Exactly two weeks before Americans cast their vote for president, the statistical forecast models looking at the race are painting an increasingly grim picture for Republican presidential nominee Donald Trump.

According to fivethirtyeight.com’s Nate Silver’s polls-plus forecast, which includes polls, economic data, and historical polling data since 1972, Democratic presidential nominee Hillary Clinton has an 83 percent chance of winning the presidency compared to Mr. Trump’s 17 percent chance of winning the White House.

Put another way, under Mr. Silver’s forecast model, if the election was held 100 times, Mrs. Clinton would win the presidency in 83 of those simulations.

The polls-plus forecast has three states as near toss-ups, Iowa, Ohio, and Arizona. Moving out of the toss-up column are Florida, which now has a 69 percent chance of going to Mr. Clinton and North Carolina which has a 63 percent chance of going to the Democratic nominee.

Based on Mr. Silver’s forecast, Mrs. Clinton would receive 326.9 Electoral College votes compared to 210.3 Electoral College votes for Mr. Trump.

Losing Florida could be the biggest loss for the Trump campaign. Losing 29 Electoral College votes from the map would likely end Mr. Trump’s campaign for the presidency on election night. He would have to take North Carolina and Ohio to get more votes than Florida has by itself, and he is currently trailing in polling in North Carolina.

Moving to the New York Times’ “Upshot” election model, Mrs. Clinton currently has a 93 percent chance of winning the presidency compared to Mr. Trump’s seven percent chance of a win. Put another way, the New York Times said the chances of Mrs. Clinton losing are approximately the same as an NFL kicker missing a 29-yard field goal.

Of course as the Arizona Cardinals and Seattle Seahawks proved, no field goal is a sure thing, even for an NFL kicker.

Princeton’s Election Consortium currently gives Clinton a win in the Electoral College by a 334-204 margin. The Princeton project looks at 184 state polls. The Princeton consortium pegged Clinton’s probability of wining in November at between 97 and 99 percent. It also gave a “meta-margin” of 4.4 percent. The meta-margin is “the amount of opinion swing that is needed to bring the electoral vote estimator to a tie,” meaning that opinion polls would have to be off by more than four points to have the candidates tied.

Finally, PredictWise, which looks at prediction markets, polling, and social media data, currently gives Mrs. Clinton a 92 percent chance of winning the election compared to Trump’s 8 percent.

The overwhelming majority of polls continues to put the race into a solid win for Mrs. Clinton. Polls can and have been wrong in the past, most notably with the recent British exit from the European Union. In the end, the only poll that will matter will be the one conducted on November 8.