The prediction markets pretty closely tracked NYT's probability estimate, which seems like the best possible analysis of the available data (partial current res

The prediction markets pretty closely tracked NYT's probability estimate, which seems like the best possible analysis of the available data (partial current results, complete past results, at the precinct level). Anything else would have been surprising.