What about the attached example? This is what I mean about my less than stellar ability to pick entries on pullbacks. If I had not scaled in, I would have reduced my overall profit per car by going all-in at the first entry, and taken more heat overall.
But you're not buying a pullback in an uptrend. You're shorting -- or should be -- a pullback in a downtrend.
Apologies, I don't mean to confuse - Pullbacks, either shorting in downtrends or buying in uptrends. Just scaling in at those pullbacks of the relevent trend. Why is it not a valid idea to scale in to achieve a better entry price? Especially if one is not great at catching the best entry initially? This is ok right, and used extensively by lots of traders. Why is ND suggesting its only relevant if your trading larger size?
That's fine if you need to put on significant size, but as a small retail trader why not simply trail a sell stop below the low of each closed price bar? Let the price action sweep you into the resumption (hopefully) of the downtrend?
Or you could test all of it, maintain meticulous records, and go with what provides the best yield. Then you don't have to rely on feelings. Or the opinions of others. Altho if you're going to rely on the opinions of others, ND is probably the best way to go.
this is a perfect example of something you can test and find out, what makes you comfortable and what you can live with. Its a good start that you recognize you are not good at picking a level, or dont feel comfortable with a certain amount of heat. If scaling in makes you more comfortable then do it. If this works for you and helps make you profitable then perfect. Down the track you might find that you can tweak this to improve things, but if this overall works for you then do it. Dont listen to others about what does or does not work for them. work it out for yourself. Trading is all about trade offs - risk v return, wins v losses, all in v scale in, short term v long term. Just recognise that there are benefits and costs (advantages and disadvantages) of doing certain things certain ways. Scaling in will you will miss some normal size on trades, and you might find it does not get you a better price - but these will be profitable ones - not the end of the world. First you have to do the things that work for you. Then ensure they work in the market of course.
Yes, you are all absolutely right. Now given that I don't have the ability to code patterns into a backtester, and that forward testing is the best method, but also the slowest in terms of identifying working strategies, is there a "best method" to backtest a pattern reliably, and without hindsight? Is replay the best and only method?
You can use the computer to backtest if you want something fast and you want to trade mechanically, but you'll have no way of knowing whether or not the computer backtest is worth anything until you actually begin trading it. So why not just start forwardtesting it to begin with? And if you forwardtest via replay, you can do dozens of charts in a week, which is hardly "slow", though it isn't instant. You're going to be trading it in real time, so it makes some sense to do the testing in the same way.
Due to DBP's advice I have been reading section 7 of Wyckoff's original course found here: http://www.traderslaboratory.com/forums/wyckoff-forum/3876-basics.html#post36469 Wyckoff details his entries, stop placements, places to add to a position and exits. You may find that this is a good starting point for determining when to buy/sell on a pullback. EDIT - a lot of answers came in before I hit submit.