jakobe 

Man...thank you for all that! It definitely is calculating everything completely differently. As you said it appears it is just multiplying VWAP by itself. And putting bands +1, +2, +3, etc out there with no "variance".. I played around with another setting called "Maximum Permissible Deviation (MPD)" .. they say "MPD is similar to the standard deviation but is calculated as (VWAP period high - VWAP period low)/2." Here's what I get for yesterday's action with 987 tick... MUCH closer but still off I believe. My platform is Quantower, not very customizable to specific needs but it gets the job done. I've been doing some testing around Sierra Chart and feel it is 100% superior. However, I've come across some issues with VWAP in Sierra Chart that is putting me off. I really appreciate you sharing these setups.. I've included VWAP in all my charts previously but this really inspires me to dive deeper into this as it really really fits my trading style. I'm doing my best to form strategies the best I can so I know what boxes to check off prior to taking a trade.

That is much closer isn't it. I am not sure if the small discrepancies then come down the resolution of the VWAP calculation.

Traditionally, tick data is required to calculate VWAP values. While this provides the most precise VWAP calculation possible, it can also utilize extensive CPU resources. Therefore, your platform may calculate slightly differently (to save valuable computing time and resources) and calculate the VWAP on a bar-by-bar basis, hence calculating once per bar.

The conventional method of calculating VWAP values is to recalculate with each incoming tick of price data. I am just taking a guess here as I don't know for sure. I will also continue to dig around some more because I don't know why the platform would say it is a standard deviation of VWAP, when clearly it isn't. If you are sticking with that platform I would certainly ask them, but it sounds like you are looking to change.

I can't comment on SC, but I have also heard it is a superb platform as well. There are many threads devoted to it here on NexusFi. Many people have moved from NT to SC and love it.

JonnyBoy 

That is much closer isn't it. I am not sure if the small discrepancies then come down the resolution of the VWAP calculation. Traditionally, tick data is required to calculate VWAP values. While this provides the most precise VWAP calculation possible, it can also utilize extensive CPU resources. Therefore, your platform may calculate slightly differently (to save valuable computing time and resources) and calculate the VWAP on a bar-by-bar basis, hence calculating once per bar. The conventional method of calculating VWAP values is to recalculate with each incoming tick of price data. I am just taking a guess here as I don't know for sure. I will also continue to dig around some more because I don't know why the platform would say it is a standard deviation of VWAP, when clearly it isn't. If you are sticking with that platform I would certainly ask them, but it sounds like you are looing to change. I can't comment on SC, but I have also heard it is a superb platform as well. There are many threads devoted to it here on NexusFi. Many people have moved from NT to SC and love it.

Just from looking at my charts real-time I can confirm the VWAP and SD's are being calculated at the close of each 987 tick bar. My PC is definitely capable of utilizing tick data and calculating everything off tick for tick without suffering any performance issues. I'm fairly confident Quantower does not allow this feature unless I were to custom code something, which sadly, I am an uber noob at and not comfortable completing.

Considering the switch to Sierra. I really appreciate your time and helping the best you can. Once I feel I have everything setup correctly I'll drop some charts here again to compare.

JonnyBoy 

Do I line up the T1 to T3 targets and wait for them to be to be tagged? No. With enough metrics collected you can determine the types of days when removing everything at T1 is the highest probability trade and days when you should let a T3 run. This is what I consider to be my edge.

It was a fair, and good answer to the question. Appreciate your time for the reply. Thanks.

Alright @JonnyBoy, I think I've got this looking correctly and SD's charting correctly! Was able to fairly easily setup VWAP to calculate
Loaded up SC and have been setting everything up. Please take a look and share a snap shot of today's (4/30) session if you don't mind! Would like to compare.

jakobe 

Alright @JonnyBoy, I think I've got this looking correctly and SD's charting correctly! Was able to fairly easily setup VWAP to calculate Loaded up SC and have been setting everything up. Please take a look and share a snap shot of today's (4/30) session if you don't mind! Would like to compare.

Your chart setup looks spot on to me.

I am very new to using the VWAP and I appreciate all the detail in this thread. Particularly from Jonnyboy. I do have one particular question though.

One can either trade toward the VWAP (mean reversion) or away from it (mean aversion?). What are some of the signals to guide in which direction to trade?

Tuglife 

I am very new to using the VWAP and I appreciate all the detail in this thread. Particularly from Jonnyboy. I do have one particular question though. One can either trade toward the VWAP (mean reversion) or away from it (mean aversion?). What are some of the signals to guide in which direction to trade?

Although all of the VWAP setups are valid, they will not all work mechanically day in and day out. I have said this previously that you need to spend lots of time studying and gathering metrics as to best types of days (and what the preceding price action looks like) to execute each VWAP setup. Doing this increases their probability of working by some margin.

The VWAP test long and test short are the simplest and easiest VWAP setups to recognise. From memory I recall studying these for about 6 months non-stop. The best way I found to do this is was to record my screen, take chart snapshots and fill a PowerPoint with marked up charts of my observations. Note that reversion to VWAP is a trickier setup to determine and I wouldn't start there.

I see you also have Jigsaw. Putting my views on "Order Flow" trading aside, you could start looking at what price does when it approaches VWAP for a test and see if you can spot something in the underlying structure of the market that could determine a VWAP test failure, else you have a good setup to take.

And just to note, if you want to tag a user so they get notified, you need to put an @ in front of their name.

Thanks @JonnyBoy. OK sounds like I will be investing a lot of screentime in this.

Yes I have a Jigsaw DOM up while trading. The way I'm thinking about it now is that I'm looking for setups on the charted VWAP, and then refine the entries using the market profile, order flow.

The std deviation lines are important but not as important as the main VWAP.

motowater 

The std deviation lines are important but not as important as the main VWAP.

100% agreed!