2024 Election Day Forecast

Election Day Updates

• 11/6/24 09:30AM: That's a wrap on the 2024 U.S. presidential election, and this model performed surprisingly well. While Harris maintained a slight advantage in the odds (maxing out at a 3 in 5 chance in early October), things tightened up as we neared election day. In the end, the model predicted that Trump's most likely winning outcome would be a score of 306 to 232, and it looks like the true outcome will be quite close to that. It will be tough to convince people of this, but it has to be said: the polls were mostly pretty good. In the swing states, Trump managed to beat his polling average by only 1%-3%. However, the polling averages were so close in the swing states that even a minor swing like that means a whole heck of a lot. In another universe where Harris beat the polling averages by 1%-3%, it would have been her sweeping all the swing states. Good luck everyone, and I'll be back with a U.S. Senate forecast in 2026.
• 11/6/24 08:00AM: Some final updates, though it wont change much. Harris takes Hawaii, Minnesota, New Hampshire. Trump gets Iowa, Nebraska. And that knocks out that 0.5% chance of a tie. Trump is now 100% to win.
• 11/6/24 04:00AM: Trump wins Pennsylvania and Wisconsin. Harris wins Nebraska's 2nd district and Maine. Trump has more than 99% chance to win with this update. Though the model is showing a 0.5% chance of a tie for whatever that's worth.
• 11/5/24 11:05PM: Last update for the night, I promise - Virginia to Harris
• 11/5/24 11:00PM: Georgia is going in the Trump column. The networks are being a little slow to call it, so I'm prepared to eat crow in this. Unfortunately, I need some sleep! I will update in the morning, but Trump has risen to a very good lead. Harris's path to victory has significantly narrowed, and if she can't win Pennsylvania, then it's over for her.
• 11/5/24 10:30PM: California, Oregon, New Mexico, Washington goes to Harris. Idaho, Kansas, Missouri, and North Carolina go to Trump. North Carolina is the first swing to be called, and Trump jumps to a stronger lead.
• 11/5/24 09:20PM: Ohio goes to Trump. Trump is now in the lead on the forecast, but it's still essentially 50-50. Don't expect much movement until one of the swing states - WI, MI, PA, GA, NC, NV, AZ - are called.
• 11/5/24 09:05PM: Utah and Montana go to Trump.
• 11/5/24 09:00PM: Colorado goes to Harris.
• 11/5/24 08:35PM: New York for Harris. Louisiana for Trump.
• 11/5/24 08:10PM: Nebraska goes to Trump.
• 11/5/24 08:05PM: Rhode Island, Connecticut, New Jersey goes to Harris. Texas, North Dakota, South Dakota, and Wyoming go to Trump. It's basically still a 50-50 toss up.
• 11/5/24 07:55PM: Massachusetts goes to Harris.
• 11/5/24 07:50PM: Tennessee, South Carolina for Trump. Illinois, Maryland, D.C., Delaware for Harris. The 'Momentum Bump' variable has now been zeroed out.
• 11/5/24 07:40PM: Doing this in batches since a lot of states have been called: Florida, Alabama, Mississippi, Arkansas, Oaklahoma go to Trump.
• 11/5/24 06:45PM: West Virginia goes to Trump. Again, not much change because these were expected results, but Harris's lead is fading slightly as her 'Momentum Boost' variable starts to disappear.
• 11/5/24 06:20PM: Vermont called for Harris, Indiana for Trump. Very little change in top line odds.
• 11/5/24 06:00PM: Kentucky has been called for Trump. Doesn't change the forecast at all.
• 11/5/24 04:45PM: No states called, but the numbers have been updated with whatever polling data came in late last night.
• 11/4/24 10:15PM: Harris begins Election Day as a mild favorite with 55% chance of winning. Since Harris was gaining in the final days of the forecast, she gets a small 'momentum boost' in her favor, but the effect of this variable will quickly go to zero as the first batch of states are called.