Main benefit of CTA type strategies is the diversification benefit particularly in tail events (so Sharpe Ratio is a poor measure as it's symmetric, and any kind of CAPM type analysis isn't much better as it assumes linear correlations). There are numerous studies explaining this.
Basically you would expect to pay for an insurance policy that will protect you when the world ends and the rest of your portfolio tanks. If a CTA actually makes you money over time then that is a bonus.
The only other quantitative strategies that made money in 2008 were short biased equity (obviously) and tail protection type strategies (Taleb style smart buying of vol essentially). [Oh and Paulson et al; but that's hardly a repeatable methodology] Both of these have much lower SR than managed futures. That doesn't mean you shouldn't have them in your portfolio, but I'd probably give them less weighting.
100% in managed futures obviously wouldn't make sense, but a 20% allocation to a classic long only 80:20 portfolio will improve things significantly.
GAT
Disclosure; I used to work for AHL (another managed futures shop), I use managed futures type signals in my own trading (alongside a long only portfolio - as that is what makes sense), and I know (and respect) Ewan and many of his guys. So yes, I'm biased.
