Wouldn't you model the future, though? That, at least I can understand (and I won't mention how every model always gets it wrong - from ice caps to temperature to...).
But you don't model the past. We have direct (or proxy) measurements for that, already. Why are we changing that data, and always in a way that hides stuff from the past that counters the climate change narrative?
Like, for example, arctic sea ice:
https://climate.nasa.gov/vital-signs/arctic-sea-ice/
Why start around 1980, when there is good satellite data from 1970?
Why model the past? because ideally you want data that is comparable with the data you collect now, except in so many instances it doesn't actually exist. A lot changes over time, so those changes need to be taken into account if you are going to attempt direct comparisons.
If you are then gonna attempt to model the future, you need assumptions, where do you get these if not from the past? So obviously what you want is to have directly comparable data in order to ascertain some kind of handle on cause and effect, and that has to come from observation.
Given we were discussing changes to GISTemp, here's a more detailed list of what changes have effected the outcomes of that modelling.
https://data.giss.nasa.gov/gistemp/history/
As for your trope on every model always gets it wrong, I'd suggest that's all it is. Modelling is generally done on a range with a degree of confidence assigned, because it's an attempt at recreating the unknown. Hence there's always qualification of the scenarios used, These are rarely reported in the media, oftentimes it's just presented as absolute certainty of worst the case scenario, without qualification. This is then repeated in the kind of criticisms that are so common in the blogosphere.
Or perhaps it's just as often the other way around, it's often hard to know from where misinformation originates.
I pretty much dismiss any argument that attempts to convey absolute confidence in a single modelled scenario where it is clear there are variables that would have been allowed for in the source material, whether that be from an alarmist media, or the pseudo science of populist denial-ism.
We could take your GIStemp post from earlier in the thread as an example, you focus solely on the fact that a graph has changed, and present that as evidence of some kind of conspiracy driven attempt at some kind of cover up?
Yet it's made plain there is uncertainty, as we see here.
https://data.giss.nasa.gov/gistemp/graphs_v4/#
Or this.........
https://data.giss.nasa.gov/gistemp/uncertainty/
