Nah, I read that but I'm not convinced. The reply to them from palijer is correct: inaccurate does not mean lower, it just means inaccurate. Fewer samples means

Nah, I read that but I'm not convinced. The reply to them from palijer is correct: inaccurate does not mean lower, it just means inaccurate. Fewer samples means less data means more room for weird sampling errors skewing the data. There's nothing inherent in that that suggests we ought to assume an upward skew. The best we can say is that the sampling skew this month is different than the sampling skew in past Januaries. That could be due to some interesting cause, it could be totally random, and there's zero point in speculating about it until we have more data.