The September 12, 2026 Classic draw has 3 even and 3 odd numbers, yet none in the high range defined below. These summaries can coexist: they describe different features of the same set. They do not explain why it was drawn or predict the next result.
Analysis revised 2026-09-14: the previous example, whose draw was not authenticated, was replaced with the 2026-09-12 draw. The original publication date is retained.
The verified Classic draw
Source: saved text from the official Loto-Québec page, consulted September 14, 2026, for the 2026-09-12 draw. Main numbers: 1, 3, 17, 20, 26, 32; bonus number: 5, excluded from calculations. This is one draw, not a frequency study. The saved text is not an archive of the original HTTP response. Official result.
The calculation uses the Classic model of 6 distinct main numbers from 49, as described in the Loto-Québec rules. Gold Ball and auxiliary draws are excluded, as are probabilities of winning any prize across all prize categories.
Parity, ranges and gaps
A successive gap is the difference between neighbouring numbers after sorting them. The ranges below are chosen for description; they are not privileged zones in the game.
| Measure | Result |
|---|---|
| Parity | 3 even / 3 odd |
| Ranges → counts | 1–16 / 17–33 / 34–49 → 2, 4, 0 |
| Successive gaps | 2, 14, 3, 6, 6 |
| Consecutive pairs | 0 |
| Sum of numbers | 99 |
| Span = sum of gaps | 31 = 31 |
This describes one draw. An empty range does not have to be filled in the next draw. The gaps sum to the largest number minus the smallest: 32 − 1 = 31.
A category is not one exact selection
Under a uniform main-number draw, each possible selection is equally likely. A category can contain many selections, so its probability answers a different question. These are three event comparisons, not a complete distribution of categories or observed frequencies.
Each fraction gives the number of combinations in that category in the numerator, divided by the total number of possible combinations in the denominator.
| Event | Exact theoretical probability |
|---|---|
| Exactly 3 even numbers | 4655200/13983816 |
| Counts 2,4,0 in the defined ranges | 285600/13983816 |
| One exact 6-number selection | 1/13983816 |
Here C(N,k) counts ways to choose k distinct numbers from N without regard to order. C(...) expressions written next to each other are multiplied.
The parity calculation is C(24,3)C(25,3)/C(49,6): its numerator counts selections with exactly 3 even and 3 odd numbers.
The range calculation is C(16,2)C(17,4)C(16,0)/C(49,6): its numerator counts selections with the stated numbers in each range.
One exact selection matches all main numbers with probability 1/C(49,6); this is not a whole-ticket probability.
Parity and ranges classify the same outcome in different ways. The category probabilities are separate; multiplying them would not give their joint probability here. These arbitrary ranges were applied after seeing the result, so the table is not a significance test.
What this draw cannot tell us
The sample is one draw. It does not establish historical frequencies, a cause for the result or a forecasting method. Uniformity and independence are assumptions of the model, not findings proved by this snapshot. Neither category makes an individual selection more likely. See Choosing Lotto Numbers: What Statistics Can and Cannot Change.