Repeater Add-on Dynamic Range anomaly?

[brucerobinson]

Hello,
I use the Ninja Repeater add-on to highlight Sessions.
Please see the attached images.
You will find charts of 2 instruments - 6B and 6E, each with a 5m timeframe and 1m timeframe.
If you look at 6B - you will see to the RHS of the chart - marked up in red pen on the chart, the range high of the London Session (my choice of timings). Note that the range high is different. It is the same chart window - i.e. simply switched timeframe from 5m to 1m
The Properties dialogue box shows Settings for Repeater corresponding Dynamic Range field 02:00:00hrs
Note that in the 6E chart, the range is the same for 5m and 1m
Note that the attached 6B and 6E are from different chart windows
The issue is specific to 6B - if a select one chart window and display 6B I experience the issue. In that same window with 6E it does not. If I choose another chart window and choose 6B I experience the issue, not with 6E.
So it is specific to 6B and not 6E (I haven't tried other instruments) but not templates etc attached to the Data Series

Please advise
Kind regards

[NinjaTrader_PaulH]

Hello brucerobinson,

Thanks for your post.

So that I can take a look at my charts, what time zone is your PC set to?

If I understand corectly, you are saying the process to replicate is:

1) Create a chart of 6b 1 minute and the Repeated set 20 dynamic Rectangle with a stat of 2:00 and an end of 11:00
2) Switch the chart then to 5 minutes? (or vice versa)?

In your screenshots (thanks for those) the image is quite small and I am not clear on what the red marking is showing. Is the Dynamic range not starting on the 2:00 AM bar?

[NinjaTrader_PaulH]

Hello brucerobinson,

While waiting for your reply I had some time to closely look at your chart and it appears your charts are based on the EST. Please advise if this is incorrect.

I created a chart of 6b 1 minute and set the dynamic rectangle for the MST at 00:00 to 09:00 AM (to match your 2:00 AM to 11:00 AM setting)

The Dynamic Rectangle appears to be displayed correctly, reaching the Highs and lows found within the window. I added two horizontal lines one for the top of the range and the other for the bottom and then switched to a 5 minute bars and there was no change in the rectangle. I did try switching to a 6E chart and then back again but saw no different results on the return.

I've attached my example below (of course with more data later in the day).

If you can provide a repeatable condition/steps I would be glad to recreate it on my end.

[brucerobinson]

Hello Paul and thanks.
Sorry, I was away from my PC. To confirm - my PC clock is set to EST.
I've attached 2 further 6B charts that I hope may assist.
I've stretched the 5m to make it larger.

You will see:
a Yellow shaded Dynamic Range which is the London Session 02:00:00 to 11:00:00
a White shaded DR which is 02:00:00 to 05:00:00 'overlaid'

You may observe when the Time Series is 5m, the White is higher than on the 1m Time Series

My gut tells me it is something to do with bar timestamps at end of bar, not beginning, as I have recollections of this.
Being a simple human I expect if I type 02:00:00 start time in DR field it will produce the same whatever the Time Series (given start time is same).
The 'anomaly' doesn't happen with 6E - today.
I expect you can enlighten me, having coded it and knowing what goes on under the hood.
If correct, does this mean I need all start times xx:xx:01 as workaround to prevent this sort of thing?
Or something else entirely?

Kind regards,

[NinjaTrader_PaulH]

Hello brucerobinson,

Thanks for your reply.

I've created a short video of my observations here: https://Paul-ninjaTrader.tinytake.co...OF8xNTk0OTYzNg

If possible, please recreate what I've shown and see if that will illustrate your observations which are difficult for me to see on your screenshots.

[brucerobinson]

Hello Paul and thanks.
Re: video, helpful, yes seems you're configured to replicate.
I've emailed duplicate screenshots of last attachments in case that helps you see larger versions. Although when I open them here, I see them full size. Having said that, it isn't the easiest of issues to describe and that isn't my strongest forte, coupled with attempting to stretch charts for scale etc.

Anyhow, whatever the reason I appreciate it isn't helping either of us if you can't 'see' hence fully understand the issue for yourself.
In that vein, may I humbly/respectfully suggest as follows (as I can't see a way around it from my end as I don't understand why you get small images & if emailed ones don't resolve, but rather than go to and fro...(and my video creation skills are woeful))

I referred to the issue as an anomaly because I don't see it happening at other 'times', and don't see it happening on e.g. 6E (or 6A, 6J - I have all 4 instruments identical charts templates etc concurrently displayed). i.e. the issue can only be seen when switching Time Series and also 'doesn't happen' i.e. at different instances. So my instinct tells me it is specifically but likely not exclusively at the time/instrument 6B/price 'structure' I've stumbled upon at 02:00:00 on 4th when by chance I switched between 5m and 1m Time Series. i.e. I expect if I scroll to 02:00:00 on 3rd Feb, 2nd Feb, 5th Feb etc etc it may (or may not) occur - in the same way it does not appear on 6E at 02:00:00 on 4th Feb, or 6B on 5th,2nd,3d and likely does or doesn't at other DR White Boxes.
When you demonstrated switching from 5m to 1m in your video you were NOT where I experience the issue per my screenshots. i.e. replicate exactly. You switched Time Series on 5th at 02:00:00 and did a comparison, not 4th where I experienced the issue, then you went to different dates at 02:00:00 but DID NOT do a time switch comparison of charts/DR box (quite possibly because you're not yet clear what you're looking for because you can't see the screenshots full size - no criticism is intended, likely my descriptive capabilities also culpable).

With respect, may I suggest/ask you do a direct comparison. I suspect that if I replicate on 5th or 2nd or 3rd I will not see the issue that occurs at this particular event, just as I don't on 6E at 02:00:00 on 4th etc. I just happened to switch 5m to 1m on 4th at 02:00:00 and happened to notice because the box 'jumps', have reference points of horizontal lines nearby the price action etc. Conversely, I expect if I were to investigate on 6B or other instruments I might be able to find same needle in a haystack if the conditions that are causing it occurred, which I suspect is where/when the high is, and Repeater bar time. I might be wrong, but worth a try, given we don't have too many options. If you don't get the same, we may have at least made some progress (although not sure where that would lead if you can't/leave us in a better place... remote in to my pc)

To add some narrative to the screenshots if you can read the emailed attachments any better (or help to interpret them/articulate):
On the 5m chart, the White DR box begins on a down bar that just penetrates Horizontal Line Triangle style Salmon and just fails to penetrate Horizontal Line Dash Dot Dot Yellow above. Time axis isn't isn't displaying 5m but consistent with 02:00:00 bar. The bar, and its predecessor are a short-term higher High in the price action. Accordingly, the DR White Box with a start time Setting of 02:00:00 has a high of the top of this bar, between the two aforementioned Horizontal Lines
On the 1m chart, the high of the DR White Box (same Repeater Setting 02:00:00) is well below the two aforementioned Horizontal Lines, marked in red ink on the screenshot.....
Hence, when I switch time series the box 'jumps' down, hence noticeable.

This - imho, isn't 'right' (correct me if you disagree). At 02:00:00 there can only be one high of this 5 minute period the 5minute bar represents regardless of time series (for lower time series). Counting the bars, the (lower) High when using 1m Time Series that Repeater is using is 5 bars (minutes) later than the 5m (the first of the 'next' 5 minute period, as it is in a downtrend in the period). Simply put by a simple soul, if price was never up at the 2 horizontal lines after 02:00:00 on the 'more granular' 1m Time Series, how could it have been higher at or after 02:00:00 according to Repeater when using the 5m Time Series? Of course if price had risen, we wouldn't see it during the DR Box time span, once it had we wouldn't see it. Hence 'anomaly'.

Anyhoo, if from the above hopefully you are now able to replicate this on your own Platform that should I hope help with matters.

Kind regards,

[NinjaTrader_PaulH]

Hello brucerobinson,

Thanks for your reply.

So what is happening with the particular instance of February 4, 2021, on the 6B is that the 5-minute candle with a bar time stamp of 2:00 AM has a higher value than the 1-minute bar at 2:00 AM. That is what would cause your observations and it would not be instrument-specific as much as it would be price movement related so you could see this on any instrument where the 5 minute bar has a higher high than the 1-minute bar of the same timestamp. It likely would be more visible if the higher time frame was larger.

When you specify the start time as 2:00 AM, the indicator will start with the 2:00 AM timestamped bar. As a reminder, NinjaTrader puts the time stamp at the end/close of the bar.

The 2:00 AM 1-minute bar consists of data from 1:59:00:001 through 2:00:00:000. In the case of the 1-minute data series, the bar high was 1.3604. The 2:00 AM 5 minute bar consists of data from 1:56:00:001 through 2:00:00:000 The 5-minute 2:00 AM bar high was 1.3613 and this is why the box would appear to change because of the underlying data (1-minute vs 5-minute changed).

If you want the rectangle to only use data after 2:00 AM, regardless of bar size then yes, you could use 02:00:01 as the start time.

I've attached a screenshot to hopefully further illustrate.