There are many successful traders who thrive by keeping their trading methods very simple. They’ll look for the same method of entry, time after time. They won’t detour from their set up and can easily adapt this set up to any form of automation. They look for the same confluence of indicators, or indications, to appear on their charts each time they take a trade.
As an example they may set their charting package to signal when the PSAR, DMI and MACD and stochastic lines all trigger a positive or negative reading. Once the alert sounds they’ll immediately take the trade without hesitation having learned, through their own trail and error, that this method of entry works perhaps 80/90% of the time.
However, this HPSU example may give superb entries, but that doesn’t necessarily mean that 80/90% of trades taken will be profitable, an overall win rate of 60/40% would be considered very high, 50/50% and a competent trader could still make those percentages work in his favour with tight money management. But what is beyond debate is just how important entries are.
There’s a body of opinion circulated through trading forums suggesting that entries are “easy and it’s all about the exit” and who could honesty argue with that pronouncement? Many of us could select several entry methods that work, but can “what gets you in get you out”? Can you use the opposite reaction to stop and reverse direction? Possibly, but that’s a discussion for another day, in this article we want to concentrate on how to develop a HPSU and a little known but obvious test that we can conduct in order to determine the effectiveness of any method of entry.
Testing out a HPSU
In theory the same HPSU entry should work on every time frame with very little modification needed to the standard settings and in many ways that’s a perfect evaluation of the potential worth of a HPSU.
Let’s suggest that we’re using a combination of; PSAR, DMI, stochastics and the RSI, all left on their standard settings, do the maths indicating a turn in the market (based on the time frames your testing) work as you ‘go through the gears’ of the time frames? For example, does the HPSU give good entries on: the fifteen minute time frame, the one hour time frame, the four hour and right up to the daily? Can you observe good entry and exit points based on this combination of indicators?
If the method works on many time frames then your method may be right, and let’s be honest here you’re not going to get 100% results given that shortly after entry the trade may go bad. By ‘working’ we should be looking for the trade to develop in your chosen direction the majority of the time for a period of time, not all of the time.
How far back should we backtest our HPSU?
The lower the time frame the smaller the test window. With a method being tested on the daily time frame it’s a simple observation exercise to test your HPSU method as far back as three years. But with lower time frames, perhaps fifteen minutes, you’ll need to employ specialist software to conduct a back test that far reaching. Traders may therefore prefer to shorten their back test, or selectively use random periods to test their HPSU. A good tip would be to concentrate on time periods where you can readily identify where the method fails. Forensically examining the reason for the HPSU failure is an exercise that many experienced traders will advise and this failure will be particularly valuable on the lower time frames. Is the failure simply part of the random distribution of winners and losers, or is there a deep lying fundamental fault with the method? Analysing the points of failure will reveal the strengths of the method far better than concentrating on the seamless entries that always result in winning trades.
Selecting a high probability set up should be a simple process for traders, back testing the method likewise. Having the confidence to then put the method into the market should be a natural consequence of the thorough testing. Keeping the risk to a minimum, as part of your trading plan, should see the HPSU reap rewards.
