|
I think we agree on more than we disagree, but we're looking at things from different points of view...
The question of biological free will aside, and setting aside also the relationship OOTP places between a player's stats and the steps the development engine takes to adjust ratings based on those stats (which, if it still works as it did some time back is actually backward), the relationship between player ratings and stats is a very valuable one to understand if you're playing with scouts on. If you play without scouts, then clearly ratings are the thing to follow (though stats are still valuable to pay attention to if your rating scale is broad enough).
When playing with scouts, you should constantly be asking yourself how accurate your scout is. I completely agree that players ratings can and do change, but stats are the only thing you have at hand to give you a read on how excellent or poor your scouting is. And predicting future stats are the goal of scouting.
We can get into a remarkably interesting philosophical discussion about what a scout is actually doing vs. what the development engine is actually doing. And we could get into even deeper conversation about how the model might be tweaked to make the system more natural (or biological?).
All that said ...
The random action of the development algorithm, when matched with the quantum dynamic nature of randomness itself, make any projection based on anything in OOTP dangerous. Barring a perfect knowledge of ratings, however, past performance _and knowledge of the full situation_ around a player is the best predictor of future performance--but cannot be completely trusted in any situation. Or, as knowledgeable poker players will tell you, there is always the chance that your best bet will turn into a bad-beat.
|