As a real investor with real money, the one metric that I am the most interested in (which I initially mistakenly thought that APD was describing) is . . .
Average (annual) Profit to (maximum lifetime) Drawdown.
This number tells me at least the minimum amount of money I have to be willing to risk in the hope of making the historical returns with any particular system, as well as the efficiency of a particular system in generating its profits relative to the losses that must be borne along the way.
I would appreciate having the option of being able to rank systems in the Grid using this metric. The good news is that no change to the website would be required, other than merely changing the definition of APD to "Annual Profit to Drawdown" defined as "the average Annual Profit divided by the maximum lifetime since inception Drawdown (and adding that definition to a new or easier to find definitions page on the site).
Thank you!
P.S. For the many systems on the site that are younger than one year old, you could simply annualize their returns, by for instance, dividing the inception-to-date rate of return, by the number of weeks since inception, and then multiplying by the 52 weeks in a year.
In addition to that I would like to see in the Statistics Section:
1) Below the "Compound Annual%" show the APR ( simple arithmetic annualized return), that would at least ameliorate the bizarre compound returns presently shown for .systems with early success.
2) Show Maximum Drawdown in dollar terms rather than, or in addition to, percentage drawdown.
APD compares average per trade profit to average per trade drawdown. Max life of system drawdown is already visible. It was aimed at traders who would game the equity curve.
The APD was to unmask systems that habitually would refuse to take losses, but allowing a trade to get a huge drawdown in hopes of getting back to profit (Hold&Hope). It also would show systems that do a lot of averaging down on trades in a bad light.
But that is why there are different statistics, to try and expose systems that had serious flaws, gaming of the equity curve, bad money management, etc.
In no particular order, I look for:
High profit factor (anything with a 1.2 is pretty useless, once slippage, commissions, system fees, risk, etc. are factored in).
Sharpe and APD, to look for systems that are reckless or use trickery for profitability.
Other systems the vendor has had (including killed systems). In fact, I was the one of several that pushed that for killed systems to be kept visible, so vendor couldn’t keep throwing systems up hoping for one to work.
Life of system, #trades, equity curve appearance.
Frankly, the pickings are usually slim for people hoping to autotrade a $4,000 account and safely get 20% a month. To semiquote the Hindenburg reporter, "O the ignorance."
But there is always a couple that seem promising…
"Average (annual) Profit to (maximum lifetime) Drawdown."
This is calmar and can be found in adv stats.
I agree in that it SHOULD be in "The Grid" column heading and perhaps in the system page stats.
gA
The max drawdown doesn’t tell the whole story. Psychologically you have to take into account positions held at the time of drawdown. This is what drives people out of the system.
What I would like to see is average annual profit / (Max drawdown + margin requirement at time of max drawdown). Call it a sweat indicator if you would like.
