My point is simple like this: I don't think any airline simply put their fleet (same aircraft on same routes and similar conditions) flying on M.84 or CI around 150 for some time, collected all significant data, did the same on lower CIs or M.83 and compared everything.
They seems to rely mostly on FMCs software fuel burn forecasts, which aren't accurate enough.
If they do it I guess they will have a surprise.