Looks like they learned a lot from PACE.
Don't do any objective tests like 6 minute walk or step test or actometers because they might reveal the truth.
Don't have a meaningful control group like the FINE trial, because that turned out to be a null trial,
Set the end point for reporting results at 12 weeks because that was the peak point for between group differences in FINE and PACE, and the between group effects might disappear by long term follow up, as they did with those trials.
Add in the Chalder Fatigue scale to primary outcomes, because that is more likely to be influenced by therapist effect than Physical functioning.
Set the significance level for between group differences so low as to be meaningless (a 5 point difference on SF36 is not, in my eyes, anywhere near clinically significant).
Continue to use a definition that will include people with, for example, temporary post viral fatigue, poor sleep habits or depression.
The most surprising finding of the trial is that, despite adding all these faults to the long list of faults of the PACE trial, they still managed to get such pathetic between group differences.