Quote from Trend Fader:
Just to clarify my earlier post...
Your system is dependent on making a lot of trades... dependent on getting a good fill.. and dependent on transaction costs per a trade..
Although your transaction costs can be minimized with a broker like IB... you are doomed to fail because you are trading illiquid stocks which is contrary to getting good fills.
--MIKE
MIKE -- are you sure illiquid stocks make it hard to get good fills? Is
there something wrong with the idea of going for possibly great --
albeit infrequent -- fills?
For example, on Tuesday, zorak lists SOI long at $2.13. On Wednesday,
he doesn't list SOI so this means we're supposed to get out. His paper
results get out at the closing price, $1.99. That's a -6.57% return,
not including commission.
Now one way to try to trade this would be to scatter some orders and see
if any or some or all of them fill. On Tuesday at 3:32pm, SOI is quoted
at 2.13 x 2.15. So bid 2.14 for 1000, 2.13 for 2000, 2.12 for 3000, and
2.11 for 4000. If I'm reading the tape right, you'll fill at 2.14, 2.13
and 2.12 but not at 2.11 ... total of 6000 shares at an average price of
2.1329 including Interactive Brokers commission.
On Wednesday at 3:32pm, SOI is quoted at 2 x 2.02. So scatter some
orders again: 3000 at 2.01, 2000 at 2.02 and 1000 at 2.03. You fill at
2.01 by 3:40, but then SOI falls to 1.99 x 2.00, so cancel at 2.02 and
2.03 and replace them with offers at 2.00 and 2.01. The 2.00 fills
around 3:55, and after waiting a couple more minutes, you cancel the
2.01 and dump your last 1000 shares at 1.99 (the bid). Average price of
1.9970 including commission and SEC fee.
Actual return of -6.32% including commissions and fees, which is higher
than the paper return (which didn't include commissions and fees). Not
a great pick by zorak, but fills weren't the problem, at least not in
this example.