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%.
The data is presented in two graphs: first the data only, allowing readers to observe the trends without the interference from my interpretation. The second one includes (1) events that might have marked the major critical changes of winning trend at specific times; (2) a projection of the trend to the election date (1/14/2012), by extrapolating the data of the last 10 days or so, to show the possible outcome if it keeps going with the current trend.
 |
| Fig.1. The estimate of vote percentage for the 2012 presidential election. Blue: Tsai Ing-wen (DPP); Red: Ma Ying-jeou (KMT); Brown: James Soong (PFP). Source: Xfuture |
 |
| Fig.2. Critical Events and Projection. A. around 5% of Ma's supporters changed to Tsai in October, followed by a seemingly non-stop fall on Ma's side. B. Soong was introduced into the competition, which comes with a second wave of Ma's rapid fall; C. The war of persimmons, doubling up the speed of Ma's fall. See details in text. Also, the data in Fig.1. is extrapolated to the election date (1/14/2012) based on the current trend.It shows that Ma could fall behind Soong 10 days before the election, reaching a final percentage profile of Tsai:Soong:Ma = 50%:30%:20% --- if the current trend remains unchanged. Note: right after point C, there's a deep fall of Ma's rate. It was categorized by the Xfuture as intentional manipulation and was corrected back to the normal trend. |
The data in this contract group shows that, before October, Tsai and Ma both gained ground with small but steady pace, with Ma leading by about ~10%.
In the mid to end of October, however, significant amount, about 5%, of Ma's supporters switched to Tsai in a very short time (Point A in Fig.2).
What happened in October that might have contributed to this swift change ?
On Tsai's side, she finished campaign tours all over Taiwan with overwhelming support, then heated up the campaign with the Piggie Bank Trio movement, signifying a common people's war against the KMT authority. Both could be seen as an effective way to consolidate green base and potentially brings a positive face to the election.
On Ma's side, he failed to respond to farmers' wish for the Old Farmer Subsidy (OFS). It is a tradition in Taiwan to raise the monthly OFS by NT$ 1,000 in presidential election, a tradition the DPP follows. But, Ma insists that farmers deserve only 316 dollars/month of increase this time.
It went wrong seriously in the beginning. In one of stories reported by a KMT local legislative candidate (note that the legislators will be voted on the same date as the president), when they went to the home of a long-time pro-KMT farmer to deliver the 316.00, the farmer just threw it back at them. Many others told them not to come at all.
Urgent calls like this from local KMT campaign workers were rushed to the headquarter in Taipei, repeatedly, only to meet repeating announcements from Ma's totally-out-of-touch campaign team that 316 is fair and square. It won't help at all when news about Ma's gov spent huge money in unrestrained way in other directions (see below), and raise subsidy to other groups of people who already enjoy much more monetary benefit than farmers did (see here, here and here).
Based on several sources from the KMT side, this arrogant blunder costs Ma somewhere around 3~10% votes. With each 1% equates to about 150,000 votes, Ma lost 450k ~ 1500k votes with this single blunder alone.
But Ma didn't stop there. In the press conference when he formally announced his "Golden 10 year" plan (黃金十年), supposedly to be a campaign-boosting ingredient, he rushed to include his plan of a peace agreement with the the communist China (see Frozen Garlic's comment for why Ma shouldn't have done so). It was a plan that he furiously denied 2 months ago when his campaign manager King Pu-tsung (金溥聰) -- allegedly the real powerful man-behind-the-scene who makes decisions for Ma to follow -- announced the peace talk in a trip to the USA. Besides, it was originally not in Ma's golden 10 years plan -- at least not to anybody's knowledge. Nobody in the entire Blue camp seemed to have any prior knowledge of it -- the KMT heavy heads, the KMT campaigners, the KMT legislators ... until Ma suddenly announced it in that press conference.
Stunned in an unimaginable manner, people from all camps threw in heavy criticism. In response, Ma's office delivered another rushed announcement in the midnight the same day --- that the peace agreement WILL be pre-approved by a referendum. But, in less than 12 hours, Ma followed it with another rushed flip-flop by saying that the referendum is just one of the means to get public opinions on the peace agreement.
Tsai responded with an approach like, "fine, lets get on with the referendum law, then."(my wording) The next Monday, the DPP law makers sent in the proposal to the Legislative Yuan. The KMT rejected it.
Ma's plan of peace agreement was thus finished and moved to the trash in less than 10 days -- along with his supposedly boosting Golden 10 Years Plan. What is left is the deep scars on Ma's image.
All the above events exploded in October, corresponding to the onset of rapid falling seen in the graph presented here. Note that the trend of falling continued, albeit in a slower rate, even before Soong was included in the contract (Point B), which sped up the gap increase between Tsai and Ma.
It was followed by even more catastrophic events (not marked on the graph) on Ma's side along the track, including the humongous spending on a musical "Dreamers" (夢想家) that was staged only twice. The contract was given to a small group of artists who are allegedly close to Ma's wife, and is the same group of people who also took several multi-million art contracts from the government after Ma took power. A large portion of government's budget for artists flow into the hands of the same group of people like fountain water. The money given to Dreamers alone equates to 50% of the yearly funding for artists.
Ma said he ""felt sorry" for how the gov money was spent. Ma's campaign companion, Wu Den-yih (吳敦義), said that he did not know the show had such a high budget and that he regretted the money was spent in that way. However, a pro-KMT commentator revealed that Ma and Wu were actually who approved the spending in a meeting. The guy who "resigned to stop the bleeding", Council for Cultural Affairs Minister Emile Sheng (盛治仁), is but a scapegoat.
With other bad (or good) news like the revelation of Ma's link to a bookie in suspicion of manipulating the election outcome, we went on to the "War of Persimmons" , by which Pro-KMT polls all said that Ma regained the lead after this by as much as 8~10%. But the data shown here (point C) indicated that Ma actually suffered even more.
In review, Ma's fall could be roughly dissected into 3 stages:
1. In October, Tsai gained ground with campaign tours, Piggie Bank Trio, etc, and Ma lost ground with OFS, the peace talk, etc --- contributed by: Tsai, Ma
2. In mid Nov, Ma went down with Soong's entry as well as Ma's Dreamers Musical and his links to bookie --- contributed by: Soong, Ma
3. In end of Nov, Ma went further down after the persimmon price debate --- contributed by: Ma
In summary, the falling of Ma is due to 3 factors: Tsai, Soong, and Ma. Among them, Ma contributed to his own fall in much larger magnitude. That is, Ma is the one who puts his own re-election in jeopardy.
This is something we have to bear in mind, 'cos, if Ma loses, many pro-KMT media, as well as many foreign media that is really foreign, are going to blame Soong for Ma's loss, while in fact Ma's his own worst nightmare.
The extrapolation of the trend presents a possible scenario that Ma's rating could fall behind Soong's in around 10 days before the election, reaching a final result of Tsai:Ma:Soong = 50%:20%:30%. Certainly, if only the trend is kept, which may not be likely.
No matter what, 40 days before the election, there's just no sign that Ma could ever recover from his downfall trend. Seeing how many screwup moves contributed by Ma and his people, Ma might be better off not campaigning at all.
More ref:
See XFuture's own report on the same data set:
未來事件交易所目前對2012年總統選舉預測-蔡英文擴大領先馬英九8%、宋楚瑜上升到12%_20111202 http://nccupm.wordpress.com/2011/12/02/1429/
13 comments:
I think you go too far in linearly extrapolating those trends into the future - but your last point looks very true! If President Ma hadn't campaigned at all, just told the nation he was too busy being president to run a campaign, he would probably be in a better position.
But how did the persimmon story actually hurt Ma? I realize certain newspapers are dreaming when they report that it helped him, but I have a hard time believing that the issue would actually reduce Ma's support.
Yea, I agree it's a bit too far :) As I mentioned, it's probably unlikely the trend would stay unchanged.
On the other hand, from info I collected, aside from those pro-KMT media polls, it seems to be a reasonable estimate that Ma's final rate would converge to around 35.
I am not sure how the persimmon story hurt Ma. There could be other factors combined. But if it does, I would say that the KMT launched all out attack on a seemingly unintentional misplace of persimmon picture on the calendar published by the DPP, twisting it into that the DPP caused the persimmon price to go down and pressure Tsai to apologize. Some go so far by describing Tsai as a evil person. That, is not a good campaign strategy.
This is a great overview.
Assuming trends lines will continue is usually wrong. If Ma has been losing support recently, he would have his least committed supporters. However, as he loses more voters, a higher and higher percentage of those remaining are hard core supporters. They are highly unlikely to defect.
If Soong were to suddenly become a credible contestant, we might hit a tipping point with droves of voters switching from Ma to Soong. However, Soong can't become credible without a lot of support, and he won't win support away from the top two candidates without being credible.
Thanks, Garlic.
You are right about the trend and the hard core thingy. The curves are, in fact, starting to level off after this article was posted.
Since then, the lines go horizontal, until the KMT launched this Yu Chan attack. Tsai up and Ma down, both slightly.
That is on the trend of "predicting vote percentage." In other two prediction contract groups, both betting on the "chance of winning", Tsai's score jumps up 3~4% -- a sharp rise -- in 3 days since the KMT started this Yu Chang attack, accompanying a fall on Ma's rating.
It's like Ma Ying-jeou is committing a suicide.
I'll give it sometimes to see how far it goes before I share the graphs.
Just checked the Xfuture prediction for Taipei. This Yu Chang conspiracy already turn Ma's rating fall below Tsai's in just 3 days. Yes, even in Taipei !!!
I think we might be seeing an enthusiasm gap. Tsai's supporters might just be more passionate, more willing to express their support in places like the futures market, and more eager to participate. Ma's voters seem a lot less enthusiastic. However, their unenthusiastic votes count the same as Tsai's passionate votes.
Everything I know about Taiwanese politics says that there is no way Tsai can win Taipei City this year. When the futures market says Tsai will win Taipei, my instinct is to doubt the usefulness of the futures market.
You are probably right, Garlic. It's just too good to be true. And green supporters are definitely more involved.
On the other hand, Chen Shui-bian did win Taipei before. It seems that if you get the message right, with right circumstances, we can't say it will never happen.
Looking at the blue and blue media --- if it really is happening, then what can they do, other than making people feel it's impossible?
I'm not familiar with XFutures, but National Chengchi University runs another prediction market, and they have been accurate in the past, according to the Economist. Their numbers were quoted as Tsai's chances at 49 per cent, on November 11, with Ma's at less than 42 per cent.
As for Taipei, that's probably a different story indeed. And it isn't election day yet. But given the KMT's apparent nervousness, the trend seems to be on Tsai's side.
JR, Chengchi University's prediction IS the XFuture one, not another one.
XFuture was originally a research project in Chengchi University. About September last year (2010), it departed from CU and became independent one. But all the people are the same, other than recruiting more new ones.
The Chengchi University keeps reporting the result comes out of XFuture.
While it is true that unenthusiastic votes count just as much as enthusiastic votes, enthusiastic people are much more likely to vote then unenthusiastic people. If enough blue voters are apathetic or disgusted, and enough green voters turn out, then who knows what could happen. On the other hand, I am not sure we can forget about the traditional polls just yet. Everyone knows the blue polls are biased, but nobody can ever speculate by just how much. Then there is the matter that those who place bets in a prediction market are still human. And people do bet on "the wrong horse" in other betting environments frequently. I still think this probably is a very close race. The main thing that encourages me is that the blue media doesn't seem to be jumping on the Yu Chang smear bandwagon just yet. Blue papers are still reporting the green side of the story. This leads me to believe that many blues probably have doubts about how this is being handled. And doubts are the last thing an incumbent administration needs to be facing in the month before an election.
Yes, this is most probably a close race. But that's much more than what I'd have expected, early this year. If Tsai wins, Ma and the KMT will have themselves to blame in the first place. Too much blind faith in a China-sponsored "economic miracle", and, above all, in that bealt-up "trickle-down" theory.
I hope that a new government will focus on the development of the industries which made Taiwan great - small and medium-sized enterprises.
I'm wondering when your next post will appear on your blog, Tommy.
The more I look at it, the more I deplore the pathetic self-comforting tone of this article. Why are die-hard DPP supporters obsessed with an inaccurate source whose credibility has already bankrupted by mis-predictions before, over-registration of green netizens and a biased leadership whose director was a renowned DPP supporter and often goes on TV shows to slash the other party? Does it really help Tsai by fabricating those fantasized numbers? Won't it blow DPP supporters' urgency to vote if they believe this overwhelming number for Tsai? Won't this trigger the KMT supporters to massively turnout since they felt the elections is at stake? The failure of Xfuture, like the failure of many NGOs in Taiwan, is mainly due to the omnipresent partisanship. An organization occupied by partisanship will almost never gain public respect and credibility.
Post a Comment