Lotto 6/49: 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 105 draws from 11.10.2025 to 10.10.2026, 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. 1,050 lines per group in total.

MatchesQuick PickHotExpected by chance
0453485457.8
1452414433.7
2120140139.0
3241018.5
4111.0
5000.0
6000.0

Wins by tier

TierMatchesQuick PickHotExpected by chance
16 Matches000.00
25 Matches + Bonus number000.00
35 Matches000.02
44 Matches111.02
53 Matches241018.53
62 Matches + Bonus number161812.93
72 Matches104122126.07

Result: the two distributions differ over this period (chi² = 9.74, p = 0.021). 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 29.11.2023 to 10.10.2026.

Matches per prediction

ObservedExpected
0 Matches134 / 130.8
1 Matches124 / 123.9
2 Matches34 / 39.7
3 Matches7 / 5.3
4 Matches1 / 0.3
5 Matches0 / 0.0
6 Matches0 / 0.0

Average: 0.723 matches per prediction – chance gives 0.735.

Deviation from the expectation of chance: chi² = 0.32, p = 0.852. The model hits exactly as often as numbers picked at random.

Model line for the next draw

152733344144

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

Latest predictions

DatePredictionMatches
10.10.2026 92632344149 0
03.10.2026 6913344546 0
30.09.2026 3611134547 0
26.09.2026 21321273645 1
23.09.2026 91320283334 1
19.09.2026 389101428 0
16.09.2026 142023354649 1
12.09.2026 5629424446 0
09.09.2026 232433363746 0
05.09.2026 91528374549 1
02.09.2026 71923273042 0
29.08.2026 51728414249 0
26.08.2026 4613333542 1
22.08.2026 91436374049 0
19.08.2026 23437394348 0
15.08.2026 236141645 0
12.08.2026 112728303341 1
08.08.2026 132933344346 0
05.08.2026 11020274041 1
01.08.2026 469162530 0
29.07.2026 4614424548 1
25.07.2026 93039404749 1
22.07.2026 31621404448 1
18.07.2026 4927283745 1
15.07.2026 6914203549 0
04.07.2026 21516284144 0
01.07.2026 171820272832 1
27.06.2026 2913394449 0
24.06.2026 141623284044 1
17.06.2026 51530333639 0
13.06.2026 152128304147 0
10.06.2026 101336444648 1
06.06.2026 132124373944 0
03.06.2026 1210204347 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 49, 0.7 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.
▲