GlossaryTraining

EV loss per 100

How many big blinds your mistakes cost, on average, for every 100 decisions graded by EV.

Last updated

What it means
The headline number for how well you play. It adds up the EV you gave away and scales it to 100 decisions, like a win rate in reverse.
How it is calculated
EV loss per 100 = total EV loss ÷ EV-graded decisions × 100. Only decisions with EV data count, so frequency-graded chart spots cannot dilute it.
Worked example
Over 250 push/fold decisions you gave away 30bb in total. 30 ÷ 250 × 100 = 12bb per 100. Next week it is 6bb per 100: your mistakes cost half as much.
Why it matters
Accuracy treats folding aces and a close mistake as equally wrong. EV loss weighs each mistake by what it costs, so it is the number to drive down.

See also: EV loss, Accuracy, Thin sample