Tuesday, December 27, 2011

Futures Market Biases and other Election Errata

Belly dancing and other imported forms of dance, including flamenco and Bollywood, are hugely popular in Taiwan. Here is a performance in Fengyuan the other day. If your wife is looking for something to do in the evenings, belly dancing is a great way to socialize and slim down. Most local community colleges offer some form of dance class.

The topic of poll divergence, which I have often lamented on this blog, led the Taipei Times today.  The paper noted that the NCCU futures market has Tsai up enormously:
The exchange’s closing “prices” on Sunday showed that Tsai received 50.4 percent of the vote compared with Ma’s 43 percent and People First Party (PFP) Chairman James Soong’s (宋楚瑜) 7.7 percent.

Almost two in three “buyers,” or 65.6 percent, said that Tsai would win the election, while 30.8 percent predicted a victory for Ma, who had been leading Tsai before Soong entered the race on Nov. 4.
....whereas a Taiwan Thinktank poll had Tsai down by less than 1% to  Ma Ying-jeou -- the Taiwan Thinktank being a pro-DPP institution. In passing, the most recent China Times (pro-KMT) poll has Ma up by five.

I've been complaining about the irrational exuberance of the prediction market, whose pro-Tsai numbers I've long considered are too high, and sure enough, Nathan over at Frozen Garlic came out with a great post this week on its predictions for the legislative election that shows in this election it is fundamentally biased (if you are following the election, you need to be reading that blog). He says:
I’m on record as not being a big fan of the futures market here in Taiwan, but rather than explain why I think it is flawed from a theoretical perspective, I’m just going to bury them with their own numbers. Welcome to fantasyland.
Frozen Garlic has long argued that the Xfutures market is flawed because, among other things, people aren't committing real money (which is what he means by the first sentence there) and thus risk nothing with their choices. Here he has the numbers....
You may have noticed that there are a lot of DPP districts.  According to the xfuture market, the DPP will win 52 of the 73 districts.  The KMT wins only 20, and the New Party wins the other one.  The DPP is predicted to win 5 of 6 in Taoyuan, 9 of 12 in New Taipiei City, and 4 of 8 in Taipei City.  Just for reference, they did not win 18 of those 26 seats four years ago; they won a mere two.  You know that phrase the DPP loves to repeat about how the KMT can’t cross the Zhuoshui River?  The prediction market takes it literally.  The DPP wins everything south of the river, and that even includes Penghu and Taitung.  There are crazy predictions up and down this list.  The DPP is supposed to win Jilong City by a comfortable 6.7%.  If that happens, I’ll eat my pink Tsai Ing-wen flag.  Maybe the most incredible result is one that gets the winner right.  In Taoyuan 6, the KMT is predicted to win by a mere 1%.  That could happen, but only if you change the “1%” to “30%.”
These numbers are fantasies. The DPP will likely perform better this time around, but not to the tune of 50 seats -- it might reach 40 with the wind at its back. Nathan has shown that the prediction market has a pro-DPP bias. Hence, its numbers are no more trustworthy than any other public poll from media supporting either party, at least for this election.

If you have trouble using the text graphic FG offers, there's a map of the districts here. Just sweep your eyes over it... even if you give the DPP everything south of Taichung/Nantou, that's only 26 districts. Assuming a split in Taichung, the DPP would have to pick up ten more districts in what has historically been Blue territory. 3 or 4 maybe, but ten? I think the DPP will be lucky to get more than 35 single member districts (remember there are an additional 34 at-large seats). Still the dumbest move ever, reducing the number of legislative districts.....

Another remark in the Taipei Times article also struck me:
Former DPP legislator Kuo Cheng-liang (郭正亮) said it was ironic that Tsai’s support rates were higher when she was attacked than when the DPP retaliated with an attack on Ma’s integrity in a bank merger case, which showed that negative campaigning might very well “have the opposite of its desired effect.”
Last week I pointed out that going negative on the Fubon merger case would likely hurt Tsai.... sure enough, Tsai's lead in the prediction market has fallen, as has the value of her shares, which are now at 65, though they were pushing 70 last week. Interestingly, this trend has not appeared in the pro-KMT polls. But I suspect that if Tsai loses this campaign by a whisker, the decision to stage a frontal attack on Ma, rather than let one of her supporting media such as the Liberty Times handle it, will have been a key factor.

Fortunately for the DPP, the KMT is continuing its attack with the Yu Chang case (love to know what their internal polls are telling them). This may help wash the effect of the DPP's negative turn. The government is planning a big nationwide biking activity for Dec 31 that looks a lot like a KMT campaign rally....
Don't miss the comments below! And check out my blog and its sidebars for events, links to previous posts and picture posts, and scores of links to other Taiwan blogs and forums! Delenda est, baby.


Okami said...

What's the point of having a futures market without betting?

I'm more interested in where the actual money is being bet on. Are there any websites listing that?

Michael Turton said...

No, but the bookies have the election at even.