A proper prediction market (virtual one) is designed to make people reveal their expectations early. That solves the problem of "waiting for discovery".

Must disagree. This - and several more issues due to real money - was well known to first generation prediction market operators since the middle of the 2000's. Almost all of them went under, which why I was so surprised to see the 1st gen model pop up again with crypto starting with Augur.

Correct, and it is just one further issue with any prediction market mechanism deployed on a real money (or a fiat currency impostering as a real money, for the avoidance of doubt).

I can confirm that for this reason, and a few more, a prediction market mechanism works much better in a well-designed virtual market economy where bets are not denominated in a monetary currency. In fact that's why I sold Redmonitor, a real money prediction market I founded in 2004 only 4 years later, in 2008 and started a virtual one.

In principle, I like the research question and the comparison above is probably the most you can make out from what is published. That said, it is the year 2022,  capabilites and methodology have advanced enormously at least with those PM firms operating successfully  in the commercial world markets. So it's the proverbial comparing  apples and oranges on several dimensions to talk about how "prediction markets" (sic) perform  for whatever. Different platform implementations  have very different capabilities suited to very different tasks. Moreover, like any advanced tool, practical application of the more advanced PM platforms  need a high degree of methodic knowhow on how to use their specific capabilities - based on real experience of what works and what does't.  

OP's claim of inadequacy of EA London COVID measures lacks an objective "effectiveness" measure. It criticises a lack of clarity of the EA measures  while lacking any clarity itself. 

I am stressing this as this same problem applies to public policy in many countries.  Australia is a severe case: its extremely harsh travel policy seems based on the premise that either the virus and its innumerable variant mutations will simply go away or that Australia can remain in perpetual lockdown without serious negative socioeconomic consequences on the medium and long term. 

For a proper decision (even just at EA Lonon) we would need a collective forecast of the objectively measured number of participants who will be infected/severe cases/die at EA London with the current policy and conditional on the one proposed by OP. Step 2 would be a vote in full knowledge of the forecast consensus and distribution. Prediki Prediction Markt offers free use of its platform for such "effectiveness" purposes. While this would be an effort, it might be well worth it, as a showcase to the world (and Australia) on how to decide such matters objectively and "effectively".