I will never acknowledge the existence of LENR/CF unless there is an experiment that is repeatable 100% of the time by reputable entities with 100% (and not just 1 part out of 200,000,000) of the energy clearly accounted for by quantities measured during the experiment.
I agree with the rest of your post but this doesn't make sense to me. If someone designs an experiment for which a clear cut positive result is defined, the probability of error in measurement is extremely low (including good calibration methods, best measurement methods and devices, reliable labs doing the measurement, accounting for or ruling out Shanahan's calibration constant drift, and so on)... if that can be accomplished even once out of many tries, I would be pretty sure there is an accounted phenomenon present, some sort of anomaly. If there is a low yield, I would certainly try to account for it, but it would not rule out concluding that *occasionally* known and unknown parameters of the experiment "lined up right" and something strange was indeed happening. Might even be LENR.
LENR advocates think they have reached that criterion but I am not convinced I have seen it demonstrated. And I resent the defensiveness with which it is all discussed. Asking for easy to read graphs and papers is "spoonfeeding" and asking for high level level results in which error would be extremely unlikely is "shifting the goal posts." Requesting high level results together with good calibration and calorimetry, long duration, and high signal to noise ratio is being overly fussy. Then, it follows, according to believers, I/we wouldn't believe it anyway unless the device was for sale (ridiculous!) Those are the problems. Not reaching 100% consistent results. In my estimation.