-bottom spread is roughly 46 appearances. The observed spread is 48. A chi-square test on all 69 counts returns 79.8 on 68 degrees of freedom, p ≈ 0.16, which is no evidence of bias at all.

Is there any point in choosing numbers deliberately?

Only for one reason, and it isn't winning. Jackpots are split among all winning tickets, so avoiding numbers other people crowd into — chiefly 1 through 31, which is only 44.9% of the pool but attracts a large share of tickets — can mean a bigger share if you happen to win. Only 1.51% of draws in this window came out entirely at 31 or below. Your odds stay 1 in 292,201,338 either way.