The Silver Bulletin win probabilities have moved even though the popular vote forecast has barely moved, because the translation from popular vote to Electoral College follows the baseball game principle.

Judged as of 2024-10-08 against Silver Bulletin's published 2024 presidential model history. True: in the period Silver is describing (roughly the preceding weeks), Harris's projected national popular-vote margin moved by a small amount (around 1 point or less), while the Electoral College win probability moved noticeably more (several percentage points or more). Mostly true: both hold, but the size of one movement is somewhat off from what the claim implies, e.g. the popular-vote forecast moved a bit more than 'barely' or the probability shift was modest but still clearly larger in relative terms. Mixed: only one half holds, e.g. probabilities moved but the popular-vote forecast also moved by a similar relative amount. Mostly false: the probability shift was minor, or the popular-vote forecast moved substantially. False: the model history shows win probabilities essentially unchanged in that period, or popular-vote and probability movements matched each other.

Set before the research began.

1. E1 Kamala Harris no longer favorite in Nate Silver model: "Literally 50/50" Newsweek · 16 October 2024

   “It comes as Harris' odds of victory in the election have dropped by about 6 percentage points since late September, with Trump gaining in 19 states, according to Silver's forecast.”

2. E2 Kamala Harris no longer favorite in Nate Silver model: "Literally 50/50" Newsweek · 16 October 2024

   “Nonetheless, Harris is still ahead nationally, with Harris ahead by 2.4 points according to FiveThirtyEight, and 2.8 points according to Silver.”

3. E3 BARONE: The Democratic Party's bad October Northern Sun / NSJ Online (Michael Barone column) · 3 November 2024

   “But it’s perceptibly below her 57% chance on Sept. 27, which reflected poll results after the Sept. 10 debate.”

4. E4 Nate Silver Says 2024 Race Still Tossup Despite Kamala Harris ... Benzinga · 19 September 2024

   “The 2.9-point lead Harris now has over her Republican rival Donald Trump compares to the 2.2-point lead she had on the debate day”

5. E5 1 MNI Political Risk Analysis (Market News) · 4 October 2024

   “Nate Silver writes: “The forecast is still in a steady phase. We’ve gotten a few new polls since yesterday’s update but there’s very little movement in the model from day to day. Harris now leads by 3.4 points in our national polling average and has a 57 percent chance of winning the Electoral College.””

6. E6 1 MNI Political Risk Analysis (Market News) · 4 October 2024

   “There has again been little change in the polling since last week’s edition of this newsletter.”

7. E7 Harris is gaining in post-debate polls Silver Bulletin (Nate Silver) · 17 September 2024 Own record

   “She’s now up by 2.9 points in our polling average, as compared with 2.2 points on the day of the debate”

8. E8 Harris is gaining in post-debate polls Silver Bulletin (Nate Silver) · 17 September 2024 Own record

   “Kamala Harris is moving up in the polls enough that the model is converging back toward 50/50 in the Electoral College.”

E3 and E5 put Harris's Electoral College chance at 57% on both September 27 and about October 4, with a 3.4-point national lead in early October. E5 and E6 say there was very little movement. E1 and E2 show a drop of about 6 points in win probability and a lead of 2.8 points, but only as of October 16, after the October 8 date being judged, so they cannot show what the model said when Silver spoke. No evidence gives the model's figures on October 8 itself; the closest, E5 from about October 4, reports a 3.4-point national lead and a 57 percent win probability with very little day-to-day movement, so it cannot be confirmed that the probability moved by several points while the popular-vote forecast barely moved in the weeks before then.