Slippage vrs minimum $ slippage

Discussions about Money Management and Risk Control.
Post Reply
Chris67
Roundtable Knight
Roundtable Knight
Posts: 1046
Joined: Tue Dec 16, 2003 2:12 pm
Location: London

Slippage vrs minimum $ slippage

Post by Chris67 » Wed Jan 28, 2004 3:15 am

I m finding that if i test a system using a 20 % slippage factor .. which i think is verging on pessimism .. that the system is still robust. However if i set the slippage facor to , say , 3 % ( obviously optimistic ) , but , set the minimum dollar slippage to say 75 dollars that the system then goes into negative territory. Surely the 20 % would be bigger than the 75 usd .. am i doing something wrong .. what is the relationship between these 2 parts of the global parameter settings ??
Thanks in advance for any help.

Chris

Forum Mgmnt
Roundtable Knight
Roundtable Knight
Posts: 1842
Joined: Tue Apr 15, 2003 11:02 am
Contact:

Post by Forum Mgmnt » Wed Jan 28, 2004 5:36 am

Chris,

Most system testers who quote a figure of $75 use it for commission and slippage per trade combined. Depending on your settings for commission, a minimum slippage of $75 is at least $150 per trade and probably more like $165 ($75 Entry Slippage + $75 Exit Slippage + $15 Commission). That's very high.

It might not be for some markets on some days but on average that's a lot of slippage.

Long-term systems can probably handle slippage settings like that but it would crush a medium-term system like the Turtle's System One.

Also you setting slippage to 0% when you set the minimum? Otherwise you're getting charged at least $75 every single time, entering and exiting, and occasionally you'd be getting charged much more if the range for a day was high and the 20% factor is higher.

As far as 20% being higher than $75, that would imply an average daily range of $750, since on average we might assume that your entry price would be the mid-point of the entry day so (20% of 1/2 day = $75). Many futures, if not most, don't have daily dollar ranges that high. Consider Corn, a $750 day is a HUGE day in corn, a 15 cent range. Most days are 3 to 5 cents lately. If the average was as high as 5 cents, then that's still only $250 per day. 20% slippage would be only $25 on average.

- Forum Mgmnt

Chris67
Roundtable Knight
Roundtable Knight
Posts: 1046
Joined: Tue Dec 16, 2003 2:12 pm
Location: London

Post by Chris67 » Wed Jan 28, 2004 6:22 am

Thanks c.f.

Basically then I can work witha slightly smaller number I guess.
So one other question .. say I'm using the bollinger band countertrend and its trades are generated over nite ? Then by the time markets opens isnth the slippage already built in .. for example veritrader is excellent at this because it will say suppose today i need to sell soybeans on the open at 836 and it opens at 825 then veritrader assumes you sell , in its backtesting idea , at 825 .. that slippage is fierce and realistic already ?
Is there an element of double continuing

Is the slippage of say 75 $ = per contract or per trade ... obviously if your trading say 100 lots as you are toward the end of the system test period if its been profitable to date .. then I would have thought per trade I was looking at at least 1000 usd of slippage .. say 10 bux per lot

Cheers for help so far ...

Post Reply