Numerically illiterate horsepuckey. For example: If 100% of climate models predicted a rise of 10-14C (mean 12C), and we observed a rise of 11C, this doesn't

Numerically illiterate horsepuckey. For example: If 100% of climate models predicted a rise of 10-14C (mean 12C), and we observed a rise of 11C, this doesn't mean that the models are 'wrong', and it CERTAINLY doesn't mean that "the observations must be wrong." It does mean that there's some potentially interesting correlation in the models' errors, I guess. [Or, as cjensen said: 'Thou shalt not make a graph without error bars']