I'm developing some strategies right now that test fine on paper, but when backtesting, introducing 1 point of slippage per transaction significantly erodes the profitability (down to nearly nothing in a newest one) ... this newer on is on the ES. So the question remains: is the only way out of this trouble on intraday only futs trading strategies to have large returns versus small losses and generally infrequent entries? My systems with the most beautiful backtesting curves without slippage generally make plenty of entries, but bail early on nonperformance. So the jist: the very technique that makes my strategies perform on a backtest (early exit, more frequent trades) is whats killing it on slippage. Note my risk/reward is usually in the 10-15 ticks risk to 30 ticks reward range. I have extended periods of nonperformance due to a lack of volatility in my backtests (ie 2005, early 2006), and this is where the slippage kills the most. So 1) is .5 or 1 tick avg per entry and exit (so 1-2 ticks total per trade) realistic slippage on futures like ZN and ES where the intraday bid/ask spreads are without any space? Is there potential for -better- performance than this? I'm planning on using market orders. 2) any way to offset this slippage impact besides reducing total trades done? Or is the best way to do it by discretionarily turning these systems on and off as volatility ebbs and flows in the markets.. (ie if ES returns to VIX of 12, turn it off until it returns) ????
Slippage works against the trend follower because the trading method buys when price values increase and sells when price values decrease. You might test some systems that buy low and sell high. Buy low sell high puts slippage in your favor. Buy low sell high can use limit orders placed in advance. Buy low sell high systems also miss big trends and risk management can be troublesome.