Let's say that over your entire trading lifetime you make 10000 buys and 10000 sells. Buys average $35 and sells average $37. Not bad. Let's say another trader makes 2000 buys and sells at the same prices and same total size. Both come out with the same profit. Scaling in is irrelevant, just adds to transaction costs. By scaling in and out of positions, you're scaling in and out of your lifetime portfolio a fraction at a time, a waste of trading fees. The idea I suppose is that you will get a better price versus the initial buy -- the notorious "I lowered my cost basis!" - but only if you average down exclusively. If you average down and average up, then your average price will in all likelihood equal your initial price over your trading lifetime - what was achieved? And if your initial buys were well-timed, then scaling in will only hurt your profits. Of course this doesn't address the whole pyramiding issue - in that case you have more information and scaling in (or pyramiding) is actually a function of trading based on the information contained in the price action. That might be a good strategy. But that's not what people mean when they talk about "scaling in/out" of a position. What they mean in that case is raising their winning percentage, which is a crutch for those who don't want to be wrong.