Hi Lionel! I think the easiest way to do this is to set the axis range to be enormous, let's say -10,000 to +10,000. Since the axis is blown out, the two numbers will look like they're colinear.

This is a really clever solution. Nice work, David!

Thanks, guys!

Regarding more elegant solutions, we can also show this in table form to alleviate the need to do the tricky trick.

Let me explain how I got here. First, I converted Day to be discrete. Then, I changed the final return calc to this:

(1+ZN(SUM([% change])))*(1+PREVIOUS_VALUE(0))-1

Note that the argument in PREVIOUS_VALUE() actually provides the value that the function will return when it's at the first index. So if you put 0 there, when you're at INDEX() = 1, it'll return 0.

I also wrapped the SUM([% change]) in ZN() to make it 0 if NULL.

Then, I just used the LAST() trick you were using before, except that I embedded 0 directly into it so it became a Boolean.

Thanks! And you're very welcome!