Sorry but this is arrogant (and condescending) hubris and is wilfully misintepreting the gp comment. Firstly there is no single bias free set of figures. Infor

Sorry but this is arrogant (and condescending) hubris and is wilfully misintepreting the gp comment. Firstly there is no single bias free set of figures. Information has been thrown away and distorted to produce a single line, without error bars. (Even error bars themselves suffer from the same issue) Then what you describe sounds like overfitting. You cannot just discard a change in a trend out of hand. It may be a blip or it may not be. Worst of all though is that graphs of complex systems do not contain all the information (obviously) about the future direction of that particular graph. It may be better to distrust other information sometimes, but thoughtful and cautious thinkers should be encouraged to be skeptical.