Mad Skillz [part 3]
Wait, did we forget something in the last post? Ahhhh yes, just hold up a minute there bro, there is something else to consider. Variance, as always is just as, if not even more important than returns. I’ve added two inputs into the spreadsheet, long term and short term standard deviation. This is a simple estimate of the annual standard deviation of each of the strategies’ returns for any given year. The strategies must now be compared using some kind of risk reward ratio and the measure that I’ve chosen is the Sharpe ratio which is literally the annual return divided by the annual standard deviation (risk free rate of return is supposed to be subtracted but I’ve omitted this for simplicity’s sake). The higher the Sharpe ratio the better the strategy. In the table below I’ve set the long term standard deviation to 20% which is close to what the SP500 is like, and the short term strategy to 15%. The results are astonishing, much less return is required to have a risk adjusted return equivalent to the long term strategy. Of course, reducing a portfolio’s variance isn’t an easy task but this illustrates how focusing on risk, and not just returns, can be extremely beneficial. It reminds me of the meat-heads at the gym that only work out their upper bodies and end up molding themselves completely out of proportion; chicken legs and a ridiculously huge V-shaped upper body. Legs build a strong foundation and support for your body and just building up huge biceps and a huge chest to look good ends up making you look silly. Not to mention the fact that your center of gravity ends up being so high you get knocked over easily. You see, that’s just like ignoring risk in your portfolio. If all you do is try to get some insanely high returns you’ll just end up looking like that V-shaped, twig leg mofo at the gym who keeps falling over. Stop trying to take short cuts and only workout half your body. Work out your legs dude.
| # of Trades | Excess Skill |
|---|---|
| 1 | 0.00% |
| 2 | -0.87% |
| 3 | -0.79% |
| 4 | -0.68% |
| 5 | -0.58% |
| 10 | 0.00% |
| 15 | 0.59% |
| 20 | 1.19% |
| 25 | 1.79% |
| 50 | 4.78% |
| 75 | 7.78% |
| 100 | 10.78% |
| 200 | 22.78% |
| 300 | 34.78% |
| 400 | 46.78% |
| 500 | 58.78% |
| 1000 | 118.78% |
| 2000 | 238.78% |
| 3000 | 358.78% |
| 4000 | 478.78% |
| 5000 | 598.78% |
Download the spreadsheet below if you would like to play around with the assumptions I made.
EdgesBasedOnTimeHorizonWithVariance
Mad Skillz [part 1]
Mad Skillz [part 2]
Mad Skillz [part 3]
