A Look at the Midterms on Memorial Day

---

---

For quite awhile now, Democrats have been salivating over the assumption that they could win back the House in November. Nancy Pelosi has been speaking in public of becoming House speaker again. Just a matter of time.

And, for awhile, that assumption included the possibility of flipping the Senate, too.

I tended to agree with the former, largely because the historical trend has been for the out–of–power party to make gains in the midterm elections. That doesn't mean the control of either chamber (or both) always changes hands. It just means the out–of–power party makes gains.

I have never been as certain that the Democrats could capture the Senate in this cycle. The 2018 map simply does not favor them for achieving that task.

Gains in the House happen whether the president is popular or not. Barack Obama was popular in 2010, the year of the first midterm of his presidency, but Republicans made huge gains in the House and seized control of that chamber. Nearly 30 years earlier, in Ronald Reagan's first midterm elections, Democrats already had control of the House, but they added 27 seats to their advantage and virtually ensured their dominance for the next decade.

Reagan, it is worth noting, did not enjoy the kind of approval ratings in 1982 that he did in the rest of his presidency, largely because the economic recovery that would propel Reagan to victory in his campaign for re–election in 1984 had not begun. In fact, his 1982 approval ratings at that point in his presidency resembled Donald Trump's, which is a compelling historical reason for thinking that the House may well flip. The Democrats need about two dozen seats to gain control of the House, and House districts are more compact than statewide seats (unless those statewide seats are in the smaller states — by population).

Midterms have often been referendums on the president — and, because Trump's approval ratings have been so lackluster, it has been only natural to expect that he would drag Republicans on the ballot down.

Clearly the cards appear to be stacked against the Republicans in the House in 2018.

For awhile, it seemed that was what was going to happen. Trump's deficit in approval polls was in double digits — but the enthusiasm gap has been narrowing in recent months. According to Gallup's most recent survey, the deficit is nine percentage points. Some polls even show Republicans pulling even with or ahead of Democrats in that generic ballot.

Likewise, the generic congressional ballot that had Democrats leading Republicans by double–digit margins only a few months ago has witnessed a decline to below 5% in many polls. A couple of weeks ago, CNN reported that Democrats held a three–point lead.

That's significant because a national margin of around 5% could be accounted for by vote totals on the East and West Coasts, where Democrats ordinarily prevail. In the Hillary Clinton–Donald Trump presidential campaign, the coasts voted heavily for Clinton, which predisposes them to vote Democratic this time, leaving little or no room for error in the interior states.

Given how the 2016 election turned out, it is not surprising that I have heard many Democrats wonder if they can trust the polls. I believe they can. The 2016 polls predicted that Hillary Clinton would win the popular vote by 2–3%, and that is roughly the margin she received.

Most polls did not, however, predict which states would vote for which candidate; consequently, there was no warning that the celebrated "blue wall" would crumble on election night.

Recent primary results indicate that a battle is being waged within the Democratic Party between its far–left wing and its centrist wing, and the outcome can have a profound effect on Senate races. Centrist Democrats stand a much better chance of winning Republican–held Senate seats in the South whereas the more leftist Democrats are more likely to prevail in Northern states.

Here in Texas, for example, an agenda that favors abortion and opposes guns is going to be a dealbreaker, even though Texas' share of the vote for the Republican ticket in 2016 was smaller than usual.

Numerically, the odds would seem to favor Democrats even more in the Senate, where flipping only two seats would give them control of the chamber. But many of those Republican–held seats are in the South and the same dynamics at work in Texas apply.

There are enough Republican–held Senate seats to give Democrats the majority if they can capture them. Two are open seats — in Arizona and Tennessee — and one is a Nevada incumbent, where Clinton won by more than 27,000 votes in November 2016.

But just winning those seats won't be enough for Democrats, who must defend seats in several states that voted for Trump, sometimes by wide margins. West Virginia Democrat Joe Manchin, for instance, faces an electorate that gave Trump nearly 68% of the vote in 2016. Missouri Democrat Claire McCaskill must overcome Trump's half–a–million vote advantage.

Indiana is a traditionally Republican state that has supported only two Democrats for president (Barack Obama in 2008 and Lyndon Johnson in 1964) since 1940. Democrat Joe Donnelly faces challenging terrain there as he seeks a second term.

There are other seats Democrats must defend in states where they fared better in 2016, but the bottom line is that a great deal of time and money must be devoted to holding them.

Thus, while the odds still favor a Democratic takeover in the House, the Senate is likely to remain problematic for the Democrats in November.

What Will It Take to Flip the House?

The Democrats are going against the tide of history in this year's midterm elections, which are now slightly more than seven months away.

Conventional wisdom says the out–of–power party outperforms the in–power party in a midterm, and that is likely to be the case in 2018 — but conventional wisdom says nothing about whether the former is likely to win control of either chamber of Congress.

And that is the prize the Democrats really seek. It would nice to narrow the gap, but it wouldn't be the same as seizing control of the chamber and being able to block any White House initiatives in the next two years.

Doing well on the state level is vital for the Democrats as well for it is in the state legislatures that most of the district boundaries for the 2020s will be drawn. High turnout for Democrats running for federal office may help with this farther down the ballot.

In the House, the Democrats need to flip roughly two dozen seats, and the ever–increasing number of Republican House retirements may very well make that a possibility. But the odds are that Democrats still will need to defeat some Republican incumbents to achieve their goal.

It's a tall order, but it can be done. There have been 16 midterm elections since the dawn of Dwight Eisenhower's presidency, and Democrats have flipped two dozen or more House seats in four of them. They did it one other time, too, in a presidential election year — 1964, when Lyndon Johnson crushed Barry Goldwater in a landslide.

Historically, Democrats need some kind of catalyst to flip seats at that rate. The main catalyst in 1958 was a recession, which contributed to the flipping of 49 House seats.

In 1974, Democrats flipped 49 House seats again, thanks primarily to the fallout from the Watergate scandal and Richard Nixon's resignation that summer.

Democrats flipped 26 seats in 1982, Ronald Reagan's first midterm, when the recession that began under Reagan's predecessor, Jimmy Carter, continued.

In 2006, Democrats flipped 31 seats, thanks to a lowered national opinion of President George W. Bush, the war in Iraq and congressional scandals.

Democrats' distaste for President Donald Trump likely will not be enough by itself in 2018, and the economy is not doing them any favors.

What Does the Future Hold for Democrats?

---

---

We don't yet know the outcome in Tuesday's special election in Pennsylvania.

Well, OK, we do know — probably — that Democrat Conor Lamb beat Republican Rick Saccone for an open House seat — but it is so close that, even after the mail–in ballots are counted, there will probably be a recount. Recounts usually confirm the initial results — but not always so there is still no winner.

But the question will remain — what does this victory mean for Democrats in a political environment that appears to favor them?

Lamb did not run as a far–left Democrat. Indeed he took more centrist positions on issues like guns, and he disavowed Nancy Pelosi, which was the prudent approach in Western Pennsylvania. Other positions would have meant almost certain defeat.

But centrists are a vanishing breed in American politics.

From the days of Jimmy Carter to the days of Bill Clinton, the area was fairly reliable territory for Democrats, and that was frequently seen in its House representation as well, although two future Republican senators represented the district in the '70s and '90s, but it has been trending Republican since the turn of the millennium.

Much was made of the fact that Donald Trump won the district by 20 percentage points in 2016 even though Trump's share of the vote in that district was the same as Mitt Romney's four years earlier. Barack Obama fared slightly better in 2012 than Hillary Clinton did in 2016 only because Libertarian Gary Johnson siphoned off 3% of the vote in 2016. John McCain received 55% of the district's vote in 2008, and George W. Bush carried 54% of the district's ballots in 2004.

An extreme–left stance probably would have resulted in Lamb's defeat, which leads us to the question of what the immediate future holds for Democrats. The assumption is that this will be the Democrats' year, and extreme positions will work in some places like San Francisco and New York, but they won't work in places like Pittsburgh.

Midterm election years are quite different from presidential election years. It is easier to nationalize campaigns when there is a presidential race on the ballot. In midterms everything is more localized. Yes, to a degree, midterms are referendums on the president, but they are more about issues that concern the voters in specific states and districts.

Obviously, what works in one place won't necessarily work in another, but the midterms will be crucial for Democrats to establish their identity for the 2020 election, when the presidency will be on the ballot. Now is when the Democrats need to decide if they are going to take a more moderate approach or veer farther to the left.

If they want to take a more inclusive approach in the hope of luring disaffected Democrats who abandoned the party in recent years, they may risk a rebellion from the radical fringe.

It will be interesting to see which direction they choose.

Lessons From the Past

---

---

Our political system is an amazing thing.

It really is. Oh, I know we all complain about things that government does or doesn't do, and we get mad at our elected officials from time to time — but nearly without exception our system has permitted us to make peaceful periodic changes in our elected leadership. We take that for granted, but we wouldn't if we lived in many other places in the world.

But our system also has its idiosyncracies.

The pendulum is always swinging, and the out–of–power party always has plenty of reasons to be energized by midterm elections, starting with the clear historical trend that favors the folks who are outside looking in. This time it is the Democrats' turn as the out–of–power party, and everything seems to point to a big year for them. The president's approval numbers remain low, and Democrats continue to hold a lead in the generic congressional ballot.

Along with that, nearly three dozen Republicans in the House have announced their intention to retire, and more seem likely. The terrain certainly looks favorable for Democrats in 2018.

But history has some cautionary tales.

Let's start with the most recent history that Democrats ignore at their peril.

In 2016 polls showed Hillary Clinton with the lead over Donald Trump — and, indeed, Clinton did win the popular vote by a considerable margin.

But the United States has never elected its presidents by popular vote. It has always elected its presidents by electoral vote, and Clinton's popular votes were too heavily concentrated in the coastal states to influence the Electoral College. (In fact, if you took California's vote entirely out of the mix, Trump would have won the popular vote as well as the electoral vote; Clinton's margin in California was about 3.1 million whereas her margin nationally was 2.86 million.)

The same thing appears to be likely in this year's congressional races. Democrats are concentrated in urban districts, and the Democrats' nominees in those districts are likely to pile up impressive margins. Nancy Pelosi, for example, routinely rolls up incredible margins in her Bay Area district. It's even likely in some places here in Texas, where Clinton carried the metropolitan counties of Dallas, Travis, Bexar and Harris by wide margins.

But all you need to win an election is a single vote. You'd like to do better than that, of course, but some Democrats are likely to roll up huge margins in some districts — when many of those votes would be more beneficial elsewhere.

In Texas, outside of the metro counties and the ones that border Mexico, Republicans still dominated in 2016 — and likely will continue to do so. Some Democrats are salivating at the thought of the open seats that have been held by Republicans, like the South Texas district that has been represented by Republican Lamar Smith for more than 30 years. Smith is retiring, and there have been rumblings of how Democrats think they have an opportunity there, but one of the Democrats seeking the seat once served on Pelosi's staff. That might help win the Democratic primary, but it isn't likely to be a general–election winner in a district that voted for Trump by 10 percentage points.

That brings me to another point. The Democrats, like the Republicans in the first midterm of the Obama years, are engaged in a battle from within. The battle is between the establishment and the extremists. At stake is the direction of the party.

As the battle plays out, the establishment will prevail in some places, and the loose cannons, who are typically the most energized in the midterms, will prevail in others.

Democrats are certain to try to nationalize the campaign, but midterms are not national campaigns. They are held in every state and every House district, but the issues and candidates vary. It is tempting to vote for the loose cannons because they typically oppose everything the in–power party does, but Democrats need to remember how some of those loose cannons worked out for Republicans in the past.

In 2012, Missouri Republican Todd Akin made his widely reported remarks about "legitimate rape" that helped politically endangered Sen. Claire McCaskill win a second term by 16 percentage points. McCaskill is back, still politically vulnerable and running for a third term in a state that voted for Trump by nearly 19 percentage points.

Similarly, Indiana Republican Richard Mourdock's remark that "even if life begins in that horrible situation of rape, that it is something that God intended to happen." Mourdock won the nomination by defeating six–term incumbent Richard Lugar in the primary.

Indiana has only voted for a Democratic presidential nominee once since 1964, but it voted for the Democrat in that Senate race, Joe Donnelly. He, too, is up for re–election — in a state that supported Trump by slightly more than 19 percentage points.

McCaskill and Donnelly were originally expected to lose in 2012, and their victories are big reasons why, when Democrats need to win only two seats from Republicans to have a majority in the Senate, they must defend more than two dozen Senate seats in November.

Democrats have a rare opportunity in 2018, but it is not a slam dunk.

Taking Back the House

---

---

Democrats face a similar situation to the one Republicans faced eight years ago. In 2010 Democrats held the White House and both chambers of Congress. Today Republicans do.

Granted, the Democrats had larger majorities in both the House and Senate — and they had a more popular president, too — in 2010. Yet they still managed to lose their advantage in the House when Republicans gained a net of 64 seats that year. They lost ground in the Senate and eventually lost that majority as well.

Today many political observers are convinced that the tables have turned — which is based on solid historical data. This is a midterm year, and midterm years almost always go against the party in the White House. That has been true whether the incumbent was popular or not.

Indeed, presidential approval ratings play an important role in midterm elections, but the responses have become increasingly polarized over the years. In the 1950s, for example, an average of nearly half of Democrats said they approved of the job Republican Dwight Eisenhower was doing as president. In the 1980s, an average of less than one–third of Democrats approved of the job Republican Ronald Reagan was doing, and in the 1990s, slightly more than one–fourth of Republicans approved of the job Bill Clinton was doing.

Clinton's successors, George W. Bush and Barack Obama, failed to average even that much support from the opposing party.

But Bush's Republicans benefited electorally from the terrorist attacks of 2001. They might have been expected to lose ground in the midterm elections of 2002; instead, they gained ground in both chambers, the first time a president's party accomplished that in a midterm election in nearly 70 years.

(It is unwise to ignore the influence that circumstances can have. At the same time, though, it is not wise to expect too much from things like scandals. The Iran–Contra scandal dropped Reagan's approval rating below 50% in January 1987, but he rebounded to higher than 60% by the time he left office two years later.)

I have a theory about that trend. When it is a president's first — and, in many cases, it has been a president's only — midterm, that president is two years removed from winning the presidency, and his supporters are complacent while his foes are energized. When it is a president's second midterm, his supporters are generally demoralized by something, a scandal or whatnot, and the rest of the country, from those who are indifferent to long–time detractors, is just weary.

It takes truly unusual circumstances for any incumbent to overcome that, and so far such circumstances have not materialized in this election. But it has been observed frequently that the 2016 elections rewrote the rules so I wouldn't rule it out.

In 2018 Democrats need fewer than half as many seats as they lost in 2010 to claim a paper–thin majority in the House. That sounds plausible — and it is — but there is something that is worth remembering.

Unlike Senate seats, which are decided every six years, House seats are on the ballot in every election. There have been 22 elections since Watergate, and a single party has gained that many House seats (or more) in four of them. The rest of the time the gains were less than 24.

It's a tall order — but not one that is impossible to fill.

As Kyle Kondik of Sabato's Crystal Ball recently noted, there is already an unusually high number of House incumbents not seeking re–election — twice as many Republicans as Democrats.

In fact, there are enough open seats in Republican–held districts for Democrats to entertain thoughts of capturing the majority in the House by winning most of them — but that would be a foolish strategy. It ignores the fact that not all districts are created equal.

Some districts have long histories of voting for one party or the other. Like mine, for instance. I live in Texas' Fifth District. It has been represented by Republican Jeb Hensarling since 2003. The only time he was held under 60% of the vote was when he was originally elected in 2002 — and he received 58% in that election. He announced a couple of months ago that he would not seek re–election.

Hensarling is not leaving because he anticipates a tough election. He is highly regarded here and would be sure to win another term if he wanted one. Whoever wins the Republican nomination will be a heavy favorite to win the general election — if he/she is even opposed.

If the voters in this district elect a Democrat to succeed Hensarling, it will be a clear indication that a wave election is underway.

Democrats are more likely to gain ground in districts like Arizona's 2nd District, which was represented by Democrat Gabrielle Giffords before she was shot in 2011 and had to retire. One of Giffords' aides was elected to fill her vacancy in 2012. Voters narrowly chose the loser of the 2012 election — Martha McSally — in a rematch in 2014. McSally was re–elected with 57% of the vote in 2016 and now is running for Jeff Flake's Senate seat.

Democrats are favored to win that House seat this November.

Other open districts are just as evenly divided — and could be prone to flip in the next election with no incumbent on the ballot. The power of incumbency, as I have noted here before, is considerable.

But it is not absolute.

Open seats do present opportunities for the party that does not occupy the White House, but Democrats have to be selective about which ones they pursue. Kondik says they need to net at least half a dozen Republican–held open seats to be on track to seize the majority in the House. The rest, he wrote, will need to be taken from the officeholders. His estimate is that Democrats will need to beat 15 to 20 incumbents head to head.

That may seem like a challenge, but Kondik insists the number is not too high by historical standards.

Time will tell.

The Minnesota Two-Step

---

---

Let's start with some stuff that is important to understand, although not enough people do:

U.S. senators are elected to six–year terms, and the terms are arranged so that one–third of the Senate is on the ballot in any given election. In 2018 the electoral map has few Republicans facing the voters as the senators who are up for re–election won their current terms in 2012, the year Barack Obama was re–elected. That was a pretty good year for Democrats.

If those senators were re–elected in 2012, their previous election would have been in 2006, which was a big year for Democrats. That was the year they seized control of both chambers of Congress for the first time in more than a decade.

The Democrat senators who are on the ballot in 2018 had favorable winds at their backs the last two times their seats were on the ballot. Their party gained two Senate seats and eight House seats in 2012 — five Senate seats and 34 House seats in 2006.

The political terrain wasn't as favorable for Republicans in those years as it was in others, and fewer were successful. As a result, fewer Republicans hold the Class 1 seats that will be on the 2018 ballot.

Senate terms are also staggered so that no state must elect both its senators in the same election year — unless there is a midterm vacancy that needs to be filled.

Sen. Al Franken's stated intention to resign in January puts Minnesota in that comparatively rare category in 2018.

Amy Klobuchar, Minnesota's senior senator, is seeking a third six–year term in her Class 1 seat. Franken's seat is a Class 2 seat that would be slated to face the voters again in 2020, but because he is leaving the Senate, his appointed successor will be on the ballot in the next election. The voters will decide who will represent them until 2020 — at which time the winner of the 2018 race will have to decide whether to seek a full term.

Such two–fer Senate elections are rare — some states have never had one — but this will be the second time for Minnesota. The first time was 40 years ago — in 1978. Then, as now, both seats were held by Democrats.

Does that 40–year–old election have any relevance to 2018?

One of the seats had belonged to Sen. Walter Mondale, who was elected vice president in 1976. Minnesota Gov. Wendell Anderson resigned so his lieutenant governor could become governor and appoint Anderson to fill the Senate vacancy for the last two years of the term.

The other seat had belonged to former Vice President Hubert Humphrey, who was elected to the Senate after leaving the vice presidency and then re–elected in 1976; then he died of cancer in January 1978. His widow was appointed to the seat until an election could be held later that year. The winner would hold the seat until 1982. Muriel Humphrey chose not to run, and Minnesota Democrats selected a rich trucking firm owner to be their standard bearer.

Minnesota has a reputation as a deep blue state in 2017, and it was quite blue in the '70s, too, although it did vote for Richard Nixon in 1972 and had voted for other Republicans for president in the first half of the 20th century; still it hadn't elected a Republican to the Senate since 1952.

Republicans used that to their advantage. They made the point that Minnesota's leading statewide offices were held by people who had not been elected to them — and the seats would all be on the ballot in 1978. That was the voters' big chance, the Republicans said, gleefully asserting that Minnesota's Democrats would face "something scary" in 1978 — "an election."

That wasn't entirely fair. Democrats had been elected to all those offices the last time they were on the ballot, and Democrats were appointed to fill the vacancies. It wasn't as if someone was circumventing the political will of the voters.

(The party affiliation may have been the same, but the philosophy wasn't always. That trucking firm owner who was nominated when Mrs. Humphrey decided not to run was more conservative than most Democrats were then or are now.)

In 1976 Minnesota gave the Carter–Mondale ticket nearly 55% of the vote. The voters knew that, if the ticket won the election, someone would replace Mondale in the Senate. They voted for the Democratic ticket, anyway. When Humphrey was re–elected that same year, it was no secret that he was sick; nearly three–fourths of Minnesotans voted for him, anyway.

What hurt the Democrats in 1978, though, was Anderson's blatant move to gain Mondale's seat. I suppose it hinted of entitlement to many, and the voters clearly didn't like that. Only 42% of them supported Anderson in November.

Muriel Humphrey, by her own admission, was no politician. She was a politician's wife, and she played that role graciously for decades, then dutifully kept the seat warm until the election. Would the magic of the Humphrey name have carried the day and kept the Senate seat in the Democrats' hands if she had not decided to step down?

We'll never know, but that businessman who won the party's nomination when Mrs. Humphrey declined to run only got 36% of the vote in November.

When the Senate convened in January 1979 Minnesota was represented by two Republican senators for the first time since the Truman administration.

Republicans completed the sweep by winning the governor's office as well, and the 1978 election came to be known as the "Minnesota Massacre."

Now, to an extent, voters throw tantrums in these special elections and vote contrary to their usual patterns. Like a fever, though, it passes, and voters return to their roots by the time the next election is held. We saw this in the early part of this decade when Massachusetts elected a Republican to serve the remainder of Ted Kennedy's term, then chose a Democrat when the seat was on the ballot for a full six–year term.

I suspect we will see that same phenomenon — albeit in reverse — in Alabama in 2020, when Democrat Doug Jones must decide whether to seek a full six–year term.

The dynamics were different in Minnesota 40 years ago, though. The Republican who was elected to complete Humphrey's term, David Durenberger, went on to win two full terms and then retired after 16 years in the Senate. Rudy Boschwitz, the Republican who defeated Anderson, was re–elected once.

If Minnesota threw a tantrum in 1978, it had staying power. It remains to be seen whether the voters of Minnesota will throw a similar tantrum in 2018.

Political tantrums require catalysts, and those catalysts vary from state to state. The circumstances that led to Scott Brown winning Ted Kennedy's seat in 2010 were different from the ones that propelled Doug Jones to victory in the race for Jeff Sessions' seat or led to the Minnesota Massacre.

At present there appear to be no storms on the horizon for Minnesota Democrats, but as I observed more than a year ago, in spite of voting Democratic in 10 consecutive presidential elections, deep–blue Minnesota wavered a few times and was a candidate for flipping to the other party in 2016.

It didn't, but it came close. A week after I posted that, Minnesota voted Democrat for the 11th straight time — but Hillary Clinton's share of Minnesota's vote was the smallest for a Democrat since her husband in 1992.

And Bill Clinton could point to the presence on the ballot of a credible third–party candidate who took nearly a quarter of Minnesota's vote.

Who knows which issues may emerge in 2018 to help or hurt Tina Smith, Franken's appointed successor? Smith, who was once regarded as a gubernatorial prospect, will become a senator as an indirect result of the emergence of sexual harassment as a political issue and a direct result of credible accusations that were leveled at her predecessor.

But what if not–so–credible accusations are made that cast a shadow over the issue? That could lead to voter backlash.

Are there any Tawana Brawleys lurking out there in Minnesota?

What Does It Mean?

---

---

Barring a reversal by recount, it appears that Democrat Doug Jones has defeated Republican Roy Moore in Alabama's special election to fill the Senate vacancy left by Jeff Sessions' appointment as attorney general.

Jones did not win a full six–year term. He was only elected to finish the unexpired portion of Sessions' term. If he chooses to seek a full term, he will have to do so in 2020.

In that regard, he is much like Republican Scott Brown, who won a special election in Massachusetts in 2010 to complete the term of Sen. Edward Kennedy. Kennedy died in office in August 2009, and Brown was elected to serve the last two years of his term.

The dynamics of the elections differ, but the overriding similarity is the fact that both states elected nominees from the parties that had not been successful in those states for a long time. It was especially shocking in Massachusetts, I suppose, since that Senate seat had been held by Kennedys for more than half a century — and Massachusetts had given nearly 62% of its vote to Barack Obama in 2008 — but it is no less shocking in Alabama, where Donald Trump received 62% of the vote in 2016.

Pundits are looking for clues to the political future, just as they did in 2010. I am inclined to reach the same conclusions now as I did then.

In 2010 I wrote the following: "I see no way that yesterday's election will not be a factor in all the other races that will be held this year." I feel that way today — but I have no more of an idea of the extent than I did then.

The circumstances could hardly be more different. A cloud of controversy hung over Moore in the last month of the campaign; the extent to which that contributed to his defeat remains to be determined, but its influence cannot be denied. There was no such cloud hanging over Martha Coakley when she lost to Brown in 2010; apparently she simply took victory for granted. She tried unsuccessfully to correct her shortcomings a few years later when she sought the governor's office and lost to the Republican nominee. Was Coakley the problem? Or was the party the problem?

The same question could be asked about Moore and the special election in Alabama.

As a harbinger of things to come, the Massachusetts election may have given Democrats — who had been anticipating seizing full control of the government after Obama became the first black man to be elected president and they achieved filibuster–proof status in the Senate with the election of Al Franken in Minnesota and Arlen Specter's party switch in Pennsylvania — a glimpse into a grim future.

The Democrats didn't lose control of the Senate in 2010, but they came close. Going into the election, the party held 58 seats (which, when combined with two independents who caucused with them, provided them with their filibuster–proof status) and emerged from it with 51 seats — still a majority but greatly diminished — and fully vulnerable to a Republican filibuster.

The 2010 midterms are not remembered in history for the outcome in the Senate races, though. What is remembered is the fact that Republicans turned a 79–seat deficit in the House into a 49–seat advantage. They have held the majority in the House ever since.

As a harbinger of things that may be yet to come, the Alabama election may have done the same thing for Republicans. The question is whether the GOP will be wise enough to heed its warning.

I don't know if Democrats can be competitive in enough House districts to pull off the kind of wave the Republicans achieved in 2010. Based on what I have seen, they may be fortunate to grab a narrow majority if they claim one at all.

Republicans don't need to lose as many Senate seats in 2018 as Democrats lost in 2010 to lose control of the chamber — and, as I have mentioned before, the historical tide of midterms runs against the party in the White House.

But until Tuesday, the Democrats faced an uphill struggle in trying to capture the Senate. Only one–third of the senators are on the ballot in a given election year — unless a special election has been called because a senator died or resigned before his/her term was completed. In this cycle, the vast majority of senators whose seats are on the ballot in 2018 are Democrats. Of those senators whose seats must be defended in 2018, only one incumbent Republican senator comes from a state that voted for Hillary Clinton whereas 10 incumbent Democrats must face the voters in states that supported Donald Trump.

With their upset victory in Alabama Democrats need only two more seats to claim a majority in the Senate for the last two years of Trump's term. That is more plausible now — but not necessarily more probable.

Regardless of whether Democrats can seize a Senate majority next year, even a short–term one, the Republicans' agenda is in jeopardy. If American politics has made anything clear in the last decade or so, it is that the voters want a change; as Republicans seemed to offer what they wanted, the voters gradually turned over the levers of government to them.

But the political pendulum is always moving in America. The Republicans' already slim majority in the Senate will shrink in January when Jones is seated, which gives them a few weeks to accomplish whatever they can with the seats they currently hold. Given their track record this year, though, getting anything done seems unlikely.

Voters seem to have developed a rather limited tolerance for political performance in recent years. Nonperformance may fare worse.

Until Tuesday, I had not seen anything in any of the special elections that surprised me. All the House seats that were vacant because their representatives had been appointed to positions in the administration were in red states. Democrats made a lot of noise about being competitive in those districts, but ultimately they lost every special election.

Then in November the first statewide races in the Trump era were held — in New Jersey and Virginia, states that elect their governors in the years immediately following presidential elections. After decades of voting for Republicans, Virginia (where governors are prohibited by law from seeking re–election) has been trending blue. Three of its last four governors have been Democrats, and it has voted Democratic in three consecutive presidential elections. Even though Republicans put up a good front about being competitive in the governor's race, I wasn't surprised when the Democrat won.

Nor was I surprised when the Democrat won in New Jersey. Term–limited Republican incumbent Chris Christie is about as popular as Richard Nixon was just before he resigned the presidency, and the Republican nominee had to carry that toxic baggage. No, I wasn't surprised by the outcome there, either.

But Tuesday's special election in Alabama did surprise me.

I realize that the circumstances had a lot to do with it. The decades–old charges of sexual harassment against Moore appear to have done enough damage to permit Jones to win by about 20,000 votes out of more than 1.3 million cast — even though no charges were substantiated and most were suspect.

Perhaps, as conservative commentator Ann Coulter observes, this outcome opens the door for Rep. Mo Brooks to reclaim the seat for the Republicans in 2020.

That wouldn't be unprecedented. Two years after Brown won Kennedy's seat, he was defeated in a bid for a full six–year term — by Elizabeth Warren, who is now considered a leading contender for the Democrats' next presidential nomination.

I'm not saying Brooks is a future presidential prospect. But in 2020 he would have Trump's coattails to ride in the general election — and that will almost certainly help in Alabama.

The Crystal Ball Is Foggy

---

---

I have observed this year's special elections to fill vacancies created by Trump administration appointments with a kind of amused bewilderment.

The special elections were all billed as referendums on the president — but all the vacancies were in clearly red states with red voting histories in good years and bad. I suppose if any of those races had gone for a Democrat, that would have been big news. But the fact that no Democrat won a special election is non–news. Kind of a "dog bites man" story. When the man bites the dog, that's news.

The same is true in reverse of the gubernatorial elections that are traditionally held in Virginia and New Jersey in the odd–numbered years following presidential elections.

At one time Virginia had the longest streak of support for Republican presidential nominees in the entire South. It was the only Southern state that voted against Jimmy Carter in 1976, but shifting demographics led it to vote for Barack Obama in 2008 and 2012 and then Hillary Clinton in 2016. Today it is regarded as dependably Democratic.

Thus, while polls in this year's gubernatorial race suggested the Republican nominee might have a chance at an upset, it was no surprise when the Democrat won.

Nor was it a surprise that a Democrat won in New Jersey. The outgoing Republican incumbent was extremely unpopular, and New Jersey has a history of shifting political allegiances. No, that wasn't a surprise.

Now the scene shifts to Alabama, where a special election will be held next Tuesday to choose a successor for Attorney General Jeff Sessions in the U.S. Senate. Republican Roy Moore has been beset by sexual harassment charges that narrowed the gap between him and Democrat Doug Jones — for a time. Recent polls suggest things are returning to form in Alabama, and Moore's lead, while closer than one usually sees in Alabama elections, is expanding.

I know there are still some Democrats, perhaps many, who believe Jones can pull off an upset, but what they fail to comprehend is that Alabama is a really red state — not pink or purple like some states but deeply red — and polls suggest that the voters in Alabama are gravitating toward issues now. My guess is it would take a sexual bombshell eclipsing everything that has come before in order for Jones to win this race.

One issue in particular seems to be driving Republicans and independents who had been wavering — abortion. Jones holds a liberal position on abortion; Moore's position has been more in line with average Alabama voters

Many folks up North like to belittle Southerners as backward, ignorant, uneducated. But that is an unfair stereotype, and the voters in Alabama are smart enough to know what the margin in the Senate looks like. They know which party and its candidates hold positions closer to their own and which do not, and this, one of the reddest of the red states, is not likely to send a Democrat to the Senate for the first time in a quarter of a century.

I know Democrats are eager to seize control of one or both chambers of Congress in the midterm elections, but they would be better advised to pin their hopes on races in more centrist states.

The Power of Anti-Incumbency

I have written here before of the advantages of being an incumbent officeholder in an election year. There is clearly a strong motivation for people to support the re–election of an incumbent. An incumbent has experience doing the job. His or her every move gets attention from the media back home, and free press coverage has a lot of value.

I have also written here about the hazards of midterm elections for the president's party (whoever the president may be).

The power of incumbency isn't always so powerful.

But often the ramifications of neither are apparent until after the elections have been held. Democrats realized too late what they were up against in 2010 and 2014; likewise Republicans didn't see the wave that was upon them in 2006.

Sometimes — but not always — you can get clues ahead of time through party primaries.

Because of their strengths, incumbents usually prevail in their parties. In a typical election, whether it is a presidential year or a midterm, most incumbent senators win renomination; at worst, one may be defeated by a challenger from within the party. Between 1946 and 2012, nearly 1,000 incumbent senators ran for re–election instead of choosing to leave office for one reason or another, and only 46 (or 5%) were denied renomination by their party's voters.

When more than one incumbent senator loses in the primaries, the problem is not confined to a single state or perhaps even region. If the number of incumbents who seek re–election and lose renomination gets higher than that, it is usually — but not always — a harbinger of things to come.

The worst year for Senate incumbents during the primaries was in the first election after the end of World War II. Thirty senators sought re–election that year, and six were denied renomination by their party's voters. That's 20%.

Harry Truman had become president following Franklin D. Roosevelt's death the year before. Even though the Allies had won the war, Truman was wildly unpopular — perhaps in part because Democrats had been in charge of everything for well over a decade but also because of his controversial handling of some high–profile postwar labor strikes — and the midterm election was, as it often is, a referendum on the president.

Truman was seen as such a liability, in fact, that he campaigned for few Democratic candidates — if any. The same phenomenon has been seen in recent years. Both George W. Bush and Barack Obama were kept at arm's length by their parties' nominees.

Republicans also benefited from what is considered a "good map" — in which the other party has to defend the majority of the seats on the ballot. Incumbents tend to benefit from such maps; open seats typically are much harder to defend.

As it was, the Republicans won seven Senate seats from incumbents who were on the ballot that fall — and captured six others, seizing the majority for the first time since 1930.

The next–worst year for postwar incumbents seeking another term came in 1950.

In the intervening four years, Truman scored his upset victory over Tom Dewey, and Democrats retook control of both chambers of Congress.

But by 1950 Truman was unpopular again, and once again the election was a referendum on him. Thirty–two senators ran for re–election and five of them (16%) were denied renomination. When the votes were counted in November, the Democrats lost ground in both chambers but still retained majorities.

In the Senate, four Republicans defeated incumbent Democrats, and one Democrat defeated an incumbent Republican. (As an historical side note, future President Richard Nixon, a Republican, flipped a Senate seat that year, too, but he didn't defeat an incumbent. The incumbent, a Democrat, retired.)

In the 33 elections that have been held since, the portion of incumbents who were rejected by their party's voters has been 10% or higher only four times. Most could be said to foretell trouble for one party or the other, depending upon who was in the Oval Office, or both — but not all. For example, 14% of incumbent senators were defeated in party primaries in 1968; while Democrats lost part of their majorities in Congress that year, they lost nearly as many Senate seats two years later — but only 3% of incumbent senators lost their primaries in 1970.

In 1978, the midterm of Democrat Jimmy Carter's presidency, 12% of incumbent senators lost their primaries, and 14% of incumbents lost their primaries in 1980, the year of the Reagan Revolution. In all, Democrats lost 15 Senate seats from the time Carter took office to the day he left Washington four years later. In hindsight, the primaries of 1978 and 1980 hinted at the voter frustration that had been building in the wake of Watergate and Vietnam and the trouble that lay ahead for Democrats in the general election of 1980.

For the next 30 years, most Senate incumbents who chose to run for another term won their primaries. Then, in 2010, 12% of Senate incumbents lost their primaries, perhaps heralding an era of discontent. The Senate remained in Democratic hands (but just barely), and the Republicans gained 64 seats to claim a House majority they retain today.

And sometimes so–called wave elections happen even when incumbents don't run into problems in the primaries, and the 2014 midterm is a great example. Republicans won nine Senate seats from the Democrats that year, and the primaries were not factors.

In short, incumbency is usually an advantage, but sometimes it is a hindrance. I recommend keeping an eye on the primaries next year to see if you can get any early clues as to the voters' mood.

The Politics of the Unusual

---

---

Tomorrow Americans in some places are going to the polls, but this being an odd–numbered year means that most elections will be held next year.

Which brings me to another point: There is conventional wisdom about everything, I suppose, but it only goes so far, and then you're in uncharted territory.

The upcoming midterm elections of 2018 are a perfect example.

On the one hand, there is the conventional wisdom that the president's party always struggles in midterm elections. This is not a recent phenomenon. This is something that has been happening throughout our history. It doesn't disproportionately affect either party. George W. Bush's Republicans suffered just as much in 2006 as Bill Clinton's Democrats in 1994 or Barack Obama's Democrats in 2010.

Sometimes it is a very modest thing, with the president's party losing little ground, if any, on Capitol Hill; other times it is quite spectacular.

There are exceptions, of course, but those elections are usually preceded by a one–of–a–kind event — such as when George W. Bush's Republicans gained ground in Congress in the election a year after the terrorist attacks of Sept. 11, 2001.

There was no precedent for that in American history.

I guess the closest thing would be the election that was held the year after the Japanese attacked Pearl Harbor, but that would not have been a good predictor for 2002. Franklin D. Roosevelt's Democrats struggled in the 1942 midterms. Quite a different voter response. But it was a different world. Americans hadn't been able to watch the attack live in their living rooms in 1941.

There were other differences, too. Most, if not all, of the casualties at Pearl Harbor were in military service. They had agreed to put their lives on the line when they signed up. Their losses were tragic, but, to be honest, they came with the territory. Most, if not all, of the casualties on Sept. 11 were civilians. The folks who were working in the Twin Towers that morning took certain risks when they took their positions, too — there are risks with every job, aren't there? — but those risks had never before included the realistic possibility of an airplane deliberately crashing into your workplace.

Back to 2018.

On the other hand, while conventional wisdom can provide some helpful clues to voter behavior, recent elections have revealed an independent streak that was never seen before in the electorate — well, it had been seen before but never in the numbers we saw last year. Maybe the voters don't like being taken for granted and decided en masse — a la Peter Finch in "Network" — not to take it anymore. Folks in both parties have been guilty of taking voting blocs for granted. So if the voters feel taken for granted, it seems to me that candidates in both parties should be particularly sensitive to their constituents' concerns right now.

There are other things that usually contribute to the incumbent party's prospects — the pocketbook issues that directly affect people's lives. Is the economy thriving or sputtering? Is unemployment high or low?

The approaching midterms should provide fascinating research and lecture fodder for political scientists. They know the conventional wisdom, and they have been watching the news. They know that the polls indicate how unpopular the president is. That makes it sound like 2018 should be a big year for the out–of–power party. All the precedents of the last two centuries point to it.

Except that there is no precedent for 2018, either. Not really.

And that, I guess, is one of the consequences of the 2016 election, an election that was widely believed to be Hillary Clinton's to lose. And then she did precisely that.

We have rarely, if ever, had presidential elections like that one — in which one candidate seemed all but certain to win and then did not — and in my studies of history, I have found only a couple of elections that came close. One was the 2000 election in which Vice President Al Gore won the popular vote but lost the vote that has always elected America's presidents — the electoral vote. The other was the 1948 election, in which Gov. Tom Dewey was widely expected to defeat President Harry Truman — and then failed to do so.

But there were precedents for those elections, even if one had to go back a ways to find them. The 2016 election was decidedly not politics as usual — which, by definition, has no precedent.

Consider this: Hillary Clinton spent most of the 2016 campaign arguing that Donald Trump was unfit for office — but he won, anyway. Clinton in particular and Democrats in general have been quick to blame this on misogyny and, implausibly, racism, but that misses the greater point. I know many people who voted for Trump (before that some of them even voted for Obama twice), and I have yet to hear any of them say anything that could be interpreted as misogynistic motives behind their votes.

Those voters consistently expressed their concerns about specific issues — the economy and jobs — and they responded to the candidate who addressed those issues. Nothing new about that. Pocketbook issues are at the very core of politics as usual.

The voters knew Trump hadn't been a saint. They knew he had said and done things they didn't like, but they chose him anyway — which is a clear indicator that modern voters are more than willing to consider unconventional candidates to solve heretofore conventional problems. We are in a period of the politics of the unusual, and the 2018 midterms will tell us just how far the pendulum has swung, just how much the voters are willing to overlook in pursuit of a larger goal.

Further complicating the situation is that the economy is humming along. Like him or not, Trump's administration has presided over some of the results the voters wanted. Democrats can argue that the pieces of the recovery puzzle were put in place by the previous administration's policies, but the voters don't tend to think of things in those terms. They remember who was president when unemployment dropped below 4% or GDP exceeded 3%.

In other words, all bets are off for next year — and for 2020 — and for elections as far as the eye can see.

My major in college was journalism. My minor was history. I studied a little political science in college, and I have studied it informally for years, but I have never heard a lecture or read an entry in a textbook that discussed how an unsuccessful candidate for public office who had been expected to win should behave.

Instinct tells me that such a person would foster considerable good will by being gracious and sincere — and mostly silent.

But rather than acknowledge her own failings as a candidate, Clinton and her subordinates have spent the last year casting blame on others, the lion's share of which has been directed at Donald Trump and his alleged collusion with Russians — which, it turns out, was based on manufactured material from a dossier that had been paid for by the Clinton campaign and the Democratic National Committee.

That revelation led to the uncovering of what could well turn out to be a veritable rat's nest of, to say the least, unsavory activity. While the full extent of it may not be known by Election Day next year, the fact that this has been causing some Democratic angst is clear in the fact that Sen. Elizabeth Warren, regarded as one of the leading candidates for the 2020 nomination, criticized the Clinton campaign and the DNC literally within hours of online publication of Donna Brazile's explosive allegations in Politico.

It is important for political observers not to get carried away with the idea that open seats — in which the incumbents, for whatever reason, are not running — represent opportunities for the parties that do not hold them. It is true in some instances, not true in others.

Take, for example, the congressional district in which I live — Texas' Fifth District, which has been represented by Republican Jeb Hensarling since January 2003. Hensarling announced recently that he won't be running for another term next year.

I have already heard some Democrats speak of how this is an opportunity for them to grab a House seat from Texas, especially since Dallas County (where the Fifth District is located) was one of the few counties to support Hillary Clinton a year ago.

I read an article in the New York Times over the weekend that didn't go so far as to say that vacancies meant easy opposition pickups but it strongly implied that the midterms will "reshape" the House.

That may be true in districts where the incumbent barely won the last time, but Hensarling has a long record of winning by wide margins. So, too, do Republican presidential candidates in the Fifth District. I feel safe in predicting that Hensarling's successor will be a Republican. The question is whether the district will choose a constitutional conservative like Hensarling or more of an establishment candidate.

Still, as I say, we're living in the age of politics of the unusual.

Will Control of the House Flip in 2018?

---

---

I've heard a lot of talk recently — mostly from hopeful Democrats — that control of the House of Representatives will flip next year after eight years of Republican rule.

Given the current party division in the House, that would require the Democrats to make a net gain of two dozen seats.

Can it happen? Historically speaking, yes, of course. It has happened before. It is mathematically possible that it could happen again.

But will it happen again? Ah, that is a different question. To answer that question in May 2017 when the election won't be held for another 18 months requires a crystal ball — after all, who, at this point in the last election cycle (i.e., May 2015), predicted that Donald Trump would be the next president of the United States?

No one knows in which kind of world voters will be living when they go to the polls 18 months from now, and that will play an important role in the elections.

Now, it is true that, historically, a president's party loses ground in Congress in a president's first midterm elections, but all midterms are not created equal. Sometimes a president's party loses ground in one chamber but not both — Richard Nixon, as disliked as he was even by many who voted for him, lost ground in the House but not in the Senate in the midterm elections of 1970. In fact, Nixon's Republicans actually gained a couple of Senate seats but remained in the minority.

Four years later voter backlash over Watergate led to a loss of 48 House seats for the Republicans.

And, while sometimes presidents lose House seats in bunches, as Obama did in 2010, other times presidents lose only a handful of seats. In 1990 George H.W. Bush's Republicans lost only eight House seats. Four years earlier Ronald Reagan's Republicans lost only five House seats.

One–term presidents — Jimmy Carter and George H.W. Bush are recent examples — only have one midterm election. For presidents who have been elected to two terms, second midterm election results have been decidedly mixed. Barack Obama's party lost control of the Senate in his second midterm after losing control of the House in his first. George W. Bush's Republicans lost control of both chambers of Congress for the first time in more than a decade in his second midterm. But Bill Clinton's Democrats picked up seats in the House and saw no change in the Senate in the midterms of 1998.

Clinton's experience was rare for presidents and seems to have been fueled by voter backlash over the impeachment proceedings against Clinton. That is what seems to be necessary for a president's party to gain ground in the midterm elections — extraordinary circumstances that offset the natural enthusiasm that comes from being the party that is outside the White House looking in.

Prior to the Clinton years double–digit losses in the House — at least at the level that Democrats need next year — were uncommon in American politics. They did happen from time to time but not as regularly as they have since Clinton came to power.

Reagan's party lost 26 House seats in the midterms of 1982, but the party of his predecessor lost only 15 seats four years earlier. In between Reagan defeated Carter by 10 percentage points.

American democracy is a dynamic thing, always shifting in response to economic, social and political conditions — and the elected officials' responses to those conditions.

Such conditions are always changing. That is why it is a disaster waiting to happen if a candidate campaigns on the assumption that simply because a party has been winning for years in a state or district it will continue to do so. History is a pretty good indicator, but it is not foolproof, as Hillary Clinton should have learned on election night.

No modern president has faced an economy as horrendous as the one Franklin D. Roosevelt inherited in 1933, but the conviction that he was trying to right the ship enabled his party to make gains in both chambers in the midterms of 1934.

It runs deep in the American DNA to reject the notion of single–party rule in which one party controls all the levers of the federal government. Such a situation existed in the first two years of Obama's presidency — Democrats even held a seldom–seen veto–proof (and also filibuster–proof) majority in the Senate.

But the passage of Obamacare led to the voter backlash that resulted in Republicans seizing the majority in the House.

As much as Americans tend to reject the concept of single–party rule, though, it is important to remember that House races usually favor the incumbent. Congressional districts are concentrated, as small constituencies are wont to be, and tend to be the perfect examples of Tip O'Neill's pearl of wisdom that "all politics is local." Most House incumbents, regardless of party, keep their fingers on the pulses of their districts — if they don't they are almost sure to lose in the next election.

A few states have populations that are small enough that they are entitled to only one member of the House; in those instances, the House members are, essentially, statewide representatives like the state's two U.S. senators. But most states have more than one House member, thus concentrating the constituents' interests. A largely rural district can co–exist next to a largely metropolitan one — and, thus, different issues will matter to the constituents in each.

Even within districts, there can be pockets where the prevailing interests are different than in the rest of the district.

Currently Charlie Cook, perhaps the foremost observer of House politics, says Republicans hold 197 solid seats. That leaves 44 Republican–held seats, of which Democrats need to win 24 to seize a slim majority, that represent far more plausible takeover opportunities.

Of those 44 seats, though, Cook says 19 are likely to remain in Republican hands, which trims the Democrats' margin for error considerably.

Based on that, if the elections were being held today, Republicans most likely would hold on to a majority in the House.

But the elections are not being held today.

Stay tuned.

Looking Ahead to the 2018 Midterms

---

---

As they survey the carnage that was wrought by eight years of lurching ever more to the left, the Democrats' only immediate hope is to regain at least part of their control of Congress before Donald Trump seeks a second term in 2020.

If they can't accomplish that in the 2018 midterm elections, they will be unable to do anything except morph into the "Party of No" that they were fond of calling congressional Republicans during the Obama years. Unlike the Republicans of most of those years, though, they will not control at least one chamber of Congress.

And without a base of power, it is unlikely that the party can find a candidate capable of defeating an incumbent Republican in 2020 — which most likely means continuing Republican dominance in at least the early years of the 2020s.

There has been considerable rending of garments and gnashing of teeth among the Democrats' ranks, but it has not been exclusively due to Hillary Clinton's victory in the popular vote and defeat in the electoral vote. At least part, I am convinced, is because many Democrats recognize the enormity of the task before them.

If Democrats are to have any influence when House districts are redrawn following the 2020 Census, they need to win control of state legislatures, most of which are in Republican hands. If they can't do that entirely in 2018, they need to have a solid start toward an objective that can be realistically accomplished in the 2020 presidential election year.

That's going to be a tall order.

The good news for Democrats is that, historically, midterm elections tend to favor the party that does not hold the White House, but to seize the majority in the House of Representatives, Democrats need to win about two dozen Republican–held seats. History suggests that, even if Trump's popularity remains below 50%, the odds are against that. It has happened before — recently, in fact — that the out–of–power party has won that many seats from the opposition party in a single election, but it is the exception to the rule.

And it almost never happens that a new president's party goes from being the majority party in the House to losing that many seats and control of the chamber in his first midterm election.

Trump is the ninth president since World War II to enter office with his party holding the majority in the House. Four of the previous eight — Harry Truman, Dwight Eisenhower, Bill Clinton and Barack Obama — saw their parties lose control of the House in their first midterm elections.

Ronald Reagan's Republicans never held the majority in the House while he was president, but they did lose 26 seats in Reagan's first midterm election.

Only one postwar president who entered office with his party controlling the House — George W. Bush — saw his party pick up seats in his first midterm. Bush's Republicans did so with the Sept. 11 terrorist attacks still fresh on voters' minds. If a similar event occurs between now and November 2018, Republicans might well add to their sizable majority in the House. They would almost certainly see gains in state legislatures.

On the other hand, if voters have a strong negative reaction to something the White House does — as they did with the passage of Obamacare in 2010 — they could punish the incumbent's party severely.

Either extreme is possible, but right now neither extreme is likely. Thus far, at least, Democrats have had no galvanizing moment, but neither have Republicans. Congressional approval is about twice what it was a year ago — still not great but about as sturdy as it ever is — and House districts, being the compact constituencies that they are, are much less likely to give their own representatives the boot — or even to shift parties if the incumbents retire.

Retail politics is what matters most in congressional districts.

To add another twist to the narrative, Kyle Kondik of Crystal Ball observes that, while Donald Trump won more congressional districts than Hillary Clinton, more Republicans hold seats in districts that voted for Clinton than Democrats hold seats in districts won by Trump. If Democrats can win those "crossover seats" that are in Republican hands — there are 23 — that would leave them only one vote behind the Republicans in the House.

"If a party can win the district at the presidential level," Kondik observes, "it's reasonable for that party to believe it can win the seat at the congressional level, too."

Before Democrats start thinking that taking back the House will be a slam dunk, it is important to remember that there are 12 districts that voted for Trump and are represented by Democrats. If Republicans can win those seats, Democrats will be, at best, only halfway to their goal.

Besides, "many of these 35 crossover districts may be more competitive on paper than in practice given that several have strong incumbents," Kondik writes, "and it's also possible that their Hillary Clinton–Donald Trump vote is not really an accurate gauge of their true partisan lean."

To seize even a one–vote majority in the House, Democrats would need to flip nearly 10% of Republican–held seats. Barring a galvanizing event or issue on the order of the 9–11 attacks or the passage of Obamacare, Democrats, as the out–of–power party, are more likely to benefit from the more typical losses sustained by the party in power — about five to 10 seats, give or take. Democrats may chip away at the deficit in the House, but, at this point, it seems that seizing the majority outright is a mountain too high in 2018.

It would seem — again, on paper — that Democrats' best odds for takeover are in the U.S. Senate, where winning just three seats from the Republicans would give them a majority. It would be a razor–thin one, to be sure, but it would still be a majority, and it would ensure divided government for the second half of Trump's term.

Capturing three Republican–held Senate seats is about 6% of Republican seats in all, which seems like a more manageable task — until you remember that only one–third of Senate seats are on the ballot in a given election. Sometimes special elections are held to fill the unexpired terms of senators who have died or resigned, but it is right around one–third in each election.

Mr. Kondik observes that flipping three Senate seats is "in keeping with the average midterm performance."

The Senate seats that will be up for election in 2018 are, for the most part, the ones that were on the ballot in 2012, when Barack Obama was re–elected and Democrats added to their majority in the Senate. Only nine Senate seats that will be decided in 2018 are in Republican hands.

Thus, winning three of the Republican–held seats on the ballot in 2018 would amount to flipping one–third — and all but one of those states voted for Trump.

Jeff Flake of Arizona: Flake was elected to his first term in the Senate in 2012 after six terms in the House. He only received 49% of the vote against two other opponents but one was a Libertarian who captured 5% of the vote, most of which probably would have gone to Flake had he not been in the race, and that would have just about matched Flake's share of the vote when he was first elected to the House in 2000. His share of the vote in his district never fell below 62% after that.

Arizona has been reliably Republican in nearly all presidential races since 1952, and it hasn't elected a Democrat to the Senate since 1988. It is true that the margins have been closer in recent years, possibly the result of a growing Hispanic population, but the margins still favor Republicans by hundreds of thousands of votes. Democratic efforts probably would be wasted there.

Roger Wicker, Mississippi: Wicker was first elected in 2006 to replace retiring Sen. Trent Lott, then was re–elected in 2012, defeating Albert Gore, a retired minister and distant relative of the former vice president. As chairman of the National Republican Senatorial Committee, he faced a challenge in 2016 that was comparable to the one the chairman of the Democratic Senatorial Campaign Committee will face in 2018. His party had to defend 24 Senate seats while Democrats had to defend only 10. Democrats picked up a couple of seats, but that was a fraction of what they were expected to win on Election Night, which has to be considered a victory for Wicker.

Mississippi has voted for Republican presidential nominees in 10 consecutive elections, and the last Democrat to be elected to the Senate from Mississippi was conservative John Stennis, who won his final Senate term in 1982.

Democrats would be wise to take a few pages from Wicker's 2016 playbook — and not even think about trying to take him down.

Deb Fischer, Nebraska: Fischer took out former Sen. Bob Kerrey in 2012 to win her first statewide election when incumbent Democrat Ben Nelson retired, thus returning the seat to Republican hands. I haven't heard whether she will seek a second term, but I presume she will. She has been described by many as a "true conservative," and her record in the Senate bears that out.

She seems like a good fit for the state she represents. In the last 100 years, Nebraska has voted for only three Democrats for president — Lyndon Johnson in 1964, Franklin D. Roosevelt in 1932 and 1936 and Woodrow Wilson in 1916.

Dean Heller, Nevada: Heller was appointed to succeed disgraced Sen. John Ensign, who resigned amid an ethics scandal in 2011. Heller won a full term on his own in 2012.

Heller could be open to electoral attack. He is conservative but less so on social issues, which seems like a good fit for Nevada. Democrats, however, may sense an opportunity. Nevada has been a bellwether state in presidential politics, voting for the winner in all but two elections since 1912 — and one of those elections was in 2016, when Nevada voted Democratic for the third straight time. That's something that hadn't happened since the FDR years.

What's more Heller has recently come under fire back home, according to the Las Vegas Sun.

Caution, Democrats. Heller won a full six–year term in 2012 while Obama was carrying Nevada.

Bob Corker, Tennessee: Trump won more than 60% of Tennessee's vote. No presidential candidate in more than 40 years received a larger share of the votes that were cast in the Volunteer State.

That should benefit Corker, whose only political experience prior to his election to the Senate was four years as the mayor of Chattanooga. He received 65% when he won his second term in 2012; Mitt Romney carried the state with 59% of the vote that year.

If you think Corker will be tough to unseat ...

Ted Cruz, Texas: ... it will be virtually impossible to unseat Cruz.

There is frequently talk of how unpopular Cruz is with his fellow Republicans in the Senate — but they won't decide whether he gets a second term. Texans will, and Texans like him. His presidential primary victory there last year kept him in the race with Trump — for awhile.

Orrin Hatch, Utah: At 82, the longest–serving Republican senator apparently is considering seeking an eighth term next year, according to the Salt Lake Tribune. If he does, that's bad news for the Democrats. Hatch hasn't been held under 60% of the vote since his first campaign for re–election — in 1982.

John Barrasso, Wyoming: Originally appointed to serve until a special election could be held to fill an unexpired term, Barrasso won that special election with 73% of the vote in 2008, then won a full six–year term in 2012 with 76% of the vote.

That's even better than Ronald Reagan did in Wyoming when he ran for re–election.

So there you have it. Eight Republican–held Senate seats that are up in 2018. Add to that one more — Jeff Sessions' old seat now occupied by Alabama's former attorney general, Republican Luther Strange. Strange was appointed to succeed Sessions until a special election could be held. That special election will be held in June 2018, then the winner (presumably Strange) will be on the ballot again in November.

Not only has Alabama voted Republican for 40 years, it has given Republicans better than 60% of the vote in the last four elections.

Alabama, like most Southern states, once routinely elected Democrats to Congress, but it hasn't elected a Democrat to the Senate since 1992 — and he switched to the Republicans after the GOP seized control of Congress in 1994.

It seems that winning the Senate is every bit as elusive as winning the House for Democrats in 2018. They would be well advised to focus on returning Democrats to the 25 Senate seats they have on the ballot next year (that includes the two independents who typically vote with the Democrats). Not all are in danger, of course, but Democrats do hold 10 seats from states that voted for Trump.

And at least one of those senators, Joe Manchin of West Virginia, has been rumored to be considering switching parties. That might not be a bad idea. West Virginia has voted for Republican presidential nominees in five consecutive elections, and it gave Trump nearly 68% of the vote.

Scratching the Six-Year Itch

---

---

For the last seven or eight years, American voters have seemed to be intent upon turning history on its ear.

They elected and re–elected a black president while taking away his party's advantages in first the House and then the Senate. Aggrieved Democrats have complained that, somehow, the system is rigged against them in midterms. Yet, while these congressional shifts were extreme by historical standards, the pattern has been unmistakable.

The party that is not in possession of the White House almost always does better in midterms than the party in power. Sometimes the Democrats benefit. Sometimes the Republicans benefit. Depends on who holds the White House.

It is Americans' way of preventing the political pendulum from swinging too far in one direction or the other. We like to think of ourselves as fair and balanced, tolerant of all and open to all — whether we really are or not — and we use the ballot to pursue equilibrium. (If we ever actually achieve equilibrium, it is short–lived.)

I have written of this before, and you can find those posts archived elsewhere on this blog.

But a lot of that has addressed congressional politics. It naturally leads to an interesting phenomenon I have observed in presidential politics — but have not written about. Others have, though, to an extent. I think political analyst Charlie Cook wrote something to the effect that, in discussions of presidential politics, whenever the conversation turns to the dynamics of a campaign, the introduction of the phenomenon is "as sure as the sun coming up in the morning."

It is called the "Six–Year Itch," and it holds that voters are inclined to look favorably upon the out–of–power party by the time the current administration has been in place for six years. This really goes beyond the midterm elections, which, as I say, almost never go well for the incumbent party, and has more to do with the popularity of the incumbent during the time of the midterms.

After all, even popular presidents see their parties lose ground in midterms, especially second midterms (which fall in a re–elected president's sixth year in office). About a week before the second midterm of his presidency, Ronald Reagan's approval rating was 63% — but his personal popularity failed to help his Republican Party maintain its grip on its majority in the Senate — a majority it had held since Reagan was first elected in 1980.

Voters, though, treat legislative elections and executive elections differently. Following the '86 midterms, Reagan's popularity took a beating during the Iran–Contra affair, but he bounced back and helped his vice president win the presidency two years later when he was constitutionally prevented from running.

Following Barack Obama's re–election in 2012, the Washington Post sought to shoot holes in the notion of a six–year itch.

"It's overrated," wrote Aaron Blake for the Post. He wrote that column, it is worth noting, less than six weeks after his employer endorsed Obama's re–election bid so you need to consider that as a counterweight to Blake's argument. I was inclined to agree with him, to an extent, when he wrote, "It's not so much that a second midterm isn't trouble for an incumbent president, as much as midterms in general are trouble. And the American public scratches that itch nearly as often in a president's second year as in his sixth year."

That, it seems to me, supports what I wrote about that political pendulum correction. So does the fact that today more Americans than ever do not identify with either party and call themselves independent.

Whether they do so consciously or not, I think most Americans are inclined to give a president — of either party — the two four–year terms in office to which he is constitutionally limited, all things being equal. I guess Americans tend to be reluctant to admit having made a mistake in electing someone a first time. But it depends on what he does with his first four years, and experience tells me that is largely a matter of perception.

If a negative perception takes hold early — if the president suffers a string of setbacks at the start of a presidency — and the perception of misfortune is allowed to harden, it can be almost impossible to overcome. If the president is perceived to have made a mess of things — as Jimmy Carter was — voters look elsewhere for leadership. If a president is perceived to have exceeded expectations, that reservoir of good will makes a landslide re–election likely.

That, I think, is a big reason why some presidents who don't seem to share the same belief system with many of their constituents nevertheless win their votes for second terms.

In those second terms, a president's popularity really is more of a concern to whoever his party nominates to replace him. That is the true coattail effect of which political analysts often speak, and it is the last (if not only) opportunity for a president to have an electoral influence. Coattails are not really factors in House and Senate midterm elections, which are not national and tend to be decided by issues that matter only within the boundaries of states and congressional districts, but they can be factors in national campaigns.

But there is a catch.

Historically, the United States has not been likely to elect candidates who are nominated by an incumbent president's party to succeed that president. Well, I guess that should be narrowed down to the post–World War II period. Prior to the war, the United States was hardly hesitant to stick with the same party in more than two consecutive presidential elections.

It elected Franklin D. Roosevelt four times, then elected the man who succeeded him following his death and just prior to the end of the war, Harry Truman, to a full four–year term of his own. Just prior to FDR's time, Republicans won three straight elections. In fact, Woodrow Wilson's two terms in office in the early 20th century and Grover Cleveland's two nonconsecutive terms in the late 19th century were the only interruptions in a period when Republicans won 14 of 18 national elections.

But since World War II and Truman's decision not to seek another term in 1952, Americans have only elected the same party three straight times once. That was in the 1980s, when Reagan won twice and then his vice president, George H.W. Bush, was elected to succeed him.

Reagan, as I pointed out, enjoyed solid approval ratings just before his party sustained significant Senate losses in 1986 — but that was on the legislative side. His personal popularity benefited Bush in the 1988 election.

Not that Bush's opponent, Michael Dukakis, didn't seem to do everything in his power to sabotage his own campaign.

And that, I think, underscores an important point about the six–year itch. It is susceptible to the dynamics that are unique to each campaign.

The popularity of the incumbent president seems to have a lot to do with the outcome, but that is no guarantee. Dwight Eisenhower enjoyed approval ratings that exceeded 50% for much of his presidency, but his vice president lost narrowly in his first bid for the presidency.

That leads me to another observation: It is also important for the president's would–be successor to take advantage of the resource of a popular incumbent. To my knowledge, Richard Nixon never distanced himself from Eisenhower, but the Republican ticket was hurt by the recession the country experienced in 1960.

Al Gore didn't embrace the popular Bill Clinton in 2000, and that was a decision that apparently cost him the presidency. Clinton's approval rating just before the 1998 midterms was over 60%, but Gore, while winning the 2000 popular vote, lost the Electoral College.

Since the advent of the polling era, few presidents have been popular with a majority of voters at the ends of their presidencies, and their would–be successors suffered for it. In 1966, after six years of the Kennedy–Johnson presidency, Democrat Lyndon Johnson had an approval rating of about 43% — roughly the share of the vote his vice president, Hubert Humphrey, received on Election Day two years later.

1974, after nearly two full terms of the Nixon–Ford presidency, Republican Gerald Ford's approval rating was around 47%, and he narrowly lost the election to Jimmy Carter in 1976.

History says the voters will have an itch to scratch next year, and Obama, like Ford, hovers below the 50% mark. Ford, of course, had the advantages of incumbency in the election year of 1976, and Obama will not be allowed to seek a third term, which suggests that 2016 will be an uphill climb for the Democrats' nominee.

It looks like it will be the Republican's race to lose.

The Fine Art of Compromise ... and Lost Opportunity

---

---

"The trusts and combinations — the communism of pelf — whose machinations have prevented us from reaching the success we deserve should not be forgotten nor forgiven."

Letter from Grover Cleveland to Rep. Thomas C. Catchings (D–Miss.)
August 27, 1894

---

I have mentioned here that I have been studying the presidency most of my life.

And Grover Cleveland has always fascinated me. He always stood out because he was — and still is — the only president to serve two nonconsecutive terms. (He was also president half a century before presidents were limited to two terms — so, presumably, he could have sought a third term in 1896, but his party repudiated him. More on that in a minute.)

I have found it fascinating, too, to observe all the different presidents in American history to whom Barack Obama has been compared.

That didn't really begin with Obama. Incoming presidents are almost always compared to presidents from the past. I don't know why. Maybe to try to get an idea of what to expect. There have been no other black presidents so Obama couldn't be compared to anyone on a racial level.

When he was about to take the oath of office for the first time, Obama was compared, at different times and for different reasons, to great presidents from American history like Abraham Lincoln and Franklin D. Roosevelt.

Lincoln, of course, was a natural, having presided over the Civil War and issued the Emancipation Proclamation. There were some comparisons, as well, to Franklin D. Roosevelt, mostly because FDR had taken office during the most perilous economic period in the nation's history, even to John F. Kennedy, perhaps because both were young and their elections made history.

Over the course of his presidency, Obama has been compared to less accomplished presidents. In recent years, it has frequently been asked if he is more incompetent than Jimmy Carter, who is generally regarded as the most incompetent president in recent memory.

Six years ago, about three weeks before Obama took the oath of office the first time, political scientist Michael Barone suggested that Dwight Eisenhower might be the more appropriate comparison, and I wrote about that.

Barone's point was that Eisenhower had done little to help his fellow Republicans, many of whom "grumbled that Ike ... was selfish.

"Eisenhower, I suspect, regarded himself as a unique national figure," Barone wrote, "and believed that maximizing his popularity far beyond his party's was in the national interest."

I was reminded of that tonight when I heard Obama's speech on immigration. Many congressional Democrats are supporting the president — publicly, at least — but some are not. Regardless of the negative ramifications of his executive order — and a poll conducted Wednesday night indicates that nearly half of respondents oppose Obama's acting via executive order — Obama seems determined to prove that he is still relevant.

Coming a mere two weeks after Democrats lost control of the U.S. Senate in the midterm elections, it seems to me a president who was more concerned about his party's future than his own would act more prudently. Bill Clinton, after all, lost control of both chambers of Congress in the midterms of 1994, and Democrats didn't regain the majority in either chamber for 12 years.

Clinton did manage to retake some his party's lost ground when he ran for re–election in 1996 and then again after surviving an attempt by the Republicans to impeach him before the 1998 midterms, defying all logic.

I've always felt that a lot of that was because Clinton was appropriately chastened by his party's massive losses in the midterms. I felt, at the time, that many of the voters who had voted Republican in 1994 believed Clinton had learned an important lesson and were more open to supporting him and the members of his party in 1996.

Obama has now been through two disastrous midterm elections, and he has emerged from the second not chastened but defiant. He appears to be entirely ready to do everything on his own, completely ignoring the role that the Founding Fathers intended for Congress to play. An opportunity to let compromise and cooperation be what the Founding Fathers envisioned in their fledgling republic is being squandered.

Once such an opportunity is lost, once such a president takes this kind of approach, it is hard, if not impossible, to establish a rapport with the other side.

Obama isn't the first to do this, which brings me back to Grover Cleveland. A little background information is called for here.

Cleveland was first elected president in 1884. He was the first Democrat elected to the office in more than a quarter of a century — in spite of the revelation that Cleveland had fathered a child out of wedlock. It was close, but Cleveland managed to pull it off.

Four years later, when Cleveland sought a second term, conditions were good. The nation was at peace, and the economy was doing pretty well, but there was division over the issue of tariff policy. The election was another cliffhanger. Cleveland again won the popular vote by a narrow margin, but his opponent, Benjamin Harrison, received enough electoral votes to win.

So Cleveland left the White House in March 1889, but he returned as the Democratic nominee in 1892 and defeated Harrison. It was the second time a major party nominated someone for president three straight times. The first one, Andrew Jackson, also won the popular vote all three times; like Cleveland, though, he was denied the presidency once because he lost the electoral vote.

Perhaps it was the experience of having been returned to the White House after losing the electoral vote four years earlier that contributed to Cleveland's messianic complex. To be fair, it would be hard not to feel that there was an element of historical inevitability at work.

But that doesn't really excuse how Cleveland approached the outcome of the 1894 midterms.

One cannot tell the story of the 1894 midterms without telling the story of the Panic of 1893 for it defined Cleveland's second term as well as the midterms. It was the worst economic depression the United States had experienced up to that time. Unemployment in America was about 3% when Cleveland was elected in 1892. After a series of bank failures, it ballooned into double figures in 1893 and stayed there for the remainder of Cleveland's term.

The depression was a key factor in the debate over bimetallism in 1894. Cleveland and his wing of the Democratic Party were known as "bourbon Democrats," supporters of a kind of laissez–faire capitalism. They supported the gold standard and opposed bimetallism, in which both gold and silver are legal tender.

The economy was already the main topic of the campaign, and a major coal strike in the spring didn't help. In fact, it hammered the fragile economies of the states in the Midwest and the Northeast. Republicans blamed Democrats for the poor economy, and the argument found a receptive audience.

Republicans gained House seats just about everywhere except the Southern states, which remained solidly Democratic, and states where Republicans already held all the House seats. Democrats went from a 220–106 advantage to a 104–226 deficit. It remains the most massive shift in House party division in U.S. history.

Under circumstances such as these, a president has two choices — he can be conciliatory and try to move to the political center, as Bill Clinton, George W. Bush and Ronald Reagan did, or he can dig in his heels and be even more intransigent.

Much as Obama is doing 120 years later, Cleveland chose the latter approach after the midterms in 1894. Perhaps he felt he had no allies in Washington anymore, but I've always felt his go–it–alone approach was a big reason why he was repudiated by the Democrats in 1896. The fragmented party chose instead to go with William Jennings Bryan, who would be nominated three times and lose each time. In fact, with the exception of the Woodrow Wilson presidency, no Democrat would win the White House for the next 36 years.

For that matter, they didn't regain the majority in the House until the 1910 midterms, but they lost that majority six years later in spite of the fact that President Wilson was at the top of the ballot. It took the stock market crash of 1929 to restore Democrats to majority status in the House in the midterms of 1930.

That is one cautionary tale that emerges from this year's midterms. Another is the exaggerated importance given to the turnout. I know it is a popular excuse to use after a party has been slammed in the midterms, but it is misleading.

In 2006, when Democrats retook the majority in both chambers for the first time in 12 years, they treated it as a mandate for change. But roughly the same number of voters participated in 2006 as participated in 2014. Granted, there has been an increase in the overall population in those eight years so the share of registered voters who participated is different, but the overall numbers are the same.

Republicans, too, pointed to low turnout in 2006. My advice to them would be not to duplicate the Democrats' mistake. They believed their success was permanent — and it never is in politics.

It can last longer, though, if you lead.