“Dear David, On market timing. Do you have a view as to the merits and performance of Moving Average crossover systems? The most successful parameters when backtested (15/30 EMA, 100/350 EMA etc)? Has your service carried out any historical testing in this area? I appreciate that an individual's personality has a major influence on what is best suited to him/her so I am referring to the pure theoretical results over a long period of time. Regards,”

David Fuller's view

Many thanks for an interesting email.  Not least as, you make a very shrewd point:

“I appreciate that an individual’s personality has a major influence on what is best suited to him/her…”

This is so true, and I would expand it slightly by citing personality and also strengths, some of which are inherited and others which are learned and developed throughout our lives.  For instance, I think emotional intelligence is hugely important in understanding not only crowd psychology but also in assessing our own reactions to pressure, success or failure.  We are not born with knowledge of emotional intelligence, but we have the capacity to discover and develop it.  I often discussed this when I conducted The Chart Seminar over all those decades.    

Re moving average crossover systems, you appear to know more about them than I do, but here are my thoughts, starting with the negatives.  Moving averages lag by definition, so their signals will follow behind what should have been useful evidence of momentum changes in the price action on charts.  Yes, moving averages will nevertheless signal all important trend changes, albeit with a time and price lag.  However, moving average crossovers will often be whipsawed in choppy markets, including after reactions or corrections which may only be of a comparatively short-term duration.  With hindsight, one can apply the moving average crossover combinations that would have worked best, but the next market cycle might be quite different.  At best, they are a confirming indicator because if we are on the wrong side of a moving average crossover, we must have missed or ignored some earlier important price signals.

For these reasons I prefer to use moving averages on a regression basis, as we do all the time with the 200-day MA.