CatLady,
<<One of the concerns I have about any mechanical shorting strategy is the possibility of
being caught short in a KTEL or ENMD type situation. Sleepy little companies that
become internet fads or discover a cure for cancer. Other recent examples, not quite as
extreme are MALL, SATH, NAVR and AWRE.>>
This has certainly been a consideration to me. I consider Float, TotalShares and TotalShort and so far stocks with the potential
to kill you short, have not found their way into the lowest
6 to 10 stocks. The worst short performer i recall, was an
86% winner (at end of holding period). I can't say what the
maximum loss was.
So this also relates to how many stocks one is long,
and how many short, if one is short at all. Certainly in the
markets of the last 13 months, being nakedly long was the thing to
do by a wide margin. Although being short would have
felt good last August - Oct 8.
Four or five stocks long is the smallest number where the
returns did not fluctuate wildly due to one huge winner
or loser. As someone else said here, I have found correlations
at the extremes of a distribution to be the most useful,
so I'm tempted to go long with 4. If I go short, I would use
at least 8.
On the long side, for a modest investor like me, I have found
even Schwab at $30 a pop gave no better fills (so far)
than a maximum discount broker like Scottrade (at $7 market orders)
which gave me 17 fills on the long side in these bigcap/big volume
stocks I get selected at totally fine prices given the bid..asked.
Doing short trades on low priced,low volume stocks is something I am not ready to yet. But the on-paper atrocious returns of such short selections gives me comfort that while the presence of what I seem to be measuring gives good returns, its absence gives poor returns.
As to stop loss orders, all my analysis has been done on the basis
of entering the market on a random day, then holding for
3, 6 or 12 months, and selling. I have no reason to think I can
outguess the market on where to sell after a loss.
|
