It seems the prevailing wisdom is that when a market starts to move in one direction, you should wait for a pullback before you hop on board the trend. I suppose the idea is that if you buy a rising market at a lower price, that is in some sense a ‘bargain’. I have never really understood this. Why does waiting for a sign that the market may be starting to move against a trend make that trend more attractive? It seems to me you would be better just jumping on board while the market is still moving up. Unsurprisingly to me, none of the backtesting research I have done on various systems over the last five years or so suggests that waiting for a pullback is a good strategy. And yet it is commonplace to see it recommended. Am I missing something?
My backtesting experience has been similar to yours... it's of course impossible to know what is a pullback and what is a trend change without further context. In such a case, you might as well go for maximum trend strength if that's your only signal. The 'theory' is something like this: Stocks tend to like to pull back a lot (mean revert). Entering on a pullback affords a more narrow stop should the wider trend persist, which makes the RR ratio more appetizing. E.g. a common occurence when entering a hyperbolic gone midcap stock is that it decides to make a 25% drop from highs before continuing. Now, if you can actually make out what is a pullback and what is a trend change in some way, it makes good sense to buy the dips. But my dumb technical systems couldn't, it was just noise fitting. Alternatively, trade without stops (risk manage/position size accordingly) and be happy because now you can't be forced out of positions. Just IMHO. I'm sure someone with an entirely different view than mine will chime in.
the prevailing wisdom is that when a market starts to move in one direction, you should wait for a pullback before you hop on board the trend ---> the prevailing nonsense is that when a market starts to move in one direction, you should wait for a pullback before you hop on board the trend ____________________________________________________________ There are thousands of ways the market moves. Sometimes, the market will retrace (pullback). After the end of retracement, you enter your trade. To make trading challenging/difficult, a retracement pattern can be - simple - complex Some stocks move very rapidly. You have to enter fast (say before the end of the day) because there wouldn't be any retracement. etc etc etc
Backtesting huh. Let me see...if your backtesting suggests otherwise, I suggest don't wait for a pullback. Lord knows I have not waited for a "pullback" sometimes, and I usually get burn since it's not part of my plan. Like maxinger said, it can be more complicated than that. Goodluck with your experience.
Simple questions receive simple answers. So. If it's early in a strong trend and the volatility is low it may not matter much where you enter. You may not even get much of a pullback. In the current market conditions you can easily end up buying a top and end up liquidating where? On a deep pullback. Trading successfully requires more than just hopping on a trend or waiting for a pullback, though. You really need to know the market you're trading - where it's at and where it's going. And also what's a normal movement within that market. When that is the case you'll know when to "jump in" and when to wait for a better spot to enter.
Thanks for this. Good to know I am not alone in finding waiting for pullbacks to not be particularly successful. I can understand that different instruments may have different quirky behaviours. That is partly why I tend to stick to very liquid instruments such as indices which seem less prone to this sort of thing. One thing my backtesting does show is that trading without stops can be very successful... until it isn't in a massive way.
TBH I would be mostly interested in hearing from somebody who system trades and actually have pullbacks codified into their rules. But it's certainly not as straightforward as it's made out to be. As for stops, I did qualify with position sizing. If you have e.g. 1% in a single name, then all you can lose due to that name is 1% of your account (assuming long). Of course, finding 100 names is a lot of work for a human, but not necessarily for an algo. Stops are absolutely needed if you're doing leveraged single name trading, or you will get the high win rate huge loss effect you mentioned.