Quote from dr_kang:
The original quote clearly said that a positive average profit must be guaranteed, in which case limited profits obviously implies limited losses (noting moreover that the standard deviation being small says even more than just that overall money is made). What intradaybill said about limiting losses is true, but it was implicit in the original quote, so I think it is unfair to call the authors stupid traders (especially since I, as a co-author of that paper, am neither stupid nor was I a trader at the time the paper was written).
The paper from which the quote came never went anywhere near saying that the Sharpe ratio is useless; that particular quote was a comment on how a competition, the Penn-Lehman automated trading competition, which everyone involved knew to be artificial, was won. There was no suggestion that Jump and Dump could form the basis of a good strategy in real life. I am not sure why the original poster trd was quoting us, but that quote is certainly not one I would have chosen as a good critique of the Sharpe ratio. There are many good critiques out there, and none of them would say the Sharpe ratio is useless!
The point may have not been made optimally, but it remains true that small but consistent profits may give rise to a large Sharpe ratio, but this number says nothing about the "capacity" of the strategy generating the returns. Obviously, small profits cannot always be increased by leverage (if the other poster can tell me how to do this, I would love to hear).
Capacity is only one issue and by far not the most important one (which is why the quote was ill-chosen): without knowing the trading strategy generating returns, one get's a rather limited picture from the Sharpe ratio. For example, by selling deep out-of-the-money options one may get a very high Sharpe ratio until the sh*t hits the fan and a massive loss is incurred.
Rahul
