Consider this theoretical trading system (performance after all trading costs): Win rate: 40% Average winner: 2% Average loser: 1% A trading system with the above metrics is not an outstanding trading system, it is pretty mediocre in fact (In the calculations below i am going to ignore the risk free rate and also ignore compounding effects of profits and losses) Based on number of trades taken with the above system per year, the results are as follows: 50 Trades per year: Average Return: 10%, Stddev 12%: Sharpe Ratio: 0.8 100 Trades per year: Average Return: 20%, Stddev 17%: Sharpe Ratio: 1.2 Based on number of trades taken per DAY (Assuming 250 trading days a year) 1 trade per day: Average Return: 50%, Stddev 27%: Sharpe Ratio: 1.8 2 trade per day: Average Return: 100%, Stddev 38%: Sharpe Ratio: 2.6 5 trade per day: Average Return: 250%, Stddev 62%: Sharpe Ratio: 4.0 10 trade per day: Average Return: 500%, Stddev 87%: Sharpe Ratio: 5.8 As you can see from these results, Sharpe ratios above 2 and 3 are possible when day trading, even when using a mediocre system. It is all about getting enough trading opportunity.

It's all about knowing HOW to trade. Knowledge is the key. With knowledge you will see enough trading opportunities. I compare it with rubic's cube. Easy to solve, but for most people however impossible to do. Looks like a contradiction, but is not in reality. Knowledge is the key. For those who find video above a piece of cake: Nobody knows where the limits really are for humans. Even in trading we don't know where the limits are.

Normally you either use the spreadsheet builtin stdev function or google the code for your favourite programming language.

The only problem is the smaller the timescale, the stronger the competition is. Everyone is competing for the high sharpe stuff...

The Sharpe Ratio is considered to be laughable by professional traders because it punishes upside volatility as if it was downside volatility. There are better metrics for the lumpy returns that are inherit with trading such as: profit-factor, closed NAV, Calmar, and gain-to-pain to name a few.

Stddev and Sharpe ratio have a place. Especially if you have an automated trading system. They give an indication of the kind of statistical dispersion you can expect over a years worth of trading.

Well if your strategy has too much upside volatility then that is a happy problem. Sadly most strategies don't fall into that case which is why sharpe is still heavily adopted. If you have a > sharpe 1 strategy that is consistent and scalable you can run at least a 1B fund already.

"it punishes upside volatility as if it was downside volatility" In general the more upside volaltiliy the more down side as well. Big rewards generally come with bigger risks. The only place where this doesn't happen is with curve fitted trading systems that have lots of upside and little downside in the back test. The sharpe ratio will help you from deluding yourself. Even if based off real trading, it might just be that you been lucky and haven't seen a big downside period yet.