COP is relevant in LENR experiments where (as is often the case) there is fairly standard calorimetry, with errors not properly quantified but supposed no more than "typical 10% or so".
I do not think this is "often the case." As far as I know, this is never the case. Which experiments do you have in mind? In what paper did you read "typical 10% or so"?
If errors are not properly quantified than the COP would not make any difference. Whether it was 110% or infinity (meaning no input power), the results would be questionable, or meaningless.