Quantitatively speaking, the number of signals at frequencies 100-1000 Hz that are shorter than 1 second and stronger than about 10^{-19} (amplitude of the metric perturbation) is smaller than 0.26 per day at 90 percent confidence level.
Some of their statistical methods to search for a signal look pretty fancy. The bounds have been improved. Only the signal is missing so far. Mother Nature is able to be as cruel as She was co-operating in the past...
Before someone is gonna blame Kip Thorne et al. for having wasted 365 million USD (the most expensive project ever funded by NSF), you should know that virtually all theoretical physicists have very good reasons to be convinced that the gravitational waves should exist, they should be visible by LIGO a bit later, and once we can detect them with some accuracy, they can tell us a lot of things about the universe.
