Predicting the NY Democratic Primary

My county model (or more accurately one of the models - as the situation is more of a continuum) of primary only states came up with a prediction of Clinton 62.0 / Sanders 38.0 for NY.

That said, the main goal was to predict county values that would let me use my real-time county swing analyzer to predict the state swing.

In previous elections (WI, IL, OH, MI, NC, FL, MO), my real-time model had generally started off being more pro-Clinton and drifting 1-1.5% to Sanders as more votes came in -- possibly as absentee ballots were counted first and went for Clinton. By contrast in NY, my model started off forecasting 56/44 (and I was forecasting 55/45 as I expected a 1% swing to Sanders that I'd seen in the other states) and ended up at 58.1 / 41.9. It's possible that NY counted absentee ballots last or that there was another factor at play.

The good news is that my model was more accurate at predicting the final result than the NYT Upshot model. For instance, I'm currently predicting a 16.44% lead for Clinton, NYT's Upshot model is predicting 15.8% and the Secretary of State numbers are 16.22%.

So it looks like my state prediction was off by 3.8 to 3.9%. By contrast the polling average (Pollster - 56.66%) was off by 1.5%.