i invented a TA indicator

[hooligan]

if Vp% is high, it's taking a lot of volume to move price.

if Vp% is low, it's taking little volume to move price.

if price movement is requiring a lot of volume, there are two reasons:
1) the market is not agreeing. there is a battle.
2) the market makers are buying or selling while holding price steady (they have this power)

if price movement is requiring little volume:
1) the market is agreeing with itself.
2) market makers are moving price around to test market psychology without committing capital.

perhaps there are more reasons as well for each scenario.

[hooligan]

This is an interesting approach, congratulations. I feel that overall far too little attention is paid to volume.

Have you compared the results of H-L/V with C-O/V ?

Notice how Vp% spikes a few times near the top of the market.

I don't doubt that this is your observation, (and mine too from that graph), but it would be useful to quantify that mathematically - the ability of the human eye to see patterns after the event is legendary.

To consider: do you trade based on an 'outlier' level (and how do you define that), or a change in level over a certain amount of time and so on. Do you use it as a standalone indicator or confirming/refuting indicator in association with other signals. Or do you use it as a leading indicator to decide whether or not to take the next long/short trade using other methods (I think this one could be worth a look). Does a different pattern invalidate or reverse the trade, does it work better on the long or short side, does it work on different markets etc etc.

I know a fair bit about this stuff - please feel free to contact me if you think it could be useful.

see my immediate above post. since high Vp% can indicate a "battle" occurring, it's another indicator that can bolster the ideas of support and resistance. so you can combine it with trend lines, fibs, moving averages, RSI, etc, to identify support and resistance.

it's a bit more dynamic though because it tells its story for every candle. rising Vp% means rising tension. Falling Vp% means slow release of tension. Sudden rise means sudden strong tension. Sudden drop means fast release.

average Vp% relative to background levels would be considered "valid normal price movement".
high Vp% would be considered "meaningful levels, support-resistance, supply-demand, be watchful"
low Vp% would be "be cautious of fakeouts" or "rare market agreement"

[jlhooter]

if Vp% is high, it's taking a lot of volume to move price.

if Vp% is low, it's taking little volume to move price.

if price movement is requiring a lot of volume, there are two reasons:
1) the market is not agreeing. there is a battle.
2) the market makers are buying or selling while holding price steady (they have this power)

if price movement is requiring little volume:
1) the market is agreeing with itself.
2) market makers are moving price around to test market psychology without committing capital.

perhaps there are more reasons as well for each scenario.

Hooligan,
If you purely think about the equation Volume is on the top and the (I will call it) PriceRangeRatio (PRR) is on the bottom, meaning VP% = Volume/PRR where PRR = (H-L)/open*100
If you think about Limit math (taking the extremes) you get
- Two ways to get High VP%
--- Volume is very large (think infinitely large), OR
--- PRR is very small (think almost 0)
--- and there is a 3rd way: both Volume is large and PRR is small
- The opposite is true for Low VP% (small volume, large PRR or both)

Real examples for COUP, 1-day time frame
1) Jun 08, 21: V: 6.6M; VP: 1.37M; PRR: 4.8
2) Nov 12, 21: V: 3.4M; VP: 813k; PRR: 4.2
3) Dec 07, 21: V: 6.5M: VP: 606k; PRR: 10.8
4) Jul 15, 21: V: 3.8M; VP: 304k; PRR: 12.5

In the above examples
- 1 and 2 have the highest and lowest volumes in the group, respectively, but with the highest VP
--- 1's VP is driven by high volume with a similar PRR to 2
--- 2's VP is higher than 3 & 4, but lower than 1 because of its similar PRR to 1 with lower volume
- 3 and 4 have volumes very similar to 1 and 2, respectively, but with significantly lower VPs for the group
--- 3's VP is ~1/2 of 1's because of its significantly higher PRR
--- 4's VP is lowest because of its highest PRR even through it has volume higher than 2

In a very long-winded reply, I agree with what you are saying, but it is a dual problem where very low PRR also drives high VP. I view low PRR as low volatility for that time period which can drive high VP as well.
I updated my script to plot PRR, but I am not sure that is enough; also added a feature to allow averaging of PRR over time.

[hooligan]

In a very long-winded reply, I agree with what you are saying, but it is a dual problem where very low PRR also drives high VP. I view low PRR as low volatility for that time period which can drive high VP as well.
I updated my script to plot PRR, but I am not sure that is enough; also added a feature to allow averaging of PRR over time.

Correct on everything, but you're needing a way to make this concrete for yourself. Highly suggest the book I mentioned. Will help understand how volume drives price and why its more meaningful to relate volume to %moved (PRR) rather than see it on its own.

Volume moves price, in either direction. No volume, no movement. High volume, high movement (though it can be oscillating and stagnant movement). Or explosive.

Volume is gas for price. NOS.

Example:
If a candle exploded up, but volume exploded up in exact proportion, then the volume change may look meaningful, but it isnt. The amount of volume required to move price was the same.

However if %moved on a candle is high, and volume is proportionately even higher, or quite lower, then its meaningful, the price move required less, or more, volume to drive it. This means something. It means price moved more easily or less easily. It means the market is reaching new ease (or tension), or that market makers are doing something fishy. .

Its predictive. If you have 5 candles in a row going up in an uptrend, all with fairly low Vp%, but then the last one has very high Vp%, that means that it took alot more gas to move it up, which means there's more tension at that price level. Signals an impending consolidation or reversal.

[SOL]

Volume is a good indicator but not as great as it once used to be as the big players are now messing in that area. For example, they can create the illusion of a strong breakout, by churning a lot of shares for a week with daily closes that are slightly higher than the day before, but each day they can produce sudden upward spikes. Then suddenly one day on very high volume most of which was rubbish, they drive the stock much higher on lower volume towards the end of the day.

for example from 9.30 to 3.00 they play around with the price (only slight price movement) but volume is already 150% higher than the past 7 days, let's say total volume is standing at 5.5 million shares. From 3 to 4 on 500K or fewer shares they can push the stock 10 to 30% higher. So to the naked eye, it appears that the stock closed up on higher volume, when in fact it did not. The last push of 500K shares is what drove it higher. If yesterday's volume was 3.5 million shares, then it only closed higher today on 500K shares and it could a head fake set-up

Having said it is still a useful tool if combined with other factors

Have you thought about using a simple moving average of volume, you can change the scale to exponential etc

Simple VMA(n) = (sum of N volume bars) / N

Instead of just using the daily figure, continually determine the average for the past 7 days or past 14 days or whatever number you want and then use this figure to replace the volume figure you are using currently

Jl are you in the AI trend service as there is a very good chance we will start a community Tool building project with pros next year

[jlhooter]

SOL and Hooligan
Both of your comments make sense and I will dig deeper. I like the idea of averaging volume but also get how it is being manipulated by your examples. I will continue to play.

I am in the AI service too and would love to be part of Tool building. I am addicted to learning TA; thanks!

[Budge]

If I understand correctly:
Vol / { [ ( High-Low ) / Open ] * 100 }
[/quote]

I plugged this into a spreadsheet, it looks the crazy cousin of Chaikin Money Flow in need of a smoothing factor (some Ritalin or Adderall maybe ). That's where the similarity ends. It doesn't work if you extend it into the accumulation/distribution line or the oscillator per Chaikin (differences in numerators and denominators).

Interestingly though, if you calculate per Chaikin Money Flow Index, chart looks the same.

jlhooter, have you looked at hooligan's chart versus other momo indicators on TOS?

[hooligan]

Then suddenly one day on very high volume most of which was rubbish, they drive the stock much higher on lower volume towards the end of the day.

for example from 9.30 to 3.00 they play around with the price (only slight price movement) but volume is already 150% higher than the past 7 days, let's say total volume is standing at 5.5 million shares. From 3 to 4 on 500K or fewer shares they can push the stock 10 to 30% higher. So to the naked eye, it appears that the stock closed up on higher volume, when in fact it did not. The last push of 500K shares is what drove it higher. If yesterday's volume was 3.5 million shares, then it only closed higher today on 500K shares and it could a head fake set-up

in this case, on an hourly or minute-based chart, Vp% would get super low relative to the earlier part of the day, indicating manipulation!

[jlhooter]

If I understand correctly:
Vol / { [ ( High-Low ) / Open ] * 100 }

I plugged this into a spreadsheet, it looks the crazy cousin of Chaikin Money Flow in need of a smoothing factor (some Ritalin or Adderall maybe ). That's where the similarity ends. It doesn't work if you extend it into the accumulation/distribution line or the oscillator per Chaikin (differences in numerators and denominators).

Interestingly though, if you calculate per Chaikin Money Flow Index, chart looks the same.

jlhooter, have you looked at hooligan's chart versus other momo indicators on TOS?
[/quote]

Budge, I looked a bit at other indicators, but still trying to wrap my head around it. I can see at very short time durations (1 day) that VP become interesting. For example, I looked at the end of the day yesterday for CHGG and tons of volume in the last 5 minutes of trading and VP accelerated. Kind of what SOL was talking about.
I made a SMA study with various levels of SMA and I notice how Chaikin Money Flow looks very similar; pretty cool stuff even if I don't understand it yet.
I notice that the Chaikin Money flow equation has a similar equation to VP too but allows negative values, so in the end very different.

[hooligan]

BTC/USD 4 hr, Dec 11 - Dec 31

relative spikes in Vp% can indicate trend reversals

Screen Shot 2021-12-31 at 12.20.39 PM.png

[SOL]

One thing to keep in mind when working on a new indicator is to make sure it does not eat your day. For example, intra charts while useful are deadly in terms of how much time you have to allocate to them. I have seen people glued from morning until the markets close and in some cases even longer if they are playing currencies and futures.

Money is useful but If I have to give too much of my day it's not worth it. this is why we hardly look at intra-day or daily charts.

So focus on the longer-term indicators for nothing in this world is more precious than time. Time is the only teacher that kills all it students.

[harryg]

I plugged this into a spreadsheet, it looks the crazy cousin of Chaikin Money Flow in need of a smoothing factor (some Ritalin or Adderall maybe ).

As a line chart it looks like the ECG of someone on Amphetamines but as a histogram it's OK.

I've had a quick look at some dailies, no combining with anything else for the moment.

QQQ: https://ibb.co/2q0GZ4J
GOOGL: https://ibb.co/hm5FhZt
DOCU: https://ibb.co/yPbHw53

Control: https://ibb.co/5BKQJ3r

The book to which Hooligan refers is an interesting one, but a lot of the interpretation of Volume Price Analysis is Art Not Science (their words not mine).

{...}
focus on the longer-term indicators for nothing in this world is more precious than time. Time is the only teacher that kills all it students.

Agreed. Also I think that big money is made on the longer term moves.

[hooligan]

I've had a quick look at some dailies, no combining with anything else for the moment.

QQQ: https://ibb.co/2q0GZ4J
GOOGL: https://ibb.co/hm5FhZt
DOCU: https://ibb.co/yPbHw53

Control: https://ibb.co/5BKQJ3r

The book to which Hooligan refers is an interesting one, but a lot of the interpretation of Volume Price Analysis is Art Not Science (their words not mine).

thanks for trying it out !

in your examples, on this large of an overall scope, 1day candles may be too granular. going to 2day or 3day or up to a week long candles may iron out some noise and produce more coherent signals. when i use the indicator i always try and find the most coherent granularity. discovering it often makes me wonder: "is this the time-frame the whales are using?"

[hooligan]

One thing to keep in mind when working on a new indicator is to make sure it does not eat your day. For example, intra charts while useful are deadly in terms of how much time you have to allocate to them. I have seen people glued from morning until the markets close and in some cases even longer if they are playing currencies and futures.

Money is useful but If I have to give too much of my day it's not worth it. this is why we hardly look at intra-day or daily charts.

So focus on the longer-term indicators for nothing in this world is more precious than time. Time is the only teacher that kills all it students.

i agree here overall. wondering though: if intra-day patterns can be spotted, do you think they can be utilized for algorithmic trading bots, etc? how do you feel about algo trading in general?