Scattered throughout the presidential debate headlines Tuesday are lines about which candidate – Hillary Clinton or Donald Trump – "won" last night, based on rapid instant polling done by different sites and organizations.
CNN/ORC for example polled 521 registered voters and found 62 percent thought Clinton "did the best job in the debate." That's compared to 27 percent who said it went to Trump. Public Policy Polling had Clinton ahead 51-40 after the debate.
Most though have leaned to the Right.
A Time.com poll has Trump ahead 54-46. CNBC's voters also picked Trump, who leads that one 67-33. He also leads on a Fortune poll, and on Breitbart's survey (which, it should be noted, is a site whose chairman is the chief executive of Trump's campaign) the Republican has a commanding lead.
These don't necessarily mean anything
Here's the thing: These snap polls after the first debate don't necessarily mean anything in terms of who might win the election.
As FiveThirtyEight points out, the CNN/ORC poll "correlates fairly well" with how the next round of polling moves afterward – basically, that survey seems to best indicate how the public will weigh in during the polls that are done over the next few days. Because of that, the site is predicting a bounce for Clinton – but they also note that doesn't mean she'll hold it through Election Day.
And we've got a very recent example to point to: Barack Obama and Mitt Romney in 2012.
In that CNN/ORC poll after their first debate, respondents said Romney won by a huge margin – 67 percent to 25 percent – over the president. That then led to a near-tie in the national polls in the week afterward, as well as a "freak out" from Democrats.
And then Obama had better showings in debates two and three, out-polled Romney, and won the election handily.
To some of these sites' credit, they note that the instant polls/survey on their site aren't scientific, and aren't a representative sample (meaning they're not made to match the demographics of America and lively voters). But that's not always in the headline, or high up in the story.
So remember: Instant poll numbers after the first debate might be fun to take the nation's temperature – but don't take them as a sign of things to come.