Of all the offices up for grabs in Taiwan’s local elections, the Taipei mayor’s race has attracted by far the most attention. The colorful and increasingly acrimonious campaign pits Ko Wen-je (柯文哲), an independent running as the de facto candidate of the green camp, against Sean Lien (連勝文), the scion of a wealthy and powerful old guard KMT family. Unusually for Taiwanese politics, neither candidate for this high-profile office has held elected office before, and so the dominant media narrative has tended to emphasize the repeated missteps and at times unconventional political strategy of both campaign teams.

As polls continue to show the KMT’s Lien trailing well behind Ko and raise the spectre of a KMT loss in Taipei for the first time in a generation, there is an equally fierce competition to shape the narrative about what a Ko victory might mean. Recent commentary has ranged from broad claims about the Taiwanese electorate’s views on cross-Strait relations, to its support for Taiwanese or Chinese nationalism, to the shifting partisan balance and each camp’s prospects in the 2016 presidential election. What is odd about this intense focus on Taipei is that the Ko-Lien race is highly idiosyncratic, and about the last place we should look for insight into these questions.

First, like all the positions at stake in this election, Taipei mayor is a local, not national, office. Neither candidate’s views on cross-Strait relations are related to their potential competence as mayor, nor will they be in a position to directly influence Taiwan’s foreign diplomacy or trade policy. Instead, much more prosaic concerns are at stake: does the garbage get picked up? Will the city improve its management of traffic and flood control? Does either candidate have a realistic plan to tackle the soaring cost of housing? And so forth. To conclude that voters are sending a message about these larger national issues is to assume that the same issues are driving voting patterns across the entire island—and while that is probably true in a presidential election, it’s much harder to justify in local elections. If we really want to know the Taiwanese public’s view on major national issues, public opinion surveys, not local election returns, are the place to look.

Second, it is worth remembering that one of the two major candidates facing off for Taipei mayor is an independent. It would be one thing if Ko Wen-je were effectively a DPP candidate in disguise, but if recent polls are to be believed, he is something else to many “light-blue” voters. Ko has campaigned with PFP candidates for the city council, and his campaign manager is a mainlander and former New Party member. Thus, we should be wary of attempts to infer the generic partisan strength of the DPP and KMT in Taipei from election returns featuring an official nominee from only one of those parties.

Third, and most importantly, Taipei is just one race of 22 for county and city executives around Taiwan. Why focus on this one rather than examine the results as a whole? If we want to try to extrapolate from a single race to all of Taiwan, the most representative election is probably in Taichung, where both parties have poured significant resources into the race, the candidates are well-known nationally and ran against each other four years ago, and the partisan composition of the electorate in 2012 was nearly evenly split—certainly more than Taipei’s. Much better, though, is to take the whole set in, and avoid developing our interpretations until we actually have the results in hand.

So, in addition to who wins and loses, here are four other more fundamental things to watch for on election night:

Which party wins the aggregate vote?

The KMT comes into this election with a significant advantage in the number of localities it runs (15 vs. six for the DPP, with one independent.)  This advantage, though, is deceptive. If we take the total votes, not offices, won by each party in the last local elections—2010 for the five special municipalities, and 2009 for the rest—the DPP actually polled ahead of the KMT (5.76 to 5.46 million votes). But the DPP’s support was highly concentrated in the south, especially in Tainan and Kaohsiung, where the party’s nominees racked up huge margins of victory at the same time that they lost closer races in Taichung, New Taipei City, and elsewhere.

A key question about this election is whether unhappiness with President Ma Ying-jeou and the tumult over the Trade in Services Agreement will produce a large enough swing toward the DPP in the electorate to flip some of these local governments in central and northern Taiwan into the green camp. Measuring the size of this swing is complicated by the fact that local executive races turn on the qualities of the candidates as much as their parties—and the personal appeal of candidates varies tremendously in this election cycle. So, rather than looking at the total number of races won and lost, we will learn more by watching each party’s aggregate vote count. The more the vote totals diverge from the previous elections (5.76 million for the DPP, 5.46 million for the KMT), the stronger will be the case that there’s a meaningful shift in the partisan landscape for 2016.

Does the party that wins the executive also win the council?

The DPP has traditionally fared much better in winner-take-all executive races than in multi-member council elections, where the KMT has usually retained more seats. One of the reasons for this divergence in the past was that elections were held at separate times, and the KMT’s local factions could have a bigger effect on the outcome in the lower-profile, lower-turnout local contests. But no longer: in this election, voters can weigh in on council, township, and village or ward races at the same time that they cast a vote for mayor or county executive.

It will be interesting to see whether the party that wins the top office also gains a majority in most of the councils: if it does, that suggests deeper consolidation of Taiwan’s two-camp partisan divide down to lower levels of government. An alternative is that we see no single-party majority in many of the councils. Several minor parties are attempting to convert the interest and engagement of social movement participants into actual seats in local councils, and how they fare will tell us a lot about their prospects at the national level in 2016.

How high is turnout?

Turnout in Taiwanese elections has varied dramatically over the last decade, from a high of over 80 percent in the 2004 presidential election to a low of 23 percent in the 2005 National Assembly election. The share of eligible voters who actually show up to the polls on Saturday can decisively affect the outcome—and it’s not clear from precedent which way turnout will go. On the one hand, the controversy over the Trade in Services Agreement and the Sunflower student movement could motivate some people to vote who otherwise wouldn’t. In addition, the current local elections combine races at three different levels: county and city, township and town, and village and ward. The county and township races have in the past seen much lower turnout: in December 2001, for instance, turnout in the local executive elections was about 66 percent, while in the election for council races a month later it was only 55 percent. We might expect the turnout on Saturday to be at the higher end of this range, since most voters who take the time to come out primarily for the executive race will cast ballots in the others, too, and since the turnout operations of candidates at all these levels will be working at the same time.

On the other hand, these are still local elections, and turnout there has rarely if ever reached the heights it did in the presidential elections in 2000 and 2004, at over 80 percent of eligible voters. The extraordinarily high numbers of voters in these two elections occurred in close, high-stakes races with highly-motivated partisans on both sides. By contrast, this election has signs of a significant “enthusiasm gap,” with core pan-Green supporters likely more motivated to vote than pan-Blue ones. As a consequence, lower overall turnout will probably benefit the DPP’s candidates more than the KMT’s, as pan-Blue-leaning voters are disproportionately likely to sit the elections out. And if the KMT’s candidates outperform expectations, it will probably be because turnout is toward the high end of the historic range.

Election administration

This election is actually nine elections in one: special municipality (1) mayors and (2) councilors, county and city (3) executives and (4) councilors, township and town (5) heads and (6) representatives, village and ward (7) chiefs, and elections for (8) district heads and (9) representatives in aborigine districts in the special municipalities. This is the first time offices at so many different levels have been elected at the same time. Individual voters are going to cast as many as five separate ballots: for county or city executive and councilor, township head, township representative, and village or ward chief. That’s a lot by Taiwanese standards, and it will be a testament to the quality of the electoral commission and the dedication and professionalism of tens of thousands of poll workers if all the results are tallied quickly and efficiently, as they usually are. Taiwan’s election administration is in many ways a model; let’s hope it meets the challenge this time around.

So, as the results roll in on Saturday, the Taipei race will probably capture the headlines. But it’s not going to tell you much about how the Taiwanese electorate feels about cross-Strait relations, competing nationalisms, the Sunflower Movement, or the 2016 elections—nor will any other single race, no matter how the media narrative develops.

(Feature photo of the town of Puli, Nantou County, by Fcuk1203 on Wikicommons, CC BY-SA 3.0)