Takeaways from the Primaries

As usual, there are lots of takeaways from Tuesday's primaries. Here are some of them:

New York Times

• A night of firsts (a transgender gubernatorial candidate won in VT, a Somali-American won in MN, etc.)
• Middle-aged white men triumphed in the Midwest, where the Democrats need to win back blue-collar workers
• A black woman was nominated for a safe seat in the House in Connecticut
• The old Republican party is still napping somewhere; it's Trump's party now

Washington Post

• Trump is king: His endorsees win (even if they may result in Democratic victories in November)
• Diversity triumphed in Connecticut, Vermont, and Minnesota (all blue states)
• Some candidates won despite bad behavior (e.g., an abuser, a brothel owner, a Confederacy supporter)
• Both parties got their man for the Connecticut gubernatorial race and it will be a real battle
• Any Republican who wants to win had better hug Trump, and pronto
• The GOP nominated the Trumpiest candidates for governor in Kansas and Minnesota, which gives the Dems hope
• The voters don't like openly white supremacists like Paul Nehlen in Wisconsin, who ran for Ryan's seat and lost
• When billionaires collide, like in the Wisconsin Senate primary, one of them loses

CNN

• Diversity rules in blue states; white men win in red states
• T-Paw's comeback failed in Minnesota
• Keith Ellison won in Minnesota, but a cloud looms over him due to abuse allegations
• Billionaire Richard Uihlein poured $11 million into the Wisconsin Senate race of Kevin Nicholson and got nothing
• Progressives surged in many states

Politico

• Democrats made history with diverse candidates in the Northeast
• Democrats bet on boring candidates in the Midwest and won
• Scott Walker's machine remains a force
• Voters don't like lobbyists like former Minnesota governor Tim Pawlenty
• Minnesota is ground zero for the battle for the House with four toss-up races

The Hill

• The establishment flexed its muscle and got its candidate in most races
• Democrats made history with diverse candidates
• Ironstache (Randy Bryce) won in Ryan's district, but his history of DUI and using pot will hurt in November
• Allegations of abuse could sink Keith Ellison's AG bid in Minnesota in November
• Democrats are warming to "Medicare for all"

The big themes are diversity won, Republicans had better kiss Trump's rear, but Trumpy candidates may not win in November. (V)

Does Trump's Endorsement Matter?

Whether or not Donald Trump has the power to influence the outcome of elections is a question of much more than mere academic significance. First, because if GOP candidates across the country believe that Trump is a kingmaker, they will be compelled to hold him closer (which means both supporting his agenda, and not challenging his problematic behaviors). The President certainly knows this, and is working hard to create this perception:

As long as I campaign and/or support Senate and House candidates (within reason), they will win! I LOVE the people, & they certainly seem to like the job I’m doing. If I find the time, in between China, Iran, the Economy and much more, which I must, we will have a giant Red Wave!

— Donald J. Trump (@realDonaldTrump) August 8, 2018

Second, and perhaps even more importantly, the Democrats are hoping to win a lot of close elections in November, and if Trump can even swing a point or two in the GOP's direction in some of those, he could affect a half-dozen seats or more (and, very possibly, control of Congress).

One way to approach this question is mathematically. Let us start by observing that there is zero evidence (and, for that matter, zero chance) that Trump can cause a swing of 5-10 points, despite his regular claims to the contrary. No politician has that power, particularly one whose approval rating is mired in the 40s. If he does have coattails, they are 1-2 point coattails, and nothing more. There are some significant difficulties with numerically proving (or disproving) an impact that small. Among them:

• Polling of local races tends to be sporadic, and is often done by second-tier firms.
• Trump doesn't make that many endorsements, which creates a small sample size issue.
• Trump tends to make his endorsements late, and often bestows them upon a candidate who is already leading in the polls, or else is close behind and is surging.

It's certainly possible to make a case that Trump really does have the magic touch, and can indeed swing elections. CNN's Chris Cillizza recently wrote a piece arguing just that, starting with the observation that, "Here's an amazing stat: In the last 14 contested Republican primaries where President Donald Trump has endorsed a candidate, his pick has won—or is leading—all 14 times." When you put it that, way, it seems like a slam dunk—clearly, Trump's endorsement is decisive.

However, Cillizza's analysis skips over the fact that less than half of those races were legitimately competitive when Trump made his pick. Further, he appears to be falling victim to the clustering illusion, giving too much credence to the fact that Trump's "wins" were consecutive, and overlooking his much spottier record before that run (for example, the failed Roy Moore endorsement, and before that, the failed Luther Strange endorsement). In other words, while Cillizza's "14-for-14" assertion is correct, it would also be correct to say this: "In the last 10 legitimately competitive races where Trump has made an endorsement, his candidate has prevailed 7 times." Not bad, but not quite as definitive as 14-for-14, especially when you consider that 7-for-10 is not substantially better than a random coin flip. And that's before we get into the fact that some of Trump's winners, like Kris Kobach, have prevailed by the tiniest of margins, meaning Trump easily could have gone 6-for-10 or 5-for-10.

The point is, the math here is ambiguous. Trump might have small coattails, he might have none, or he might even have negative coattails. Polls and voting totals are too imprecise to reach a conclusion based on the data we have available, and the sample size is too small to compensate for our uncertainty.

There is another way to approach this, however, which we might call the "common sense" approach. Here, let us point out some things that make Trump very unusual:

• Those who love him, generally speaking, are particularly strong in their feelings
• Those who hate him, generally speaking, are particularly strong in their feelings
• It is particularly easy for candidates to signal their allegiance to Trump. Beyond an embrace of his distinctive political platform, someone running for office need only say "Lock her up!" or "Build the wall!" or "Make America Great Again!" or any of half a dozen other things, and it is instantly clear to everyone that they are on Team Trump. It is concurrently very easy for a candidate to signal opposition to Trump, by talking about gold star families, or separating families at the border, or the evils of Muslim travel bans, or any of a half-dozen other instantly-recognizable controversies.

In contested GOP primaries, there are three possible configurations vis-a-vis Trump. Keeping in mind that he stays away from "losers," all that really matters are the top two candidates in the race. One possibility, very rare this year, is a race where the top contenders are both anti-Trump Republicans. In these cases, however, Trump stays out of it, since he doesn't support either. The second, much more common, configuration is an anti-Trump Republicans facing off against a Trump supporter. The President does sometimes wade into these contests, but it's hard to see how he could have much of an impact. That is to say, those who vote the Trump party line would already know whom to vote for, and likely don't need the Donald's input. And those who don't vote the Trump party line ignore his input. The third, and very common, configuration, is when two of his supporters are facing off against one another. In these cases, the President's endorsement probably does matter, because it helps his supporters decide which candidate is the Trumpiest.

Of course, the general election (outside of Louisiana, the only GOP-controlled "jungle primary" state) isn't going to feature races with two candidates competing to see who is the Trumpiest. It's going to feature, at most, one very Trumpy candidate, and one who is either much less Trumpy (say, a red-state Democrat like Sen. Joe Manchin, D-WV), or one who is avowedly anti-Trump. And when those are the potential configurations, it is much harder to see how Trump's endorsement could possibly have a meaningful positive impact for the GOP.

Consider that there are two ways in which an endorsement can theoretically influence an election. The first is that it can cause a voter who might have voted for candidate "X" to switch their vote to candidate "Y." But in the case of Trump, given how beloved/hated he is, the number of fence-sitters will be few and far between. The Trump lovers will know exactly which candidate to vote for, the Trump haters will know exactly which candidate to vote for, and it is implausible that the President will be able to drag many people from the latter candidate to the former.

The second dynamic, which is generally much more common, is that an endorsement can cause folks who otherwise would have stayed home to get to the polling places and vote for a favored candidate. However, Trump's base, while relatively small compared to that of, say, Ronald Reagan, is unusually enthusiastic. They are going to get to the polls to vote with or without an endorsement. Meanwhile, a lot of the groups who tend to be strongly anti-Trump (minority voters, young people) don't have the best record of showing up to vote in the midterms. A ringing Trump endorsement plausibly could motivate them to get out and vote for the endorsed candidate's opponent. The point is, Trump's "get out the vote" power is likely to be unusually small for a president, and may well have a negative impact for the GOP.

We should note an exception, of sorts, to the preceding analysis. The basic presumption running throughout this article is that Trump can command his supporters to do his bidding, but that he has virtually no impact with any other voters. While Trump probably can't do much to get his base to the polls (since they'll already be there), he could probably get some of them to stay home, if he wanted. Similarly, he probably can't do much to drag anti-Trump voters into the Trump fold, but he could probably get some of his supporters to defect, if he so commanded. Point is, pro-Trump candidates need to do everything that they can to keep him close, and to kiss his ring. If Trump were to tweet, "I guess Rep. Jim Jordan (R-OH) doesn't really support my agenda" or "I would enjoy the opportunity to work with Sen. Heidi Heitkamp (D-ND) for the rest of my term," it's plausible he could move the needle away from a Trump-friendly candidate who angered him. With most presidents, it would be inconceivable that they would cut off a member of their own party at the knees, but Trump isn't most presidents.

Trump has made a handful of endorsements for the primaries coming up in the next two weeks, and quite a few more for the general election in November. For those who wish to keep track, the New York Times is keeping a running tally. It could be that we eventually collect enough numerical data to reach some sound conclusions, but until then, it is safe to assume that Democrats, at least, have little to fear from Trump's coattails in November. (Z)

Trump Revokes Security Clearance of Former CIA Director John Brennan

One of Donald Trump's guiding principles is: "Use the power of the presidency to punish your enemies." Yesterday, he demonstrated this by revoking the security clearance of John Brennan, who formerly ran the CIA. Brennan's "sin" was criticizing Trump. Next in line for revocation are Director of National Intelligence James Clapper, former national security adviser Susan Rice, and more.

No previous president has ever done anything quite like this, not even Dick Nixon or LBJ. Those two were surely the most thin-skinned presidents of the 20th century, but even they didn't have skin a micron thick, as Trump does. Previous presidents (1) were not so vindictive, or else (2) understood that in a national crisis, asking top ranking former officials for their advice on matters affecting national security might be helpful. Trump is never interested in advice, not even from his own appointees, let alone those of a previous president. So far no national crisis requiring tough choices has come up, but it could, and then Trump could find himself quite isolated. What Trump is probably thinking is that lack of a security clearance may make it difficult for Brennan and the others to get good-paying jobs with defense contractors. Trump, who places a high value on money, as you may have heard, frequently tries to punish his enemies financially, the most egregious case being firing Deputy FBI Director Andrew McCabe one day before his pension would have kicked in.

The revocation is likely to be counterproductive. Brennan, Clapper, and the others aren't going to tuck their tails between their legs and go gentle into that good night. They will continue criticizing Trump, perhaps even more than before since their criticisms don't depend on getting classified information (V).

Manafort's Trial Ends

Both sides made their closing arguments yesterday in the trial of Donald Trump's former campaign chairman Paul Manafort. The prosecution took 2 hours to go over the evidence that Manafort had lied to his bookkeeper and accountant in order to evade taxes he owed and then, when his business of consulting for foreign authoritarians fell apart, lied to banks to get loans to support his extravagant lifestyle. While he mentioned his star witness, Rick Gates, prosecutor Greg Andres told the jury that the real case was made by the documents, not the witnesses.

After lunch, the defense got its chance. Manafort's lawyer Richard Westling said that the prosecution selectively pulled together bits and pieces or evidence that don't show proof of a crime. He also said that to the extent crimes were committed, they were committed by Gates, without Manafort's knowledge.

In his final rebuttal to the jury, Andres pointed out that he had brought in 27 witnesses and 399 exhibits with documents and that the bottom line is that Manafort is fundamentally a dishonest person. He brought in $60 million from foreign sources, hid it in 31 offshore accounts, and paid for $15 million in personal expenses using it. The defense's claim that Gates did all this without Manafort's knowledge is absurd, especially since multiple witnesses said Manafort is a detail-oriented guy and was on top of every penny his firm spent.

The judge then delivered lengthy instructions to the jury, which is expected to begin deliberations this morning. (V)

Republican Midterm Strategy Is to Play Nice for a Few Months

Senate Majority Leader Mitch McConnell (R-KY) is very good at counting—at least to 50—and has developed a midterm strategy to make sure he has at least that many Republican senators in January. What he wants to do is push Mr. Hyde into the closet and play Dr. Jekyll until November. No more partisanship and obstruction and being nasty to Democrats. For the next three months, the keywords are "bipartisanship" and "playing well with others." He wants to work with the Democrats to pass laws that both parties agree on, such as fighting the opioid epidemic and helping farmers. He wants to show the independent voters, especially college-educated suburban voters, that Republicans are perfectly reasonable and can govern nicely. In January, Mr. Hyde can come out again.

Not all Republicans are on board with this plan. Some want a more hard-edged campaign that emphasizes the differences between the parties, in order to turn out the base. For example, some of them want more tax cuts. What previous majority leaders have sometimes done is schedule votes on bills they have no interest in passing, just to force the other side to make tough votes. If McConnell tries this, he can forget the bipartisanship thing.

The problem with making nice with the Democrats is that Minority Leader Sen. Chuck Schumer (D-NY) has his priorities, too. He might say things like: "I'm OK with combating opioids—but only as a part of a bill to strengthen and enhance the ACA." McConnell won't go for that. So the big question is whether the Democrats will play ball with McConnell, knowing his strategy is to seem so very reasonable for a few months. (V)

Democratic Midterm Strategy Is to Go Local

The Democrats have a completely different plan for the midterms: Not stand for anything. It's not such a crazy idea, though. There will be a bare-bones national agenda on which all Democrats can agree: lowering health-care costs, increasing worker pay, and fighting corruption. After that, it is every man (or woman) for himself (or herself). This avoids rerunning Bernie vs. Hillary, Part 97, but it means the party doesn't actually stand for much. Or maybe it is a return to Tip O'Neill's motto: "All politics is local."

The master plan here is that if Alexandria Ocasio-Cortez wants to run as a Democratic Socialist in a D+29 district, it's go, girl, go. But if Conor Lamb wants to run on a platform supporting the Second Amendment and Donald Trump's tariffs, go for it, boy. In other words, each candidate is expected to tailor his or her platform to their district (or their state), since what works in a D+29 district will most definitely not work in an R+11 district, and vice versa.

The downside of going local is people will (rightly) ask: "What do Democrats stand for?" And worse yet: "What will they do if they attain power?" There is no easy answer for the blue team. Nevertheless, it could work. Voters in southern Pennsylvania are interested in what their representative will fight for, and not so much what one in Queens will fight for. (V)

Researchers Show that Votes Can Be Hacked in Nearly 30 States

At the recent Defcon Conference in Las Vegas, the oldest and largest conference on hacking, researchers showed how in 2 hours they could set up a fake server that could intercept votes sent by e-mail and change them before sending them on to the proper server. One of the researchers, Lyell Read, said point blank: "Ballots sent over e-mail are not secure." David Jefferson, a researcher at Livermore National Laboratory, had this to say about e-mail voting: "Anyone who controls a router can change a ballot. It's just insane. It's like attaching a $100 bill to a postcard and mailing it and expecting it to get there." Nevertheless, dozens of states allow voters to e-mail their ballots.

Mark Neary, an assistant secretary of state in Washington State, pointed out a different risk, other than someone changing votes. An attacker could attach a virus to a vote, and when it arrived at the election server, could corrupt the server and plant a backdoor, allowing an attacker to gain access and change much more than the one vote that had been captured.

Pulling off these attacks isn't that hard. Yonatan Lensky, an 11-year old attending the conference, showed how easy it was to tweak some code on a mock-up of the website used by Iowa. He gave a fictitious politician, Mark Albert, 999,999,999 votes. Lensky said: "I just thought it was really easy."

In short, security experts are all convinced that Internet voting is a horrible idea and must be abandoned immediately. yet the trend is in the other direction: More of it rather than less. That doesn't bode well for the security of the upcoming elections. (V)

Defeated Democrat Says He Was Targeted by Hackers

Given how much discontent and chaos they managed to sow with their 2016 election meddling, the chances were roughly 100.00% that the Russians would be back for another go-round in 2018. There have already been a handful of incidents confirming this, with (possibly) the latest coming out of California, where Democrat Hans Keirstead claims he was targeted during his (unsuccessful) run for Rep. Dana Rohrabacher's (R) seat:

They wanted full control of my email. They were fairly sophisticated. Mimicking personalities in order to extract money and info. It was pretty amazing how sophisticated it was.

Keirstead is a neuroscientist, so it is not likely he misunderstood what was going on. And he's not a career politician, and is not seeking redress, so it is not likely he is making it up. Especially since the company hosting his website ultimately found evidence of 130,000 separate unsuccessful attempts to gain administrator access.

The FBI and the DNC have both refused comment, arguing that they don't want to give more attention to the issue, and to encourage even more hacking. Keirstead says he only went public so that Americans would understand the threat is real. That said, there are currently two big questions that have no answer (at least, not one that is publicly known). The first is: Who was behind the attacks? The Russians are certainly the obvious culprit, since they've done it before, and since Rohrabacher is famously very Russia-friendly (nickname: "Putin's favorite Congressman").

However, that leads to the second question: What was the motivation here? In some states, a candidate who gets 35% of the vote wins if it is more than any of the other candidates, and that's the end of the story. Under California law, though, the top two primary finishers advance to the general election regardless, so there was no way that a hacker could "elect" Rohrabacher in the primary. All they could really hope to do was influence the choice of opponent. And that is the one hole in the "the Russians did it to help Rohrabacher" theory, that there's not a clear motivation for their involvement. It's possible they wanted some practice for the general, or that they somehow felt that Harley Rouda was a more beatable opponent than Keirstead, or even that Rohrabacher was in danger of finishing third and being eliminated (although he actually finished in first by a comfortable margin). Any of these explanations is plausible, but none is particularly satisfying.

There is another way of looking at this, similar to the way in which police sometimes approach murder and other crimes. We could ask, "Who had the most to gain from harm done to Keirstead?" There is a clear answer to that: Rouda. While Rohrabacher was never in any real danger, the race for second place was very, very close. So close, in fact, that Keirstead appeared to be the winner on election night, only to have Rouda rally when the absentee ballots were counted, and to win the right to advance by just 125 votes (30,099 or 29,974). Note that there is zero evidence that Rouda or his campaign did anything, beyond mere supposition. Of course, there is zero evidence that the Russians did it, either, beyond mere supposition. If it did somehow turn out that one Democrat cyber-attacked another to win a primary, though, then it would unleash a scandal of epic proportions. Particularly if it turned out that the DNC knew the truth, and kept it to themselves. (Z)

Email a link to a friend or share:

---The Votemaster and Zenger