I tried estimating the chance that a new variant would arise in the UK in the next couple of months:
I think the risk of a new superbad variant arising in the UK is 6%. We’ve seen two game-changing variants so far (Alpha and Delta) out of roughly a billion Covid cases (extremely crude estimate). The UK will have roughly 10 million cases in the coming three months (extremely crude estimate). That would be 1% of the total so far, so a 2% prior for a new game-changing variant (since this happened twice already). Conditions for evolving vaccine evasion have never been better, which adds at least a factor 2, I would think. It’s more in expectation, so maybe 3.5. (Also, the Alpha variant happened in the UK, so maybe conditions are particularly favorable for virus evolution here for reasons I don’t understand.) OTOH, I may be ignoring that the virus has run out of low-hanging mutations. Overall, I’m going with a 6% chance.
Note that I don’t necessarily predict a new variant to be more deadly by itself. (But it would be more deadly given better resistance to vaccines.)
It’s indeed scary that the same experiment will be run across many countries, so in absolute terms, the odds are much larger than what’s correct for the UK for the next couple of months.
But the risks per country are heavily correlated (are there low-hanging mutations that increase transmissibility?), and overall I’m not sure I’d go above 40% for a new superbad variant in 2021. I think this is partly also influenced by having read some experts express a lot of confidence that the antibodies to the spike protein, especially from the Pfizer and Moderna vaccines, are fairly hard to circumnavigate when you’re the virus, because probably all Covid viruses need some kind of similar-looking spike protein. Even so, you could get a variant where infection is reduced by 50-70% after two shots of Pfizer, instead of the 15-30% we see currently. That would basically guarantee that nearly everyone gets exposed to long Covid risks of having to go through one illness.
I tried estimating the chance that a new variant would arise in the UK in the next couple of months:
Note that I don’t necessarily predict a new variant to be more deadly by itself. (But it would be more deadly given better resistance to vaccines.)
It’s indeed scary that the same experiment will be run across many countries, so in absolute terms, the odds are much larger than what’s correct for the UK for the next couple of months.
But the risks per country are heavily correlated (are there low-hanging mutations that increase transmissibility?), and overall I’m not sure I’d go above 40% for a new superbad variant in 2021. I think this is partly also influenced by having read some experts express a lot of confidence that the antibodies to the spike protein, especially from the Pfizer and Moderna vaccines, are fairly hard to circumnavigate when you’re the virus, because probably all Covid viruses need some kind of similar-looking spike protein. Even so, you could get a variant where infection is reduced by 50-70% after two shots of Pfizer, instead of the 15-30% we see currently. That would basically guarantee that nearly everyone gets exposed to long Covid risks of having to go through one illness.