Shocks, Polls, And Outcomes
On Things That Can Happen
51 weeks ago today was the Qualifying session for the 2021 Belgium Grand Prix - the only “real” session of that weekend, and the one that ended up setting the results, after the Sunday was rained out. It’s notable as the half point race, where they did two laps where nobody could overtake, and then everyone got half the points they usually would - and therefore induced Lewis and Max having half points in their standings all year, which was apparently quite triggering (remember, I’ve only started watching this year).
It’s also memorable as the George Russell Showcase, after a truly stonking lap led him to start the race that never was in 2nd in a Williams, outqualifying Lewis and cementing, to whatever extent it was still up in the air, that he deserved that Merc seat. It was a truly preposterous lap from the lad, and other than one other wet session in Sochi, it was a feat that wouldn’t be replicated again that season. But holy shit it was a great moment.
What it wasn’t, however, was a sign that that should become the standard. The reason that George’s lap was so incredible wasn’t because it was a reasonable outcome, but because it was so fucking wild as to be shocking. For George to put that car in that place, on the front row, was a fucking miracle - and nobody’s response to that was “well, he should be doing that all the time now”, because that would diminish the sense of accomplishment and amazement at what he managed.
That’s the thing about miraculous events - they’re memorable because of their uniqueness, not in spite of it. Why do golf fans remember Go Get That more than the eagle putt that missed two holes later? Because making a 50 foot eagle putt in a Major 2 holes after you fucking bricked it into no man’s land is fucking memorable. If people made 50 foot putts regularly, nobody would remember or care.
It’s the same with politics - we remember Susan Collins’ win not because it was normal, but because it was fucking wild! And to pretend otherwise is to deny what made her performance so incredible.
…
If you haven’t been able to figure out what this column is about yet, it’s about the persistent claim from some Republicans who are having a very bad 6 weeks that even though Dr. Oz and Doug Mastriano (and other Republicans, but mostly the Pennsylvanians) are losing, and losing badly, right now, they’ll win when it matters. It doesn’t matter that the polling miss in Pennsylvania has mostly been smaller than the rest of the upper Rust Belt or that the current leads are well outside the historical error band, or that GOP pollsters are showing some of the biggest Democratic leads, because to them, they’ve found their evidence of races with big Democratic leads in the summer which the GOP ended up winning.
It’s funny, because there is actually an argument that this might happen in Wisconsin, where the two polls this week have Mandela Barnes up 7% and 4%, and yet nobody really thinks Barnes is winning an election held today, let alone would win in November. In Pennsylvania, the obsession with Collins or Maryland Governor in 2014 makes an amount of superficial sense - it shows that Oz and Mastriano can come back, and even that not having led at any point doesn’t mean they can’t win. But it’s worth remembering that the whole point of Collins’ shock win was that it was a fucking shock.
You know what’s not shocking? “Candidate everyone expects to win wins by amount they’re supposed to win by”, because it happens all the fucking time. Our brains are biased to remember notable things, good or bad, so we remember Collins winning or Feingold blowing it because they were true shocks. In the same way, we forget Sinema having a 1.8% lead in 2018 and winning by 2.3%, because it was par for the fucking course. It’s all textbook stuff that feels way too simple to actually have to explain, but too many people are conflating that Oz and Mastriano can come back by saying that the existence of times when comebacks have happened means it will.
I’ll be honest and break the fourth wall here and let everyone in on a secret of doing this - you can make intelligent sounding, entirely factual cases for almost any take you want. If you want to argue that George is going to win this upcoming week at Spa, you can make a very coherent case that it’s a track he likes with a car that should be better. If you want to make the argument he’s going to come fifth, you mention that it’s a Ferrari comfort track, he’s come behind Lewis for 5 straight races, and Max is still Max. You can make the case either way. In the same way, very little argument about political predictions is about a disagreement about factual matters, it’s a difference in interpretation.
In prepping for this column today, I went back and read everything I wrote about the 2021 Virginia’s Governor’s race last year. Everything I wrote as a fact was correct - everything about the 2018 polls and the historical tendency to over correct was true - as far as it went. As a statement of what had happened before, it was true. The predictive statement that Youngkin would lose by 6 was, obviously, dogshit. This is a soft science - politics is a land of the mushy middle, of no hard and fast rules applying because every hard and fast rule has some exception. How did Joe Manchin win West Virginia but Joe Donnelly lost in Indiana, despite a ~25% gulf in Trump win margin? Shit’s wild.
Do I think Dr. Oz will win? No. Would I be shocked by it? God no. I think it’s super unlikely for 1000 reasons that I laid out for TheLines two weeks ago, but it’s not gonna be shocking if it happens. But pretending he’s anything but a huge underdog is insanity of the purest order - and pretending so because other wild upsets have happened before just proves that you don’t get what made those moments so shocking.
I mean, unless you think Alex Albon’s starting on the front row next weekend, in which case, you might just be delusional about everything.