The May contracts for options on federal funds futures prove that when a probability is at 98% it can't go much higher.
The July numbers are just a bit more interesting. The CPI report last week took some of the steam out of the "no change in June" sentiment. The probability of no change and the probability of 50 basis points are about the same now, and both are dark horses indeed. The betting on another 25, though it slipped just a bit in favor of the more aggressive posture, is still above 70%.
For those who asked, here is the data that these charts are based on...
Download may_pdfs_043505.xls
Download july_pdfs_042505.xls
... and here is a paper by John Carlson, Ben Craig, and Will Mellick that takes a more in-depth look at the methodology and how it can be used: