I had a BS seed time of 1:50.8 (though I had split 1:51 point a few times) for the JMU last chance meet this year. Despite the fact that maybe two people in the whole meet ran that time, I was in the 3rd heat. While it took times under 1:52 to get into the third heat, after I finished, I turned around to see Duke's coach congratulating some kid..."you did it! I had you at under 1:55! Great job!"
It seems to me that his sub 1:52 seed time was excessive BS, considering he was thrilled to crack 1:55. But that's Duke.
Not as bad as morgan st or choppin. Go to penn st's national invite in indoor, you will see 50-60 guys seeded in each event faster than what the 10th place finisher ends up running.
We should force all times to be submitted to a public database. Then, when an athlete is seeded for a race, the database could automatically produce a seed time for him. If he hasn't run the event yet, there is a way for the database to interpolate his time using other performances (say assign quality points to each event, look at events on either side, average the quality points and produce a time of that averaged quality).
It would create a public domain performance list, and if the hosting meet had computer access and the right software, it could eliminate ridiculous seed times. It's getting too hard to guess how much you have to lie just to keep up with the other guys that lie worse.
Hell, I could write the software for it. Any takers?