different pivot points on different time frames

[tomek001]

For some reason I'm getting a different pivot point on three minute chart for TF than I do on 6RB. Today 3 min was 649.3 and 6 RB was 649.7 Not very large ( relatively speaking ) but large enough for missing a trade or exit.

What might be causing it?

th

[Harry]

- False Data (who is your data supplier and what is your time zone)
- False Settings (Dailybars or Intraday)
- Indicator bug

If you want to find out, give us more details and a chart.

For some reason I'm getting a different pivot point on three minute chart for TF than I do on 6RB. Today 3 min was 649.3 and 6 RB was 649.7 Not very large ( relatively speaking ) but large enough for missing a trade or exit.

What might be causing it?

th

[tomek001]

I get the same data zen fire, the same settings for charts 9:30 - 4 PM.

th

[NinjaTrader_RyanO]

Hello,

Thank you for your post.

Can you please let me know who your data provider is?
Also, can you please clarify the manner that you are calculating the pivots, whether using previous day data to calculate or intraday data.

With that information I will test this issue on my end.

[tomek001]

Hello Ryan,

Data is from zenfire. I'm not sure i understand your question i.e I don't calculate pivot point myself, simply get it from ninja 6.5 from hte list of indicators.

th

[NinjaTrader_RyanO]

Hi,

Right, within the indicators menu when the applied pivots indicator is highlighted is a parameter titled PriorDayHLC. With this parameter you tell NinjaTrader how to determine those values, which it then uses for the pivot calculations.

With Zen-Fire that parameter will need to be set to CalcFromIntradayData for both the 6RB and the TF.

[tomek001]

Thaks Ryan,

It seems I had priorday HLOC selected on one chart and not the other charts ( I still don't understand why it would make a difference for the same symbol ). Once I included those on the other chart, the values for PP seem consistant.

thanks

[NinjaTrader_RyanO]

Hi,

It makes a difference because Zen-Fire does not offer daily historical data, which is what would be needed for calculating pivots on daily reported HLC values. So, rather, with Zen-Fire you use the intraday historical minute data to calculate the values.

[Harry]

Specific problem with range bars and CalcFromIntradayData

I have reedited and modified this post, because it was in contradiction with the previous post and would create confusion.

But I want to reexpose the problem with range bars and pivots. The selected session end (chart properties) does not cut off the range bar, but you get an overlapping range bar between two sessions.

Therefore the pivots indicator will - if set to CalcFromIntraday Data mode - take the close of the last range bar that closes prior to the session end and take it as a proxy for the session close. If you look at the example below the close of the last range bar within the session is 1.3705. This is ten pips off the session close of 1.3695, which is actually hidden within the overlapping bar. The close 10 pips off causes an error of 3.3 pips for the main pivot PP.

Maximum error induced by the false session close would be one third of the chosen range, the expected error 1/6th of the chosen range. For the chart with a 20 pips range this means

maximum error (if not closing near high or low) 6.6 pips
expected error 3.3 pips
observed error 3.3 pips (rounded to 3 pips)

So with Zenfire and no daily data, if you use pivots with CalcFromIntraday data on range charts, do expect an error.

The same applies to minute charts if they do not fit exactly into the session and exceed the session break. A 37 min chart may create overlapping bars.

However NT cuts off volume and tick charts at the session end (as per chart properties), so Pivots are accurate on volume and tick charts.

This has been changed for NT 7. Hope I confused everybody now, LOL.