Supporting limit orders involves a *lot* of complications. JBT would
need to know not just how to place limit orders, but also how to move
limit orders and cancel them. The strategy code would be a lot more
complicated, too. Consider this: your strategy has a long profitable
position and it wants to sell. It places a limit order to sell, but
the market never reaches that limit, but moves lower instead. What
should the strategy do? It has to have some sort of code that would
instruct it to cancel the existing limit sell order, then place
another limit sell order. And what if that second limit is never hit
either? Additionally, partial fills will be a problem, too. With the
limit order, it's quite possible that you managed to sell half of your
position, and left with the other half to sell. So, again, at some
point, the strategy would need to decide somehow whether to wait for
the other half to be filled at the existing limit order, or to cancel
the remaining order and to place a new one. All of this becomes pretty
complex. None of these issues exist with market orders.
One would think that you'd get some benefit for the trouble of using
limit orders, but it doesn't work quite the way you calculated
(0.25*50*1000=$1250). That's because you are not accounting for the
trades that were not filled at the limit and lost profit because of
that. What I am saying here is that using limit orders instead of
market orders does not give you an extra edge. You'll gain an extra
0.25 points on the orders that were filled, and lose some amount on
the trades that were not filled. Overall, it would probably balance
itself out.