2012 Election Controversy: Official Vote Rigging?

The 2012 election is over. The incumbent Ma Ying-jeou of KMT party, after a hard fight, won with a comfortable margin of 6% (see the table in the end of this article) over the challenging candidate, Chairperson of the DPP, Dr. Tsai Ing-wen.

To many of Tsai's supporters, however, the election outcome is questionable. That mind set mainly (my observation) comes from two factors: First, the popularity of Tsai prior to the election was so high, making people feel that there's no way Tsai could have lost, let alone lost with such a margin. Secondly, the vote counting process is way too problematic, when comparing to what people experienced in the past elections. People reported all sort of different abnormalities and law violations of the vote counting behaviors in many counting stations all around the country. Just a couple of examples to let readers understand why voters doubt the result,

* When counting Tsai's votes, two tickets were read once, but when counting Ma's, one ticket was read twice;

*  Tsai's ballot was read correctly, but was recorded as Ma's when entered into the record;

* At the time of counting process started, people were blocked by the police for 10 to 15 minutes before they were allowed to enter the counting room. When they did enter, the ballot box already opened (this is against the law) and counting already started;

* More common complaints: many stations rushed the counting in a super speedy way such that people who watched were not able to tell if the counting was correct. For example, one station with 700 voters finished the counting of all ~2100 ballots ( each voter has 3 ballots: president, legislator, party)  in less than 20 minutes, averaging more than ~105 ballots/minute. If the counting proceeds as required by the law, one ballot would have taken at least 3 seconds. In one minute, you can only count ~20 ballots at most.

These are just the tip of iceberg among overwhelming complaints that I heard.

Because that voters are not allowed to carry any electronic device into the voting room (well, in the location I went to, this rule was violated, too -- people were only asked to turn off the cell phone. Not only was cell phones were allowed, but also there was no check on any other devices), any voting violation can't be caught on camera. And because the vote counting started right after the voting was done at 4 pm sharp, many people who voted and waited to watch the counting didn't have a camera with them. So most of the counting violations reported so far didn't come with evidence. This is a hard lesson to learn for democracy.

However, that doesn't mean the vote counting is completely safe from scrutiny. At the time of counting, the counting progress was recorded and published live online on the official website by the Central Election Commission (CEC),  the official unit in charge of all things related to elections. One guy took screen shots, roughly one in every 3 minutes, and made a slideshow out of them. There were total of 68 slides starting from 17:48 (note: the counting started at 16:00) to 21:59 (when all the counting was completed) on Jan 14th.

Since I am usually obsessed with any data of time series nature, I tried to write down the votes on each slide along with its time and total counted stations,  to see how the counting proceeded. Much to my surprise, there is one abnormality that puts me in disbelief : on slides from 21:00 to 21:06, when the total counted stations remains unchanged for 6 minutes -- which means that the vote count should have been unchanged -- Tsai's votes mysteriously dropped, and Ma's increased. See the screenshots below.

The screenshots, summarized in the following table,  indicate that in the official ballot counting, Tsai's votes dropped and Ma's increased when there's no new counting stations added.

Time	Tsai	Ma	Soong	Diff
21:00	6088958	6881307	369012
21:03	6088958	6881307	369012	0
21:06	6088278	6882220	369024	Tsai -680 Ma +913

It seems that the official election commission is automatically adjusting the vote counts in favor of Ma Ying-jeou in the middle of counting. The adjustment we are able to see is not large: Tsai down by 680 and Ma up by 913. But that might be an indication of something much more serious in the official counting which could have been carried through the entire course of counting. We don't know if the final result given to us is the true count or the count after such a possible systematic adjustment.  

The numerous counting violations in local stations in favor of Ma as seen by many voters might not account for the winning edge of Ma's (~ 800,000 votes). However, if such an auto ballot adjustment does exist right inside the CEC, it could have more damaging effect. Comparing to potential ballot rigging on local counting stations, in which the outcome is unpredictable, a direct adjustment on the numbers in the official final steps ensures Ma's win with a predefined ratio -- a ratio large enough to exceed the maximal vote difference required by law for a recount.

Interestingly, the doubt on Ma's winning based on a predefined ratio does circulate based on different channel of info. Recall that Ma got total of 6,891,139 votes. In Taiwan, large numbers are described in shorter form on  base of 10,000 (called 萬). For example, Ma's votes would be called >6 million in English, but "689 萬" in Taiwan's language. So the number "689" is meaningful.

Two days before the voting, a netter named v99999 posted on the forum of xFuture website with a definite tone of prediction of the election outcome with the number 689. I translated the content, with most of original format below:

~Please remember~689~Don't be shocked Posted by v99999 at 2012-01-12 17:54 #

評價： 39

~Please remember ~689~ Don't be shocked ~Please remember ~689~ Don't be shocked What does it represent ?? You will know it the day after tomorrow,,,There are many things that can't be learned by deduction and statistics.,,,Is it something not yet occurred? In fact, it already occurred,,,Many people always wait till the time of seeing it to ask why. In fact there's no "why" to it,,,It is just the way it is,,, 689

In fact, some reported that the same netter already predicted the number 689 on Jan,7th, one week before the voting. 

Is it a coincident ? A prediction of powerful fortune teller ? Or, a leak from someone who knew something behind the scene ? In a followup post to the above one, v99999 claimed that he/she actually saw the number before the election. The tone in the post makes people wonder  if the election outcome was already defined before the voting. 

There are many other reports of potential ballot rigging. The more you heard, the more you start to doubt if the election is indeed rigged. Many people against this ballot rigging theory claim that it's impossible to rig, 'cos every party would send people to each single counting station to monitor the votes and do the vote count themselves to ensure the counting is correct.

But here is where it gets even more bizarre -- the DPP didn't send people to watch a large number of counting stations !! Many voters reported that when they went to watch the counting voluntarily, they didn't see anyone from the DPP.  In as many as 1600, 2000 or even 5000 stations, according to different sources, out of the total of 14806 counting stations, the DPP didn't assign anyone to monitor and collect the vote counts. 

How on earth could this have happened? The DPP has a long history of experience fighting against the KMT, a party that has long been recognized by people with vote rigging. Fighting against such a rigging-prone party, you would imagine the DPP will definitely monitor the counting closely. In fact, in the past elections, especially ones as critical as the president election, the campaign team of DPP always mobilized and trained people to watch the counting very closely to prevent the vote rigging. Why did they skip it and just let the KMT did whatever they wanted to do this time?

If there indeed was ballot rigging by Ma Ying-jeou's government and the KMT, the DPP campaign team must be seen as partly responsible, at least on the basis of negligence. There might be something more to it. My instinct tells me that someone inside the DPP didn't want to see Tsai elected. I can go on for several articles on this but we will stop here.To people interested in the DPP politics, especially how Tsai Ing-wen struggled to guide the DPP through trouble waters within different fighting forces inside the DPP, this is a topic worth of study.

Anyhow, this article is about the vote rigging of this election. Some more studies on the vote counting might reveal more of the details. Or maybe sometime in the future, some might come out with more evidence. No matter what, this election must be recorded in the history book with not just the result but all the vote counting of law violating nature.


Update: 


1/31/12: See more counting violations compiled by Claudia.
1/31/12: The DPP releases a statement saying that they had nearly the same amount of vote-count monitors as the number of counting stations. 
2/1/12: The CEC releases a statement saying that it's an input error. 

Reference:
Summary of the 14 January 2012 Republic of China presidential election results from wiki

Party		Candidate		Votes	%
		Pres	V. pres
	KMT	Ma  Ying-jeou (incumbent)	Wu  Den- yih	6,891,139	51.60%
	DPP	Tsai  Ing-wen	Su  Jia-chyuan	6,093,578	45.63%
	PFP	James  Soong Chu-yu	Lin  Ruey-shiung	369,588	2.77%
Total				13,354,305	100%

Tsai will get at least 47% votes -- UDN Editorial

Observers have been highly confused by polls for the upcoming 2012 presidential election --- blue media claims that Ma keeps leading by a margin around 2~8%, and pro-green media say 50-50, which will likely be translated, according to past experiences, into that Tsai's the winner by a margin as large as 10~15%.

Oddly, the United Daily News (UDN), one of the can't-be-bluer KMT propaganda-like media, published an editorial today, claiming that Tsai will get at least 47%. Their reasoning, in fact, indicated that the 47% is a bit underestimated:

The premise of this framework is that Tsai gets 47% votes. Two references support this idea:

1. Frank Hsieh got 42% in 2008, and the momentum of DPP now is stronger than the DPP in 2008, and support to Tsai is also stronger than the support to Hsieh in 2008. So Tsai's votes will go up from 42%;

2. Chen Shui-bian got 51.11% in 2004. Although it was interfered by the 319 shooting, but whichever you look at it, it shows that the DPP has the potential to get 50% support of the voters. Now the momentum of DPP is stronger than that in 2004, and Tsai is better than Chen, therefore Tsai has the potential of approaching 50%.

Based on these two references, it's convincing to argue that Tsai can get 47% of the votes.

So what's going on here ? Do they suddenly come to sense and realize that the momentum of DPP is un-ignorable this time ? How do they explain those polls they did, that Tsai was around 33% ?

Polls by UDN on 2012 President Election
Date	Tsai	Ma	Soong
2011/11/27	35	41	10
2011/12/ 3	32	39	10
2011/12/10	33	40	10
2011/12/12	35	42	10
2011/12/17	33	41	10

This editorial from the UDN seems to bn an admission that:

(1) Blue polls always underestimate green candidates by 10~15% (47-33 = 14%). But this is certainly not new. A famous pro-blue writer Nan Fang Shuo (南方朔) said in an article that :

正因為綠色群眾較不表態，懂選舉的人都知道，國民黨要贏，投票前半年，他們必須領先至少15%，投票前3個月至少要領先10%
Because green supporters are more likely not to display their voting behaviors, whoever knows about the election knows that, the KMT must lead by at least 15% and 10%, half year and 3 months before the election, respectively.

See another admission in a video by a TV commentator who is deep-dark-black blue.

(2) Green supporters DO NOT display any stronger willingness to tell the truth in polls than ever, a point what many observers ( example ) claims without support of substantial evidence. In a response to the above article cited, I wrote:

Many people claim that (that the green supporters are more willing to tell the truth in polls this year), but none can provide anything to back that up other than pure speculation. Such speculation most probably came from a sense that the green supporters are higher in moral this year, as seen in new report or TV or any campaign activities.

That sense is correct. However, converting that observation directly into green’s willingness to provide faithful answer during polls conducted by pro-blue media is a serious mistake. Many green netters I saw on forums reported how they intentionally told poll calls that they want to vote for Ma.

The blue poll center knew this very well, too. Here is a paragraph in a recent interview of the vice manager of ChinaTimes poll center, conducted by a journalist (?) from China, posted on 12/27/2012 :

【对我们来说，一般加权之后，绿营会上升，因为考虑到愿意接受中时访问的可能是蓝营较多，绿营的可能直接把电话挂了。

To us (China Times), the green support rate generally will go up after weighting. Because we considered that blue supporters are most probably more willing to take polls conducted by China Times. The green supporters might just hang up the phone.】
访中时民调

Yes they did take that into considerations and make weighting outta it. But, they refused to publicize how much adjustment they made:

【在民调数据之后要进行加权，但是加权数不会公布。
After we took the polls, the data need to be weighted, but we won’t reveal how much we adjusted 】

From all sources, “more pan-green supporters were willing to express their preferences” seems to be nothing but a rumor that many took as a fact.

In the mean time, less than 2 weeks before the voting, the xFuture has been predicting that Tsai lead Ma by 6~ 12% for 4 weeks ...

UDN's Editorial:

選宋楚瑜就是選蔡英文

【聯合報╱社論】
2012.01.02 03:10 am

有人說，「選宋楚瑜就是選蔡英文」，宋楚瑜揚言提告；然而，「選宋楚瑜就是選蔡英文」，已是昭然若揭的事實。

這是以數字推演出來的結論。一般的說法是：若以蔡英文的得票率為四十七％當做準據，則馬英九若要贏蔡英文，就至少必須得票四十八％。四十七加四十八，餘下的空間只有百分之五；也就是說，在此一架構下，宋楚瑜的得票若超過百分之五，馬必落選。

此一架構的假設前提是蔡英文得票四十七％。有兩個參考點，可以支持此項假設。一、謝長廷在二○○八得票約四十二％，而此時民進黨的氣勢勝二○○八的民進黨，此時蔡的聲勢勝二○○八的謝，因此蔡英文的得票至少可從四十二％起跳。二、陳水扁在二○○四年得票五十‧一一％，當年雖有三一九槍擊案影響，但無論如何皆顯示民進黨有得票近五十％的潛力；而此時民進黨的氣勢勝二○○四年，此時的蔡勝二○○四的扁，因此蔡的得票率亦有向五十％趨近的實力。從這兩個參數作出蔡英文得票四十七％的推論，具有說服力。

如前所述，在這個架構下，宋若得票超過五％，馬即落選。那麼，宋的得票會不會有五％呢？答案是有可能。因為，五％就是六十五萬票，若以全國一萬五千個票匭來計算，只要宋楚瑜在每一個票匭開出四十幾票，就能達到那個票數，怎謂沒有可能？

也就是說，宋要拿到五％選票，從統計的常態分布來看，是有可能的（他在二○○六台北市長選舉得票四‧一四％）；反過來說，如果宋楚瑜拿不到五％，則那種選民的表現所透露的對世局國情的深刻成熟思考，才是統計學及社會學上的異態，反而可以令人驚異不置。

此次選舉唯一可以確切預言之事，就是宋楚瑜絕對不會當選。因為，仍以蔡得票四十七％為準據，宋若要當選，須獲四十八％選票，也就是要將馬壓至五％以下；但如今宋的民調在六至七％徘徊，其看好率則在一％以下，甚至有時低至○‧一％。宋之絕對不會當選，是當然與必然之理，亦為宋楚瑜自己心知肚明之事。

此時宋能拉到的選票，絕對大多數應是泛藍的選票。因為，宋標舉的是極統觀點，不可能有泛綠選民把票投給宋；所以，宋拉到的每一張泛藍選票，都會造成使馬落選的壓力，亦將成為使蔡當選的助力。因此，選宋楚瑜就是選蔡英文，這個論點可謂理所當然，毫無疑問。

問題在於：宋楚瑜執意倒馬挺蔡，卻陷於自我矛盾之中。宋楚瑜的中心觀點仍在國憲認同與兩岸政策，他公開肯定馬的ＥＣＦＡ是「重大努力」，又公開指蔡之否定九二共識「後果嚴重」；倘若這是宋的真信仰，他卻要拉下「重大努力」的馬，而要拉上「後果嚴重」的蔡，這豈不是自相矛盾？

沒有人比宋楚瑜更知道他絕對不會當選，也很少人比宋楚瑜更知道蔡英文若當選，將在國憲認同及兩岸關係上發生的「嚴重後果」。有人問起「五％生死門」的問題，宋楚瑜說：「重要的不是五％，而是誰能不讓台灣趴下去！」但是，宋楚瑜如今正在做的事，就是要拚命拉下馬英九，將台灣綁在蔡英文的台獨戰車上，親眼目送台灣就這樣趴下去！

宋楚瑜如果有一丁點為國為民的不忍之念，即使他有千般百般的私仇私恨，也沒有理由裹惑五％的選民與他一起去製造「嚴重的後果」，「讓台灣趴下去」。宋楚瑜明明知道他自己是在摧毀國脈傷害台灣，但那些迄今仍支持他的泛藍選民，尚以為是在伸張正義救國家；那些泛藍選民難道知道：宋楚瑜現在要利用他們去做的，其實是等於要他們扶助蔡英文做中華民國總統？

選情緊繃，間不容髮。除非蔡英文可衝破五十％過半選票，或蔡英文的選票低於四十五％；否則，宋楚瑜的得票高低必定是馬英九能否當選的決定性因素。一％十三萬票，莫說五％是生死門，一％都可能決定馬蔡二人的生死。

宋楚瑜必須說明的是，「棄馬保台」難道就是他的救國方略？他說，現在如果不選了，「連人都做不成了」；但當他在一月十四日若與蔡英文共同實現了「棄馬保台」的台獨大業，莫說「人做不成」，當他揹上「汪精衛／吳三桂」的罵名，恐怕另日也無顏見他那為捍衛中華民國而浴血苦戰的亡父宋達於地下了！

非「讚」不可！歡迎加入udn粉絲團，新聞你HOLD住！
【2012/01/02 聯合報】@ http://udn.com/

全文網址: 選宋楚瑜就是選蔡英文 | 社論 | 意見評論 | 聯合新聞網 http://udn.com/NEWS/OPINION/OPI1/6818432.shtml#ixzz1iH3HmaUC