The MAD Ichimoku Squeeze

­­In this post we’ll discuss Ichimoku trend and momentum setups may be improved by using narrow range analysis in higher timeframes. For information on how the Ichimoku trend and signals are calculated, please refer to this post.

Qualifying Ichimoku signals by applying higher timeframe narrow ranges is similar to the squeeze setup. Basically, we’re looking to identify situations when the market is building up momentum for its next major move higher or lower. However, instead of using the Bollinger Bands with standard deviations to determine breakouts, we’ll use a Rolling VWAP with the mean absolute deviation calculation.

Consolidation as requirement for directional setups

Range Analysis in Higher Timeframes

The Range Analysis indicator from our NinjaTrader 8 Library detects the narrow range patterns, as described in Toby Crabel’s book “Day Trading with Short Term Price Patterns and Opening Range Breakout”. Specifically, Crabel states that these scenarios generally precede trending days. Identifying narrow ranges in higher timeframes, may improve the probability of certain momentum and breakout setups.

Crabel has a few ways of determining narrow ranges. First, he introduces the Inside Bar which is defined as having a range which is completely encompassed by the previous bar. The prior high is higher than the current bars’ high and the prior bar’s low is lower than current bar low.

In extension, one may combine this condition with a lookback period, identifying for example the narrow 4 bar pattern. As such we’ll look for an inside bar with a range narrower than the previous three bars’ ranges compared individually.

As mentioned, our Indicator Library Range Analysis indicator will enable you to identify these types of scenarios and although Crabel originally developed this type analysis for daily charts, one may can also be apply it for intraday charts. Below is a 30 min. chart of the ES contract and the the yellow bars indicate an Inside Bar Narrow pattern (IB4). We see an inside bar with a narrower range than the prior three bars.

Later on there’s a double inside bar (DI5) pattern, a fairly rare pattern telling us that there’s an even a tighter range following an Inside Bar Narrow pattern (IB4). Finally, note the slight difference in the coloring here between the up/down closes. Up closes will plot with gold and yellow signifies a down close.

Crabel also identified additional contraction patterns, for example the Narrow Range 7 pattern (NR7)as displayed below. It is defined by a range which is narrower than any of the previous 6 bars. However, it is not required conclude with an inside bar, but will be displayedas such when the last bar fulfils this specification, i.e. being a Narrow Range 7 inside bar pattern (IB7). Below is an example of the NR7 and IB7 patterns on a 240 min chart for the NQ.

Again, the concept discussed in this post is similar to the Squeeze setup, namely identifying situations where there’s a lull before the storm, or scenarios where the market is waiting for a catalyst to bring about the next major move higher or lower. Crabels assumption was that following a period higher timeframe narrow range, the new trend is likely to align with short term momentum.

Ichimoku Setups and the VWAP Value Area

However, one should always define a value area from where breakout setups should depart. One way to establish a value area where is to use a daily, weekly, monthly or N-monthly Volume Weighted Average Price (VWAP) indicator. This calculation plots the volume weighted arithmetical mean of all transactions taking place during the session. Effectively, prices for all transactions are added, dividing by the total number of contracts traded during that time. A value area can then be established using the first upper and lower standard deviation bands from the VWAP, as the majority of transactions (70%) are likely to take place within this area. It is well known that institutional traders use this area to find opportunities where there’ll be plenty of supply / demand in order to keep execution costs in check, i.e. submitting large orders in plain sight.

Still, a significant limitation with conventional VWAPs, is that they’re anchored to the beginning of each session. Therefore, one will have to wait some time to make useful readings from the standard deviation bands. Above is a weekly VWAP on a NQ 60 min. chart, anchored at the first day of the week. But one may at the earliest start using a weekly VWAP midway through the Tues. session.  That is because on Mondays the Current Week VWAP will display the same information as a Current Day VWAP, seeing that they’re both using the same anchor point. For directional setups , the general idea is then to take positions from inside the Weekly VWAP value and consequently, one would miss the signal identified by our premium Ichimoku indicator.

A MAD Rolling VWAP

For this reason, we’ve developed a Rolling VWAP taking data from the prior session and moving it forwards. We then see a wider and more stable value area, enabling you to consider setups at the beginning session. Again, the idea is then to take trend setups from value, towards the outer standard deviation bands. Once the 2nd standard deviation band is reached, institutional participants are likely to take a step back and momentum usually stalls, as seen below (NQ 60 min. chart). Those areas may therefore be used for profit targets.

One observation we‘ve made however, is that the standard deviation calculation might not be the best approach for calculating the deviation bands. This is particularly true as it relates to the outer bands which again, may be used for exit timing.

Briefly, standard deviations does not deal well with outliers, a.k.a. fat tails or black swans. The assumption behind the standard deviation model is that outliers do not occur frequently, but in reality this is not the case. Below is a NQ60 NQ chart showing two events beyond the 3rd standard deviation bands in only one week, which is clearly not representative for this model.

The Mean Absolute Deviation

So to avoid pre-mature exits, and hang on to profitable trades longer, we’ve started looking at using the Mean Absolute Deviation (MAD) which is available with our premium VWAP suite. Technically speaking, it is the Residual Mean Absolute Deviation, which applies a slightly different calculation.

In any event, it is indisputable that the MAD calculation plots a wider value area and that the outer Bands bands are farther removed from the VWAP when compared to the standard deviation bands. The idea is therefore to determine whether the probability for momentum trades can be improve by using the MAD calculation to establish the value and overbought/oversold areas.

Basically, the MAD calculation provides more information contained in extreme events (think 1929 crash, Black Monday of 1987, Dot-com bubble, etc). It does not seek to reduce the effect of outlier events but instead, include them as part of the analysis. This makes the Residual Mean Absolute Deviation bands more stable, providing more information about the tails of the probabilitydistribution.

A closer look at the statistical impact of this is available in follow-up presentation, available by registering in the form below. It explains how one may compare the probability Ichimoku setups by applying rolling standard deviation vs. MAD VWAP value areas and outer bands.


Summarizing the above points, we first went over how to identify consolidation scenarios, i.e. sideways markets in higher timeframes. Such scenarios, where the market is waiting for a catalyst to bring about the next major move higher or lower, are generally beneficial for identifying momentum setups.

Secondly, we’ll want to align Ichimoku signals with a VWAP value area, similar to the squeeze channel, spotting setups that may see institutional participation. However, instead of using the standard deviation bands as a method to determine value we use the Mean Absolute Deviation, i.e. MAD.