Blog

Friday, October 11, 2024

What's the Matter with Michigan?

The majority of election-hawks seemed to have settled on election forecast models being the ideal approach to understanding elections and what polls are saying. These models undoubtedly have several characteristics that commend them--weighing polls based on several factors such as historical accuracy, introducing error to make the predictions more probabilistic (75% chance of winning for candidate X instead of predicting a winner outright), and using non-poll data to fill in some gaps from polling such as a convention bounce or undecideds.

What people should be wary of, though, is being overly confident in these models and attributing to them an objectivity or accuracy that doesn't exist. Partisanship has led to an over-confidence in these models. It's worth bearing in mind how the modelers' choices can affect the ultimate outcome and how accurate the models are, at the end of the day.

A good way to evaluate a model is by comparing to a simple average. RCP has fallen way out of favor for most election-hawks, particularly those who lean left, but it's much simpler and can be a good measuring stick even if it's not as data-driven. Also, it performed better in 2016 and 2020 than 538 did.

Why is Michigan so Different?

Right now, Nate Silver has Harris up by 1.1% (48.4 to 47.3). RCP, on the other hand, has Trump up 0.9 in Michigan (48.5 to 47.6). How can they be getting such different results? I presume most reading will reflexively think "Because RCP sucks" or the more diplomatic "Because RCP is run by conservatives who bias the results." Actually, that's not the case with Michigan.

Difference in Timing

The first difference is that RCP has a more recent cut-off. RCP considers only the 10 most recent, non-repeated polls. This effectively, for Michigan, goes back to 9/19, as of today, so 3.5 weeks. Silver's model considers polls going back to 8/30--6 weeks, double RCP. The effect of this is to add weight to older polls. If trends have moved since then, it will produce biased results. This has definitely been the case. Taking all of the polls in Silver's model, and estimating a 5 poll moving average shows movement toward Trump recently.

If Silver cut off the polls at the same time as RCP, it would move the polls about 0.2% toward Trump.

Differences in Poll Choice and Weighting

One of the smart decisions Silver made was to weight polls according to their historic accuracy along with other factors. Contrarily, RCP has a more subjective and simple weighting method. RCP's editors choose which polls to include, which usually are the most high profile ones. After that decision is made, all polls are weighted equally. On the other hand, Silver includes all polls and weights them using historic data, a measure of transparency, and sample size.

For the time period that RCP is using, it includes 10 polls, while Silver includes 16. Of the 10 that are in both 6 of them have Trump winning. Of the 16 Silver includes, Trump is winning in 7 of them. Meaning RCP is disproportionately leaving out polls which have Harris ahead. The ones in Silver's average are disproportionately less famous.

The table below shows which polls made it into the averages. It also shows Silver's letter grade for each pollster. For the most part, the cut-off for RCP corresponds to a letter grade of B-. The only exception is Mitchell Research (Trump +0.5), Morning Consult (Harris +3.5), and some combinations of polling firms.

Blue denotes polls that were included both in Silver's average and RCP's average. Taken 10/11/2024.

If all firms were included in RCP's average, weighting them all equally, however, Trump would be up by 0.9 instead of 1.0, their poll choice doesn't have much effect on the outcome.

What Else?

Beyond these explanations, admittedly, I can't figure out what's driving the difference. I don't know exactly how Silver calculates his average, but I tried accounting for house effects and weights, and can't reproduce it. I don't have the ability to plug in the non-poll fundamental effects, so it's some combination of that and his secret sauce on polling calculations that I just can't crack.

Using the weights that he provides, before accounting for house effects etc., I calculate, from his data in Michigan, that Trump is winning by -0.2%. Then adding in the house effects, I still have him winning by -0.2. (It's not that they didn't change anything; they certainly did. It's just that they cancelled each other out.

At the moment, I cannot explain how Silver's result is so pro-Harris. It's hard to believe that the fundamentals give her a 1.2% bump, but that's the only thing I can think of.

Friday, August 18, 2023

Let He with Reservations Cast the First Vote

Disclaimer: This is not intended to support or malign any specific candidate. It is general advice promoting voting when candidates are unlikeable. Any promotion of specific candidates is a result of current polling on race, not policies or characteristics of the candidates.

Dominic Pino said something interesting on the 8/15 edition of The Editors Podcast. He argued that while a majority of Republican primary voters may support Donald Trump, a majority of Republicans probably doesn't, and almost certainly, a majority of people who might consider voting Republican also wouldn't support Trump.

This is a unique election. In my lifetime, these are the two worst candidates ever. Arguably, both are unqualified to be president, but for very different reasons. In elections before 2020, many people would complain about the quality of candidates, and it's true that the two parties' nominees are typically pretty poor representatives of the greatest America has to offer, but to a large extent, that's what the system, as it is currently constituted, can produce. That's also a very complex topic.

The difference with the 2020 election and the two front-running candidates for 2024, is that they're not just sub-optimal people, neither of them should be in the White House with any power whatsoever to make decisions for the country. Many, many people detested Hillary Clinton, but the idea that she was unqualified is ludicrous. The same can't be said for Biden and Trump. It would be easy to make an argument for either one of them that they are unqualified from holding this office.

For Biden, there's not much that can be done in a primary election, when no viable candidate wants to challenge him. For Trump, however, there is an opportunity. This will make practically everyone angry, but the most American thing a person can do is to vote in the Republican primary, for the candidate who's most likely to beat Trump.

This will make practically everyone angry, but the most American thing a person can do is to vote in the Republican primary, for the candidate who's most likely to beat Trump.

Many moderate Republicans, who despise Trump, have also developed a strong aversion to DeSantis because of his stances and focus on cultural issues. Even if they don't think DeSantis should be president, they should still vote for him in the Republican primary, assuming he stands the best chance of beating Trump. It is un-American to believe that Trump is a mortal threat to democracy, but allow him to win the primary and have a chance at becoming president. Right now, polls are pretty close. In fact, Trump has a better chance at beating Biden than DeSantis. Anything can happen between today and the 2024 election, and if you truly believe Trump is unfit for office, doing nothing but hoping things work out is a dangerous gamble.

If you truly believe Trump is unfit for office, doing nothing but hoping things work out is a dangerous gamble.

But you don't like DeSantis, and you're afraid he might win the general and become president. If you're a pro-democracy patriot, you should be okay with that, even if it's not your ideal outcome. In fact, you should want a strong contest between qualified candidates, even if it doesn't go the way you want. Whether you're a full-blooded Democrat, an ambivalent moderate, or a disaffected Republican, the most civic-minded thing you can do is vote for the Republican who stands the best chance of beating Trump.

Perhaps you think DeSantis himself is unqualified, just like Trump, so why vote for him. Here's a great test as to whether you believe a candidate is qualified. Ask yourself, why shouldn't this person be president. Come up with as many answers as you like, but try to be honest with yourself as to why you don't think they should be in office. If the reasons are all policy-related, then the candidate is qualified, but just has different positions. Shouldn't voters make that decision? Don't you want an election where voters can talk about issues and choose who better represents them and not a "my guy's awful, but their guy will usher in Ragnarok" election?

Sunday, December 4, 2022

2022 Mid-Term Post-Mortem

It's no secret that the Republicans greatly underperformed expectations in the 2022 mid-terms. On average, the out-of-power party typically gains more than 25 seats in a mid-term election. As of writing, Republicans gained 9 seats, and there is one seat left to call. In a year where the President is as unpopular as any predecessor, and inflation is at a 40 year high, Republicans were expecting much more than they obtained.

What went wrong? Why did they underperform so spectacularly? There have been many theories posited, but none have been enormously satisfying. To work toward an answer, Republicans have announced that they will review the 2022 election to determine why they performed so woefully.

This is a necessary and welcome step, but for several reasons, I have little confidence that it will help clarify the 2022 election. For one, the people involved are all politicians. Kellyanne Conway did get Trump elected, so she has some knowledge and experience she can bring to bear but I'm concerned that she's too beholden to Trump to fully consider the possibility that Trumpism was to blame. Blake Masters lost his election which could provide insights into what he did poorly, if he is a clear-headed person, but is more likely to lead to mis-emphasis and mistaken conclusions.Secondly, Republicans did a post-mortem after Romney lost and decided they needed to moderate on the immigration issue. Then Trump was elected after notoriously not moderating on that issue.

Here are the theories I'd like the committee to consider. For each, I'd want a section of the report that lays out the theory, the evidence for it, the evidence against it, how important the committee believes it was to the results, and what can be done in the future to prevent it.

Was it Issue-based?

Some argue that Republicans don't stand for anything and that hurt them with independent voters who didn't have anything to vote for. I've always thought Republicans should push a positive agenda, but I've also come to understand that Republicans aren't good at defending their agenda, that much of their agenda is complicated and hard to defend, and that running on nothing has worked for them in the past. This is a really interesting question because Republicans have actually been gaining on the issues (and this) Inflation and crime are two issues where Republicans had a commanding lead and were top of mind for voters, yet Republicans still lost.

Two issues where Republicans did not have an advantage were abortion and election "denialism". Abortion is the issue that many analysts are claiming made the difference. Especially for independents. I can believe that. Some have also made the argument that it was the democracy issue. If either of these is true, it represents a massive failure of Republican messaging. For abortion, while a majority of Americans supported Roe v. Wade, when polled more precisely about abortion policy, a majority supports banning late-term abortions with a cutoff somewhere between 12-15 weeks. The Mississippi law, which was the basis of the challenge under Dobbs, because it was questionable whether it would be allowed under Roe, would've banned abortions past 15 weeks. Neither side's base is in line with popular opinion on abortion, yet Democrats have convinced the public that they're the majority and Republicans are extremists. Republicans need to push back harder on Democrats on abortion. Most Republican-led states have laws in line with majority opinion, while most Democrat-led states are pushing for unrestricted abortion. It is the Democrats who are the extremists.

The democracy issue is similar to abortion in that Democrats and the media have effectively painted Republicans as extremists. There's obviously something to that because of Trump and his accolades' claiming the election was fraudulent or that they wouldn't abide by results. But Democrats aren't guiltless here: Hakeem Jeffries, Jamie Raskin, etc. Not to mention Biden's student loan forgiveness diktat, his continuing use of Covid emergency powers, his attempt to continue the eviction moratorium that the Supreme Court warned him not to do and then struck down when he did, the vaccine mandate end-run he tried. There are many examples Republicans could use against Democrats to blunt the impact of the "election deniers".1

The risk of running on issues is that even a hint about maybe cutting any money from any government program from a single person while no one else wants to talk about it and says nothing about it gets turned into 'Republicans have decided to cut Social Security'. Every reasonable person knew that the NY Times and Biden were blowing up what Rick Scott said to scare seniors and independents for Democrats' political gain, but both are adept at selling molehills as mountains. If your agenda is based on reducing the debt and the size of government and any hint of any reduction, no matter how minor will be magnified to 'Republicans have already decided in secret to zero-out your retirement and deprive you of medical care when you turn 65', it's no wonder that Republicans want to stay silent on issues.

Money

I haven't spent as much time looking into these issues, but were Republicans outspent? If so, how much of the differences came from that? Open Secrets reports that Republicans actually outspent Democrats by a few hundred million. Though they allocated their spending differently. I do not have the expertise to judge whether Republicans' allocation was suboptimal. I will say, though, that Democrats outspent Republicans on media by over $100M (~25%), which seems problematic.

I've heard it argued that a lot of Republicans' money came from PACs whose ads cost three or four times more than ads purchased directly by a candidate in the final days of an election. If money was even, but ads cost more for Republicans by virtue of the way they were financed, that's a problem that needs remedying. Some have also already started talking about how Republican money-raisers are raising money for themselves while claiming to raise it for other candidates.

Other Topics

Candidate Quality - This has been discussed frequently by analysts, and certainly should be reviewed by committee. It will be more difficult to address than others, though, because to some extent, Republicans are stuck with who runs and whom voters choose. Many of the candidates were clearly sub-par (for Democrats, too). I will look forward to seeing evidence on how some candidates performed poorly, and what characteristics determined that.

Gerrymandering - Did Democrats win the gerrymandering game and shave off a few seats. Conveniently, 538 has already considered this and concluded that, yes, Democrats gained a couple seats because of gerrymandering.

Turnout - Did Democrats turn out more than Republicans. The very early indications were that young people made the difference and turned out in huge numbers that won the election for Democrats, but those conclusions have been walked back. In fact, Republican turn out was very good, and Democratic turnout was slightly lower. The big question here, though, is whether voters have realigned in a way that mid-term electorates will be consistently more Democratic than before (more upper income, high propensity voters)

Why were the polls off?

Lastly, the committee should consider why the polls were so far off. The generic ballot seemed to perform well, but many of the state contests were pretty far off. New Hampshire, in particular, but also Laxalt in Nevada, Walker in Georgia, and Oz in Pennsylvania. All were leading in the final days, and Laxalt had been ahead for weeks. Incorrect polls lead to misallocation of funds and election efforts and need to be closer to reality.

Notes

1 For what it's worth, this is another example of the uphill battle Republicans have against the media. There were undoubtedly some deeply problematic candidates Republicans ran. Foremost among them was Mastriano, but the media, as it is wont to do, expanded the definition of election denialism to capture more behavior and Republican candidates, and then with help from Democrats including Biden himself, amplified the concern to convince Americans that democracy was teetering on the edge. In the end, nearly all of the candidates who were "election deniers", accepted the results of the election as candidates have always done. Democracy was saved.