The trigger finger — precision and recall.
Two officials, the same accuracy, opposite habits. One number can't tell them apart. Two sharper ones can.
The confusion matrix gives you two questions far sharper than accuracy. Precision asks: when the VAR chalks a goal off, how often is it truly a bad goal? — of all its NO GOAL calls, the share that were right. Recall asks: of all the goals that should not have counted, how many did it actually catch?
They are not the same, and a model can be strong at one and weak at the other. Here are two officials judging the same twenty appeals. Both score 70% accuracy — identical. Yet one is trigger-happy and one is cautious, and precision and recall pull them apart at a glance.
High precision means few good goals chalked off — you can trust a NO GOAL call. High recall means few bad goals slip through — invalid goals rarely stand. As you will see, each VAR has bought one by giving up the other.
Switch between the two officials and watch accuracy hold at 70% while precision and recall trade places.
Precision and recall see what accuracy hides.
Same 70% accuracy, opposite officials. Precision — of the NO GOAL calls, how many were right — rewards not crying wolf. Recall — of the goals that should not have counted, how many were caught — rewards letting none slip. The trigger-happy VAR wins recall and loses precision; the cautious one does the exact reverse.
So report both, never accuracy alone. A single number can look healthy while the model is quietly making the one mistake that matters to you. Which of the two you should favour — precision or recall — is the last question of this lesson.