I saw a poll showing that Hillary lost 7 points from that debate, with Annoying Orange losing as well, but ending up where he was before the hot mic thing. I'll have to see if i can find it, but I firmly believe these polls are cooked, and for them to show Clinton dropping by 7 points must mean something big.
First of all, that is not how statistics works at all. You cannot base your conclusions off of a single data point when a much larger sample is available to you. That is like tossing a coin once, seeing that it's heads, and deciding that the coin can
only flip heads, without looking at any other trials.
Second, the poll you probably saw was from the LA Times/USC, and it's well-established by this point that something in their methodology is biased towards giving Annoying Orange higher numbers. Why? Because
literally every single other poll gives completely opposite figures.
Let me demonstrate the importance of looking at a larger sample of data by showing the overall curves for the popular vote over time:
(full disclosure: polls on this graph are weighted and trended by 538, but that's irrelevant to what I'm about to say)
This relatively smooth curve is constructed from wildly varied data, as shown below.
Now observe as I cherry-pick 8 polls (!!! even bigger sample size than yours !!!) and falsely conclude that Annoying Orange is winning the election.
In other words,
single polls are useless information that is too varied to derive any meaningful conclusions. Only when you look at the averages of
dozens of polls can you start to make any reasonable prediction about the election. And no, dozens of independent polling firms with no incentive for partisan bias are
not skewing the election towards Hillary. Annoying Orange is actually losing. Badly.
Poll