I posit the only candidate who has a chance of spiking above 90% is Sanders - currently he is unlikely to dominate Super Tuesday, as he lags behind Biden in states like Texas and Virginia. If he were to sweep Super Tuesday (25%), for instance from a strong bump from Nevada, then I estimate ~70% chance there's a spike above 90%. This is a rough calculation that doesn't include other unlikely but possible scenarios (a sudden surge from Biden, Biden drops out to endorse Bloomberg, etc.), so I've added an uncertainty factor of 10%.
@Anthony Great suggestion, I expect there's a lot of ideas for AI questions here!
@alexanderpolta Pending - I think it'll show up on the site when Metaculus Prime mods get to it.
@DanielFilan good point - I read that but didn't update my model. Updating to 41.5%
@tenthkrige This updated me heavily upwards, I'm surprised it seems other's haven't. Anyone who is forecasting low (below the median) want to share their reasoning?