Ok, so you think the gold jump to 1310 was just driven by participants wandering around trying to find a price and it fell subsequently because those traders didn’t like that price, all speculation, no real change?
6/23 was brexit, which drove the first jump under reallocation and the drop came as people rotated out on 10/4 when the ECB made comments about gradually winding down bound purchases on the same day the dollar jumped 0.5% on Fed commentary pointing towards a rate hike. The move accompanied broad based market declines. It’s chasing your tail trading, which I can’t imagine is profitable for anyone. The charlatan idea that these BS near term strategies have any payoff is being taken out behind the woodshed as evidenced by the hedge fund industry contraction and the fact that over the past 10 years about 20% of mutual funds beat the S&P after fees.
First, I seem to have mislead as to the utility of volume profiling. It is not a signal producer, it is contex. Huge difference. I use volume profiling in my trading everyday to answer certain questions like “where do I see price going?” “Where might price get rejected” “Is price likely to range today or break out”. After that I use discretion, not signals, execute trades. I’m not trying to get to the pot first, I’m just trying to join in the flow with others who are seeking it. Volume profiling does not turn on the lights in the room… but at lest I have a flashlight 
I know the futility of trading from signals. Backtesting and I had a heated affair over the summer. It ended badly… which is why I am now focused on discretionary trading.
BTW… I am sharing my progress as a trader in water cooler. My second month review will be up in a couple days.
No, how I interpret what I see in the gold chart is that for a few months, the intrensic value of gold existed somewhere within that upper range. Given that set of fundamentals, two times when price went to 1310 price was promptly deemed way too cheap and it rejected. At the time of the breakdown two things happend. Something was brewing in the set of fundamentals for gold. Price again went down to 1310, but this time it was accepted. The fact it was not rejected is the auction saying something is different. So that opens the auction for exploration below. Traders have many approaches to how to takes trades but many of these motives ate different ways of noticing the same phenomenon. Value shifting to the lower part of a range can also be witnessed in a bearish moving average crossover. 1310 could also be seen as a major support line. I could go on and on, but the auction is the bottom line.
As a context, volume profiling is useful, but it doesn’t seem to be very stable. The market hates 1310 now, but will it hate it tomorrow? Will it hate it in 3 months? None of this seems all that stable.
It would be interesting (and I’m sure people have done this) to break out the volume profiling by parts of the trading day. The open, the first hour, midday, the last hour, the last 5 minutes, since it seems likely that the behavior of traders at the open and close is likely to be different than the behavior. Many traders need to close their inventory for the day. ETF adjustments are made at different times, etc. At that point one might see some kinds of patterns.
If you’re not trying to get to the pot of gold first, what are you trying to do? You talk about going with the flow, but that only seems to make sense if you are on the trade floor or have a colocated server such that the bid-ask spread is your profit rather than your cost. Perhaps there’s some clever thing you can do with options to hedge, but the transactions costs on options seem like they’d be hard to overcome.
Maybe the people you’re working with have that position (colocated servers, or being a floor specialist, or such), in which case it seems like the volume profiling might be more useful. But is it more useful than modeling it as the open price +/- the last 5 days intraday standard deviation.
A lot of what you’re describing is not so different from traditional technical analysis, with its trend channels and support and resistance levels. I don’t think that stuff is so crazy at the intraday scale, though I often don’t have the patience to go through with it.
Comparing the volume profiles at different time frames can be useful, where you compare the profile of the last month with that of the last few days and look for patterns in the differences.
Often times in technical analysis, the key moments for action are the changes in technical patterns rather than the assumption of continuity. These seem to be the moments when what is happening is a change in perceptions of risk and so reacting to risk changes may be the better way.
It’s been a while since I’ve worked with this, so I don’t remember the details. But it is indeed fascinating.
So basically it works until it the minute it doesn’t work but you can create backwards rationale? Sounds super useful and not at all contrived. The first time I played blackjack I was 20 or so, basically blackout drunk and thought I’d solved the system after some early wins. Discussing my losses the next morning, it turned out I had no idea how the game even worked…
BS… it doesn’t always work. It is just was I use to undestand what I’m looking at and what my best guesses for what will happen next. The fundamentals are key to the big picture. I just believe the the process that allows for them to be priced in is an auction. Volume profile, I believe, depicts that auction.
You are spot on. Some of the reason the market has such “memory” is that large players will enter the market in certain spots… form a volume node there. Then when price returns to that place there is again activity as the large player needs to manage the position.
Also, as this is all context, it is never an if/then kind of thing. 1310 in gold was interesting for the reasons I explained to BS. If it goes back to the area, my analysis of what is going on may be different. It is always like I am detective building a case and Volume profile is just one place I look for evidence. Sometimes volume profile is really muddy and I will just ignore it all together.
Wuuut? You make technical analysis sound like voodoo or something?! 
If the threshold is that “it doesn’t always work but sometimes it does” how do you not question whether it’s purely a mental construct of a mind instinctively programmed to manufacture patterns grapples with randomness (which factually will have patterns). A math professor used to have his students all flip coins while he left the room and write down the first 25 results on paper while one of them writes a fake random sequence. He comes back in and always picks the random one out because it lacks runs since people always underestimate the propensity of random variables to go on runs or exhibit noise patterns. The study of heuristics covers this in depth by looking at our minds desire to fool itself into creating a narrative where there is none. Alternately, read Taleb.
Ok BS, I take it you think markets are random then. At what point do they become producers of an efficient reflection of fundamentals( a non-random result)
For 25 coin flips, the ~95% confidence interval for the longest run (of heads or tails) is between 1 and 7 (as a rule of thumb, for N coin flips the longest run will be +/-3 within log(N/2)/log(2) with ~95% certainty). There is no way anyone generating a fake random sequence will fall outside that interval and get exposed by the professor. Maybe your professor asked them to write down 250 coin flips instead, which means he wasted a good 20 minutes of his class just to be a smarta$s with little educational impact, but he probably didn’t care if he was on a tenure track and getting some good grant money.
Worth taking 10 minutes to drill home a lesson.
Noise typically increases with sample frequency it doesn’t seem to turn over at a single discrete point. So longer more disciplined holding periods with reasonable expectations would typically help. Also, assuming there’s a lot of random behavior, being extremely selective in limiting your number of trades to compelling cases helps.
Well, it’s pretty clear that markets can’t be a pure random walk (or walk with drift). One of the features of a random walk is that given enough time, it will diverge an arbitrarily long distance from its starting point, which in practice would mean that prices would be completely disconnected from fundamentals.
I can accept that things can get pretty stretched one way or the other, but it seems ridiculous to think that stocks on a companies with positive earnings would have a PE ratio of 10,000 and not be subject to any pressures to mean revert. So clearly there are at least some processes at work that act on prices other than pure randomness. Within a range, there is randomness, but outside of a range, there is mean reversion (barring identifiable exogenous shocks). Within the range, there may even be momentum.
I dare you or your Berkley prof to repeat the expirement with 25 coin flips
Right Bchad. I wanted to answer you in regards as to your curiosity by what I mean by looking for trades where I can “go with the flow”. I was just now doing my analysis for my session later tonight. I will show you what I am looking for in oil this session.
WTI crude oil (CL) looks particularly intersting at the molment. I’m looking forward to this one. I start with the bigger picture. Here is a 1 hour candles stick chart with a profile featuring the price action in the upper range it has made since the OPEC meeting.
As you can see, after 4pm, the market took a dive (right to a value node I might add!). This was fairly significant because it broke the major support levels for the upper range. One thing I will look for tonight is for price to get up above the node it is currently in and go into that area with low volume. If there is a rejection there, I will interpret that as the old support has now become resistance. I will look for a opportunity to go short and join the other participants back into the node. With that sort of trade I will risk 3 or 4 ticks to make at least 10 ticks. Another scenario I will look at is if price goes up above the support/ resistance area and holds it. Then I will used yesterday’s intraday volume profile as a template to give me context on where price will likely go .
That is just one example of the kind of hypothetical situations I will look for. I will go through each of my 3 products before the session starts and determine what I think is significant and how I may try to get a good risk/ reward trade out of it.
Meh, you “dare” us? Ooooooh, what do I get, a pack of baseball cards? Missing the forest for the trees here in your eagerness to be the kid waving his hand to be called on by teacher. I said 25 coin flips to make an illustration to a broader point, not promote some kind of thesis. Also, you spelled Berkeley wrong despite it being right in front of you while we’re being four year olds.