Admit that you were wrong or prove that you were right

There are entire threads when folks all predicted at once... you should search for 'em.

I saw one about a year ago, where the majority of the board said 15-20% or so (within a few points), but there was a good chunk going more or less (many said flat). Yes, there were a couple, like perfitz and steveF, who claimed as much as 15% up (amazing, when you think about it)

Some of the more interesting prediction quotes get pulled out from time to time, just watch 'em.

You have a very short horizon. This isn't day trading. It's a very slow-changing market that's tremendously driven by emotional attachments.

A "bad time to buy" call at the beginning of 2009 is still the correct call, even though prices are down only a lot. The standoff between buyers and sellers continues, but it's clear who has to win -- an unprecedented crazy cheap-credit cycle being what it was.

oh, i missed the horizon thing. Yes, you need to do a little more history than that.

alright, makes sense, scratch the 6 months comment

Admit it?
OK
I hammered in the nails myself.

The problem with predictions based upon present data is that it is difficult to see where you're going by looking backwards. Two completely reasonable people can look at the same data and arrive at opposite conclusions. One of them will eventually be revealed as incorrect, but only in hindsight will the correct prediction appear prescient. Since Alan's right about the short horizon, how long are we going to take this out into the future? Six months ago some were predicting 50% off. Some went so far as to claim within the year. We're still six months from finding out the answer to that question. I just wonder what you want people to "prove."

Another problem with "now" is that it takes a few months (couple of years in the case of new construction) between when price agreement occurs and when the deal closes. Then it takes weeks before the data is aggregated. Then, as evnyc pointed out, analyses differ wildly. And market segments are picked at to support this position or that; as big a city as this is, the sample size for each segment is potentially too small to be meaningful.