Futures allocation strategy

[ctd]

It's been a few years since I've been involved in futures markets, but a colleague proposed an allocation strategy in futures that seemed questionable and I wasn't sure what I was missing.



The strategy boils down to being longer futures contracts/markets that were in backwardation, more or less neutral to those that were flat and short those that were in contango. He claimed that in back testing he saw 3-4% yearly 'alpha' (relative to current strategy excluding this allocation bit). It seems to me this is a simple exploitation of the roll spread, which I remember being a common pitfall when getting started in futures trading (though off the top of my head, can't recall why at this point).



Can anyone fill in the holes of why and in what way roll spread needs to be accounted for in back testing, and/or shed some light on what I'm missing as far as how this would be a viable strategy for alpha generation?

[svquant]

Adding that much "alpha" is open to question but the basic strategy is reasonably sound but beware of the drawdown. There have been a dozen academic and research papers published over the last few years. You can find most of them via google or at ssrn itself.

[TonyC]

google the works of geert rouwenhorst