To help us all keep our sanity — and keep Trump's campaign for Il Duce, err, President, in perspective, it's time to look at the race as it really is, not as anyone fears (or hopes) it to be. Modern polling has in fact become very, very good, at predicting the outcome going into Election Day. It’s not uncommon for polling aggregators to correctly predict 49 or 50 of the 50 states headed into the last few days — and for similar accuracy in predicting Senate outcomes as well. And with multiple online aggregators, each with their own algorithms for crunching the numbers, our confidence in the state of the race should increase if they’re telling the same story.
We’ve had enough time since the conventions that bounces should be fading and the underlying, enduring sentiments of the electorate should be emerging. I will be compiling results from multiple polling aggregators and hopefully updating them weekly as we proceed to the election. And although the national numbers are regularly reported, we of course don’t elect Presidents by national vote — we do so through the state-by-state electoral vote. So, although I’ll include the national numbers, I will focus on the statewide polling from the dozen or so states that by consensus are “swing” states. It is really the vote within these states that will determine the election’s outcome. Finally, I will also track predictions of the Senate balance of power, which will impact the winner’s ability to shape the judiciary and her ability to pass any part of her agenda. We’ll have more to say about the snapshot below and each week to come, but first take a close look for yourself.
| state | electoral-vote | fivethirtyeight | pollster | realclearpolitics | sabato | |
|---|---|---|---|---|---|---|
| co | C, 41-32 | C 49.2-41.2 | C, 42-37 | C +9.5 | Likely D | |
| fl | C, 43-39 | C, 49.3-44.4 | C, 44-41 | C, +2.7 | Leans D | |
| ia | C, 40-39 | C, 48.4-43.7 | C, 46-43 | C +0.5 | Leans D | |
| nv | C, 42-41 | C, 48.5-43.3 | tie, 43-43 | C +2.3 | Leans D | |
| nh | C, 47-32 | C, 48.0-42.4 | C, 42-37 | C +7.0 | Leans D | |
| nc | C, 46-42 | C, 48.8-45.5 | C, 44-42 | C +0.5 | Leans D | |
| oh | Tie, 41-41 | C, 47.9-43.7 | Tie, 43-43 | C +0.8 | Leans D | |
| va | C, 49-37 | C, 49.4-41.9 | C, 44-38 | C +7.0 | Likely D | |
| mi | C, 42-32 | C, 51.2-40.9 | C, 42-34 | C +6.6 | Likely D | |
| pa | C, 47-39 | C, 49.4-42.5 | C, 46-40 | C +8.0 | Leans D | |
| wi | C, 43-37 | C, 51.0-40.9 | C, 45-36 | C +5.6 | Likely D | |
| az | C, 44-43 | T, 46.5-46.1 | T, 46-45 | T +0.3 | Leans R | |
| ga | Tie 42-42 | C, 46.7-46.5 | T, 44-43 | T +2.4 | Leans R |
In future posts I’ll discuss in more detail the algorithm used by each aggregator to make their predictions. But here’s the story they seem to be telling us:
(1) The first 8 states listed are consensus ‘swing’ states, those that have swung back and forth in the last several elections. Clinton leads by every measure in 6 of them and four of five in NV.
(2) The next 3 states (MI, PA, WI) have been identified by the Trump team and supporters as part of a ‘winning’ strategy to take the Rust Belt states away from the Democrats. It’s obviously not working, as Clinton comfortably leads in all of them.
(3) The final two states, AZ and GA, are where Trump has managed to change the electoral map — just not in a way that’s good for him. These reliably Republican states appear to be swing states this year.
Although of less importance, the national numbers clearly favor Clinton as well. The 538 estimate of popular vote percentage shows Clinton ahead 49.3 — 41.8; HuffingtonPost’s Pollster shows Clinton ahead 44 — 36; and RCP has Clinton up +7.2.
But this has not yet translated, apparently, into the Democrats retaking the Senate. Electoral-vote.com currently predicts D 49, R51 for the Senate. RCP is more cautious with D 47, R 44, and 9 “tossups.” Notably, they’re in agreement that Democratic pickups in Illinois and Wisconsin are done deals. So the Senate is within reach, but there is much to be done.