Wall Street vampire squids suck the life out your compounded returns with their high fees. How else do they pay for offices on Central Park; hot receptionists; zebra hide sofas; and thoughtful paintings on every wall?
But, you know, if fund managers were really vampires they would live for centuries and make a lot of money on boring long term investments.
They might show off incredibly long backtests (Vamp Cap has amazing data resources!) and would guarantee amazing returns over super long future tie horizons.
As I wrote in the 'Murdering Hitler Portfolio' post, long backtests are better left for the immortals amongst us.
What do successful hedge funds actually do though? They leverage.
Sure it ratchets up the risk, but real hedge funds actually do hedge! They home in on tiny opportunities, hedge out the extraneous risks and leverage (cf. microscopic momentum).
Using maths in this way is a like bringing an Ironman suit to a knife fight.
Any small opportunity can be specifically homed in on and leveraged to the maximum.
This is why a strategy's average returns don't matter at all!
And, why, when you are comparing strategies you should equalise average returns by applying leverage.
Fixed income and equity exist on different plains. Fixed income is the most reliably quantifiable area in finance, and yet the returns are often miniscule.
Now we can make a comparison, just by eyeballing the data.
Seeing as the returns are equal, we only need to compare volatilities in order to calculate a Sharpe ratio.
SHY is far less volatile.
SHY has a Sharpe of 1.5 over 10+ years, whereas the S&P 500 has a Sharpe of 0.3.
If you wanted the data to tell you an even clearer story you should demean both time series, which we can do now because they have the same total return.
Now we have two Sharpe Trajectories and not only are both strategies directly comparable, but returns are intertemporally comparable.
It is now far easier to understand the story what the data is telling us.
Statistics is all about making fair comparisons between data sets and intertemporally, and this is exactly how leverage, demeaning and Sharpe Trajectories help us understand backtests better.
[I updated this after a mistake was spotted, see here]