Keltner Channel Wedges

Demo of the Final Version of the Indicator

Here's what the final version of the "Overly Extended" indicator looks like:

You should notice the "Boolean Wedges" (that's what ThinkOrSwim calls them), indicating that price action is too extended. Our definition of "extended" involves using the Keltner Channels, and specifically measuring price action's relationship to the upper and lower Keltner Channel bands.

It will plot both long and short wedges, using a very simple boolean condition, which we will write in the thinkScript section of this tutorial.

 

Understanding the Keltner Channels

Before we start writing our "Overly Extended" indicator, we need to understand the Keltner Channels code.

Let's start with the user inputs.

While there are a few different inputs available, the one input that we care about is the "Factor" variable. The "Factor" variable is what determines how elastic the Keltner Channel upper and lower bands are (the higher the number, the "more extended").

A 3.0 factor means we're increasing the elasticity of the Keltner Channels. As you continue to increase the factor, and if the "Overly Extended" boolean wedges continue to plot, the further extended we are.

This "Factor" input is something we'd like the user to be able to control from the studies menu in our version of the final indicator. With the value that the user inputs, the upper and lower bands of the Keltner Channel will adjust accordingly (and thus, our wedges).

Using your newfound knowledge about the Keltner Channels, we are now ready to start writing some custom thinkScript code.

 

Writing thinkScript Code

Now, let's start writing some thinkScript code. We can give our new indicator a name, calling it TI_ExtendedKeltnerChannels.

We can start by first defining our one input variable, which is the Keltner channel "Factor." We make this an input variable, so the user can change the value directly from the study's menu, instead of needing to edit code. We can define this variable as "KCfactor" and give the variable a default value of 3.0.

Next, I can define our extended variables, which is what will contain the logic behind what plots on our chart.

Let's start defining the long-side extended variable, which we can call "extendedLong." Our condition checks if the closing price of the current candle is greater than the Upper Band of the Keltner Channels. If it is, then "extendedLong" will return a 1, plotting true.

Now, we can copy paste this one more time, for our short side.

The reverse of the above condition is true for the short side. In this case, we are checking if the closing price of the current candle is less than the Lower Bands of the Keltner Channels, using the same "KCFactor" variable.

Finally, we can add in the appropriate formatting, by defining the painting strategy. We can set the painting strategy for extendedLong and extendedShort to wedges, using:

• "PaintingStrategy.BOOLEAN_WEDGE_UP"
• "PaintingStrategy.BOOLEAN_ARROW_DOWN"

And, finally, we can also define the color, with our overly extended on the long side wedges painting as pink, and the short wedges as light_green.

Here's an example of what the final version of the indicator looks like, plotting on a chart of the S&P 500 ETF (SPY) on a daily time frame chart:

In terms of how you can continue to improve this indicator beyond what we currently have, you can start to make these conditions a bit more intricate. For example, you may also want to include an oversold/overbought indicator, such as the RSI (ie. RSI > 70).

You can layer on other indicators that you may prefer trading with, to continue to build out the "extendedLong" and "extendedShort" variables, and start to make these conditions a bit more complex.

The goal of the indicator is to not only help save you chart space, but also the mental bandwidth to automatically alert you when we are overly extended.