The two mistakes — false positives and false negatives.
Accuracy lumps every error together. But there are only two ways to be wrong — and they cost very different things.
When the umpire gets a call wrong, it is wrong in one of exactly two ways. It can give OUT when the batter was not out — a wrong dismissal, punishing an innocent batter. Or it can give NOT OUT when the batter was out — a missed dismissal, letting a wicket slip away.
Sort every decision by what the umpire said and what was actually true and you get four boxes: two right, two wrong. That grid is a confusion matrix, and it is the tool the whole rest of this lesson is built on.
Now the technical names — simpler than they sound. Treat a verdict of OUT, the finger going up, as the “positive”. A false positive is then a wrong OUT: the finger went up but the batter was not out — that is the wrong dismissal. A false negative is a missed OUT: the finger stayed down but the batter really was out — that is the missed dismissal. The word false just means the umpire got it wrong; positive or negative says which way the finger went.
Send a set of appeals through the grid and watch the four boxes fill — then see which two are the mistakes.
OUT
NOT OUT
Two ways to be right, two ways to be wrong.
Every decision lands in one of four boxes, and the diagonal is where the umpire was right. The other two are the mistakes — the false positive (a wrong dismissal) and the false negative (a missed dismissal). Accuracy just sums the diagonal, so a single number hides which of the two you are actually making.
That matters because the two errors rarely cost the same. A wrongly given wicket and a wrongly spared one have very different consequences — so the next step is to measure each on its own, which is exactly what precision and recall do.