Historical test
Every strategy was tested on official Spor Toto weeks as if it had been played with the data available at coupon close: how many columns got 12, 13, 14 and 15 right, what it cost and what it returned with the official prizes.
Data at close
opening odds + form
Columns are played
scored against the official result
Weighed by prizes
return ÷ cost
Before you read: the return is calculated with the official prize per winner and is an optimistic upper bound; had our columns entered the pool, the amount per winner would have been lower. Probabilities were produced with today's model calibration. Past results do not indicate future winnings.
Weeks evaluated (total): 93 · column price as of each week's own date; 30 weeks with an unknown price are excluded from the return · measured Oct 5, 2026, 10:30
Weeks tested
20
Last 20 weeks
Strategies tested
22
4 groups
Covered their cost
3 / 22
return of 100% or more
Highest return
334.8%
Analysis Column 3 — Differentiated
Last 20 weeks
Each bar is return ÷ cost. The vertical line marks 100%: left of it is a loss, right of it is above cost.
Scale: the longest bar is 334.8%. Prizes are paid per column; a single week with a high prize can lift the ratio on its own — the cards below show in how many weeks each strategy hit.
The three system columns published on the site every week, applied to past weeks. The old fixed-distribution version stands alongside for comparison.
The site's 3 published system columns applied to past weeks: the single/double/full-cover split comes from the model probabilities and the budget each week (default full-system cap); full system.
Return
57.9%
20 weeks
830 columns/week
12+
10
13+
5
14+
1
15
0
Columns right (12·13·14·15): 123 · 17 · 1 · 0
The site's 3 published system columns applied to past weeks: the single/double/full-cover split comes from the model probabilities and the budget each week (default full-system cap); full system.
Return
55%
20 weeks
830 columns/week
12+
11
13+
5
14+
1
15
0
Columns right (12·13·14·15): 132 · 18 · 1 · 0
The site's 3 published system columns applied to past weeks: the single/double/full-cover split comes from the model probabilities and the budget each week (default full-system cap); full system.
Return
334.8%
20 weeks
830 columns/week
12+
5
13+
4
14+
3
15
0
Columns right (12·13·14·15): 220 · 45 · 4 · 0
The system columns before Phase 2: a fixed 7 single / 5 double / 3 full-cover distribution, full system. For comparison only.
Return
72.4%
20 weeks
864 columns/week
12+
11
13+
6
14+
1
15
0
Columns right (12·13·14·15): 161 · 26 · 2 · 0
The system columns before Phase 2: a fixed 7 single / 5 double / 3 full-cover distribution, full system. For comparison only.
Return
96.4%
20 weeks
864 columns/week
12+
11
13+
6
14+
1
15
1
Columns right (12·13·14·15): 223 · 59 · 11 · 1
The system columns before Phase 2: a fixed 7 single / 5 double / 3 full-cover distribution, full system. For comparison only.
Return
131%
20 weeks
864 columns/week
12+
2
13+
1
14+
1
15
0
Columns right (12·13·14·15): 88 · 20 · 2 · 0
The full system of a single / double / full-cover distribution: every column inside the selection is played.
Return
72.4%
20 weeks
864 columns/week
12+
11
13+
6
14+
1
15
0
Columns right (12·13·14·15): 161 · 26 · 2 · 0
Return
83%
20 weeks
1,728 columns/week
12+
11
13+
8
14+
1
15
0
Columns right (12·13·14·15): 242 · 33 · 2 · 0
Return
72.1%
20 weeks
288 columns/week
12+
9
13+
3
14+
1
15
0
Columns right (12·13·14·15): 90 · 19 · 2 · 0
Return
51.4%
20 weeks
96 columns/week
12+
5
13+
1
14+
0
15
0
Columns right (12·13·14·15): 19 · 2 · 0 · 0
Return
31.4%
20 weeks
48 columns/week
12+
3
13+
0
14+
0
15
0
Columns right (12·13·14·15): 6 · 0 · 0 · 0
Return
40.6%
20 weeks
24 columns/week
12+
2
13+
0
14+
0
15
0
Columns right (12·13·14·15): 3 · 0 · 0 · 0
Return
0%
20 weeks
1 columns/week
12+
0
13+
0
14+
0
15
0
Columns right (12·13·14·15): 0 · 0 · 0 · 0
The same distributions reduced with a covering set: fewer columns, a lower lower bound.
This guarantee does not guarantee match results. If your selections cover the actual results of the 15 matches, it is mathematically verified that at least one column of the generated set will get at least N matches right.
Return
39.7%
20 weeks
92 columns/week
12+
6
13+
2
14+
0
15
0
Columns right (12·13·14·15): 13 · 3 · 0 · 0
Return
86.7%
20 weeks
23 columns/week
12+
4
13+
1
14+
0
15
0
Columns right (12·13·14·15): 4 · 1 · 0 · 0
Return
0%
20 weeks
8 columns/week
12+
0
13+
0
14+
0
15
0
Columns right (12·13·14·15): 0 · 0 · 0 · 0
Return
155.8%
20 weeks
171 columns/week
12+
10
13+
4
14+
1
15
0
Columns right (12·13·14·15): 19 · 4 · 1 · 0
Return
4.6%
20 weeks
36 columns/week
12+
1
13+
1
14+
0
15
0
Columns right (12·13·14·15): 0 · 2 · 0 · 0
Return
27.3%
20 weeks
72 columns/week
12+
4
13+
1
14+
0
15
0
Columns right (12·13·14·15): 9 · 1 · 0 · 0
Return
4.5%
20 weeks
24 columns/week
12+
1
13+
1
14+
0
15
0
Columns right (12·13·14·15): 1 · 1 · 0 · 0
The highest guarantee level that fits a fixed column budget; the rest of the budget is filled with the most likely columns.
Return
52.6%
20 weeks
94 columns/week
12+
5
13+
1
14+
0
15
0
Columns right (12·13·14·15): 19 · 2 · 0 · 0
Return
78.5%
20 weeks
486 columns/week
12+
7
13+
4
14+
1
15
0
Columns right (12·13·14·15): 87 · 14 · 1 · 0
Use the formula tool to build columns from your own selections.
Spor Toto formulas →18+ · Play responsibly
Cookie preferences
We use strictly necessary cookies to make the site work. Analytics and advertising cookies run only if you allow them — if you decline, those services are never loaded. For details see our Cookie Policypage.