Welcome to NexusFi: the best trading community on the planet, with over 150,000 members Sign Up Now for Free
Genuine reviews from real traders, not fake reviews from stealth vendors
Quality education from leading professional traders
We are a friendly, helpful, and positive community
We do not tolerate rude behavior, trolling, or vendors advertising in posts
We are here to help, just let us know what you need
You'll need to register in order to view the content of the threads and start contributing to our community. It's free for basic access, or support us by becoming an Elite Member -- see if you qualify for a discount below.
-- Big Mike, Site Administrator
(If you already have an account, login at the top of the page)
Recently, I implemented a more mechanical system to describe the above strategy. Most parts are customized intraday strategies.
There are three conditions - buy, sell and no position. This is why an anti-buy condition would not necessarily always produce a sell strategy.
Although this is not automated or an algorithm that runs trades, the system is able to provide a snapshot of the market. Before this, some time was needed to analyze the strategies before providing a market summary.
The added simplicity allows me to focus on the method and wait for the system to generate information. Granted, this method is not completely blinded, but doubled checked with the original signals.
The exit condition is the nullification of the current condition after the bar close. After some time, a trailing stop should be implemented to capture profit.
This method needs back testing in current market conditions like today and ranges - Perhaps also during high volume news events to see what else might break.
Prior to implementation, the discretionary trading has been spotty. Although today, with the large movement, I captured a handful of points, I would like to gain confidence in the system.
A longer term wish would be to fully program and back test with large data sets to really check the robustness. There seem to be many parts to build such a system as well as time separate failures from bugs.
This is an issue with every trend-following method or system: there are both trends and not-trends, and the not-trends will kill you, when your objective is to get into a trend and hold it. The range reverses just about then the trend would have started.
You can always go to a lower timeframe, but you may find that the same trend/range phenomenon exists there, too.
I wish you success in this, but be aware of the issues. They are basic to the choice of using a trend-following approach. "If I could just not trade during these ranges" has been said, with feeling, by trend traders for a very long time. (And I include myself here . )
I won't suggest an answer because I don't have one yet. I should point out that a pure system approach has some pluses, in that trade executions are always according to the rules, but also that a totally rules-based approach means that you have to have all the possible variations already figured out and incorporated into the rules. Which is something else that is intrinsic to the basic choices that have been made, this time in going for a system instead of a discretionary approach. (Pure discretionary traders face their own issues, and still have the trend/range one.)
Good luck. I hope you manage these issues successfully.
Bob.
When one door closes, another opens.
-- Cervantes, Don Quixote
LT can be daily/hourly but is not practical for a day trade. This does not discount the utility of the LT effect on day trades, however. A strong LT signal will override any ST signal and should be considered the main signal, like a strong current. And honestly, the daily, weekly, etc. as well.
I move to a lower time frame that allows room for a day trade - basically how long are you willing to wait for the trade to develop. A patient, borderline swing day trader can hold for hours, but with the random news, sometimes even 5 minutes is a while.
For this reason, I also use ticks as a frame, I can time how long it takes for a tick cycle to complete and use that as a basis for a time frame as well.
The most immediate HI/LO on the day is the extension. This comes from the section in the elliot wave book that caught my eye on measured moves. I do not count waves, but liked the extension idea there.
AJ Frost, Robert Prechter - Elliot Wave Principle, p 72-5, Chapter 21, Wave Multiples.
There are a lot of rules in the book, I just took away one idea, that a pullback can be used as a measured target for an extension. This is interesting because, you can wait for the pullback cycle to complete, and get in at any level during the previous pullback and set a target for the previous high, and a second target at extension.
This also implies that a larger pullback would offer a larger extension.
Retraces are also interesting, someone AKA TTMcGee on stocktwits observed CL algos tend to complete 50% retraces. This does not mean that an immediate entry should be at the 50% retrace, but one can allow expectations for something similar. The issue here is that once CL stops trending, sometimes the trend does not restart with the expected extension in the same day at least.
Thanks Bob, yes the existential crisis of the trend/range.
This is interesting because by nature, I revert and traded ranges initially. The reason stems from starting out trying to chase breakouts when oftentimes price reacted in a range.
And then once in a while, a trend started destroying a range approach.
Very disheartening if you have to keep having to discern between apples and oranges.
----------------
This seems to suggest a hybrid approach, to use a ranged-trending system that is both systematic and discretionary.. if only things were so universal.. but the middle ground approach might be better than going to extremes.
So the immediate issue to resolve is to identify trend quality.
A range is also a trend, it is one that does not move much within the time frame.
And a strong trend is on the other end, A+ 100%, maybe even 200% like a flash crash.
If one imagines the order flow in such cases, the bid or offer is very one sided, imbalanced.
During the overnight Tuesday (Jan 7) the bids were hit with size, and very little on the offer.
But even this downtrend was preceded by a pull back, nothing moves completely vertically.
Which suggests that even trending patterns contain ranges.
The danger here is that a range trader might see a reversal and enter too early.
Something should say, hey, wait a minute, this is still trending strong, please wait.
So, perhaps an attempt to use a range trading approach, while eyeing trend quality may help.
For example, one waits for the low, and then finds trend confirmation supporting entry.
The quality of the trend might also help for exit, but whether price holds in the original range or an extension, the concept of the range still holds throughout.
And the range concept is really just, buy low, sell high in essence. This does not mean attempting to pick bottoms, but enter after the trend quality stops in one direction and picks up in the other. Usually this ends up being somewhere in the early-mid range of the move, with the trader usually kicking themselves for not getting in earlier in hindsight. Which, is probably better than kicking themselves for taking risk/heat from getting in too early.
This range/trend struggle will persist, but my reversion tendencies and frequency of ranges treat my developing trend system as a method to quantify trendiness while continuing to the reversion approach.
In this case, trend quality would suggest not reversing a strong trend, and possibly maintain holding a position that is trending. This is an advantage of reversion trading, the risk taken from a reversal is rewarded by obtaining a favorable position in the event of a developing trend - already bought low or sold high.
just adding a quip, reading some livermore while I test strats. chapter 5, the semisucker rule - don't be a sucker. and basically how? become a sucker and do the things that suckers do, until you no longer do them.
the jumpiness of nasdaq is no longer of interest and throw off signals. until there is a filter, the instrument is no longer of interest.
observing a double tap that is more prominent and extended in nasdaq, this jumpiness might be useful actually, and should not be filtered, but the spikes do provide early signals (which might be useful)
deciding to use the second test as an extra warning, an attempt to take continuation is low probability here. basically, you get two outs, but on the third one, you strike out.
this is actually valuable, as the first warning can serve as the first target and the second, the close. this prevents early exit and also temptation to retake a position due to lack of position.
the third test may not actually strike out. in fact, a successful third test resumes the trend and should be considered a breakout. so maybe the first warning sets the initial stop, the second a target, leaving the third to be ridden out.
now realizing that taps are actually divergences and the counts can move further. this is more clear on the nasdaq so the jumpiness may actually be an advantage.
this concept is now morphing into a dual wave pattern, where initial entry provides the early wave risk and secures a position that allows adding into a second wave. the second wave amplitude is unknown and initial entry in the second wave is mid-late, so having a secured position here is advantageous.
also, have tuned strat to more comfortably trade ranges. in the event of an actual trend, the entry from range would provide continuation. strong trends might afford entry, but ranges seem to provide best risk entries.