MetEye has a disclaimer that it is an experimental service. You're a bit tough basing your criticisms on a service that they declare to be experimental (and is therefore likely to include errors as they work out the bugs).
Forecasting is an inexact science. I suggest you pay a visit to your local met bureau and see how it's done. I found that very interesting, and they were quite candid about when they can get good results, and what weather systems still give them trouble. Here in Adelaide, they have problems with forecasting the impact of troughs which sometimes produce large variations in precipitation. As a pilot, I rely on their forecasts whenever I take to the air and I find they are pretty accurate most of the time.
Coop

