A short YouTube conversation (Adrian Reid with Marcel) analyzing the suitability of the Parabolic SAR indicator as an exit/trailing-stop for trend-following systems. Reid argues SAR tightens (accelerates toward price) as a trend advances, which is opposite the proportional leeway trend following requires—so SAR can prematurely exit large trends while performing acceptably on small swings. He prefers simple exits (e.g., close below 200-day MA) as a benchmark for long-term trend systems.
Chapters
I can tell. I have spent months of my life using Parable Xar to try and create a amazing exit that captures the profit but doesn't give it all back. Like probably years of thinking all up. And when I first started, I was obsessed with it because I was like, oh my God, this indicator is amazing. But here's the problem. Well, a, the problem is it doesn't work. But why? It doesn't work because it's completely the opposite to what people. Sorry. It's doing exactly what people want to do, which is the opposite of what works. If you think about trend following, when a move progresses in your favor, what happens to the amplitude of the swings in that move? They get smaller or bigger?
They get bigger.
They get bigger, Right. So we get in here and the trend's going in our favor and then goes in our favor and then it goes, shit, we gave back 25% of our profit of the peak price. But what you miss is the fact that you're up a bazillion percent from where you started when I hit your trailing stop.
Right?
What? And the way trend following works is you've got to, as it progresses, you've got to give it room to move proportional to the price of the instrument you're trading. That's why it gets wider and wider. We start to freak out more and more and our account swings bigger and bigger as the big trends progress. Because in order for trend following to work, you've got to give it leeway proportional to the cost of the asset. But what SAR does parabolic, sir, is it accelerates towards the price as it moves up.
Which is doing exactly the opposite. As it moves up is giving it less leeway so that when it pulls back, it exits quickly. So it does a really great job of accurately catching a really small price swing.
So if what you're trying to do is accurately catch a really small price swing, I think it's probably a perfectly legitimate indicator. But if you're trying to catch a big trend, not so much.
Yeah, I've done that many times. You know, you have an idea in your mind and you say, oh, this is going to work. And then you go to all kinds of trouble to put it into a code and, or, you know, you test it manually or something like that and then after all that work, you see, no, it doesn't work.
Yeah, it's like, oh, shit, just closing a day, moving average beats it. Okay. It's kind of my benchmark, right. Particularly for long term systems. I'll have whatever fancy stuff I dreamed up and I'll compare it to the close crossing below the 200 day moving average for trend following. And if it doesn't drastically beat the stupid exit of close below a 200 day moving average, then I just throw away my complexity.
Now the other party equation, Marcel, was about this concept of parabolic SAR and applying it to like a trailing stop, saying okay, over time, if it moves in our favor, can I tighten it so I give up less? I've played with it and I mean not in amibroker but in other platforms and coded it and whatever. And let's say we did this, I'll do it in, I'll do it in blue. Let's say we had our trailing stop and it's like, oh, we're in profit, so let's tighten it. And oh, we made more profit, so let's tighten it. And oh, we made more profit, so let's tighten it.
What happens is it looks really good on some trades but you miss out on. Actually that's probably not the best example. But let's say this trade actually went on to not get stopped out and we exited up here.
And we exited there. The tightening often gets you out earlier in the trend, so you might get out with a more certain profit but the expected profit is less.
So you have to deal with less drawdown from less sort of end of trade drawdown, whatever you call it, like sort of let's give back but you also get less profit in total because it doesn't allow you to stay in for those big moves.
