While they publicly say that they can and poe items for trade change droprates this is probably very rare. This is done so that if there is ever a case where a drop rate has to be adjusted and it is somehow proven by the community then GGG cannot be held accountable as it is in their right to so.
On all accounts I do not expect GGG to silently change droprates without notice since it would only negatively affect them as the community always finds these things out sooner or later.
People forget that things like Breaches, Abysses, Stupid Beasts and whatnot add high variance to the drops when considering just 500 map runs. Sincerely, someone needs to analyze all these things separately and then we will have a definitive answer to all this bullshit.
The recent case of oni-goroshi could be an example of this although that was a very publicized case and overall I expect that old poe items and currencies are probably not really changed ever and only newly introduced items are in case of an error or misjudged power level (such as bisco). The droprates of uber atziri long ago (gloves and axe switched) is another example.
You're not understanding the point. The test is invalid because the comparison was not 90% IIQ vs 270% IIR. The method, as Lagwin1980 stated and slipperyjim confirmed himself was alch and go. Therefore, it was more like comparing say 130% IIQ/30% IIR with 40% IIQ/300% IIR. A more precise method would have to take this additional IIQ/IIR from the maps themselves, recording it for each map that was run, and at the very least take an average of them over 500 maps.