How to use this glossary

This page collects terms that appear most often in discussions about lottery odds and probability. Each entry includes a plain-language definition, an everyday example, and a lottery-specific note. Terms are ordered to build on each other, starting with foundational concepts and moving to more nuanced statistical ideas.

Odds

Definition: The ratio of favourable outcomes to total possible outcomes. Often expressed as "1 in X."

Everyday example: A standard deck has 4 aces in 52 cards — odds of drawing an ace are about 1 in 13.

Lottery context: A 6/45 game has top-division odds of 1 in 8,145,060. Oz Lotto's 7-from-45 format yields approximately 1 in 45,379,620. Comparing odds across games reveals what each ticket represents.

Probability

Definition: A number between 0 and 1 (or 0% and 100%) that represents the likelihood of an event. A probability of 0 means impossible; 1 means certain.

Everyday example: The probability of a fair coin landing heads is 0.5, or 50%.

Lottery context: The probability of winning a 6/45 jackpot with one ticket is approximately 0.0000123%, or 1.23 × 10⁻⁷. If every person in Sydney bought one ticket, the probability that at least one wins would still only be about 65%. Probability gives you a precise handle on likelihoods that intuition struggles to grasp.

Combination

Definition: A selection of items from a set where the order does not matter. Denoted C(n, k) or "n choose k."

Everyday example: Choosing 3 books from a shelf of 10 (where you don't care about the order you pick them) is a combination problem: C(10, 3) = 120.

Lottery context: The number of unique ticket combinations defines the total outcomes in a draw. Adding one ball to the pool — from 6-from-45 (8,145,060) to 6-from-46 (9,366,819) — increases combinations by over 1.2 million. The formula makes the relationship between game structure and difficulty transparent.

Independence

Definition: Two events are independent if the occurrence of one does not change the probability of the other.

Everyday example: Flipping a coin twice: the result of the first flip does not affect the second. Each flip is independent.

Lottery context: Each draw is independent of all previous draws. A number that appeared last week is neither more nor less likely to appear this week — the draw mechanism has no memory. Understanding independence is the key to seeing why "overdue number" strategies have no mathematical basis.

Variance

Definition: A measure of how spread out the outcomes of a random variable are around its expected value. High variance means outcomes are widely scattered; low variance means they cluster closely.

Everyday example: If you commute for 30 minutes most days but occasionally face a 90-minute traffic jam, your commute time has moderate variance.

Lottery context: Lottery outcomes have very high variance — most tickets win nothing, a few win small prizes, and an extremely rare few win large prizes. Tracking 100 tickets would likely show total returns well below expected value, punctuated by occasional small wins — a direct manifestation of high variance.

Expected value (EV)

Definition: The long-run average result of a random process, calculated by multiplying each outcome by its probability and summing the results.

Everyday example: A vending machine gives a $2 item 90% of the time, nothing 10%, for $1.50 per press. EV = (0.9 × $2) + (0.1 × $0) − $1.50 = $0.30 gain per press.

Lottery context: A typical Saturday Lotto ticket (~$1.35) has an EV of roughly $0.55 to $0.75. The gap represents the cost of participation.

Distribution

Definition: The pattern describing all possible values of a random variable and how often they occur. Can be uniform, normal (bell curve), skewed, or many other shapes.

Everyday example: Human heights follow a normal distribution — most cluster near the average, with fewer very tall or very short individuals.

Lottery context: Lottery outcomes follow a heavily right-skewed distribution: an enormous spike at zero, a small bump at the lowest division, and an almost invisible tail to the jackpot — illustrating why the average experience differs dramatically from the headline prize.

Sample size

Definition: The number of observations collected in a study or analysis. Larger sample sizes produce more reliable estimates of the underlying parameters.

Everyday example: Asking 5 people their favourite restaurant gives a rough sense of preferences. Asking 5,000 gives a much more reliable picture.

Lottery context: If a number appears unusually often over 20 draws, that is a very small sample — likely explained by normal variance. Over 2,000 draws, the same deviation would be far more noteworthy. Many online frequency analyses use sample sizes too small to support their conclusions.

Standard deviation

Definition: A measure of the typical distance between individual data points and the mean. It is the square root of variance. Low standard deviation means tight clustering; high means wide spread.

Everyday example: Two students averaging 75% — one scores 70–80% consistently, the other swings 50–100%. Same average, very different standard deviations.

Lottery context: In a 6/45 game over 200 draws, each number's expected frequency is about 26.7, with a standard deviation of roughly 4.7. Seeing any number appear between 17 and 36 times is within the normal range. A number appearing 35 times may look suspicious, but falls within the bounds of what randomness routinely produces. Standard deviation provides the ruler for measuring whether an observed frequency is genuinely unusual.

Regression to the mean

Definition: The tendency for extreme measurements to be followed by measurements closer to the average — not because the process corrects itself, but because extreme values are less common by definition.

Everyday example: A cricketer who scores 150 is unlikely to repeat it next match — not because they have worsened, but because 150 is an outlier.

Lottery context: If a number appears in five consecutive draws, regression to the mean predicts its future frequency will likely be closer to the expected rate — not because the draw suppresses it, but because sustained extremes are statistically uncommon. This is often confused with the gambler's fallacy. The fallacy says the process compensates for past outcomes; regression to the mean says extremes are unlikely to persist because they are, by definition, rare. The probability per draw does not change.

Confirmation bias

Definition: The tendency to seek and remember information that confirms pre-existing beliefs, while overlooking contradicting evidence.

Everyday example: Believing a car brand is unreliable, you notice every breakdown on the roadside while overlooking thousands that drive past fine.

Lottery context: A player who believes number 23 is "lucky" vividly recalls draws where 23 appeared and forgets the many where it did not. Confirmation bias makes ineffective strategies feel effective by filtering evidence. The antidote is systematic record-keeping: tracking all predictions and outcomes reveals the true hit rate.

Law of large numbers

Definition: As the number of trials increases, the observed average converges toward the expected value. It requires a large number of trials and guarantees nothing about small samples.

Everyday example: Flip a coin 10 times and you might get 7 heads. Flip it 10,000 times and the proportion will be very close to 50%.

Lottery context: Number frequencies even out over thousands of draws, but not because "cold" numbers are due. Convergence happens because new data swamps old deviations — the evening-out occurs through dilution, not compensation. The past deficit is not erased; it becomes trivially small relative to the growing total.