Taiwan’s Presidential Election Trends, according to the Prediction Future Markets

I’m not sure if Taiwanese opinion polls benefit experienced Taiwanese readers – to someone like me, who is no close observer, most of them seem to be of little use.

Look at those included in the chart underneath, for example. They are all pan-blue sources: the China Times, United Daily News, and Apple Daily. The Apple Daily may not be very China-friendly, but according to Wikipedia Chinese (as of today), they are said to have close connections to president Ma Ying-jeou‘s staff, and Apple Daily’s numbers of December 3 would seem to confirm that.

Opinion Polls, all Pan-Blue Sources, Nov. 16 to Dec. 7, 2011

Opinion Polls, all Pan-Blue Sources, Nov. 16 to Dec. 7, 2011

Yes – 47.04 per cent of the interviewees would vote for Ma Ying-jeou, 36.9 per cent for Tsai Ing-wen, and 12.11 per cent for James Soong Chu-yu, according to Apple Daily, which published a poll completed on December 3 (see Ma’s best number ever in the chart above – if Wikipedia quoted them correctly).

For the numbers included in the chart, and beyond (back to July 1, and including sources other than pan-blue), see “Three-Way Race, Wikipedia”.

Polls seem to be much more part of the “spin-doctoring” in Taiwan, than they are in most European or North American countries. 47.04 per cent for Ma Ying-jeou – not even a distant watcher can take that forecast serious.

Evidence so far suggests that prediction markets are at least as accurate as other institutions predicting the same events with a similar pool of participants, Wikipedia (English) of today suggests.

So there might be alternative sources. And there are prediction markets forecasts for Taiwan’s presidential elections in January, too. Here, the picture is very different: there, Tsai has been more likely to win the elections of the time, ever since October 26, 2011 – the  market forecast is run by the National Chengchi University (國立政治大學). What looks particularly plausible there is that Ma’s chances fell, just as Soong Chu-yu’s were rising. This would seem plausible because Soong’s People-First Party is “bluer” and closer to China than Ma’s KMT, and unlikely to draw support from any other party than the KMT.

The Economist, not necessarily a fan of Tsai Ing-wen, quoted the Chengchi University numbers, too, on November 19:

A prediction market run by National Chengchi University, accurate in the past, says the probability of his winning the election dived from over 59% on October 16th to under 42% on November 14th; Ms Tsai stands at 49%. Opinion polls in the island’s media, which usually leans towards the KMT, also show slumping popularity, though Mr Ma still leads by a few percentage points.

[Update, Dec 13: XFuture and the National Chengchi University prediction markets are basically identical, according to Echo Taiwan]
Echo Taiwan has also turned to a future market for clues (there are links within his paragraph – see there:

Xfuture, the future market website, claimed to be more accurate than most opinion surveys conducted by media in Taiwan, is conducting surveys in the form of stock exchanges for the upcoming legislative and presidential elections. There are 3 contract groups for the president election. I am sharing the timeline of one of them, The Estimate of Vote Percentage (2012總統選舉投票率預測), for all three candidates: Tsai Ing-wen (蔡英文, DPP), Ma Ying-jeou (馬英九, KMT) and James Soong (宋楚瑜, PFP). The data covers the period from 9/1/2011 to 12/2/2011 for both Tsai and Ma. Soong was not included into this contract group until 11/4/11. At the time of preparing this post (12/2), the data show a profile of Tsai : Ma : Soong = 51.0% : 37.1% : 12.5%.

Here, too, Ma’s rate is falling, as Soong’s is rising.

Echo Taiwan’s post also contains guesses and clues as to which events of the past two months may have led to the shift in Tsai’s favor, plus links to further posts by other bloggers. Just as there, the Economist’s Nov 19 article attributes some cause for Ma’s troubles to events prior to Soong entering the race:

But Mr Ma’s popularity was falling even before Mr Soong’s formal candidacy. He dropped a bombshell on October 17th by saying that he favours signing a peace treaty with China within the next decade, provided the public and parliament supported it. It was the first time that Mr Ma had given a timetable for negotiating such a hugely sensitive issue, and it has whipped up alarm in the media and among a China-wary public. The DPP accuses Mr Ma of steering the island towards unification. Mr Ma later backtracked, suggesting, among other things, that a treaty would need a referendum.

Only to backtrack once again, shortly after that. Ma had apparently become dizzy.

Early in October, political commenter Wong Chong Xia warned the KMT that

Ma Ying-jeou’s support rate never exceeds a ten-percent lead over Tsai Ing-wen, and the pan-green camp’s voting rate has always been stronger than the pan-blue camp’s, and past experience shows that when it is a one-on-one race, and the pan-blue camp’s lead isn’t better than ten per cent, it is the loser when the ballots are counted on election night.

I don’t know if the future markets include reflections of the phenomenon observed by Wong, and I can’t tell Wong’s observation itself is correct – but at the moment, Tsai looks like the more likely winner of next year’s presidential elections.



» Closing in on the Presidency, Nov 25, 2011



» Election Campaign Coverage in China, Taipei Times, Dec 12, 2011
» Presidential Debate, PTS TV / Youtube, Dec 2, 2011



One Comment to “Taiwan’s Presidential Election Trends, according to the Prediction Future Markets”

  1. The prediction market today has Tsai winning by nine points over Ma. That feels too high to me. Shares of Tsai are selling for $20 more than those of Ma in predicting who will win. These gains are steady, as you note, Tsai has been slowly inching up for months, steadily eroding Ma’s lead. She’s a tough opponent for Ma, since she was never involved in the rough and tumble of local politics like many other DPP heavyweights and hasn’t made all sorts of compromising political deals with local factions. The KMT’s negative campaign machine thus doesn’t have a whole lot to grab onto. Couple that with the inexplicably awful campaign the Ma camp is running…..

    I blogged briefly on a round up of polls from Now News:


    Apple Daily’s polls appear to be outliers, but I do not know why. I don’t think it is because some individuals on staff are close to the Ma Administration and they want to pimp the KMT. I think there is something grossly wrong with their sampling methodology since they are always an outlier.

    One thing I’ve found fascinating so far is how little gender matters, at least on the surface.



Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out /  Change )

Google+ photo

You are commenting using your Google+ account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )


Connecting to %s

This site uses Akismet to reduce spam. Learn how your comment data is processed.

%d bloggers like this: