Random does not mean shapeless
A single die roll is unpredictable. A thousand die rolls are not: the counts settle into a recognisable pattern. The same is true of sums, card draws and waiting times. Individual results scatter, but the spread of results has a shape, and knowing the shape tells you what is likely, what is unusual and what is nearly impossible.
This does not let you forecast the next independent result. It helps you set sensible expectations about a collection of results.
Probability trees: listing the possibilities
A probability tree draws each step of a process as branches, with a probability on each. Multiply along a path to get the chance of that sequence; add paths that lead to the same result.
Take two cards from a standard 52-card deck, without replacement, and note the colour. There are 26 red and 26 black cards.
- Red then red: 26/52 × 25/51 = 650/2652, about 24.5 percent.
- Red then black: 26/52 × 26/51 = 676/2652, about 25.5 percent.
- Black then red: 26/52 × 26/51 = 676/2652, about 25.5 percent.
- Black then black: 26/52 × 25/51 = 650/2652, about 24.5 percent.
The four numerators add up to 650 + 676 + 676 + 650 = 2652, so the tree is complete. Notice that two cards of the same colour are slightly less likely than two of different colours (1300/2652 against 1352/2652). Removing a red card makes the next red a little rarer, which is the effect of drawing without replacement. The tree makes that visible in a way a single formula hides.
Counting up: the shape of repeated trials
Flip a fair coin three times. The tree has 2 × 2 × 2 = 8 equally likely paths. Group them by number of heads:
| Heads | Paths | Chance |
|---|---|---|
| 0 | 1 | 1/8 |
| 1 | 3 | 3/8 |
| 2 | 3 | 3/8 |
| 3 | 1 | 1/8 |
The counts 1, 3, 3, 1 come from Pascal’s triangle, and they follow the binomial distribution. The middle values are most common because there are more paths to them. Extend the idea to ten flips and the central counts become much more likely than the extremes, while the chance of all ten heads is 1/1024, below 0.1 percent.
Sums of dice: a built-in hill
With two dice, ways to make each total from 2 to 12 are 1, 2, 3, 4, 5, 6, 5, 4, 3, 2, 1, adding to 36. That triangle peaks at seven. Adding more dice makes the hill smoother and more bell-shaped, a pattern described by the normal distribution. Totals of many independent pieces tend to bunch near the middle, with the tails thinning quickly.
How wide the hill is can be measured by standard deviation. A wider spread means results wander further from the average, even if the average is the same. That is the distinction behind two games with equal averages feeling very different.
A skewed shape: waiting for a first success
Not every distribution is symmetrical. Count how many rolls it takes to see a first six.
- First roll: 1/6, about 16.7 percent.
- More than six rolls needed (no six in six rolls): (5/6)^6 = 15625/46656, about 33.5 percent.
The single most likely answer is one roll, yet the average wait is six rolls, because a long right-hand tail of unlucky waits pulls the average up. Where a result is bounded below but can stretch upwards, the shape is lopsided, and the typical case and the average differ. Expected duration questions often have this feature.
Using shapes to judge what is likely
Shapes make several judgments easier:
- Is this result unusual? If results bunch near the centre, a value far in the tail deserves attention, though over enough trials an occasional tail value is normal.
- How much should totals vary? Totals over many independent events vary less in proportion than single events.
- Which assumption am I making? If the events are not independent, the simple shapes above may not fit.
The results of conditional probability are another useful check: when new information arrives, redraw the tree rather than adjust by instinct.
What shapes cannot do
Knowing a distribution describes how outcomes are spread across many plays. It does not tell you which outcome arrives next, and it does not make a game with a house edge profitable. In casino-style games, the shape of results shows how a player can be ahead in the short run, but the average still favours the house. If play stops being enjoyable, or you notice you are staying in to recover losses, consider stopping and speaking to a local support service.