Let's talk about the studies that don't show efficacy. Consider this one:
Small Trial Suggests Antimalarial Drugs Not Effective For Treating Coronavirus
Here's a quote from an article about the study:
"Like the Marseille study, the Molina trial was also a small pilot study. Molina and colleagues used the same dosing regimen as Gautret. In contrast, however, to the Gautret study, eight of the 11 patients had underlying health conditions, and 10 of 11 had fevers and were quite ill at the time the dosing began."
So, the study used only 11 patients, all of who were already showing very serious illness and 8 of whom had underlying additonal problems. What the study showed, if you can believe a study with so few participants, is that HCQ when given too late in the disease's progression may not be particularly helpful. Even the proponents of the treatment would agree.
I submit that this study was undertaken with the expressed purpose of giving the earlier HCQ studies a black eye. Both the original 20 person study in France and this 11 patient rebuttal study have problems. What one needs to do is look at the entirety of the data to gain a clearer picture.