taking measurements from inaccurate thermometers and scant coverage from over a century ago, and claiming we know global average temperatures in the 19th century is beyond ludicrous.
The coverage was a lot smaller than today's but it was so much cooler then, that the error is less than the change.
No amount of massaging of data can make credible comparison to today's grid of sensors.
Yes it can. It's just that the error bars are larger back then.