It's good presentation, thx for posting it.
I tend to use classic peak/trough analysis to determine the trend on the higher / trend time frame ie HH & HL in an uptrend LL and LH in a downtrend...(I notice James Chen uses moving averages and a rough visualistaion of the price action to determine his trend.)
In identifying where a pullback may end and proceed to turn back 'with trend,' ... I have a filter that in re-entering the trend (that exists on the higher/trend time frame,.. howsoever measured/identified) at that pullback , the macd histogram must still be above it's axis on that higher/trend t/f in an uptrend and below in a downtrend. This pullback will also be showing hidden/reverse divergence in the intermediate t/f oscillators (James Chen looks for oversold stochastics turning up in an uptrend and overbought in a downtrend turning down, on the intermediate t/f.)
In timing the entry on the trigger/shorter t/f James Chen uses a 'trailing entry' technique as the intermediate stochastics are turning which is very valid, ...Hereto, I personally look for regular divergence or extreme readings on the shorter/trigger t/f oscillators, as the intermediate t/f oscillators are showing hidden/reverse divergence.
To ensure that you are acting at the highest probability set-ups you can also seek to identify that the pullback is at an area of previous support on the higher/trend t/f, in it's downtrend, or an area of previous resistance on that higher/trend t/f, in it's uptrend. This technical phenomenon of support becomes resistance -and- conversly, resistance becomes support is a very useful extra filter and increases the overall probability of the set-ups.
You can even go a step further by ensuring that the trend exists on the time frame up from your higher/trend t/f too, in this way you are entering a trend that exists on at least the next 2 time frames up from your intermediate chart, thereby making it a potentially stronger trend.
