Why doesn’t a result become “due” just because it has not happened lately?

That feeling can come from expecting a sequence to balance out immediately. But in a process with independent trials and fixed chances, previous results do not change the next trial’s chance. Treating an outcome as more likely simply because it has been absent is the gambler’s fallacy. [S1]

Imagine a fictional coin-flipping process. Assume each flip is independent, with a one-half chance of heads and a one-half chance of tails. If the sequence so far is tails, tails, tails, tails, the next flip is still assumed to have a one-half chance of heads. The streak does not make heads “due.” These are assumptions for the example, not claims about any particular real coin.

A longer sequence may look more balanced without the next result being forced to correct the earlier run. In the fictional example, consider the fifth flip: the four tails are part of the history, but they do not change the assumed chances for that flip. Heads is not guaranteed or made more likely by the streak; each side still has a one-half chance under the example’s assumptions.

That distinction is about the next trial, not about whether a long sequence will look balanced. The example illustrates a reasoning error, not a betting method or a way to predict a particular result. It also does not establish that every real-world process has fixed chances or independent trials.

Ask yourself what, specifically, would change the next trial’s chance. Separate the observed history (“tails appeared four times”) from a possible mechanism that affects the process. If your only reason for expecting heads is that it has not appeared lately, that expectation does not follow from the independent-trials model.

A quick check:

1. Name the outcome you think is “due.”
2. State the reason you think its chance has changed.
3. Ask whether that reason describes how the process works, or just how unusual the streak feels.
4. If you are assuming independence and fixed chances, keep the next-trial chance separate from the past sequence.

For a real process, do not assume those conditions just from seeing a streak; this example does not determine its chances.

You could say: “If each trial is independent and the chances stay the same, a streak doesn’t make an outcome due. In our fictional example, even after four tails, the next flip is still assumed to have a one-half chance of heads and a one-half chance of tails. That describes the chance, not a guarantee about what happens next.”

Keeping the assumptions in the explanation helps avoid turning a model into a prediction. The course discussion supports the distinction between independent draws and the mistaken idea that a long-unseen outcome is more likely next; it does not establish the properties of any particular real-world process or offer betting advice. [S1]