Let (\Omega,\mathcal{F},\mathbb{P}) be the probability space of interest.
It is given that the random variable X\sim \text{Unif}(0,1).
The random variable Y is defined as Y(\omega).=\begin{cases}X(\omega) & \quad\text{ if }0
And, if we multiply CDF by the power of 0.2 to obtain its e-value, it turns out that it is exactly the e-value of F(1/2)! This was a great experiment by John to show that if we use the right tools, we can figure out most of the relationships between these weird functions and make a lot of interesting measurements. And I suspect he is right about the value of e: the correct value for e is 1/2. But, to get a better understanding of what we just did and why we did it that way, you might want to read the article here.