Every strategy, no matter how clever, comes down to a few statistics: how much it makes on average, how much it swings, and how often it wins. The trick is knowing which of those matter, and which one lies.
About 2 min of reading
The average return is how much money a strategy makes per year on average. It's the first number everyone looks at, and it's the worst place to stop. Without the risk, it's just a number that looks good on a slide.
Volatility is how much the value swings up and down. A strategy that jumps 30% in a day is risky whether or not it wins. Most of quant work is really just the search for good returns that don't make your stomach drop.
Strategy A: +10% per year, but swings of 30%
Strategy B: +6% per year, swings of only 5%A is better at making money. B is much safer. Most people wouldn't sleep well holding A, because it can vanish fast. Higher returns come with higher risk, the two are glued together, and it's usually the calmer one you want.
Average yearly return: 8%
Annual swing (volatility): 18%Volatility is how bumpy the ride actually is. An 8% return on a strategy that swings 18% a year feels very different from an 8% return that barely moves. The reward only starts to look worth it once you hold both numbers in the same hand.
Strategy wins 60% of the time
But its losses are bigger than its gainsA 60% win rate sounds great until you realize it's still losing money. How often you're right barely matters if the losses you keep eating add up to more than the wins you celebrate. Win rate is the number everyone notices; the size of the losses is the number that decides your fate.
Each one is defined here on this site. Open the entry for the full definition and the other words it points at.