His model was predicting a different outcome than The New York Times' model, which was heavily leaning in favor of a Donald Trump victory by 10:30 p.m. EST.

Statistics guru Nate Silver has thrown in the towel early this election night.

Silver, 46, shared an update to his popular Substack just before 10:30 p.m. to announce he was pulling his prediction model, in part because it wasn't "capturing the story of this election night well."

"Something like The New York Times Needle is a much better product," he conceded.

Prior to the abrupt take-down, Silver's prediction model gave Kamala Harris a 53 percent chance of beating Donald Trump as of 9:05 p.m. -- a sharp contrast from The New York Times' voting needle, which by 10:30 p.m. gave Trump an 81 percent chance of returning to the White House for another four years.

Silver, who founded FiveThirtyEight but now runs a Substack, wrote that part of his reasoning for bowing out of the prediction game was due to the fact that his team of two simply didn't have the resources to keep up with election results.

"We're spending a lot of time trying to fix code that isn't working right -- and it's distracting from our ability to cover the election for you," he wrote. "We think we took on one too many things, and we appreciate your patience."

Eastern time zone swing states like Georgia, North Carolina, and Pennsylvania are still far from being called, but they appeared to be leaning in Trump's direction hours after a majority of polls closed in those states. Michigan, meanwhile, appeared to still be a toss-up. Tens of thousands of votes -- if not more -- still needed to be counted in each state, however, at the time of Silver's shock announcement.

Election experts have said it's unlikely there will be an outright winner determined by Tuesday night as mail-in ballots and Election Day votes -- which are coming in even later after bomb threats were called into dozens of voting sites in swing states -- may take days to tabulate.