OMG…How to see if you are guaranteed to fail.
I just looked at a system that has a beautiful long term track record. This is a system that people told me is a golden one.
At the statistics window i pressed the right arrow and at the bottom of the window it told me P/L per trade it was under $50.
Mythbuster said "Example: mini S&P per contract - $5 commission, $10 in autotrade fees/costs, and say 2 ticks slippage ($25) in and out - which is enormous for that market - gives you $5 + $10 + $25 + $25 = $65 in C+S."
His comment was focused in the market that is probably more liquid than most of the contracts this unnamed system trades and the per tick costs are different. So when we add this all up lets call it cost per trade.
If the cost per trade in terms of slippage and other cost is over the amount in the P/L per trade. Failure is virtually guaranteed.
Since I am quoting you…Mythbuster…i will permit you to post again in my forum ;-))
Regards
Ron[LINKSYSTEM_50288258]
One note, the if you trade stocks and futures then the P/L per trade will not be applicable (as I have in my system). It’s good you are thinking these things but, in general, you have to look at the C2 stats in some detail. There is, also, a personalization for the equity curves. This too may not give the results one might expect. If one trades both stocks and futures because one can trade futures on margin whereas the stock trades are scaled.
Example, on my system I traded the SPY as a proxy for the ES mini during the beginning, this makes the returns for a smaller account somewhat less then they would be if one were to trade those with the futures.
The c2 score should, also, be taken in context in that it must be largely based on relative out-performance and not really the risk adjusted returns. Again, taking my results as an example, I could have achieved (to date) a roughly 30% return at a better then 1:1 R:R ratio, had I merely traded more contracts. Example, a system that returns 20% with 10% drawdown (and uses instruments that can be leveraged like the futures) could just have easily produced a 40% return with 20% drawdown. C2 score will reflect the latter with a higher score.
I’ve suggested this before but the only way to “truly” compare systems would be to use the minimum contract margins and assume a 1 contract* or to simply learn a bit about the system. Such a reporting measure would allow the subscriber to select their own risk to reward based on their risk appetite versus requiring the developer/trader to guesstimate what subscribers may desire. although no method will be perfect.
I’d suggest a developer recommendation while also allowing the subscriber choose and see a range of scenarios, as you suggested. I’d also suggest, especially with the current way of systems reporting, requiring the developer to state up front the max drawdown expected and being penalized (in some fair that doesn’t advantage c2 even more) should their system exceed that drawdown (i.e returning a portion of fees to subscribers).
Hmm, interesting concept, your last paragraph. But the devil would be in the detail. A close analogy in the managed fund world would be the 2 & 20 charging structure.
That’s where a fund manager charges a 2% fixed fee to cover costs (and running a regulated fund is expensive) plus 20% of any profit made over a year. Even better, in some investment firms the 20% performance fee is charged only if the investors’ equity is above the high water mark. This means that the investor does not pay the performance fee after a drawdown until the drawdown is fully recovered.
Therefore the 2 & 20 plus high-water-mark means the fund manager has an incentive to (a) make profits long term, and (b) take care to minimise drawdowns. Not perfect, but better than the “get paid whatever happens” model, or even worse “get paid to over-trade and/or gamble recklessly” model.
Now, if one dispenses with the 2% fixed fee altogether and just keeps the performance fee plus high-water-mark… well, I have never come across a “professional” money manager who would even consider such a radical step!
"I’d suggest a developer recommendation while also allowing the subscriber choose and see a range of scenarios, as you suggested. I’d also suggest, especially with the current way of systems reporting, requiring the developer to state up front the max drawdown expected and being penalized (in some fair that doesn’t advantage c2 even more) should their system exceed that drawdown (i.e returning a portion of fees to subscribers)."
I have recommended a few times in the chatter area to include maximum risk for open trades and maximum drawdown expected in the system details setup and possibly for trades to be automatically closed when this maximum is hit. I have seen a few subscribers complain bitterly about sudden huge losses in their accounts and this would possibly deter reckless trading and prevent huge losses .
"Now, if one dispenses with the 2% fixed fee altogether and just keeps the performance fee plus high-water-mark… well, I have never come across a “professional” money manager who would even consider such a radical step!"
Totally agree with a perfomance based model based on high water mark ONLY. There are a few “professional” money managers with that setup.
[LINKSYSTEM_38885500]
I’m pretty sure tying developer fees to performance (like 20% of profit, with or without high water mark) would violate US NFA and CFTC rules for publishers. Then, every developer would have to be registered.
Maybe Matthew can clarify.
Dean, Great Picture.
Good point. Although, if such a charging structure were available at C2 it would still only refer to the hypothetical returns as published, not applied to each subscriber’s personal circumstances. I suspect you’re probably right though.
Ha, thanks Ron!
That was from a series of photos we took around London to put on a calendar as a present for my sister who was moving to Australia. Hence the kangaroo 
