Dg> A Great Post fm: Walkingshadow
Message 21836904
Many think volume is relatively unimportant. I am not one of those. I think it is tremendously important, because it can provide very important clues as to what the big boys are doing. And it is not the retail investor who determines market trend, but rather professional traders and institutional investors of various types with diverse styles and agendas (ranging, e.g., from conservative stodgy types methodically building huge portfolios for big pension funds who may hold positions for many years to hedge fund managers with the attention span of of a hummingbird and all the ethics and morals of a shark in bloody waters. Just by sheer size, these forces move markets. And, much of what they do is computerized and quantified (i.e., so-called "program trading", which accounts for the majority of all trading on the exchanges). So, when you see big surges of volume, you can be sure that does not reflect the activity of retail investors.
Now, markets move essentially in response to discordances between buying pressure and selling pressure, but this is a dynamic relationship, not static. These discordances force price efficiency when they are equal or nearly equal, and as they deviate, price efficiency erodes, and so price can move. IMHO, one important factor that allows rapid price movement is big discordances between buying and selling pressure, and this tends to be greatly amplified by sudden decreases in volume. This is what I mean when I say that the relationship between buying pressure and selling pressure is dynamic---time is a factor, and so is absolute quantities of buying and selling pressure, not just the relative quantities.
There's a couple of basic facts that should be understood about volume.
1. The markets are fully fungible. This has important consequences. What this means is that for a firm to make a market, it must find a seller for every buyer, and if it cannot do so, it must maintain sufficient inventory to supply equity to anybody who wants it. Similarly, it must find a buyer for anybody who wants to sell the equity, and if cannot do so, the firm must buy the equity. And, the firm is required to execute within a reasonable period of time, so they cannot just let the order sit indefinitely. The effect of this policy is that it forces a tremendous amount of efficiency into the markets and thereby narrows the bid/offer.
Further efficiencies are forced onto the market by the various ways that orders can be routed and filled, e.g. the ECNs. So often, just because a MM or specialist can receive and order doesn't necessarily mean he will, and if he does receive an order, that doesn't always mean he will fill it---it can be rerouted, for example. The upshot is that this tends to allow a lot of flexibility in order flow, and enhance the chances that the order will get filled at the best price possible. Well, that's the intent, anyway. Everybody knows that intent and practice may not always be the same, and there are ways for firms to skim profits anyway. An example is when a firm pays a broker for order flow. In that case, an order that could have been filled at a more competitive price is not, because it never finds that competitive offer. Instead, it is given a one-way ticket to a particular firm making a market.
2. Professional traders buy into weakness, and sell into strength. This alone is frequently sufficient to distinguish buying volume from selling volume, provided there are marked differences in absolute magnitude of volume (which there usually are). Still, often this may not be easy to judge, because total volume is comprised of several components, including retail volume. But retail volume is not important, or if anything, is a contrary indicator.
Just as market internals tend to give the most information as the extremes are approached, so too volume gives the best information when it gets very high, and especially when it gets very high within a very short period of time.
Now, if the rate of exchange of shares remained constant, volume would not be a significant source of information. Luckily, the vast majority of volume tends to occur in just a fraction of each session. This is really ballpark, but my guesstimate is that roughly 80% of all trading volume occurs within perhaps 20% of the duration of any session (the open, the close, and for very brief periods in between at intraday highs and lows). The open is traditionally "reserved" for retail investors and those professionals who "feed" on retailers. Hence the old maxim, "amateurs in the morning, professionals in the afternoon."
3. Large, sudden, short-duration spikes in volume represent predominantly program trading, and are highly significant. Often these will occur at market highs and lows. Why? Because they have crystal balls and can discern the highs and lows? Nope. Because those very spikes cause the highs and lows because they are anomalies. This can be quite striking at time. It is not black and white, because there are many things going on at once, many sources of volume, but it is when all other sources are dwarfed that volume becomes a powerful predictor of price movement---actually, a determinant of price movement. That's because relative proportions of buying and selling pressure are so perturbed that pricing inefficiency is suddenly introduces, and the market must react to "reign in" that pricing inefficiency. And that can take considerable time.
Let me give you an example. Let's say NDX is trading along during the middle of the day, and has been drifting down all day long. Volume from minute to minute is not varying much, averaging say 800,000 shares a minute (that's about what it actually trades). Now suddenly a program kicks in, and in a 2 minute period you get a trading volume of 6 million, or a 3 million share/minute average. This actually happened today, at just after 11:00 ET. That's almost 4 times the average, and is a tremendous stress on the system that the market must adapt to. Also, it is a huge amount of money. The NDX is market cap weighted, and I am not sure how much it costs to actually purchase all 100 companies, but if the NDX is trading at 1545 (approximately correct), then 6 million shares divided by 100 would be close to $100 million. So I think this is approximately how much stock was traded during just two minutes of trading----almost $100 million. This is not retail trading, obviously.
But what happened next? Volume evaporated, and the next hour or so saw much lower volumes, generally about 1 million/min or less, sometimes much less. That means that the bid and the offer can be moved because of the sudden discordance between buying and selling pressure, and that's exactly what happened. So, price proceeded upwards from there quite sharply to reach an intraday high over the next 50 minutes or so.
In a nutshell, what it boils down to is this: when you see a big surge of volume as a stock or index is declining (particularly if it is declining very rapidly), that represents predominantly buying volume from professionals. When you see a big surge of volume as a stock or index is rallying (particularly if it is rally sharply), that represents predominantly selling volume from professionals. The bigger the surge, the more confident you can be in coming to that conclusion. When it happens again and again at particular price levels, we call this support (in the case of buying volume) and resistance (in the case of selling volume). Why? Because this volume tends to stop the advance or decline of the market. The bigger the spike, the more likely it will be to halt or reverse the market.
This works out in many time frames, and the market can "store" and "process" volume in various time frames also; it does not always react immediately. You can think of this like potential and kinetic energy: water behind a dam in a lake has potential energy; when the floodgates open, that potential energy is dissipated, but is turned into kinetic energy, and the kinetic energy possessed by the flowing water can be harnessed to do work---turn a turbine in a hydroelectric plant and convert the kinetic energy to electrical energy. Or, in the market, "potential" volume can be stored to a limited extent, then when "released", it does useful work: price moves (instead of a turbine).
As it turns out, the reason we remain in an uptrend is because of huge buying volume that occurred in July and October of 2002 and March of 2003:
139.142.147.19.
So far, nothing has sufficiently counteracted this, and so the long-term trend will remain in place until that volume is dissipated and counter-volume stops the advance and reverses it.
Sorry to be so long-winded, but these are not simple matters. Anyhow, hope this answers your questions. |
