Earlier quoted context omitted.
You’re absolutely right, but I didn’t want the overall message to get lost in the details. I’ve been seeing people parrot the “AZ is actually much better than it looks because they included asymptomatic patients” for months now and it’s just incorrect.
I’m honestly confused! Can you point me to your source for the excerpts in your sibling comments? Edit: reading your lancet link now (sorry about that). My understanding from reading the description of the Moderna results was “XX / YYY people developed covid; but we didn’t test everyone in the YYY set” (so people who definitely got it / went to a hospital / whatever are in XX, but asymptomatic cases may not be). Agre…
As the other commenter pointed it may be possible that someone had mild symptoms they were ignoring then tested positive on the weekly screening and said "oh yeah I do have a sore throat" but the numbers should still be roughly comparable. It's not worth diving into 94% vs 95% efficacy but I think we can be pretty confident that the 95% vaccine is more effective than the 60% vaccine.
This is also muddied because the AZ trials were a bit of a mess with dosing error and varying times between shots. So now we have a number of different subgroups with their own statistics but not a big enough sample size to say anything confidently about half dose vs full dose or 2 weeks vs 12 weeks. The confidence intervals are a mile wide, but they still get talked about (and even drive policy) adding a lot of noise to the discussion about how effective the vaccine really was.