The building blocks of lottery probability
Lottery probability begins with counting the possible outcomes of a draw. The count depends on whether the order of the selected numbers matters and whether numbers can be repeated. In a standard draw without replacement, each set of numbers is an unordered combination, while the binomial coefficient provides a compact way to count such sets. The hypergeometric distribution then describes the chance of matching a specified number of selections. These tools give a mathematical basis for calculating outcomes rather than relying on numerical patterns.
Permutations and combinations
A permutation is an arrangement in which order matters; a combination is a selection in which order does not matter. For a draw that records only which numbers were selected, choosing 2 numbers from 5 produces combinations, not distinct outcomes for each possible ordering. The combination count is C(5,2) = 5! / (2! × 3!) = 10. If the two picks were instead recorded as first and second, the ordered count would be 5 × 4 = 20.
| Counting method | Example outcomes | Count |
|---|---|---|
| Unordered combinations | {1, 2}, {1, 3}, {1, 4}, {1, 5} | 10 |
| Unordered combinations | {2, 3}, {2, 4}, {2, 5} | 3 shown; 10 total |
| Unordered combinations | {3, 4}, {3, 5}, {4, 5} | 3 shown; 10 total |
| Ordered arrangements | (1, 2) and (2, 1) count separately | 20 |
The table groups the ten unordered pairs across three rows: four pairs in the first, three in the second, and three in the third. Each pair has two possible orders, so 10 × 2 = 20 ordered arrangements. In lottery probability, use combinations when the result is defined by the set of numbers drawn; use permutations only when a draw assigns significance to sequence or position.
The binomial coefficient
The binomial coefficient, written C(n,r) or “n choose r,” counts the ways to select r distinct items from n when order does not matter. Its factorial formula is C(n,r) = n! / [r!(n-r)!], where n! means the product of all positive integers from n down to 1, and 0! = 1. For example, C(5,2) = 5! / (2! × 3!) = (5 × 4) / (2 × 1) = 10.
The division by r! removes the repeated orderings of the same selection: selecting 2 and then 5 describes the same pair as selecting 5 and then 2. This coefficient is the basic counting unit for combinations in lottery calculations. If a draw takes k numbers from a pool of n, C(n,k) gives the number of distinct possible drawn sets, provided each number can appear only once and the sequence is irrelevant. The coefficient counts outcomes; a probability requires comparing the relevant outcomes with the total.
The hypergeometric distribution
The hypergeometric distribution gives the probability of a specified number of matches when items are drawn without replacement from a finite pool. “Without replacement” means a number drawn once is not returned to the pool, so it cannot be drawn a second time in that same event. For a lottery-style calculation in which a ticket and the draw each contain k numbers from a pool of n, the probability of exactly m matches is:
P(X = m) = C(k,m) × C(n-k,k-m) / C(n,k).
Here, C(n,k) counts every possible set of k drawn numbers. The numerator counts the sets with exactly m matches: choose m of the k numbers on the ticket, then choose the remaining k-m drawn numbers from the n-k numbers not on the ticket. Dividing this count by the total number of possible draws gives the probability for that exact match count.
For example, let n = 10, k = 3, and m = 2. A ticket contains 3 numbers from a pool of 10, and the draw also selects 3. The number of favorable draws is C(3,2) × C(7,1) = 3 × 7 = 21: choose 2 of the ticket’s 3 numbers and 1 of the 7 numbers outside it. The total number of draws is C(10,3) = 120, so P(X = 2) = 21/120 = 7/40 = 0.175, or 17.5%.
This formula describes an exact match count, not “at least” that many matches. To calculate at least 2 matches in the same example, the outcomes for exactly 2 and exactly 3 matches would be added. The hypergeometric model is appropriate because the pool is finite and each draw removes a number; a model that assumes independent draws with replacement would describe a different process. For lottery mathematics, the practical choice is to define the pool, ticket size, and required match count, then apply the formula to those values.
Prize tiers and matching numbers
In a 6-of-49 draw without a bonus ball, a prize tier is defined by how many of the six drawn numbers appear on a ticket. Each exact-match count corresponds to a set of possible ticket combinations, and the size of that set determines its probability. The table and calculations below show how the Lottery winning odds differ across the three-, four-, five- and six-match tiers. These figures describe one ticket selection in a single draw.
Matching three of six
For exactly three matches, a ticket must contain three of the six drawn numbers and three numbers from the 43 that were not drawn. The same counting method gives the combinations for the higher tiers in this 6-of-49 example.
The odds column expresses the chance for one ticket selection; rounded figures are shown where the ratio is not a whole number.
| Matches | Combinations | Odds |
|---|---|---|
| 3 | C(6,3) × C(43,3) = 246,820 | About 1 in 56.7 |
| 4 | C(6,4) × C(43,2) = 13,545 | About 1 in 1,032 |
| 5 | C(6,5) × C(43,1) = 258 | About 1 in 54,201 |
| 6 | C(6,6) × C(43,0) = 1 | 1 in 13,983,816 |
The exact count for three matches is C(6,3) × C(43,3): there are 20 ways to choose three drawn numbers and 12,341 ways to choose three undrawn numbers, giving 246,820 combinations. Against 13,983,816 possible six-number tickets, that is approximately 1 in 56.7. A three-match result is consequently much more frequent than the higher exact-match tiers, although each individual ticket still has only one specified combination.
Matching four and five of six
The count for exactly four matches is C(6,4) × C(43,2). A ticket must select four of the six winning numbers, which can be done in 15 ways, then two of the 43 numbers outside the draw, which can be done in 903 ways. Multiplying those choices gives 15 × 903 = 13,545 combinations. Dividing the 13,983,816 possible tickets by 13,545 gives approximately 1,032.4, so the odds are about 1 in 1,032.
For exactly five matches, the calculation is C(6,5) × C(43,1). There are six choices for which drawn number is omitted from the ticket and 43 choices for its one undrawn number. The product is 6 × 43 = 258 combinations. Of the 13,983,816 possible tickets, 258 meet this exact condition, giving odds of approximately 1 in 54,201.6, or about 1 in 54,201.
The factor C(43,6-m) matters because the remaining ticket numbers must come from outside the six-number draw. Choosing only the matching numbers would count incomplete tickets; choosing numbers from all 49 without separating drawn and undrawn numbers would also include tickets with a different match count. The multiplication counts each valid ticket once: its selected drawn numbers identify the matches, and its selected undrawn numbers complete the six-number entry.
Why smaller prizes are far more likely
Lower-match tiers usually cover more possible tickets because there are many ways to combine a smaller subset of the six drawn numbers with numbers outside the draw. In this example, exactly three matches account for 246,820 combinations, compared with 13,545 for four matches and 258 for five. The three-match tier therefore contains about 18.2 times as many combinations as the four-match tier, while the four-match tier contains 52.5 times as many as the five-match tier.
The jackpot is a single combination: all six drawn numbers and none of the other 43. A five-match ticket can omit any one of the six drawn numbers and replace it with any one of the 43 undrawn numbers, producing 6 × 43 = 258 possible tickets. As a result, a five-match outcome is 258 times likelier than the jackpot in this draw format.
That comparison concerns two distinct outcomes, not an “almost-win” that moves a ticket closer to a future jackpot. A ticket matching five numbers has one number different from the drawn set; its probability comes from the 258 combinations that satisfy that exact condition. The broad pattern follows the combination counts: tiers allowing more possible selections are encountered more often, while the exact six-number match remains one specific arrangement.
Expected value of a ticket
Expected value (EV) puts the possible payouts and the cost of a lottery ticket on the same mathematical scale. It combines each outcome’s probability with its prize amount, then accounts for the ticket price. This gives a long-run average across repeated plays under the same rules, not a forecast of what one ticket will win. A hypothetical calculation can show how a jackpot contributes to EV before real-world payout conditions are considered.
What expected value means
Expected value is the sum of each possible outcome’s probability multiplied by its prize, minus the price of the ticket. In compact form, EV = Σ(probability × prize) − ticket price. The outcomes include winning prize levels and receiving no prize; each must be weighted by its own chance and payout. This calculation measures the average net result per ticket when the same lottery rules and ticket cost are considered.
EV is a long-run average, not a prediction for one ticket. A single draw produces one result, while the calculation combines all possible results according to their probabilities. Even if a prize outcome has a substantial contribution to EV, that does not mean an individual ticket is likely to receive it. Lottery probability describes the weighting of outcomes, not a guaranteed return or a schedule for when a prize will occur.
For a complete calculation, include every prize tier and subtract the purchase price once. A positive gross prize contribution alone is not net EV, because it leaves out both other outcomes and the cost of entry. In practice, most tickets have negative expected value: their probability-weighted payouts are lower than their price. EV is therefore useful for describing the average mathematical return, but it cannot identify the result of a particular draw.
A worked illustration with hypothetical figures
Hypothetical example: Suppose one ticket has a jackpot chance represented by one winning combination among 13,983,816 equally likely combinations, and the jackpot is $10,000,000. The jackpot’s probability-weighted contribution is $10,000,000 / 13,983,816 ≈ $0.72 to the EV of one ticket before other prizes, taxes, or sharing. This is the jackpot component only, not the ticket’s complete expected value.
Several real-world details can reduce the amount represented by that headline calculation:
- Shared jackpot: If more than one ticket matches the jackpot combination, the prize is split among winners, so a ticket’s expected jackpot payout is lower than the full advertised amount.
- Taxes: Tax on a prize reduces the amount retained from a winning payout; the advertised jackpot therefore does not necessarily equal the winner’s after-tax proceeds.
- Cash option: When a winner selects a cash payment instead of the advertised annuity, the cash amount can be lower because it reflects the present value of future payments.
- Ticket price: The purchase price is subtracted from the total probability-weighted prize value, so gross prize contributions must exceed that cost for net EV to be positive.
A full EV calculation would also add the probability-weighted value of every other prize tier, using the lottery’s actual payouts and rules. The $0.72 figure does not include those prizes, and it does not account for the effects of sharing or the cash option. Once the ticket price and payout reductions are included, most tickets have negative expected value. The illustration shows how a large jackpot can contribute to EV without making the jackpot amount a likely return on a single ticket.
Common probability mistakes
Lottery probability is easy to misread when ticket counts or past results are treated as stronger evidence than they are. Two common errors are confusing the chance from several tickets with the chance from one, and assuming previous draws influence the next result. For a 6/49 jackpot, each specific six-number combination has the same probability in a fair draw. Understanding how outcomes are counted helps keep statistical analysis focused on what the numbers actually show.
Odds for one ticket versus several tickets
For a 6/49 jackpot, the 13,983,816 possible combinations are equally likely. A ticket covers one combination, so the chance of matching the jackpot is 1 in 13,983,816. When comparing several tickets, count the distinct combinations they cover: probabilities can be added when the winning outcomes are mutually exclusive.
- One ticket covers one jackpot outcome, giving a probability of 1/13,983,816; the other possible combinations remain uncovered by that line.
- Two tickets with different number combinations cover two separate outcomes, so add their counts: 2 favorable combinations out of 13,983,816.
- The resulting chance for those two different tickets is 2/13,983,816, or 1 in 6,991,908; this is exactly twice one ticket’s chance.
- Two identical tickets still cover just one combination, leaving the jackpot chance at exactly 1 in 13,983,816, not two chances in the draw.
- For three distinct combinations, count three covered outcomes and use 3/13,983,816; this counting method applies to any collection of unique lines.
- Multiplying one ticket’s probability by another’s would describe both specified outcomes occurring, not either ticket winning; that is a different event.
The useful rule is to add the counts of distinct jackpot combinations, or their probabilities, rather than multiplying them. More unique tickets increase coverage, but the chance remains small and the purchase cost rises; participate only if you meet the legal age where you live, and set a fixed gambling budget.
Treating past draws as information
Independent draws have no memory: in a fair lottery, the numbers drawn last time do not alter the probabilities for the next draw. If a 6/49 draw is independent, any particular six-number combination remains a 1-in-13,983,816 jackpot outcome, whether or not it appeared in an earlier result. A combination that has not appeared for a long time is not “due,” and a recent repeat is not less likely solely because it just occurred. Past results can be summarized as frequencies, but those records describe completed draws rather than provide a dependable forecast of the next one. For example, seeing one number appear more often in a sample does not by itself show that the lottery favors it; random variation can produce uneven counts. Predictive modeling or numerical patterns cannot turn independent outcomes into a reliable sequence forecast. Before interpreting a streak or applying a pattern, check that the draw process is independent under the lottery’s rules and mechanism. Do not treat historical results as predictive evidence unless independence has been checked.