UK49s predictions, measured instead of guessed
UK49s predictions are the most searched thing about this game in South Africa, and they are also the one thing about it that cannot be done.
Each draw is independent of every draw before it. That is not an opinion about UK49s, it is how the machine works, and it means the six numbers that came up yesterday tell you precisely nothing about the six coming up today. What the history does tell you is what has happened, and that is worth publishing properly rather than dressing up as a forecast.
What the draw has actually done
Counted across the last 365 Teatime draws, to Wednesday, 2 September 2026. A fair 49-ball draw gives each number about 44.7 appearances over this run.
| Number | Drawn | Expected | Share | Against chance |
|---|---|---|---|---|
| 24 | 67 | 44.7 | 18.4% | extreme |
| 3 | 59 | 44.7 | 16.2% | notable |
| 42 | 59 | 44.7 | 16.2% | notable |
| 25 | 55 | 44.7 | 15.1% | ordinary |
| 5 | 54 | 44.7 | 14.8% | ordinary |
| 6 | 52 | 44.7 | 14.2% | ordinary |
24 has come up 67 times in the last 365 draws against about 44.7 expected, a gap of 3.6 standard deviations. Over enough numbers and enough windows, a gap this size turns up by chance alone. It says nothing about the next draw.
Edges publishes what has happened. It does not publish what will happen, because on a random draw nobody can, and any site that tells you otherwise is selling you something. Every number above is computed from our own checked archive and you can recount it from the results pages.
The numbers drawn least
The other end of the same table. These are not "cold" in any sense that affects the next draw; they are simply the balls that have come up least often in this window.
| Number | Drawn | Expected | Share | Against chance |
|---|---|---|---|---|
| 14 | 32 | 44.7 | 8.8% | notable |
| 43 | 32 | 44.7 | 8.8% | notable |
| 7 | 34 | 44.7 | 9.3% | ordinary |
| 8 | 35 | 44.7 | 9.6% | ordinary |
| 39 | 35 | 44.7 | 9.6% | ordinary |
| 28 | 37 | 44.7 | 10.1% | ordinary |
The longest current gaps
How many draws ago each of these was last seen. Long gaps are completely ordinary on a 49-ball board: with six balls drawn from 49, the average wait for any given number is about eight draws, and waits of forty or more turn up constantly without anything being wrong.
| Number | Drawn | Expected | Draws since | Against chance |
|---|---|---|---|---|
| 4 | 39 | 44.7 | 45 | ordinary |
| 39 | 35 | 44.7 | 31 | ordinary |
| 15 | 45 | 44.7 | 27 | ordinary |
| 49 | 48 | 44.7 | 22 | ordinary |
| 23 | 41 | 44.7 | 21 | ordinary |
| 26 | 39 | 44.7 | 18 | ordinary |
Numbers that have come up together most
Nobody else publishes this and it is the most interesting table on the page. Over the last 90 Teatime draws these pairs appeared in the same draw most often. As with everything else here it describes the past, and a pair that has landed together five times is not a pair that is going to land together again.
- 5376x
- 16176x
- 34486x
- 255x
- 7165x
- 16345x
- 16485x
- 33485x
Lunchtime against Teatime
The two draws use the same machine and the same 49 balls, and they are entirely independent of each other. Over the last year the most-drawn Lunchtime number has been 17 on 61 appearances, against 24 on 67 for Teatime. That the two lists disagree is the clearest demonstration on this site that neither list means anything about what comes next.
Does any of this predict the next draw
No, and it is worth being exact about why rather than just asserting it.
Across the last 365 Teatime draws the chi-square statistic for the 49 numbers is 54.5 on 48 degrees of freedom. For a fair draw that figure should sit near 48, and anything below roughly 66 is inside what randomness produces. In other words the machine is behaving exactly as a fair machine does, the spread you see in the tables above is the spread randomness makes on its own, and there is no bias in it to find or to bet on.
That is the whole finding. Edges publishes these tables because people want to see them and because the counts are genuinely interesting, and it publishes this paragraph next to them because leaving it out is how a data page becomes a tip sheet.
Questions about UK49s numbers
- Can UK49s numbers be predicted?
- No. Every draw is independent, so the result of one draw carries no information about the next. Any site publishing tomorrow's numbers is either guessing or selling something. The frequency tables on this page describe what has happened and make no claim about what will.
- What does a hot number actually mean?
- Only that it has been drawn more often than the others inside the window being counted. It is a fact about the past. A hot number is no more likely to be drawn next than a cold one, because the balls have no memory of having been picked.
- Is a number that has not come up in a long time due?
- No, and this is the most expensive mistake in lottery betting. It is called the gambler's fallacy. A number absent for forty draws has exactly the same chance in the next draw as one drawn yesterday. This desk publishes the longest current gaps because people want to see them, and labels them absence rather than anything resembling due.
- How many draws are these Teatime figures counted from?
- 974 confirmed Teatime draws, from Monday, 1 January 2024 to Wednesday, 2 September 2026. Each one was matched against an independent publisher before it entered the archive, so the counts are computed on checked data rather than on whatever a scraper caught.
- How do you tell a real bias from ordinary randomness?
- By comparing the count against what a fair draw produces and measuring the gap in standard deviations. Anything inside two is ordinary. Across the last 365 draws the chi-square statistic for this game is 54.5 on 48 degrees of freedom, which is what a fair 49-ball machine looks like. If that ever stops being true, this page will say so.