If the question is "is the world getting better?" then yes, treat everything the same. But that question isn't as useful as others that could be asked, such as

If the question is "is the world getting better?" then yes, treat everything the same. But that question isn't as useful as others that could be asked, such as "which policies make the world get better?" Hinkel's argument is that Pinker uses "is the world getting better?" as a proxy for "are my preferred policies making the world better?". If policy evaluation is their goal, the ad-hoc exclusion of China is overly simplistic, but so is drawing policy conclusions from overall trends without examining what's driving the trends. Both Pinker and Hinkel should be measuring the degree to which various policies were applied and evaluate poverty-reduction against that quantity.