I am glad to hear that you are able to trade with a method which improves on MF system. I am not familiar with sessionNoise bands. Will look up today to see if I can find useful information about it. I will appreciate it if you can point me towards a link/source where I can read about them.

FatTails, for the 6E, what is your start time for the initial balance opening ? Is it 2am or 3am EDT?

You cannot read about them. The idea is rather simple. Over a lookback period of N days, you just look how much price moves from the open to the high and from the open to the low of the session.

For each day the smaller of (high - open) and (open-low) is considered as noise. Then you calculate the N-day average of all noise values. This gives you an expectation for a threshold, where a breakout from the open may either succeed or fail.

The noise bands use two lookback periods, for example 10 and 20 days to take into account varying volatility. It is understood that the noise bands only represent the range calculated from the average failed breakouts. I may add standard deviation bands at a later stage to get a confidence interval.

Chart attached with A and C values and noise bands for 6E.

You cannot read about them. The idea is rather simple. Over a lookback period of N days, you just look how much price moves from the open to the high and from the open to the low of the session.

For each day the smaller of (high - open) and (open-low) is considered as noise. Then you calculate the N-day average of all noise values. This gives you an expectation for a threshold, where a breakout from the open may either succeed or fail.

The noise bands use two lookback periods, for example 10 and 20 days to take into account varying volatility. It is understood that the noise bands only represent the range calculated from the average failed breakouts. I may add standard deviation bands at a later stage to get a confidence interval.

Chart attached with A and C values and noise bands for 6E.

That sounds too complicated. My problem with the Fisher method is that the breakout points seem to be set somewhat arbitrarily. Sometimes they are very close to the opening range and sometimes they are far away.

So I have settled for a different concept, based on failed breakouts. A failed breakout is the smaller move made by price as counted from the open:

Failure move = Mininum of (High - Open) and (Open - Low)

The smaller of these moves can be considered as noise produced by the market while it was in its typcial bipolar state of mind. From this I am calculating noise bands that define the levels beyond which a price move becomes meaning full. I have produced the noise bands indicator, and will in a nex step add standard deviation bands to the noise levels to get significance levels along with it.

Options implied volatity would be more interesting for adjusting daily range targets, but that is another story.

Fat Tails,

Can you post an example as to how you use noise bands? if you trade WTI crude, that will be nice example. I typically have trouble figuring out failed breakouts ( before its too late) . Is noise band indicator available for down load?

Thanks,

Fat Tails,

First I want to thanks for your work.
I would want to apply those indicators on the ES and I've tried to replicate to see If I was doing well, your CL chart, but I can't put the noise bands just for the american time, they appear extended to the all day. I dont know if it is cause you are working with the V37 and I have the V34. And I also do not have the V38 volatily band, could be for that that it doesnt matchs with yours?

Since the ranges on the ES are smaller, do I have to change something on the Options/Parameters of the indicator?
Could you sugest me something please?

Thanks a lot by now,

Felix.

PS: And could you also send me the page of those indicators (volatily Bands) ? I couldn't find those.

The Session Noise is the smaller of (High - Open) and (Open -Low), which I consider the failed move from the open. The Session Noise Bands calculate the average size of the failed moves over the last n days. The indicator is available for download here:

Download: SessionNoiseBands V39 Exported using NT Version 7.0.1000.10 This indicator is an old version and there will be no more maintenance. All the features have now been integrated with the SessionPivots indicator that can be downloaded here: https://nexusfi.com … personFat Tails

Let us have a look at the statistics:

The average noise over the last 20 days for CL was 65 ticks. The average daily range was 345 ticks. This means that id you use the noise bands as a tool to enter a trade in the direction of the breakout, when it retraces to the noise bands, your expected target is about 345 - 65 - 65 = 215 ticks. Of course this is oversimplified, as you need to apply confidence intervals to calculate the expectancies, but it shows the way to go.

The chart shows the noise bands for the last 3 business days.

Tuesday: CL10-11 broke out upwards from the noise bands and retraced to the upper band. The pullback was the ideal entry for a long position. The target band - representing the average daily range (ETH) - was reached later during the session.




Monday: This was a holiday session. CL 10-11 had made a larger move during the European morning. The US session was quiet, as CL 10-11 mostly stayed within the noise bands. At the end of the regular session the daily range target (ETH) was reached after a weak breakout.

Friday: There was no valid breakout as CL 10-11 stayed within the noise bands during the US session. Nevertheless the daily range target (ETH) was reached as well, as there had been a large move prior to the opening to the US session.