Loading…
Reading the draws and working out what chance predicts.
Loading…
Reading the draws and working out what chance predicts.
GB · ball game
Latest drawThursday, 22 October 2026
Jackpot£100,000
SourcesOnly Primary (commercial feed) reported this draw, so there is nothing to check it against. Settled 23 Oct 2026, 01:20 UTC.
5 numbers from 1–50, plus 1 from 1–50. 3,387 draws held here since 2011-10-08 — 3,274 under the current rules.
Health Lottery changed its number pool then, which started a new era. Only the 3,274 draws since count towards number statistics — mixing draws from before a rule change would make any frequency meaningless.
Your numbers, read by every tool
Type the numbers you always play. Has the line ever come up? How close has Health Lottery come? What has each number done, what does the line look like, and has the same set turned up anywhere else in the world?
Every tool, for this game
Each one opens with Health Lottery already selected. The line under every title is true of this game’s rules right now — not a promise about its next draw.

In 3,274 draws every number is expected about 327 times, and a ball favoured by more than 22% over its share would already have shown — see which ones strayed, and how far chance allows.

A number waits 10 draws on average; 5% of the time it waits 30 or more. It is never “due”.

The 5 numbers add up to about 128, give or take 31. Most tickets sit in that bell — and so do most draws.

35.3% of draws contain two numbers side by side. That is the rules talking, not luck.

1,225 possible pairs, each expected together 26.7 times so far — and about 12 will look “special” by chance alone.

All 50 numbers as a constellation of who appeared with whom — indistinguishable from a shuffled sky.

42% of draws repeat a number from the draw before. A whole draw repeating: 92.0% likely by now.

Sorted, the numbers sit about 8.5 apart on average — and every “spread” you can name is exactly as likely as every other.

It takes about 44 draws for every one of the 50 numbers to appear at least once. The last few are the slow ones.

2,118,760 possible tickets. Every one has a face — and a typicality score you can look up.

Seven independent tests of randomness run live against all 3,274 draws. Any of them can fail; watch whether they do.

The lowest number lands near 9, the highest near 43. Position is a property of sorting, not of the balls.

5/50 rescaled onto the same axes as every other game here. One law, many machines.

Six documented cases of a beaten lottery. Not one involved patterns in past draws.

What the world believes about 7, 8, 13 and 17 — and what every machine says back.

At 30 million tickets, an average line keeps about 7% of a jackpot it wins. Unpopular numbers keep more — and still lose money.

A full wheel of 10 numbers here is 252 tickets. A guarantee is real; a profit is not.

Five folk claims about 5/50 games, each answered with the closed form that kills it — never a tip for the next draw.

This is a without-replacement ball game; the Digit Laboratory is for Pick 3 and Pick 4.
Not for this game
The Chernoff Map reads the straight tier of a digit game, which counts tickets on one exact number; a ball game has no such row.
Not for this game
This game draws once a day, so it has no second draw to compare its first with.
Not for this gameThis draw, read by every tool
What the site’s explorations say about the numbers drawn on 22 October 2026 — each beside what chance predicted. None of it says anything about the next draw.
Scored 71 of 100 across six shape features — more typical than about 96% of every possible ticket.
Times drawn before — chance expected 327 of each
Every number was expected about that many times after 3,273 draws, and they scatter around it. Warm means more than expected, cool means fewer — and a cold number arriving is exactly what cold numbers do.
Frequency Explorer →Shares 3 of 5 numbers (10, 39, 46) with the draw of 18 December 2024 — its closest relative in 3,273 draws. Chance gave a match this close a 100% likelihood somewhere in that history.
From the draw just before (21 October 2026): 39 carried over. Chance says 42% of draws repeat at least one number.
Carry-over / Déjà Vu →The numbers add up to 140 — the 65th percentile of all possible tickets, around a middle of 128. Side-by-side numbers: 22–23. 35% of draws have at least one such pair.
Sum Distribution →23 had not appeared for 13 draws. A number waits about 10 on average, and a wait this long was not a signal — the machine has no memory of it.
Gap & “Overdue” Tracker →A line like this is played about 0.9× as often as an average ticket — about as popular as most. At 30 million tickets sold, a winner on these numbers would keep about 8% of the jackpot.
Illustrative model, calibrated to published studies — not live sales.
The Unpopular Numbers Edge →What the top prize has been, draw by draw. It climbs while nobody wins and drops to the floor when somebody does — the sawtooth every rollover game draws.
Open any draw to read it by every tool — the same six readings the latest result gets above, each against what chance predicted at the time.
SourcesOnly Primary (commercial feed) reported this draw, so there is nothing to check it against. Settled 23 Oct 2026, 01:20 UTC.
Scored 71 of 100 across six shape features — more typical than about 96% of every possible ticket.
Times drawn before — chance expected 327 of each
Every number was expected about that many times after 3,273 draws, and they scatter around it. Warm means more than expected, cool means fewer — and a cold number arriving is exactly what cold numbers do.
Scored 21 of 100 across six shape features — more typical than about 3% of every possible ticket.
Times drawn before — chance expected 327 of each
Every number was expected about that many times after 3,272 draws, and they scatter around it. Warm means more than expected, cool means fewer — and a cold number arriving is exactly what cold numbers do.
Frequency Explorer →Shares 3 of 5 numbers (39, 47, 49) with the draw of 30 September 2026 — its closest relative in 3,272 draws. Chance gave a match this close a 100% likelihood somewhere in that history.
From the draw just before (20 October 2026): 47 carried over. Chance says 42% of draws repeat at least one number.
Carry-over / Déjà Vu →The numbers add up to 206 — the 100th percentile of all possible tickets, around a middle of 128. No two numbers side by side. 35% of draws have at least one such pair.
Sum Distribution →42 had not appeared for 22 draws. A number waits about 10 on average, and a wait this long was not a signal — the machine has no memory of it.
Gap & “Overdue” Tracker →A line like this is played about 0.4× as often as an average ticket — less popular than most. At 30 million tickets sold, a winner on these numbers would keep about 19% of the jackpot.
Illustrative model, calibrated to published studies — not live sales.
The Unpopular Numbers Edge →Scored 42 of 100 across six shape features — more typical than about 25% of every possible ticket.
Times drawn before — chance expected 327 of each
Every number was expected about that many times after 3,271 draws, and they scatter around it. Warm means more than expected, cool means fewer — and a cold number arriving is exactly what cold numbers do.
Frequency Explorer →Shares 4 of 5 numbers (12, 32, 34, 47) with the draw of 16 January 2025 — its closest relative in 3,271 draws. Chance gave a match this close a 29% likelihood somewhere in that history.
From the draw just before (19 October 2026): 50 carried over. Chance says 42% of draws repeat at least one number.
Carry-over / Déjà Vu →The numbers add up to 175 — the 93th percentile of all possible tickets, around a middle of 128. No two numbers side by side. 35% of draws have at least one such pair.
Sum Distribution →34 had not appeared for 14 draws. A number waits about 10 on average, and a wait this long was not a signal — the machine has no memory of it.
Gap & “Overdue” Tracker →A line like this is played about 0.4× as often as an average ticket — less popular than most. At 30 million tickets sold, a winner on these numbers would keep about 16% of the jackpot.
Illustrative model, calibrated to published studies — not live sales.
The Unpopular Numbers Edge →Scored 31 of 100 across six shape features — more typical than about 10% of every possible ticket.
Times drawn before — chance expected 327 of each
Every number was expected about that many times after 3,270 draws, and they scatter around it. Warm means more than expected, cool means fewer — and a cold number arriving is exactly what cold numbers do.
Frequency Explorer →Shares 4 of 5 numbers (19, 45, 46, 50) with the draw of 10 June 2023 — its closest relative in 3,270 draws. Chance gave a match this close a 29% likelihood somewhere in that history.
From the draw just before (18 October 2026): nothing carried over. Chance says 42% of draws repeat at least one number.
Carry-over / Déjà Vu →The numbers add up to 204 — the 99th percentile of all possible tickets, around a middle of 128. Side-by-side numbers: 44–45, 45–46. 35% of draws have at least one such pair.
Sum Distribution →45 had not appeared for 43 draws. A number waits about 10 on average, and a wait this long was not a signal — the machine has no memory of it.
Gap & “Overdue” Tracker →A line like this is played about 0.4× as often as an average ticket — less popular than most. At 30 million tickets sold, a winner on these numbers would keep about 18% of the jackpot.
Illustrative model, calibrated to published studies — not live sales.
The Unpopular Numbers Edge →Scored 10 of 100 across six shape features — more typical than about 0% of every possible ticket.
Times drawn before — chance expected 327 of each
Every number was expected about that many times after 3,269 draws, and they scatter around it. Warm means more than expected, cool means fewer — and a cold number arriving is exactly what cold numbers do.
Frequency Explorer →Shares 3 of 5 numbers (4, 8, 9) with the draw of 8 February 2026 — its closest relative in 3,269 draws. Chance gave a match this close a 100% likelihood somewhere in that history.
From the draw just before (17 October 2026): nothing carried over. Chance says 42% of draws repeat at least one number.
Carry-over / Déjà Vu →The numbers add up to 29 — the 0th percentile of all possible tickets, around a middle of 128. Side-by-side numbers: 8–9. 35% of draws have at least one such pair.
Sum Distribution →4 had not appeared for 29 draws. A number waits about 10 on average, and a wait this long was not a signal — the machine has no memory of it.
Gap & “Overdue” Tracker →A line like this is played about 4.1× as often as an average ticket — more popular than most. At 30 million tickets sold, a winner on these numbers would keep about 2% of the jackpot.
Illustrative model, calibrated to published studies — not live sales.
The Unpopular Numbers Edge →Scored 66 of 100 across six shape features — more typical than about 88% of every possible ticket.
Times drawn before — chance expected 327 of each
Every number was expected about that many times after 3,268 draws, and they scatter around it. Warm means more than expected, cool means fewer — and a cold number arriving is exactly what cold numbers do.
Frequency Explorer →Shares 3 of 5 numbers (25, 31, 48) with the draw of 18 December 2025 — its closest relative in 3,268 draws. Chance gave a match this close a 100% likelihood somewhere in that history.
From the draw just before (16 October 2026): 16, 33 carried over. Chance says 42% of draws repeat at least one number.
Carry-over / Déjà Vu →The numbers add up to 153 — the 78th percentile of all possible tickets, around a middle of 128. No two numbers side by side. 35% of draws have at least one such pair.
Sum Distribution →48 had not appeared for 14 draws. A number waits about 10 on average, and a wait this long was not a signal — the machine has no memory of it.
Gap & “Overdue” Tracker →A line like this is played about 0.8× as often as an average ticket — less popular than most. At 30 million tickets sold, a winner on these numbers would keep about 9% of the jackpot.
Illustrative model, calibrated to published studies — not live sales.
The Unpopular Numbers Edge →Scored 54 of 100 across six shape features — more typical than about 55% of every possible ticket.
Times drawn before — chance expected 327 of each
Every number was expected about that many times after 3,267 draws, and they scatter around it. Warm means more than expected, cool means fewer — and a cold number arriving is exactly what cold numbers do.
Frequency Explorer →Shares 4 of 5 numbers (5, 16, 17, 49) with the draw of 4 March 2026 — its closest relative in 3,267 draws. Chance gave a match this close a 29% likelihood somewhere in that history.
From the draw just before (15 October 2026): nothing carried over. Chance says 42% of draws repeat at least one number.
Carry-over / Déjà Vu →The numbers add up to 120 — the 40th percentile of all possible tickets, around a middle of 128. Side-by-side numbers: 16–17. 35% of draws have at least one such pair.
Sum Distribution →17 had not appeared for 20 draws. A number waits about 10 on average, and a wait this long was not a signal — the machine has no memory of it.
Gap & “Overdue” Tracker →A line like this is played about 0.8× as often as an average ticket — less popular than most. At 30 million tickets sold, a winner on these numbers would keep about 9% of the jackpot.
Illustrative model, calibrated to published studies — not live sales.
The Unpopular Numbers Edge →Scored 63 of 100 across six shape features — more typical than about 80% of every possible ticket.
Times drawn before — chance expected 327 of each
Every number was expected about that many times after 3,266 draws, and they scatter around it. Warm means more than expected, cool means fewer — and a cold number arriving is exactly what cold numbers do.
Frequency Explorer →Shares 3 of 5 numbers (20, 32, 35) with the draw of 30 May 2026 — its closest relative in 3,266 draws. Chance gave a match this close a 100% likelihood somewhere in that history.
From the draw just before (14 October 2026): nothing carried over. Chance says 42% of draws repeat at least one number.
Carry-over / Déjà Vu →The numbers add up to 136 — the 60th percentile of all possible tickets, around a middle of 128. No two numbers side by side. 35% of draws have at least one such pair.
Sum Distribution →35 had not appeared for 45 draws. A number waits about 10 on average, and a wait this long was not a signal — the machine has no memory of it.
Gap & “Overdue” Tracker →A line like this is played about 0.6× as often as an average ticket — less popular than most. At 30 million tickets sold, a winner on these numbers would keep about 11% of the jackpot.
Illustrative model, calibrated to published studies — not live sales.
The Unpopular Numbers Edge →Scored 52 of 100 across six shape features — more typical than about 49% of every possible ticket.
Times drawn before — chance expected 327 of each
Every number was expected about that many times after 3,265 draws, and they scatter around it. Warm means more than expected, cool means fewer — and a cold number arriving is exactly what cold numbers do.
Frequency Explorer →Shares 3 of 5 numbers (11, 22, 37) with the draw of 19 November 2025 — its closest relative in 3,265 draws. Chance gave a match this close a 100% likelihood somewhere in that history.
From the draw just before (13 October 2026): nothing carried over. Chance says 42% of draws repeat at least one number.
Carry-over / Déjà Vu →The numbers add up to 111 — the 30th percentile of all possible tickets, around a middle of 128. No two numbers side by side. 35% of draws have at least one such pair.
Sum Distribution →37 had not appeared for 13 draws. A number waits about 10 on average, and a wait this long was not a signal — the machine has no memory of it.
Gap & “Overdue” Tracker →A line like this is played about 0.9× as often as an average ticket — about as popular as most. At 30 million tickets sold, a winner on these numbers would keep about 8% of the jackpot.
Illustrative model, calibrated to published studies — not live sales.
The Unpopular Numbers Edge →Scored 29 of 100 across six shape features — more typical than about 9% of every possible ticket.
Times drawn before — chance expected 326 of each
Every number was expected about that many times after 3,264 draws, and they scatter around it. Warm means more than expected, cool means fewer — and a cold number arriving is exactly what cold numbers do.
Frequency Explorer →Shares 3 of 5 numbers (8, 9, 15) with the draw of 16 July 2025 — its closest relative in 3,264 draws. Chance gave a match this close a 100% likelihood somewhere in that history.
From the draw just before (12 October 2026): 8 carried over. Chance says 42% of draws repeat at least one number.
Carry-over / Déjà Vu →The numbers add up to 85 — the 8th percentile of all possible tickets, around a middle of 128. Side-by-side numbers: 8–9. 35% of draws have at least one such pair.
Sum Distribution →24 had not appeared for 10 draws. A number waits about 10 on average, and a wait this long was not a signal — the machine has no memory of it.
Gap & “Overdue” Tracker →A line like this is played about 3.1× as often as an average ticket — more popular than most. At 30 million tickets sold, a winner on these numbers would keep about 2% of the jackpot.
Illustrative model, calibrated to published studies — not live sales.
The Unpopular Numbers Edge →Scored 35 of 100 across six shape features — more typical than about 14% of every possible ticket.
Times drawn before — chance expected 326 of each
Every number was expected about that many times after 3,263 draws, and they scatter around it. Warm means more than expected, cool means fewer — and a cold number arriving is exactly what cold numbers do.
Frequency Explorer →Shares 4 of 5 numbers (7, 12, 13, 31) with the draw of 1 September 2026 — its closest relative in 3,263 draws. Chance gave a match this close a 29% likelihood somewhere in that history.
From the draw just before (11 October 2026): 31 carried over. Chance says 42% of draws repeat at least one number.
Carry-over / Déjà Vu →The numbers add up to 71 — the 3th percentile of all possible tickets, around a middle of 128. Side-by-side numbers: 7–8, 12–13. 35% of draws have at least one such pair.
Sum Distribution →13 had not appeared for 24 draws. A number waits about 10 on average, and a wait this long was not a signal — the machine has no memory of it.
Gap & “Overdue” Tracker →A line like this is played about 2.7× as often as an average ticket — more popular than most. At 30 million tickets sold, a winner on these numbers would keep about 3% of the jackpot.
Illustrative model, calibrated to published studies — not live sales.
The Unpopular Numbers Edge →Scored 51 of 100 across six shape features — more typical than about 47% of every possible ticket.
Times drawn before — chance expected 326 of each
Every number was expected about that many times after 3,262 draws, and they scatter around it. Warm means more than expected, cool means fewer — and a cold number arriving is exactly what cold numbers do.
Frequency Explorer →Shares 3 of 5 numbers (29, 31, 49) with the draw of 14 July 2026 — its closest relative in 3,262 draws. Chance gave a match this close a 100% likelihood somewhere in that history.
From the draw just before (10 October 2026): 10, 31 carried over. Chance says 42% of draws repeat at least one number.
Carry-over / Déjà Vu →The numbers add up to 151 — the 77th percentile of all possible tickets, around a middle of 128. Side-by-side numbers: 31–32. 35% of draws have at least one such pair.
Sum Distribution →32 had not appeared for 9 draws. A number waits about 10 on average, and a wait this long was not a signal — the machine has no memory of it.
Gap & “Overdue” Tracker →A line like this is played about 0.9× as often as an average ticket — about as popular as most. At 30 million tickets sold, a winner on these numbers would keep about 8% of the jackpot.
Illustrative model, calibrated to published studies — not live sales.
The Unpopular Numbers Edge →