This is not the 1st time you have presented an idea on ET looking for approval. As previously, I suggest you try it out with your money or at minimum use a demo- account.
There are those who blabber. There are others who can't read carefully other people's posts and go on the attack. Then there are those people who manage to do both.
This is like the strategy in "Stocks on the Move: Beating the Market with Hedge Fund Momentum Strategies," by Andreas Clenow.
+1 Can this sort of thing work? Absolutely. If it ends up working for you, is it because of it? Probably not, my guess would be that over a sufficiently long time span it isn't adding alpha (discriminatory ability to allow you to do better than random/market average). To the extent it is useful to you is as a a part of your decision making framework, e.g. because it lets you periodically review your choices and force you to drop sour bets, in a way that you otherwise might not have on a discretionary basis. One man's fecal matter is another man's treasure. So better test it yourself, with your own edge appended (because again my hunch is you are going to need one beyond the described method). For one, the method would depending on frequency incur significant transaction costs that need to be overcome.
How are you letting your profits run when you are exiting on pullbacks? Stocks do have periods of consolidation or pullbacks. Does not mean the trend is over. Not even close. So, you end up ditching stocks that could very well end up the big winners. What you are doing is not trend following the way it is intended. And I subscribe to trend following as my method of trading.
This is a pretty simple system with weekly rebalancing based on momo/ relative strength.. Now all you need to do is get off your ass and backtest it