Anyone notice how small the infection rate is for the placebo groups?
Yes, it's a feature, not a bug. You can't have been infected to start and the trial has only been going since summertime. Also, all of the participants get education in how to stay safe.
But that's why there is a control group. So there were 95 infections. Five were in the study arm and 90 were in the placebo arm.
Hypothesis (null hypothesis): the vaccine doesn't work and this difference in infection rate was solely due to chance. OK, so we run the statistics (there are well-established methods to run these statistics) and we find out that there is a 1:1,000,000 chance that an ineffective vaccine could produce this result, or a p value of 0.0001. By convention, it is accepted that the null hypothesis is "disproven" if the p value is lower than 5. So we have now disproven the null hypothesis.
So how effective is the vaccine really
given these numbers? Well, there is a 95% chance that the actual efficacy is somewhere between 82% and 99%. Now, given what we saw from Pfizer, which uses similar technology, I think we can exclude anything below the high 80s.
Now, how about severe cases? There were 11 severe cases, all in the placebo group. Does this prove that the vaccine is 100% effective at preventing severe COVID-19? No. When you run the statistics, it shows that it probably doesn't make people more likely to have a severe case, but if you get COVID-19 after having been vaccinated, we don't have enough data to say for sure that it will protect you from severe disease. However, given the way that vaccines work, my prediction is that we will see eventually that even if you do get COVID-19 after having been vaccinated, that you will be much less likely to get severe disease.
In any case, this is fantastic news. I think it's time to get to work on distribution. Over 600 people in the USA died yesterday alone.