Forget the value added for grammars and just look at the baseline results
Goodbye, I know that progress8 and its ilk are imperfect measures - and as Talkin has already said on this thread / another, there is a statistical influence of %PP and %SEN on results, which mean that any measure is only really valid if applied to schools with similar intakes.
However, if you think about the quality of a school, and what it does with the intake it has, are raw results or value added the better measure of the SCHOOL, rather than the INTAKE?
Take selective school A, with 95% of its intake high attaining at the end of KS2. Then take comprehensive school B, with 15% of its intake high attaining at the end of KS2.
If School A has higher raw results than school B, that is surely not a measure of the quality of the school, rather a measure of its different intake? It is only when you start looking at a value added measure that you can start to get any kind of grip - however imperfect current measures are - of the balance between to what degree the raw results are due to intake, and how much to the school.
You would expect average grammar schools in Kent - accepting 25% of the pupil population - to have lower raw results than super-selective grammars in partially selective counties which only accept a couple of %. that doesn't mean that the schools are worse, just that they have different intakes.
It's like the common [but really hard to eradicate] fallacy that grammar schools are 'better' than true comprehensives 'because their exam results are better'. Anyone thinking about it for a moment would realise how ludicrous that statement is, but you hear it all the time.