|
PureBytes Links
Trading Reference Links
|
I've also been concerned with the best way to handle unables on limit
orders in the emini. My ES system over the last 1000 trades has
only about 4% unables on the limit entry, however even that few
is significant for the bottom line.
I've been debating using 2 methods: enter on a limit 1 tick early so
no unables occur but I give up 1 tick, versus using an MIT run on a
tick by tick basis which I assume will usually also give up a tick on
most fills. One added bonus I see in getting in 1 tick early is getting
a few 'free trades' when I get filled but the sys entry pri does not
get hit so my sys is still flat. Yesterday this occurred for a customer
of mine who trades my sys and has been using that method for a few
weeks. That trade went 24 ticks before retracing so he could have
easily taken 20 ticks which is my normal profit target. That rare but
occasional 'free' 20 ticks may make up for a big chunk of the 1 tick
give-ups. I'd be interested to hear how the real slippage of the MIT
method would compare.
regards,
Rich
On 5 Dec 2008 at 12:19, Ray Gurke wrote:
> Hi, Bob.
>
> I'm curious as to what the fills have been like using the pseudo MIT order
> compared to a limit order. I would imagine you may get as much positive
> slippage as negative. Limit orders are great ...except when they are the
> exact high/low of the bar and your order is too far down in the queue to get
> filled ;-) So.. I'm wondering if you prefer the MIT approach over the limit
> order for overall better fill prices as well as mitigating the problem of
> how/where the orders are stored?
>
> Thanks,
>
> Ray
>
> ----- Original Message -----
> From: "Bob Fulks" <omegalist@xxxxxxxxxxxx>
> To: <omega-list@xxxxxxxxxx>
> Sent: Wednesday, December 03, 2008 12:35 PM
> Subject: Re: Russ2K trading
>
>
> > If you are using the TradeStation brokerage and having TradeStation enter
> > trades automatically, the situation is more complicated than that.
> >
> > If you are using a stop order that does not move more than once a minute
> > and no limit orders then you can choose to have the stop stored on the
> > TradeStation stop servers. If the stop moves more often than once a
> > minute, it stays on your machine.
> >
> > But if you also are trying to bracket your position and also have a limit
> > order in place, TradeStation applies and removes the stop and limit orders
> > dynamically as the price gets close to one or the other level. This can
> > cause you to lose your place in the queue on the limit order.
> >
> > I usually keep the stop order in place on their servers and program the
> > profit target exit as the equivalent of a market-if-touched order using
> > intrabar order generation:
> >
> > if Close >= MyTarg then Sell next bar at market;
> >
> > Using intrabar order generation this is executed on every tick. This is
> > fine for trading but lousy for backtesting unless you test using 1-tick
> > resolution inside the bar. For backtesting with no inside-the-bar testing
> > it will show the entries and exits at the highs and lows, distorting the
> > results.
> >
> > There is a way to use these market-if-touched orders for trading and limit
> > orders for backtesting. This is the most accurate combination I have
> > found. It is described at:
> >
> > https://www.tradestation.com/discussions/Topic.aspx?Topic_ID=83381
> >
> > (I was the unnamed person he was replying to in the message.)
> >
> > Bob Fulks
> >
> >
> >
> >
> >
> >
> >
> > At 10:21 AM 12/3/2008, Tom wrote:
> >>Hi Jan,
> >>
> >>I wasn't aware of stop orders being held on our computers for TFZ / ICE
> >>stop orders. Thank you for mentioning it in your post. I agree that it
> >>is a problem but also like yourself, I've noted good performance with the
> >>Russell as compared to at least ES with the system that I am using.
> >>The risk for me is also out weighted by the reward.
> >>
> >>I have a couple of questions for you, or anyone else that might care to
> >>jump in, if you don't mind.
> >>
> >>How did you know or discover that TFZ stop orders are held on our
> >>computers vs. ES stop orders being sent at once to the CME?
> >>
> >>I use profit target orders generated automatically by TS 8.3, for example
> >>sell or buy TFZ at Five handles above entry price. Do you know if these
> >>automatically generated "profit target" orders are handled or held in the
> >>same fashion as the Stop orders that you are speaking of?
> >>I'm assuming the answer is yes but wanted to be sure.
> >>
> >>Finally, When Trade Station 8.3 automatically enters these "profit target"
> >>orders on one of my computers at the point when the original position is
> >>created: If I shut down that particular computer, but then I quickly log
> >>on to a different computer with TS, I see the original order As well as a
> >>duplicate order created by the new log on. This may be a repeat of my
> >>question above but are both of these orders still being held on my
> >>computer?
> >>
> >>Thanks!
> >>
> >>Tom Brun
> >>
> >>
> >>
> >>
> >>JHP wrote:
> >>>I switched from the Merc to ICE and the only issue I have with them is
> >>>that they don't accept
> >>>stop orders. When I set it ("synthetic stop order") it's held on my
> >>>computer and released
> >>>as market order resulting in average of 20-30 pt slippage. That problem
> >>>is made up many times
> >>>by RL's $1.00/pt value and good performance, much better for me than ES
> >>>or NQ.
> >>>
> >>>Jan Philipp
> >>>
> >>>----- Original Message ----- From: "Liam" <46p771v02@xxxxxxxxxxxxxx>
> >>>To: <omega-list@xxxxxxxxxx>
> >>>Sent: Tuesday, December 02, 2008 7:39 AM
> >>>Subject: Russ2K trading
> >>>
> >>>
> >>>>Since leaving Russ2K trading for NQ recently, I've been wondering where
> >>>>other traders wandered off to.
> >>>>
> >
>
|