Slippage

[LeeDanello]

I’m just wondering what slippage if any is being used in coding for entry and exit signals. I’ve read that 0.5% each way is reasonable ie LEPrice = O*(1+slippage), slippage = 0.005, maybe some other measure might be more appropriate like 0.1xATR(10) or maybe (O+H)/2? Reason why I ask is because I’m playing with a mean reversion system which was buying and selling on the open and it showed good profit potential until I started using slippage on the entry and it absolutely destroyed it.

[StephaneFima]

Hi,

On my side, in order to be sure that a system is profitable, I set my commission to $5.0 per trade, which takes into account the real commission (it is usually $1.0 per trade with Interactive Broker) + $4.0 of slippage (which I admit is important).

[SaidBitar]

another idea is to add average of 3 cents to buy and subtract 3 cents from sell

i.e.
BuyPrice = O + 0.03;
SellPrice = O – 0.03;

normally i do not check for slippage in my backtesting (may be i need to start to) because i try to trade liquid stocks.

[Nick Radge]

You need to use some common sense. Ask these questions:

1. What stocks am I trading
2. What is my account or trade size
3. What type of order am I using

If you’re trading S&P 500 constituents and using LMT orders with a $50,000 account then slippage will never be an issue.

If you’re trading a $1m account, on Russell 3000 constituents under $5 using MKT orders, then yes, slippage will be an issue.

I would suggest that you start with the base strategy first with those questions in mind then use a common sense application of a slippage factor. As you quite, rightly state, if you overstate the slippage when it’s not warranted, then you may be throwing away a perfectly good system for the wrong reasons. Obviously slippage will also show itself in real time trading.

[Author]

Posts