Price movements are random and are not random, at the same time.
High Sharpe ratios are predictive of huge blowups.
The more stable the return, the more likely the blowup.
Volatile funds lose money; but not as much as nonvolatile ones.
Because of the past tense use of the word “traded” I’m left to assume this system doesn’t trade anymore...
“When you're a trader you get a lot of calls from folks who found relationships that can produce a 10 Sharpe ratio. That means it's almost impossible to lose money on the trade. Sure enough, when you start trading you realize that the relationship was not there. Trading has less biases than statistics.” - Nassim Taleb from
https://merage.uci.edu/~jorion/oc/ntaleb.htm
Yea, price movement can be more predictable when signal emerges, in the case of the news of trump or China or whatever it was you linked above. But you’re drowning in noise majority of the days of the year. Prediction becomes easier when there is signal. Still, it is induction and you cannot predict the future where such prediction would matter the most even if you had a million data points. Hence why (maybe) your “successful” system that trades 64% win rate doesn’t trade anymore.
Don’t get fooled, of all things, by Sharpe ratios!!!!
(Below is simply for more info)
From:
https://merage.uci.edu/~jorion/oc/ntaleb.htm
DS: Are all correlations suspect?
NT: You can find a relationship between any two items if you look hard enough. It will be entirely spurious and have no predictive power but you will find one. To give you an idea, you'll always find what we call data miners who will show you that there is 100 per cent correlation between his great aunt's blood pressure and the back month Nikkei volatility. When you're a trader you get a lot of calls from folks who found relationships that can produce a 10 Sharpe ratio. That means it's almost impossible to lose money on the trade. Sure enough, when you start trading you realize that the relationship was not there. Trading has less biases than statistics.
DS: What are the most common mistakes you see traders and risk managers making?
NT: As a trader, my job is to understand biases and trade on them. There are all kinds of biases. The most common is the small sample bias. Let's say you have 1 to 1000 odds you will come home every day with a dollar and once in a while you lose $1000. Many traders show very steady incomes but they could be fooling themselves because they don't have a long enough period of time to chart their performance. Their Sharpe ratio will not be indicative. In option trading, there is a similar bias. Short premium option traders, typically those who sell out-of-the-money options, are more likely to make money on a daily basis and then blow up. Likewise the yield hogs, those traders who would take any risk for a few basis points. You can fool yourself with your Sharpe ratios, and you can fool all of the financial engineers, but you can't fool an old Chicago trader who went bankrupt twice.
Another bias is what I call the size bias. If you have twenty thousand traders in the market, sure enough you'll have someone who's been up every day for the past few years and will show you a beautiful P&L. If you put enough monkeys on typewriters, one of the monkeys will write the Iliad in ancient Greek. But would you bet any money that he's going to write the Odyssey next? You know that because of the sheer size of the sample, you're likely to find a lucky monkey once in a while. But the same applies to traders.
A third bias is the survival bias. Everybody will tell you that stock investing is a great idea because it's been back-tested by some serious Guru and if you bought one share of some stock during the revolution you would have owned the GNP of some banana republic. But you forget that your back testing is only on stocks that are alive today and did not cover stocks in imperial Russia that a rational investor would have bought at the beginning of the century. Many continental stocks were recycled into wallpaper. When you look at markets you are only looking at the remnants, the parts that have survived. Or take real estate. People always say it goes up. But that works only if you always bought in places that became fancy.
Regarding your comment about your A grade in statistics. Congratulations on that, that is good for you.
P.S. don’t allow win rates to fool you. Stating 64% win rate, by itself, means nothing. & luck absolutely had a factor in it, same way Jim Simons attributes the role of luck to Medallions Funds success & to his 2 year 12x return fundamental analysis driven fund in his early days. But of course luck didn’t play a role with your returns...
Source: