If you think about it with a little less cynicism you should be able to understand why they are more likely to be over inflated than under inflated.
I mean, I study these lines for a living. My default assumption is that they're wrong because there's no revenue for Disney to gain by increasing the accuracy of posted wait times.
I don't think the reason is "Guests are happy when a posted wait of X ends up being Y."
That policy prevents guests from exercising their own judgement to decide the best use of their time. It's an incredibly patronizing example of "we know what's best for you so don't ask questions."
Here's an example from Slinky Dog last week.
As the day drew to a close, Slinky Dog's posted wait was 70 minutes at 8:15 pm. It had been at 70 minutes for about half an hour.
Twenty-two minutes later we timed an actual wait of 13 minutes. And we got actuals of 10 and 14 in rapid succession within a few minutes of that.
It's almost impossible to get a 57-minute reduction in actual waits in just 22 minutes of clock time. I'm sure I could come up with a scenario where literally nobody got in line, wheelchairs caused an extended delay, extra trains were added to the track, etc. But that almost certainly didn't happen here, and I think we all know that.
To spell out the problem with "Posted of X ends up being Y", where Y < X, there are tons of people who would've got in a 10- to 20-minute line at Slinky, if they had been given accurate information about their wait. But they skipped riding because they didn't think it was worth 70 minutes. So they missed out on an enjoyable experience because of bad information.
Sorry for the long post. What do you think the reason is?