This thread is interesting given my efforts in growing my CTA. You would think that a high Sharpe is a good thing, but as it turns out that isn't the case with larger and more sophisticated clients. I've had periods of 4+ Sharpe in the past, but overall my program is around 2.5. I thought I would share some experiences here on risk metrics since it seems like there are some people following here with an interest in CTAs. First, it is actually rare for any due diligence team to ask about any of the risk metrics. They don't find them useful at all. In the end, they already have your performance capsule and can quickly do the calculation if they want to. But even then, it is not at the top of the list of things they are looking for. The only people who inquire about Sharpe ratios are usually HNW individuals who aren't really all that sophisticated. If the topic does come up in a meeting with the due diligence team at a family office or fund of funds, it is either as a 20 second tangent, or it is a loaded question. Knowing what I know now, I would never want to suggest to an experienced due diligence team that my Sharpe is even close to 4, let alone higher than 4. To most of them, a Sharpe above 4 indicates either fraud or a pending blow-up. The due diligence process then shifts from a standard interview into a full blown rubber glove probe. Even at 2.5 I get a slightly more "intense" and awkward discussion. I can't blame them really. I mean look at what that is suggesting. Depending on what period we are tracking, the S&P 500 has a Sharpe of about 0.4. So even a Sharpe of 2.5 is suggesting almost 6 1/2 times the returns for the same amount of risk. That already raises some eyebrows. In the end, the good teams don't care about Sharpe, they care more about expectancy. A quantifiable and demonstrable positive edge. Then of course there is the pedigree question and a whole ton of questions regarding how I expect to retain the edge into the future while competing with teams of Ph.Ds.
Thanks for sharing your experience of facing due-diligence teams. I guess you are correct that they will start becoming sceptical on hearing very high Sharpe. However, I find it absurd to compare S&P sharpe with sharpe of a systematic futures fund. S&P sharpe is simply buy and hold sharpe whereas sharpe of a systematic futures fund is a combination of sharpe of many diversified strategies, each exploiting an edge in the market in a very sophisticated form. Its obvious that Sharpe of even trendfollowers would be at least twice that of S&P. If that is the case, its no wonder that sharpe of a program, say which consists of 5 different strategies on multiple markets could be 4-6 times that of S&P. Also, due to a slightly different reason, it makes no sense to compare sharpe of S&P with say sharpe of a HFT operation or sharpe of a special situations fund, because the distribution of returns in both these cases is strongly non-normal.
Thanks for sharing your prospective and experience on Sharpe. I would encourage others to share their own experience. Personally I feel max DD peak to valley based on daily return is a much more important factor to assess account stability and blowup risk then sharpe.
I agree, but you have to keep in mind that you're assuming the manager of these strategies actually has an edge. If he doesn't, over the long term he will under-perform the broader market, typically by an amount very similar to what he paid in commissions and slippage. To demonstrate, what if your strategy was to buy ES at the open and sell at the close each day? This is day trading, but would likely mirror ES over the long term, less any commissions and slippage. On the same vein, if there is no quantifiable edge, a due diligence team must assume that it is a "feel" system. At least 95% of these "feel" systems turn out to be short term luck systems. IOW, they might as well just buy and hold SPY, because a luck based system is likely going to under-perform in the long run. So I guess my point is that risk metrics only mean something if the expectancy (edge) is quantifiable. But if you can state a quantifiable edge, and you know the MAX DD then the standard risk metrics are pretty much worthless.
Sharpe is back above 5. Sharpe is 5.33, max DD peak to valley is 3.38%. Also my return is reaching 50% for the first time this year. Will keep updating. I am thinking about taking money from some accredited investors in October, without having to launch a fund (my plan is not to charge a management fee, just a performance fee 40-60% for any monthly returns exceeding 1%) . What would the best way to set it up legally?
If you're trading futures, then CTA is the way to go. Lots of resources from the NFA examining the testing and disclosure requirements. If you're trading securities, the road is long and complicated, with or without a fund structure.
Yep. And if you want to use something out of academic finance (though it's not necessary), the Sortino Ratio is better than Sharpe IMO. I don't like how Sharpe punishes both upside and downside volatility the same way.
Whoah!!! 40-60% performance fee is VERY HIGH! So high in fact that your investors lose almost all the advantage of your program. Even though your returns are approaching 50%, your stated returns to the prospective investors would only be around 15% with that kind of fee. Is the risk still worth it for only 15%? Go ahead and calculate the Sharpe net of fees at 60% carry. You can't claim a 5 Sharpe if you charge a 60% incentive fee. Look at it from their perspective. 6% monthly with a 3.5% Max DD 3% monthly with a 3.5% Max DD The latter is not nearly as comfortable. Anyway, make sure that you fall within the appropriate exemptions. I don't know what you trade, so it is hard to say which regs apply. I will agree with Heech though. If you are trading securities, things get difficult. Even if you are exempt from federal regs, State regs can be difficult. If you only trade futures, then things get much easier.
BTW Macintash, I'm not trying to be a jerk here and I hate it when people here throw out speculative criticism, but now that you are suggesting managing OPM... Your posted equity curve and risk metrics are highly suggestive of a premium selling system, e.g. short option straddles, OTM credit verticals, etc... Any due diligence team that sees a 4+ Sharpe and an equity curve that is flat for a few weeks and then suddenly rises sharply just leading up to the 3rd friday of the month, won't touch that system. Ditto if each large drawdown is followed by an immediate large spike. These systems are notorious for long periods of incredible and stable growth followed by a single unfortunate blow-up about 10-20X larger than the previous Max DD. These losses are also notoriously difficult to manage in real time, both technically and psychologically. I'm not saying one way or the other, but if your system falls into this category, I would highly suggest that you reconsider managing OPM for now, especially as an unregistered fund. You might literally be weeks away from a >30% drawdown, accompanied by a lawsuit. Again, I wasn't gonna say anything, but just in case you fit that description...