Looking more closely at "negative" poll numbers.

[kentuck]

As of today, both Clinton and Trump have "negatives" close to 60%. Trump is still leading by 2 percentage points but Hillary is gaining fast. Where do they get these poll numbers?

We would assume that the people polled are balanced between Democrats and Republicans and Independents? If so, it is easy to see why Hillary Clinton's poll numbers are equal to Trump's.

If 60% of the people view her unfavorably, then we assume that 40% have a favorable view or no view at all. If the poll sample has 33% Democrats, 33% Republicans, and 33% Independents, then 40 percent "favorable" would be pretty good. That is why both Parties are usually guaranteed 40% before the race even begins.

So, who are the 60% "unfavorables"? It is probably safe to say that most Republicans see Hillary as "unfavorable", as most Democrats see Trump as "unfavorable".

What do the negative poll numbers tell us?

[Hav]

1. I'm not sure where you are going with this

but "If the poll sample has 33% Democrats, 33% Republicans, and 33% Independents" is most likely a wrong assumption because you want a sample that is representative of the population and not a "fair" distribution of the outcomes you intend to measure.

[kentuck]

2. True.

What would be a poll that was more "representative of the population"?

I'm curious as to how these "negative" polls have much benefit?