Premise: I do not see volume as anything other than a random distribution histogram. Objective: Show me that volume is useful in short term analysis. Discussion: I have heard the following, yet I have never seen any of them to be consistently true: 1) High volume signals the beginning of a trend (everyone is getting excited and is getting aboard) 2) High volume signals the end of a trend (this makes sense, too - all the suckers are getting in too late) 3) Increasing volume means (insert whatever) 4) Decreasing volume means (insert whatever) I will now address these. 1) High volume signals the beginning of a trend Let me first ask "what is high volume?" Is it a spike? Is it a significantly higher distribution over x bars of data? Looking at this picture, I think we can all agree that the area I've circled is "high volume." But is that enough to count as "high volume?" It's next to some pretty average height bars. In fact, the 3rd, 4th, and 5th bars that follow it are below average. So if we were to take an average volume would this still count as "high volume?" Let's look at that. This is the same volume histogram except it now has Simple Moving Averages (SMA) across it for 4, 10, and 30 periods. I used a wide variety of periods so no one says "hey you didn't use (insert number here)." I think we can all agree 4 is pretty fast, 10 is pretty average, and 30 is pretty slow. So looking at it again, it now only seems like "high volume" on the 4 period volume SMA: So I would summarize this as "the spike is 'high volume' but that area is not 'high volume.'" But I am usually wrong when it comes to trading. Objective 1: Correct me if I am wrong here and include a quantifiable explanation of why I am wrong and what is correct. So now that we have (kind of) defined "high volume," let's get back to number 1. Does high volume signify the beginning of a trend? Well, let's take a look at a trend. Here's a good one: What is volume doing during that trend??? Let's take a look!!! (I left the volume MAs on the chart) Well on the first bar, no. It's excessively low. One the second and third bars, I would say yes, volume was high (volume spike). I would further define this as relative to the preceding bars, volume spiked here So far, the argument could be made that volume does increase at the beginning of a trend. But remember that we're looking at these charts in hindsight. When all you see if the most recent bar in a choppy period coupled with a high amount of volume, does that mean there's going to be a trend in that direction??? Let's look at the same picture again. Imagine that you're only seeing the bars in question on the right edge of the chart: (edit - sorry I misspelled "trends" as "thrends" and I can't go back and edit it) It looks like volume spikes are random and don't signal the start of a trend. I could be wrong, tho, let's look at more data and see if high volume accompanies the beginning of trends: What do you think? It looks to me link high volume sometimes accompanies the beginning of a trend, but it also appears in the middle of chop. This seems to me like an MACD that sometimes signals big trends but sometimes gives you false signals as well. Objective 2: Correct me if I am wrong with reference to volume spikes NOT consistently signaling the beginning of trends and include a quantifiable explanation of why I am wrong and what is correct. (and I haven't even mentioned the volume spikes in the middle of the trend that pretty much negates the assumption that volume spikes might signal the end of a trend, but that will be in a future post) I'm going to go out of order here. I would like to talk about something found in items 3 and 4: Increasing/Decreasing volume means (insert whatever). Can someone please explain to me how you know if volume is "increasing" or "decreasing?" Are you going by a MA? Are you going by a visual stimulus? I see pictures posted that have trend bars drawn over volume, except the look like this: How is that decreasing??? That seems so arbitrary to me. Why wasn't it drawn like one of these, instead? They all seem about as objectively correct: They all look about the same and, in hindsight, could all be used to support some point. Remember you're only seeing this on the right edge of your chart. I know I still have items 2-4 to discuss, but I don't think I can until I've been given an objective definition of how to tell if volume is increasing or decreasing. All comments/questions/discussion is welcome, but let's keep it conforming to two rules: 1) no fortune cookie replies. I know some posters on ET love these, but ambiguity like "do what the volume tells you to do" or "do what your chart tells you to do" is not helpful at all. 2) please keep analysis quantifiable and objective. If you can interpret the same chart in two ways, then it's not quantifiable or objective. Remember that you're making decisions on the right edge or your chart with live money. You cannot do this without quantifiable and objective analysis. Upcoming in future posts: items 2-4 above. More evidence why I don't think volume is useful. Volume divergences. Also, one volume derivative indicator I created that was almost helpful, except it's not (kind of like MACD... sometimes it correctly picks huge moves, and sometimes it chops you to death). Bonus Challenge: To all the volume masters, please draw in this chart what you think price is doing based only on volume. I don't expect a line chart or anything, but identify the points where key things are happening, such as trends beginning or ending, or choppy price action, or etc. To prevent cheating, I made all the volume bars the same color and picked a random value for the chart. I've also eliminated the scale on the right. If volume signals significant events in price, you should be able to identify them here. Go go go!
Hi IF, Your view of volume is too simplistic. If it were that simple, everybody would be rich. The most valuable use of volume is in discerning what Smart Money (aka Market Makers, Floor Traders, the Floor, Strong Hands) are up to. A frequently quoted statistic is that 90% of retail traders loose money. Where does that money go? It doesn't disappear into thin air. It flows into the pockets of the MM. If you want to know how to use volume, study Richard Wyckoff's work and a modern off shoot called Volume Spread Analysis. This is a non trivial topic and will require substantial time and effort to learn how to use it. Bill
I googled "Volume Spread Analysis" and found a link to a 154 page thread over at TL :eek: Looks like I have some reading to do. So are you saying that volume by itself is not useful?
Volume is useful in combination with prices but the patterns are not repeated exactly because the MM try to disguise their activity and simplistic rules like "volume increased on an up bar so prices are going higher" don't work consistently because there is a lot more to it. Bill
dbphoenix has devoted a lot of time and work to this topic. i assume his posts are still available here, and he has an entire section over at his new virtual home. hth
exactly... MM has the ability to sell phantom shares to cool a uptrend/breakout or MM can buy to slowdown a fast descent/breakdown-pretty much depends on how many people are playing that stock-those volume factors are more important on a longer time frame-only then does the trickery/trend reveal itself-plus you have to take into consideration what price did with that volume!color code those bars also so one can see the buying or selling-run several timeframes of the same stock-but alas MM can see also what you see and strategize on how to get your money into their pockets that is if his computer sees you buying on crossovers of MACD ADX or bottoming or peaking a oscillator -depends if volume can hide your footprints etc. etc.-Buying on weakness is a good trend reversal omen so is selling into strength-means someone with deep pockets is supporting a price or fading the strength of said stock-and-------------------what volume acompanied that action-even a MM wont fight a strong trend(that is only if many are buying or selling)-mabye this helps or adds confusion-TTFN
In spite of conventional wisdom about volume, I say "volume is whimsical.. therefore volume considerations are unreliable and should be disregarded"... at least for the stock indices (no opinion about individual stocks).
I am not a volume expert...but volume on a tick chart is too "even out" for me. Volume spikes, increasing/decreasing volume are more visible to me on a minute chart.