Korea Lotto 6/45: AI prediction and backtest – what are "statistical" lines worth?

Two honest experiments on the real draws: we compare random Quick Pick lines with "hot" lines built from the most frequent numbers, and we let a learning algorithm (k nearest neighbours) predict each draw, always using only the data known before the draw. Both methods are compared with the exact expectation of pure chance.

Backtest: Quick Pick lines vs hot lines (12 months)

For each of the 53 draws from 18.10.2025 to 17.10.2026, each group plays 10 lines: random Quick Pick versus lines built from the 15 most drawn numbers over the previous 100 draws. The Bonus number are random in both groups so that only the main numbers are compared. 530 lines per group in total.

MatchesQuick PickHotExpected by chance
0231193212.3
1212219224.8
27510080.3
3121611.9
4020.7
5000.0
6000.0

Wins by tier

TierMatchesQuick PickHotExpected by chance
16 matches000.00
25 matches + Bonus number000.00
35 matches000.01
44 matches020.72
53 matches121611.89

Result: the two distributions differ over this period (chi² = 8.29, p = 0.040). Across many repeated tests, a gap like this turns up from time to time; it is not proof of a better method, as every combination remains equally likely.

AI prediction (k nearest neighbours)

For each draw, the model looks for the 7 past moments whose preceding 3 draws most resembled the current ones and plays the numbers most often drawn next. It is reapplied after every draw and never sees the future. 300 draws evaluated from 23.01.2021 to 17.10.2026.

Matches per prediction

ObservedExpected
0 Matches125 / 120.2
1 Matches125 / 127.2
2 Matches44 / 45.4
3 Matches6 / 6.7
4 Matches0 / 0.4
5 Matches0 / 0.0
6 Matches0 / 0.0

Average: 0.770 matches per prediction – chance gives 0.800.

Deviation from the expectation of chance: chi² = 0.36, p = 0.835. The model hits exactly as often as numbers picked at random.

Model line for the next draw

21617202743

For entertainment only: this line has exactly the same chance as any other.

Latest predictions

DatePredictionMatches
17.10.2026 111318212738 1
10.10.2026 51315212241 1
03.10.2026 32130333444 1
26.09.2026 152128313244 1
19.09.2026 132023273539 0
12.09.2026 1910193844 0
05.09.2026 41825273133 1
29.08.2026 61020304145 0
22.08.2026 12729343641 0
15.08.2026 1815162140 1
08.08.2026 31316263339 0
01.08.2026 168243638 1
18.07.2026 3416222428 0
11.07.2026 91723262835 0
04.07.2026 13783241 0
27.06.2026 3618212729 1
20.06.2026 1316202324 0
13.06.2026 173637414245 0
06.06.2026 132527333545 0
30.05.2026 4823314145 1
23.05.2026 42028333744 0
16.05.2026 6813192328 0
09.05.2026 1313243240 1
02.05.2026 349131924 1
25.04.2026 101214253637 1
18.04.2026 71314153740 0
11.04.2026 3622232439 1
04.04.2026 71516172738 0
28.03.2026 12330313738 1
21.03.2026 8912243945 1
14.03.2026 11318233840 1
07.03.2026 367122738 1
28.02.2026 11617212430 0
21.02.2026 349102740 0
07.02.2026 121623243233 0
31.01.2026 6722233334 0
24.01.2026 81213142729 1
17.01.2026 3926283740 0

How it works

A nearest-neighbour model looks for points in time whose last draws resembled the current ones and picks the numbers that followed most often. It is reapplied draw by draw without looking into the future; a backtest additionally compares Quick Picks with lines made of frequent numbers.

Example

For 6 from 45, 0.8 matches per line are expected on average; the model lands exactly there.

Limitations

No model can predict random draws. The page shows this transparently with a chi-squared test.

Frequently asked questions about this tool

Can AI predict lottery numbers?
No. With independent, uniformly distributed draws there is no pattern a model could learn.
▲