Once you concede that development has skewed the data, obviously you need to remove ALL the bad data from the data set.
Oh for God sake... If I was trying to take the temperature change caused by ambient air temp to liquid in beakers, I wouldn't count the ones with the Bunsen burner going under it now would I? Or are you going to complain that I'm artificially skewing the results if I move that beaker away from the burner and turn it off lest it skew my experiment?
You probably would.
When that happens the temperature of the older remaining data in the set from the time when the heat source was first introduced goes down because the warmer anomalies are removed.
No, you start fresh if you can. Since the data cannot be trusted it is invalid. You must find another way to gather it that hasn't been tampered. And since we're also talking history with unobserved results, and you can't trust computer models and corrections as we've been learning, you're kinda screwed till you find a new way.
When the now more accurate data shows a more pronounced warming, the deniers claim that making the data more accurate is "manipulating" the data because the more accurate data does not support their bias.
Okay, so we treat the bad data as valid data because it was not corrected, and therefore hotter than the new corrected readings and factor them in as signs of warming? What kind of bullshit are you trying to run here?
You THROW OUT THE BAD DATA and get new data from another untampered source if possible!
That's science. But we're not talking science, we're talking history, and that's going to be filled with guess work and assumption. Two things very un-science-like.
but since this isn't science, it's religion for you, why are we talking science?