United Kingdom: Thunderball

Thunderball Autocorrelation Analysis

Thunderball: do draws have "memory"? Are results correlated between draws?

Autocorrelation shows whether the results of draw N are related to draw N-1, N-2, and beyond. If significant autocorrelation is detected, it's a valuable signal for forecasting. If not, it confirms the randomness of the "Thunderball" lottery.

Analysis based on 20 draws from 04/24/2026, 07:00 PM to 05/27/2026, 07:15 PM

Max lag:

Draw sums autocorrelation

Correlogram with 95% confidence intervals

Observations

Significant lags

±0.4383

95% confidence interval

No autocorrelation detected

All ACF values are within the 95% confidence interval. The sequence is statistically random.

ACF(1) for All Numbers

Autocorrelation at lag 1 — quick overview of each number's "memory"

Ball	ACF(1)	Status
1	-0.2625	Normal
2	0.0500	Normal
3	0.1125	Normal
4	0.0500	Normal
5	-0.0553	Normal
6	-0.0553	Normal
7	0.0000	Normal
8	-0.1167	Normal
9	-0.0553	Normal
10	-0.1167	Normal
11	-0.1167	Normal
12	-0.0167	Normal
13	-0.0553	Normal
14	-0.1853	Normal
15	0.2069	Normal
16	-0.1853	Normal
17	-0.1167	Normal
18	-0.0553	Normal
19	-0.1853	Normal
20	-0.2000	Normal
21	-0.0553	Normal
22	-0.1853	Normal
23	-0.2000	Normal
24	-0.2625	Normal
25	-0.0553	Normal
26	-0.0489	Normal
27	-0.1167	Normal
28	-0.1265	Normal
29	0.3625	Normal
30	-0.1167	Normal
31	-0.0611	Normal
32	-0.0553	Normal
33	-0.0553	Normal
34	-0.0553	Normal
35	0.1833	Normal
36	-0.0026	Normal
37	0.0500	Normal
38	-0.0553	Normal
39	-0.1167	Normal

About Autocorrelation

Mathematical foundations

The autocorrelation function (ACF) measures the linear dependence between values of a time series separated by k steps (lag). In the context of a lottery: is the result of draw N related to the result of draw N-k?

ACF Formula

ACF(k) = Σ(xₜ - x̄)(xₜ₊ₖ - x̄) / [n · Var(x)]

ACF values range from -1 to +1. If |ACF| exceeds the confidence interval ±1.96/√n, the correlation is statistically significant.