The Cavalier Daily
Serving the University Community Since 1890

LETTER: A more complete view of Crystal Ball predictions

To the Editor:

We appreciate The Cavalier Daily writing about our Crystal Ball newsletter (“Sabato’s Crystal Ball starts predicting 2016 election results,” Feb. 3), but the story’s subheadline — “Falls short of picking Iowa winners” — is not accurate. While we did say Donald Trump was a modest favorite on the Republican side (he finished second to Ted Cruz), we did accurately pick Hillary Clinton to win the Democratic caucus. We were not alone in giving Trump a small edge heading into the race — the last 13 polls showed Trump leading — but we also stressed to our readers in our final report before Iowa that the race was very much in flux and that “we still have questions about Trump’s ability to get his supporters to caucus locations on Monday.” We also noted that “Cruz’s devoted followers know the caucus process intimately and often carry the day,” which turned out to be the case. While the Crystal Ball has a strong record of picking election winners, we can’t promise we’ll always be right, particularly in primaries when polling is often unhelpful. However, we feel that we gave Crystal Ball readers a good sense of the state of play in Iowa leading up to the caucuses, and we hope to continue to do so throughout the primary season and into the general election.


Kyle Kondik and Geoffrey Skelley

Feb. 4, 2016

Kyle Kondik and Geoffrey Skelley are, respectively, managing editor and associate editor of Sabato’s Crystal Ball at the University of Virginia Center for Politics