MillionDAY: 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.

🤖 AI lab (LSTM) ›

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 27.12.2025 to 22.10.2026.

Matches per prediction

ObservedExpected
0 Matches168 / 182.7
1 Matches109 / 99.3
2 Matches23 / 16.9
3 Matches0 / 1.1
4 Matches0 / 0.0
5 Matches0 / 0.0

Average: 0.517 matches per prediction – chance gives 0.455.

Deviation from the expectation of chance: chi² = 3.53, p = 0.171. The model hits exactly as often as numbers picked at random.

Model line for the next draw

624394347

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

Latest predictions

DatePredictionMatches
22.10.2026 2027333955 0
21.10.2026 1229394145 0
20.10.2026 410282954 1
19.10.2026 24192953 1
18.10.2026 1923242527 1
17.10.2026 67111439 0
15.10.2026 16101946 1
14.10.2026 1536394250 1
13.10.2026 14263451 0
12.10.2026 16224250 1
11.10.2026 1016192443 1
10.10.2026 815374144 0
09.10.2026 69324654 0
08.10.2026 510121954 1
07.10.2026 2024254042 1
06.10.2026 1119394551 0
05.10.2026 133394345 1
04.10.2026 914182853 0
02.10.2026 1219363848 1
01.10.2026 417304345 0
30.09.2026 611222534 0
29.09.2026 2022234255 1
28.09.2026 1719263845 1
27.09.2026 26384554 0
25.09.2026 19174347 1
24.09.2026 12143255 1
23.09.2026 618212233 0
22.09.2026 416275154 1
21.09.2026 122253041 1
20.09.2026 916193031 0
19.09.2026 722253035 1
18.09.2026 2940434950 0
17.09.2026 1314162532 0
16.09.2026 132404554 0
15.09.2026 1281043 0
14.09.2026 146830 0
13.09.2026 49173445 1

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 5 from 55, 0.5 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.
▲