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)
Exactly, I was never suggesting there was some complex coding involved, I just found the concept very interesting. I'd like to see it in SC.
Yes i'm familiar with how it works. Often i'll draw profiles from the initial swing high/low, or from an impulse wave that starts a trend. However those techniques are not the same as using a dynamic market profile, from my experience what it lets you see is a change in momentum or a divergence between value area extremes. I was curious to see if there were others out there who had experimented with a dynamic market profile( DMP).
Yeah I agree, it's of no use when price keeps trending up in the 3rd STD, however when it comes back down again and tests other STD's, without moving to the lower STD3, then that could be useful in determining if the current trend is still in place...
The guy doing the presentation is a systems engineer, he's not here to teach you about market profiling or sell you some course. He's doing a presentation to try and demonstrate the concepts of his creation, and using a "braindead" system you can back-test seems like the most logical way to do it. You can't exactly call it curve fitting either, since he's basically back-testing random data sets using the same parameters. If he focused on how it should only be applied as a discretionary tool and then pulled out impressive equity curves of it working during open drives/ open rejections, then that would be curve fitting.
I think you're right about how using a rolling period/ DMP could have tremendous value. I think this is what the presentation is trying to demonstrate- to me it seems like the innovator behind it realizes it could be a powerful concept, however hasn't exactly figured out how to best apply it. He's not some profiling guru, or some guy trying to sell you a trading system... he's just an innovator.
Awesome! When did you start experimenting with it? Did you hear about it from the presentation or has the concept been around for awhile? I've struggled to find anyone else using it, at the moment i'm looking at implementing it into my intra-day trading.
I started work on it in earnest about 3 years ago. i've been a student of MP for last 7 years - it was the only methodology i had learned that made any kind of sense. I don't believe in fibs, elliot waves, indicators with fixed periodicities, cycle theories or any other of the weird and wonderful subsets of technical analysis. AMT (and its implementation in MP) was the first method that seemed rooted in the real world. It's really hard to trade a method if you don't believe its in-built theory of the market and, with all other forms of technical analysis, I just don't.
However, although I believe wholeheartedly in AMT, I became disenchanted with MP for several reasons.
Firstly, there's an element of magical thinking in MP: e.g. that the POC and the VAH/VAL have some mystical significance and can be used as perennial reference prices to lean on. If you think about the Value Area on a trend day, what does it represent? There will be a Value area but only because the MP algorithm produces one regardless of how prices are distributed, but what use is it if it were produced on a trend day? It's not just useless, it's worse than useless if you use it as a guide to make trades.
This is one of the problems of MP and it is because the profiles are static: you have a profile for the day session and maybe a separate one for globex. If the profiles were dynamically created by market action then this deficiency would be eliminated and analysis would be much simpler.
I had no idea about the presentation you referenced at the time and was not influenced by it. I don't particularly like it as it is just a moving profile average essentially and I don't like moving averages because it implies there is a fixed cyclicality to the markets.
My view is that we should let the market tell us when it changes from a phase of balance to one of imbalance and base our profiles on the market action rather than starting them at arbitrary times.
How to define 'x' (period)? There is no right or wrong, but at least with MP/VP it's a clear answer of how it is defined.
I can also make an argument for a developing VA.
For VWAP, I use weekly and monthly. I use weekly and monthly profile's as well. I am not sure where I would begin on a dynamic profile based on something other than a day, week, or month.
How about letting the market tell you where to begin a profile? A profile (and its Value Area) is only a useful tool (at least I believe so) when it defines a balance, so start it when the market starts balancing. End it when it moves out of balance. Obviously, you need the market to have shown signs that it is coming into balance and then you can apply it retrospectively.
I am very happy you said this, which I will elaborate on below:
More simple, but not better--only a different view. Taking the value area of a trend day is meaningless, as you say; but it's not the fault of the tool. It's the misunderstanding of what value is by the user of the tool. The issue is not whether the dynamic profile or the static profile is better -- it's whether the trader using them can be effective with his tools. We should no more become disenchanted with a profile because it's misused than we can with a hammer because someone tries to use it on a screw instead of a nail.
I also at one time became quite turned off by profiling--it was becoming popular, and the drones began to treat it like a cult, and they tried to trade it mechanically. What you will invariably see in new profile traders is that they try to mechanically fade LVNs and VPOCs. Well, 90% of my trades are incorporating LVNs and VPOCs. But they don't understand the behavior of the auction and so they wind up making the profile a mechanical activity. It just doesn't work that way, which is why it is a futile attempt to try to make something simple, when it cannot be made simple. Easy, perhaps, but simple--it can't be, because it's market-generated, and markets are not simple.
I abandoned and came back to the profile as my primary tool on about three separate occasions, because I had to really make it my own, and I was unable to do that the first couple of times around. Part of that had to do with just what you said--others' use of it. But it really doesn't matter how other people misuse a tool--if it's a good tool, and a good methodology for you, that's really all that matters right?
Agree. In particular, I just don't get using fixed-time moving average that spans several sessions (like a 200 period MA on a 5-minute day-only chart). I do, however, find value in using profiles of a single day, week, and even month, because there is an open and close to the trade. However, one of the more powerful ways to use profiles is to identify areas to profile that are not bound by a certain time period.
edit: After I wrote the above, I just saw you posted this:
Exactly what I was getting at. Thanks for your post, I enjoyed reading it.
This sounds like we have taken a similar journey. We both have taken time to understand the underlying principles of MP rather than the superficial setup gimmicks and have deconstructed it to sift the useful information from the dross. We have our own unique ways of trading MP (or rather, AMT) and both work. The important thing is that we have taken the trouble to understand the underlying principles rather than lazily trying to duplicate someone else.
Just addressed this in the last post; and I agree, just having to define the 'X' is what turns me off to moving averages. The market does not define 'X' and so we superimpose our view when we have to define it. The market does, however, define a day, and a week. And given the way the financial system is set up, quarters are quite important, as are months (just note the unusual activity on month/quarter end). But taking the time out of the picture, the market is sometimes balanced, sometimes imbalanced. Aligning our profiles around those structural parts of a market that are time-independent can give a very clear picture of what's going on. Since we're talking about it, here's an example of a couple of trades in FB I took last week based on that:
A view of the overall balance area (see a daily chart to see the clear move up out of balance or zoom out further) shows a poorly auctioned area whose bounds served as absolutely picture perfect trade location for two days. It really doesn't get much nicer than that.
If you look at a monthly profile of FB, you will see a similar picture, but this is simply coincidence that the part after the push up (after 6/10) is all part of the same month. If the month had ended a few days after this move up, you would not have been able to see the same clear picture if only using a monthly profile (it is also a very different view looking at daily and weekly as well). BTW, I think it's also helpful to begin the VWAP at the beginning of a certain "event" or structural realignment.