clock menu more-arrow no yes

The unexpectedly close Kansas House race shows that targeting is overrated

Overconfident “experts” are worse than wasting money on long shots.

Democratic Party leaders, stung by criticism for having done so little to help their candidate in an unexpectedly narrow House special election in Kansas, have swiftly converged on an explanation for their actions. They’re now arguing it wasn’t a mistake — they didn’t just underrate how much Kansas Gov. Sam Brownback’s unpopularity would matter, or overrate how good state Treasurer Sam Estes, the Republican who ultimately won, was as a candidate. No, they argue, staying largely out of the race was a strategy.

“Sometimes the best thing for a candidate is to allow them to get press and support in their own district,” writes Christina Reynolds, a former top communications aide to Hillary Clinton in a Medium post endorsed by the current communications directors for both the Democratic Congressional and Senatorial campaign committees, “without drawing a lot of national press attention to it, and make no mistake  —  the DCCC playing in a race draws national press attention.”

This is true in the sense that it probably would have been counterproductive to airlift Nancy Pelosi into Wichita to speak at a rally or even to have Joe Biden start vocally tweeting about Kansas. But the idea that national Democrats couldn’t do a single thing to help a long-shot candidate in a red district — advance him a fundraising email list, cut him a modest check, or route a little money his way through a quiet vehicle — is borderline absurd. The simple truth is the DCCC didn’t get involved in the race for the same reason that virtually nobody was paying attention to it — they didn’t think the race would be competitive.

That error in judgment is no great sin on their part. The fundamentals of the Wichita-based district are awful for Democrats, and the Republicans had a strong candidate on paper. The deeper question is whether the DCCC should do more to acknowledge that weird, unpredictable things happen in politics, and that forecasts from Washington-based experts are often wrong (mine included).

The DCCC’s business model is based on targeting

The current approach to party investment basically involves narrowcasting. A small number of vulnerable incumbent Democrats are deemed “frontline” members whom the DCCC and party leaders are committed to protecting. Then a handful of Republican districts are deemed pickup targets, where the party believes Democrats could win based on past election results. The party then focuses heavily on recruiting its idea of “good candidates” in those districts and setting them up with approved consultants and campaign staff.

If you do those things and demonstrate strong personal fundraising potential, you will be blessed with help. If you don't, you won't.

This is a largely sensible approach, but it’s based on the premise that the top political operatives have very good judgment about which races are winnable, which candidates are strong, and which consulting and campaign teams are effective. This premise is critically important because no matter how strong the abstract case for targeting is, it only really makes sense to narrowcast if you can target effectively.

Overconfidence can be a huge problem

In many areas of human endeavor, it pays to admit that you are uncertain. Looking back over my web traffic statistics, I could probably cut out 50 percent of my posts with little impact on total readership. But to do that, I would need to be able to reliably predict in advance which articles will be hits and which will be duds. Yet despite years of experience in the content game, I can’t pull this off. My best method for reliably generating hits requires me to also churn out plenty of duds.

This is true in many fields. Economists Cade Massey and Richard Thaler have found overconfidence to be a perilous factor in the professional football draft.

According to their research, teams systematically overpay for first-round draft picks. It’s not that they’re mistaken for believing players drafted in the first round are, on average, better than players drafted in the third or fourth round. But they are not as much better as often as they would have to be to justify the premium they command.

The problem is overconfidence. Drafting well requires skill in identifying promising players, but it is very hard to do this reliably. Teams would do better to recognize that their ability to identify the best players is limited, and spread their bets more widely across a larger number of lesser-rated prospects.

The lesson of the KS-04 race isn't that the DCCC should have clairvoyantly recognized a close race in this overwhelmingly Republican district. It's that political insiders aren’t clairvoyant and ought to develop a strategy that takes into account the fact that weird things tend to happen in politics. Nobody — probably including the candidates themselves — thought Donald Trump would win the GOP nomination, or that Bernie Sanders would be a fundraising juggernaut.

Targeting creates benefits for insiders

Targeting also has ancillary benefits for its proponents that may not be in the party’s best interests. It maximizes the power and influence of Washington-based staff, since they are the ones who decide what the target districts are and who the good recruits are.

It also ensures that the lion’s share of resources flow through the hands of the blessed inner circle of consultants and campaign managers (Amy Sullivan’s classic 2005 article on the consultant class remains vital reading). The presumption that fundraising “has to” primarily come from a quasi-fixed roster of K Street players also serves the personal interests of many Hill staffers, who can take advantage of a boss’s perceived need to have a good relationship with business groups to build their own good relationships — and an eventual exit ramp to a more lucrative job.

The risks of a new approach are large. In particular, party leaders worry about burnout. They worry that the same grassroots who this morning are frustrated that the party didn't invest in a 5-point loss in Kansas would be even more frustrated today if a massive effort had resulted in a 2-point loss. That asking the same grassroots brigade to trudge toward what's still a long-shot race in Georgia would be counterproductive. Better, the thinking goes, to temper expectations, count overperforming in Kansas as a win, and look ahead to the relatively winnable race in Georgia.

Party people also worry, sensibly, that a strong message for national fundraising may not be a good fit for actual districts. The national outpouring of grassroots enthusiasm for Wendy Davis’s support of abortion rights is a cautionary tale here. There are some very real trends making Texas more Democratic, but nobody (including Davis’s campaign) really thought abortion was the best issue — as opposed to Medicaid expansion, say, or school funding — for Texas Democrats to highlight.

The specter of a bunch of amateur-hour pundits and online organizers ginning up enthusiasm for a handful of lovable long shots and firebrands with weak teams and poor district fit, only to walk away when the whole thing crashes and burns, makes party insiders nervous with good reason.

All that said, if your key asset is expertise at identifying winning candidates and helping them win elections, then a record of losing elections becomes a problem. Having lost power at all levels of politics, the strategic acumen of the Democratic Party’s leadership is bound to come into question.

Failure to identify a potential victory in Kansas is hardly the greatest sin in the world, but it reenforces the basic reality that targeting decisions are made by people who are more fallible than they would probably like to admit. Spreading resources thinly would, necessarily, entail a certain amount of waste. But it would also put less pressure on expert judgment to be infallible and give more due to the fickle nature of reality.