This says it all - and certainly makes the possibility of deliberate data manipulation much more likely

The point is simple: It isn't even that the existing data set used to create the hockey stick is valid despite inconvenient starting period and ending period curves being excluded.

The data used as the basis for the hockey stick itself is a subset of available data:

There are 2 much larger data sets which actually overlap the Yamal data set; Yamal is what was used as the basis for the 'hockey stick' graph popularized by Gore et al.

Why were these data sets excluded? Why were even subsets of these data sets excluded when the other 2 data sets had trees (and tree rings) literally overlapping Yamal trees?

This is not science - it is data mining.

Read the whole gory story: