I'm forced to admit that I haven't spent a whole lot of time with the software. The lack of documentation then, turned me off a bit. I wasn't looking for a way to trade, and thus wasn't willing to bleed the stone to get answers to my questions. Lornz and I discussed the topic quite a bit and filled in the blanks ourselves.
Bracketing, as it relates to VS - if memory serves me, is simply a user-determined +/- entry value above or below the price of immediate interest. Example: +/- 4 ticks in either direction to discern when the train is leaving the station; so to speak. Also, Pete doesn't subscribe to the idea of stop-loss, at least not price-based ones. He would tell you that you need a market-generated-information reason to enter or exit any particular trade. In the case of VS, price alone would never be a clear determinant of being 'right' or 'wrong', as it were. For him, it's now all about 'price discovery risk', which, of course, implies not using price as a foundation for making most high-gravity trade-related decisions.
Don't burst a lobe on this one. It's not as mystical as it might seem. The core idea, I believe, is all that's necessary to make use of the concept. The details are of clearly secondary importance. I would sum up the macro hypothesis as this: relatively high-volume, in the very recent past, is a magnet and/or repellent of price (barbell distributions). You want to trade near these island distributions as anywhere in between are potential pitfalls of price discovery risk. In this, he's stating that markets (now) have goldfish memory (I am aware that this goldfish myth has been debunked). Anyway, It is up to each trader to decide what volume levels are worth noting, however. Deductively-speaking, we know such values are necessarily dynamic. Unsurprisingly then, I've seen nothing from them of which suggests how to make such decisions, although I haven't checked-in in a bit.
The following 10 users say Thank You to zer0 for this post:
A few quick notes as I want to post more but later:
- it looks like the presentation was delivered on June 25, 2012 (from the opening slide of the accompanying PowerPoint presentation); the link from the index page shows the presentation as being posted June 28, however the date shown on the page linked above is March 27 ... is their anyone in Chicago who actually attended the presentation?
- it seems that "Volume Strips" are out and "Time Variable Profiles" are in; the software picture appears even muddier close to a year after he last made his presentation (either September or August, 2012)
- appears to me to be a much bigger focus on time (or lack thereof) than on volume and despite the title of this presentation
- interesting and a much easier presentation to follow than his last
- disappointing as well, particularly as it relates to the part of the presentation on the "software" which I found to be both confusing and at times misleading
- still worth the view (or in my case it will be a few views)
- in any case, version 0.2 of my own take on 'volume strips' is production ready and working in real time with the CTS T4 API ... if anyone, is interested I could post more on what I am doing on this front, but just to say in advance that I have no interest in licensing or selling my 'software' to anyone else; having been there and having got that 't-shirt'.
- I will also start a trading journal here on Big Mike's as I attempt to use my software: the "POP.calculator" to navigate the ups & downs of the E6.
Looking forward to hearing from others on their thoughts re Steidlmayer & Volume.
The following 6 users say Thank You to tpbyyc for this post:
It’s fascinating to see how most of Peter Stedlmayer’s recent webinars (*) are of poor audio/visual quality. Perhaps GS is sabotaging them? It would be a hint that the content is valuable! (joking)
I am not sure to have understood everything. Let's be more honest: I am sure that I have not understood everything!
Some notes, which are clearly not summarizing the webinar...
More a retranscription of my understanding of the most "concrete" parts...
I would be glad to be corrected by other futures.io (formerly BMT) members.
Daily MP is not any more a objective foundation for trading.
Peter Steidlmayer uses "time-variable profiles". I understand that they are MP split at convenient times, when price leaves a balance to look for new value.
Trades could be placed when price is leaving a former balance (curve bell completed?) in order to look for this new value.
Only if supported by volume.
Trade is left open as long as price moves with "speed" in the right direction.
In my own words... try to capture a VPOC shift.
Once more, this is partial, personal and perhaps wrong.
Additional comments would be highly appreciated.
Excellent effort capturing some of the meat from the webinar, Nicoloas! And thank you for sharing them here.
There is at least one more webinar but I don't have it. On June 20th, Pete gave a webinar that was presented by PFG. They are now part of eSignal, but that content is not available. For some reason, I thought I at least had the audio from it, but can't seem to locate it if I do have it.
Luck is what happens when preparation meets opportunity. ~ Seneca
The following 2 users say Thank You to omni72 for this post:
Due to the rise of the quant, the market has become much more focused on the short-term. Movements away from "value" now have the potential to become a positive feedback loop, wherein movement begets more movement, and fading breakout moves can lead to disaster, as prices can simply keep going without laying down much volume (translation: if you trade size, you can't get out easily).
-Going with initiating activity away from a high volume node. If he isn't stopped out, he holds the trade until a certain amount of volume at price is laid down to indicate traders have accepted a new level/opposing activity has stopped the breakout move/whateva.
-Not using time parameters to divide his profiles (one per day, etc), but separating the profile into distributions.
Also, this isn't the first time he's significantly reformed his methods. He originally created the market profile method back in the 60's or 70's, when the market was much more prone to ranging activity. So, he created a proxy for a volume-at-price distribution using TPOs, which he could update in real time, and would then play the edges - always a decent strategy in range bound markets.
He has spoken before about how large commodity funds trading trend-following strategies created more movements in the market and made it necessary for him to adapt. Now, he's had to adapt again because of the influx of yet another type of participant (quants) who trade big size.
Interestingly enough, if his premises in these webinars are correct, and the bell-curve shape is due to the influence of locals and other market makers, then it's pretty obvious that the quants aren't really adding liquidity if their influence has caused the opposite type of activity.
The following 5 users say Thank You to ValueFocused for this post:
Thanks for your summary @ValueFocused - I have watch the last two webinars of Peter's on the time-variable profiles and I don't yet understand how his software decides to end one profile and start the next profile. I am guessing that each profile contains a constant volume and thus time (# of TPOs) and price range (height) are the variables? I'm wondering what your understanding is on this point.
Seek freedom and become captive of your desires. Seek discipline and find your liberty. - Frank Herbert
@Surely, I watched one of his webinars a while back, my understanding of how profiles are split was the following.
"The system he describes in the video was simply to start a new profile when the price broke the range of the last 9 bars, of a profile which is at least 12 bars old (although these parameters are changeable of course).
It is very simple algorithm (much simpler than I was thinking of), but seems to work remarkably well."
I had a go at coding this up. Actually, it seems to give quite good results. See attached chart (which wasn't a cherry-picked example).
((link to picture removed since I didn't reach the required post count yet! lol))
It seems to split the action up reasonably well into p,b,D and B shaped profiles.