I'm glad you can agree some comparisons across games are meaningless. That's a great starting point. So maybe start by asking 'why is it meaningless to compare numerical scores across sports'? I think the answer is that numerical scores are not independent of the games being compared. What other attributes aren't independent of the games being compared? IMO - Most! What are some examples of independent attributes - the facts that both games have positions, that a numerical score is kept, that there are referees that often have to make judgement calls around whether something is a foul, etc.
Some players move from rugby to American football to Australian Rules football. But I think few players move from rugby to cricket. Why is that? The explanation is found in a comparison of how the games are played, what sorts of skills and abilities they require, etc.
Cricket is more like baseball than it is like football. This is because of features like bowling/pitching and batting; the setting of a field; running after hitting the ball; etc.
Field hockey and soccer can be compared in many respects, and we can look at ways in which (say) soccer more closely resembles hockey or more closely resembles Australian Rules.
Etc.
Even when it come to scores, we can in fact compare particular games as being (for instance) high scoring or low scoring, and relate this to factors like team ability, weather (a wet ground can make for low scores), etc. We can also compare whether games are systematically high scoring and fast in play (say, basketball and Australian Rules) or involve less frequent scoring and "slower" play (eg rubgy). We can also compare the degree to which play might be fast yet scoring low, and consider why that is the case (eg soccer).
One of the first steps to a good comparison is being able to identify the attributes that are meaningful to compare. I'm not sure you've carefully addressed that step. It seems to me more like you are assuming agency can be meaningfully compared across RPGs and then you compare agency across RPGs and start drawing conclusions from that comparison. But if that premise is flawed then the conclusion doesn't follow.
I'm quite satisfied that agency can be compared in the ways I'm comparing it. That satisfaction arises from having had the various experiences I'm describing - experiences of agency - and comparing them.
you use ‘familiarity with a game’ to invalidate others opinions.
Well, generally if someone isn't very familiar with a thing, that is a reason to be doubtful about their opinion of a thing. Isn't it?
You should be speaking on merits of the comparison or the opinion and not using a blanket rejection based on the posters familiarity with anything
Suppose that a particular RPGer had only ever played DL-style railroads, and
did not even know that any other sort of RPGing was possible. Why would I regard that person's view as to what sort of player agency might be possible, in RPGing, as having much value?
Conjecture is not always worthless. But generally valuable conjecture follows from some degree of familiarity with whatever is being conjectured about. Someone who tells me that it is impossible to compare (say) the Prince Valiant scenario The Crimson Bull to (say) the AD&D module Dead Gods is making a ridiculous claim. I've read both scenarios. I can tell you, in some detail if you like, what it is that makes one a masterpiece that supports high player agency
despite its intricate, and temporally unfolding, framing; and what makes the other, as written, an utter railroad. Not only is the comparison possible, but it's very illuminating, both of Jerry D Grayson's talent for scenario design and also about quite subtle possibilities in scenario design that are often ignored in discussions of what is and is not a railroad.
For the record, I think games can be compared in some ways - for me it's never been about no comparison ever - it's about noting that there are certain comparisons that just don't make sense - much like you agreed that comparing numerical scores in different sports is meaningless.
But given that our hobby has a word for particularly low-agency play -
railroad - and that there have been essays discussing the degree of player agency for almost as long as the hobby has existed (I think I already mentioned Lewis Pulsipher's White Dwarf articles from the late 1970s),
comparisons of player agency across different approaches to play is manifestly
not something that doesn't make sense.