Slippage Over Time
By Dave Mabe
In yesterday's post, I showed how you can compute slippage on a trade-by-trade basis and then sum it up for the entire day.
The next step is to view slippage in aggregate over time.
The ability to use your backtest as a reference is extremely valuable.
Your measure of success goes from "did I make money today?" to "how well did I follow my plan?"
When you have a backtest going back years and a nice, smooth equity curve, the anxiety of any particular trade or day melts away.
The way you get confidence from this point is by gathering proof that the live trades won't be too far off the backtest.
And when you can generate true confidence, you can scale your strategy over time.
Just like any particular trade isn't going to make or break your strategy, the slippage from any particular trading day isn't the full picture.
But start stacking several days together and looking at slippage in aggregate, and it starts to become clear where the real problems are.
Then you can devote resources to things that will move the needle for your strategy.
Here's a slippage TODO list you can use to put this into action:
Generate a list of your live executions by strategy
Run your backtest to generate a list of trades for the days your executions cover
Create a Python script to compare the two, categorizing the slippage into 3 buckets: entry, exit, and missed trades/partial fills
The output should be an easily viewable report showing:
Slippage per trade
Aggregate slippage across all the trades
Once you have this, you can start to address each slippage type.
Tomorrow, I'll discuss how to tackle the most important and most common type of slippage.
-Dave
P.S. Do you wish you had a column library that would tell YOU how to make your strategy profitable? My column library is now included with MabeKit