The entire commentary seems to rest on the idea that the effect size for Dean Radin's meta-analysis on staring experiments is too low. The staring weighted mean effect size is about 0.06, which is quite small. However, interpreting this as noise seems to go against how small effect sizes are interpreted in other areas of science. For example, Daryl Bem says this when comparing the effect size of ganzfeld experiments with a meta-analysis on the effects of aspirin on heart attacks:
"Taking aspirin reduces the probability of suffering a heart attack by only 0.008. The corresponding effect size (h) is 0.068..."
http://dbem.ws/Does%20Psi%20Exist%3F.pdf
Now, I don't know how Cohen's h compares to Radin's measure but I would expect them to be pretty similar. In other words, they are both "just noise" according to Novella but one of them resulted in the termination of a large scale randomised controlled trial because the effect was so reliable.
Although I'm not a statistician. Maybe the aspirin and staring effect sizes are not comparable after all.