Total return is the least informative number you can quote about a strategy. It says nothing about how much risk was taken to get it, how long the bad stretches lasted, or whether the result came from one lucky trade.
These are the measures that fill that gap. Each answers a different question, and each can be gamed, which is why they are read together rather than one at a time.
The raw profit and loss over a period, usually shown as a percentage of capital and annualised for comparison. On its own it is unusable for judging quality: a 40% return from one leveraged trade and a 40% return from four hundred small ones are completely different businesses.
Beta measures how much of a strategy's movement is explained by the market it trades. Alpha is what is left over. A crypto strategy that is long most of the time will have high beta to Bitcoin, meaning most of its return is market exposure you could have had by simply holding, not skill.
Return above the risk-free rate divided by the standard deviation of returns. It asks how much return you got per unit of volatility. Its weakness is that it punishes upside volatility exactly as hard as downside, so a strategy with occasional large gains scores worse than one that grinds.
The same idea as Sharpe but dividing only by downside deviation, so profitable spikes are not treated as risk. For most trading strategies it is the more honest of the two, since nobody has ever complained about an unexpectedly good month.
Annualised return divided by maximum drawdown. It answers the question traders actually care about: how much did I make relative to the worst loss I had to sit through. A Calmar below 1 means the peak-to-trough loss exceeded a year of returns.
Gross profit divided by gross loss. Above 1 is profitable; below 1 is not. A profit factor near 1.1 is fragile, because small increases in fees or slippage can push it under, which is why execution cost matters so much to high-frequency strategies.
The percentage of trades that closed profitable. It is the most quoted and least meaningful metric in trading. A 90% win rate is trivially achievable by taking tiny profits and letting losers run, and it is exactly how many accounts are destroyed.
Average win times win rate, minus average loss times loss rate. This is the number win rate should have been: it tells you what you expect to make per trade. A strategy winning 35% of the time with a 3:1 payoff has positive expectancy and will feel like losing most of the time.
The largest peak-to-trough fall in account value. It is the single most useful risk number because it is the one that determines whether you abandon a strategy. A 60% drawdown requires a 150% gain to recover, and almost nobody keeps a strategy running long enough to find out.
Net profit divided by maximum drawdown. It measures how efficiently a strategy digs itself out of holes. Two strategies can share a return and a drawdown while one recovers in weeks and the other in a year.
The full shape of returns rather than their average. Crypto returns have fat tails, meaning extreme outcomes occur far more often than a normal distribution predicts, which is why risk models built on averages understate what can actually happen.
Metrics recalculated over a moving window instead of the whole history. A strategy with a good lifetime Sharpe can have been flat for two years inside it, and only rolling windows reveal that.
Measuring a strategy against the obvious alternative, which in crypto is usually holding Bitcoin. If a strategy underperforms buying and holding after fees, its complexity is not earning anything, however good the standalone numbers look.
Breaking results down by source: which strategy, which asset, which market condition produced the return. Without attribution you cannot tell whether you have five working strategies or one working strategy and four that are along for the ride.
Every backtest and every live strategy reports these rather than a single profit number, because the whole point of automating is being able to judge a system on evidence instead of memory.
Log in, pick a strategy, and watch it run against live markets.
