China Times questions pre-election poll of Liberty Times

Added to the laughable, extremely inaccurate polls carried out by the pan-blue's camp, an article showed up today, questioning the way Liberty Times did their polls by asking (as the article title):

真的這麼準？自由時報應公布民調方法
Really so accurate? Liberty Times should tell public how they did the polls

This article started with that all pan-blue pre-election polls consistently showed that KMT's Huang led by more than 10%, but the outcome was that he lost by a small edge. Only the poll conducted by Liberty Times show accurate prediction.

The article asks why, but instead of reviewing why it's so wrong with pan-blue's polls, they went straight ahead to question why it's so right with Liberty Times' polls, and even shamelessly went so far to say that Liberty Time's polls was "false positive":

自由時報民調陳菊支持率（三四．○九％）並未高出其他媒體太多，甚至還低於ＴＶＢＳ的三六％，只是因為測得的黃俊英支持率偏低，以致差距僅有○．二九％，造成「神準」的假象.
Liberty Times poll shows that Chen Chu's support rate is 34.09%, which is not too much higher than what other media got, and even lower than TVBS's 36%. It's only because that in Liberty Times poll Huang's support rate is low such that the difference between Chen and Huang was only 0.29%. This results in the "false impression" of its extremely accurate prediction.

How come a group of poll managers who always deliver wrong results go public to accuse an accurate poll as being wrong? I can only say, these people (China Times Poll team, 中時電子報民調組) are simply shameless.