On or around 22 March 2016, a rumor began circulating on social media that popular poll-cruncher and statistician Nate Silver had predicted that Vermont senator Bernie Sanders would sweep eight upcoming primaries and thereby snatch the Democratic presidential nomination away from Hillary Clinton. For example, one 21 March 2016 blog post maintained Silver had expressed that "Bernie Sanders is likely to win all of the upcoming 8 states":
According to polling expert Nate Silver, Bernie Sanders is likely to win all of the upcoming 8 states in the race to become the Democratic nominee. Silver boasts a long record of accurate predictions which includes successfully calling the outcomes in 49 of the 50 states in the 2008 U.S. Presidential election, for example. As a result of that particular achievement, he was named one of The World's 100 Most Influential People by Time in 2009.
Using his new method of predicting, Silver would have been able to predict the Michigan caucus far more accurately, as he explains, "Our demographic 'retrodiction' for Michigan still has Clinton winning, but only barely — by 3 percentage points, compared with the actual 2-point win for Sanders. Especially under the Democrats' proportional allocation method, that's a pretty minor difference." The difference is even more minor when you incorporate the fact that most polling companies, in the run-up to Michigan, had Sanders losing to Clinton by almost 20%.
That claim was suspect due to its source (various blogs) and its difference from the way Silver generally presents his analyses, which is to eschew making sweeping declarations and expressing future events as outright certainties in favor of explicating trends and probabilities. The post reproduced above linked to a 15 March 2016 post that Silver published on his FiveThirtyEight blog, published on the day of the Ohio Democratic primary (which Bernie Sanders lost by 42.7% to Hillary Clinton's 56.5%,) In that post, Silver projected that Hillary Clinton was running ahead of Sanders in Arizona's 22 March 2015 primary but pegged the subsequent seven (not eight) states as potential Sanders wins:
That post from Silver didn't claim that Sanders would sweep eight states following the primaries of 15 March 2016, nor did it say that Sanders would definitively — or even probably — seize the Democratic nomination. The closest Silver came to saying anything like that was this portion of his post:
Our demographic "retrodiction" for Michigan still has Clinton winning, but only barely — by 3 percentage points, compared with the actual 2-point win for Sanders. Especially under the Democrats' proportional allocation method, that's a pretty minor difference. The model's retrodictions in Vermont and Arkansas are also pretty far off, as you can see, but that makes sense given potential home-state effects for Sanders and Clinton in those states.
Aside from straight reporting of the Democrats Abroad primary (won by Sanders with 69% of the vote) on 21 March 2016, Silver's most recent Twitter mentions of Sanders again predicted neither a multi-state sweep nor a lock on the Democratic nomination:
Silver lining for Sanders tonight: seems to be making progress among black voters. https://t.co/21odfhRGip
— Nate Silver (@NateSilver538) March 16, 2016
On average between 5 states voting tonight, just 17% of Dems were late deciders. Split evenly between Sanders & HRC. https://t.co/iQfLtzq2e0 — Nate Silver (@NateSilver538) March 16, 2016
After the 15 March 2016 FiveThirtyEight posting, Silver tweeted:
Guys I think it's gonna be Clinton. https://t.co/I8rgYFGjMk — Nate Silver (@NateSilver538) March 16, 2016
We contacted Silver to determine whether he made the prediction attributed to him in any other medium or context but have not yet received a response.