I suppose that depends on whether statistical analysis is science or guesswork. What TP does is take years (more than a decade) of data, analyzing the wait times at each attraction throughout the day to predict what will happen this year. For the Spring season, they will look at the data they found for the exact calendar date; the overall trends for the first week of April; the data for second week after Easter, (irrespective of when Easter falls); and specifically the data for the first week of April when Easter fell in the third week of March. Using all of that "past performance" data, they make a prediction for this year. I am not a statistician so I don't know if this amounts to "science." But it seems that it is more than guesswork. Regardless, the first week of April seems to be outpacing prior years.
Question: Are the increased crowds being experienced at all 4 parks or mainly at MK? Could the fact that DHS is half-closed be moving people over to the MK more so than predicted? It might be that WDW has no more people than predicted, but that they are redistributed differently. And how does one assign a crowd level number to DHS with half the park closed? There could be half as many total people there but the wait times could be longer than normal after you close down crowd eaters like LMA and the Backlot Tour.