That depends on where you are in regard to the level at which you're trading. If you're just beginning and you're trading it mechanically, yes, particularly if you have issues with fear. Once you ease into it, though, you'd wait to see if the last swing low was broken before entering a short, or if a lower swing high were printed, or both. As it is, you have a series of higher highs and higher lows, so no particular reason to exit and short unless you're trading it mechanically. Judging how far price has to break the line before the trade is in terminal trouble is the sort of work that lajax and Gozilla are engaged in, but it's well past the primer level.
The following user says Thank You to DbPhoenix for this post:
The losses are minimal, which is the point. As you can't know the future when trading this in real time, you have no choice but to follow the rules and play the hand you're dealt. You can estimate the winrate and P&L if you like.
As for the bar interval, it doesn't matter. The same rules apply regardless. The only difference is the number of trades.
The same rules apply regardless of the bar interval, but an hourly chart is going to have fewer trades than a 1m chart, just as a monthly chart will have fewer trades than an hourly.
Plot the weekly trend channel for this period beginning with the first swing low in June '13. No trades. No lines other than the trend channel lines.
No matter what the bar interval or the timeframe, the trading session must be approached. This is what examining context is all about. To do otherwise is no different from watching a play beginning with Act 3. Are the dailies and hourlies "hindsight"? By the time the session begins, of course they are. But that's what preparation is, studying what has happened up to the point where the trading session begins, just as an actor prepares the backstory of his character before he walks onstage (or at least the better ones do). Without it, the trader has no context within which to make choices, which means that the odds of his making the wrong ones are pretty much 50/50.
Last edited by DbPhoenix; July 14th, 2015 at 10:00 AM.
I'm still not sure why you said that at the beginning, though.
Also, why would you be taking longs at the limit of the Weekly trend channel when you're trying to trade a mean reversion? Doesn't "approaching" from the Weekly down to the hourly pretty much rule out longs?
Every time I think the fog is clearing, a new front moves in and I can't see a thing again.
In this case, "approaching" means plotting the weekly trend channel for this period beginning with the first swing low in June '13. No trades. No lines other than the trend channel lines. One chart. This will address your question regarding longs. Or you can if you prefer begin with the trend channel that begins in June of the previous year.
Given that this is the second-most-viewed thread at the moment, I should point out to all those who visit the thread that this isn't supposed to be torture. Manual backtesting is a learned skill, not something that is intuited. And it can take a long time to learn if one doesn't know how to do it, much less if he has learned how to do it the wrong way. This is why I discourage people from backtesting the SLA unless they are going to modify it somehow; there's no need to backtest what's already been backtested. Even so, there's no need to go back beyond the current trend, which began in June '13. However, I will be happy to go back as far as the OP wants to.
Last edited by DbPhoenix; July 14th, 2015 at 12:20 PM.