Latest news

Oct. 21, 2020

President Trump is running out of time. Joe Biden leads by double digits in national polls, and state-level polling is only slightly closer. In fact, Biden’s lead is so large that traditionally red states like Georgia, Iowa, Ohio and Texas might now go blue. We’ve entered the last two-week stretch before Election Day, and Trump needs the race to tighten — we’re way past the point where a normal polling error could let Trump close the gap. Still, he has a meaningful chance per our forecast — a little worse than the chances of rolling a 1 on a six-sided die and a little better than the chances that it’s raining in downtown Los Angeles. And remember, it does rain there. (Downtown L.A. has about 36 rainy days per year, or about a 1-in-10 shot of a rainy day.)

  • Want to test out our forecast and see what happens if Trump wins Florida or Biden wins Texas? Well, now you can! We’ve built an interactive forecast that lets you explore the ways Trump or Biden could win the election. But there are limits. (Yes, we’ve seen the Fivey memes.) If Biden or Trump has less than a 1.5 percent chance of winning a state, you can’t select that candidate, as ultimately we’re hoping you’ll use this tool to explore some very plausible scenarios. That’s certainly how we plan to use it in the lead-up to the election — and beyond if it’s not clear who the winner is on election night.
  • The last two weeks of the campaign can be an incredibly stressful time, so editor-in-chief Nate Silver offers eight tips to stay sane, including paying attention to polling averages (not individual polls) and not putting too much stock in how either campaign “feels” (they usually do not have a better read on the race than the polls).

2020 Election Coverage

To put all these numbers in context, check out our coverage and subscribe to the FiveThirtyEight Politics podcast!


We simulate the election 40,000 times to see who wins most often. The sample of 100 outcomes below gives you a good idea of the range of scenarios our model thinks is possible.

Trump win
Biden win

No Electoral College majority, House decides election
Don’t count the underdog out! Upset wins are surprising but not impossible.

Every outcome in our simulations

All possible Electoral College outcomes for each candidate, with higher bars showing outcomes that appeared more often in our 40,000 simulations

More bars to the right of the 270 line means more simulations where that candidate wins. Some of the bars represent really weird outcomes, but you never know!

The winding path to victory

States that are forecasted to vote for one candidate by a big margin are at the ends of the path, while tighter races are in the middle. Bigger segments mean more Electoral College votes. Trace the path from either end to see which state could put one candidate over the top.

Maine and Nebraska’s congressional districts are shown separately because those states split their Electoral College votes, allotting some to the statewide winner and some to the winner of each district.
We call this the 🐍 snake 🐍 chart! This makes it easy to see why some states are hotly contested and others mostly get ignored.

How the forecast has changed

The forecast updates at least once a day and whenever we get a new poll. Click the buttons to see the ways each candidate’s outlook has changed over time.

As the election gets closer — and as we get swamped with new polls 😬 — the forecast will get less uncertain.

Weird and not-so-weird possibilities

The chances that these situations will crop up

Who’s ahead in national polls?

Our model relies mainly on state polls, which it combines with demographic, economic and other data to forecast what will happen on Election Day. If you want to see a snapshot of what voters are thinking right now — with no fancy modeling — check out the national polls.

Want more stuff like this? See how the race is shaping up in individual states, or watch our Election Updates on YouTube.
Congrats, you made it to the bottom! If you’re looking for the nitty-gritty of how our forecast works, check out the methodology.