I want to repost some thoughts from TrumpCoat. This was posted yesterday.
“Hi, I’ve been lurking your comments because I love crunching numbers from early votes as well. I predicted 48 out of 50 states on another forum - the other 2 were MI and WI which came up as a tie in my model so I gave to Hillary.
I’ve loved Methos8’s spreadsheet and analyzed it.
A few questions that have came up - which way do I’s vote? To answer that question, I did a little digging of my own. I took the early votes from 2014 and 2016 by county in FL, and assigned an estimated I percentage support for Trump by taking Trump’s final margin of victory for that county and multiplied that by the early vote totals, assigning all the R’s to the Trump side, and dividing the leftover number by the I early vote turnout.
These are the graphs I came up with, feel free to verify and reuse them. https://m.imgur.com/a/I4977Z6 As you can see, I support tends to lean with the county’s overall leaning. I also made 2 sets of graphs based on total turnout, not just I turnout, showing that counties that support Trump are coming out in force.
Anyway what the graphs say is we’re comfortably smack between 2014 and 2016 in “redness” and I’s probably leaning that way.
Wait, that’s not all! We have a few outlier counties in terms of high turnout relative to 2014. Here’s what I found: FL passed a law recently that all universities must have early voting stations. Alachua is a college county, so no surprise that. Also, Broward, Orange, and Osceola had the largest influx of Hurricane Maria refugees - but Osceola is the only majority Puerto Rico county. Take these counties out and Bay, and that’s a more accurate trendline for where the rest of the nation that isn’t so affected by hurricanes or college voting laws may be going. I’s might be just as red as 2014!”
That’s a lot to take in. What’s the gist of this?