Slippage for orders on the open and close

Discussions about the testing and simulation of mechanical trading systems using historical data and other methods. Trading Blox Customers should post Trading Blox specific questions in the Customer Support forum.
Post Reply
Chris Murphy
Roundtable Fellow
Roundtable Fellow
Posts: 84
Joined: Thu May 29, 2003 12:11 pm
Location: Eugene, OR
Contact:

Slippage for orders on the open and close

Post by Chris Murphy »

The system that I am currently coding does not hold positions over night. What kind of slippage should be assumed when trades are only entered on the open and close? It seems to me that ZERO slippage should be assumed as this is the actuall price that orders were filled at that specific time. The only time slippage would come into play would be when the stop loss is hit.

Chris
damian
Roundtable Knight
Roundtable Knight
Posts: 815
Joined: Tue Apr 15, 2003 8:43 pm
Location: dusseldorf

Post by damian »

yes, but just because your data file lists $10 as the close, does not mean you were filled at $10. Likewise for the open. This opens up a lot of study into how an individual commodity (or exchange) conducts price discovery during the 'pre' session leading into the open and how they also manage the reality that the last traded price is not necessarily the closing price. This is an area of study that I so far have not spent much time on.

There is also still the possibility that old fashioned slippage can still get you. Example: at 10 seconds before the open the market is at 10/12, 1,000x15. Some guy called Gavin has an order to sell 1 on open, his mega-ultra-fast broker hits 10 for 1 and he is done. You also have an order to buy 20 on open so your very fast broker pays 12 for 15 then lifts the next offer of lets say 12.5 for your remaining 5. In this instance, the market opened at 10 and you were filled at 12 and 12.5

Some markets will match orders in the pre session. If you know how they do this then it can help in determinig what order type to place such that you are filled instantly on the open and at the price that will be printed as that days official open.
Chris Murphy
Roundtable Fellow
Roundtable Fellow
Posts: 84
Joined: Thu May 29, 2003 12:11 pm
Location: Eugene, OR
Contact:

Post by Chris Murphy »

Sounds much more difficult than I had anticipated. For my purposes I suppose that I will assume some slippage, if the system works on this then I should be in good shape. Beside if the slippage kills the system then the system was in poor shape to begin with. It appears though that I could estimate slippage to be less than would a typical break out system that trades during the day if the price is hit. As a greater percentage of the slippage would be favorable than would a buy/sell based on a break that indicates the market has momentum. Does this logic make sense to you?
damian
Roundtable Knight
Roundtable Knight
Posts: 815
Joined: Tue Apr 15, 2003 8:43 pm
Location: dusseldorf

Post by damian »

It makes some sense to me.

There will be others here that know far more about how markets open and close than I. Electronic markets will have some interesting features.

One thing could well be true, that is you could enjoy positive slippage, particularly when trying to get a fill on the close.
edward kim
Roundtable Knight
Roundtable Knight
Posts: 344
Joined: Sun Apr 20, 2003 2:42 pm
Location: Silicon Valley / San Jose, CA USA
Contact:

Post by edward kim »

Chris,

We talked about opening "ranges" for commodities under another topic of this forum. An opening range means that there is a range of prices that a particular market can open at, and they print only one price as the opening price. So you have to include some sort of slip in your testing because you may not get the price that's printed on paper.

Also, HAD you been trading that particular market at that time, YOUR orders might have influenced the opening or closing price. So if you are trading several hundred or several thousand contracts, or several thousand shares, you might have had the effect of "changing history" - e.g., the printed price on the open or close for your historical data.

Edward
Chuck B
Roundtable Knight
Roundtable Knight
Posts: 481
Joined: Thu Apr 17, 2003 6:34 am

Post by Chuck B »

One interesting study would be to look at those markets where an opening rotation occurs, like most markets in New York, and then construct a data set based on the "second open", the time when all the market's expiration months are released to trade freely for the rest of the session. Sometimes this can occur as much as 10 to 15 minutes (or more) after the front month did it's "open", and a lot of realignment is possible after the opening orders are completed in all contract months.
Post Reply