Pathfinder 2e designer Michael Sayer made a post describing the game's assumptions of how spellcasters are intended to play, which I found completely baffling, and totally at odds with any game I've ever played.
…What [the game] does assume, though, is that the wizard will have a variety of options available. That they'll memorize cantrips and spells to target most of the basic defenses in the game, that they'll typically be able to target something other than the enemy's strongest defense, that many of their abilities will still have some effect even if the enemy successfully saves against the spell, and that the wizard will use some combination of cantrips, slots, and potentially focus spells during any given encounter (usually 1 highest rank slot accompanied by some combination of cantrips, focus spells, and lower rank slots, depending a bit on level).
So excelling with the kind of generalist spellcasters PF2 currently presents, means making sure your character is doing those things…
In 15 years of playing D&D-like RPGs, neither I nor anyone I play with have ever heard of any of this. None of us have ever chosen spells to target an enemy's weak saving throw, nor have we chosen our repertoirs based on the number of defenses being targeted. None of us even knew that monsters were designed with weak or strong saving throws. No D&D-like book I've ever read has ever advised players to target weak saving throws. In fact, the idea that any of us could even LEARN what an enemy's saving throws were, has never even occurred to us, nor has it even been supported by the Pathfinder 2e rules until the recent remaster.
I heard further comments from Pathfinder 2e designers saying that attack-based spells' hit chances are balanced around the assumption that the player will cast True Strike first, and that this is, in part, why spell attack rolls should not receive item bonuses. This just further confused me. None of my players has ever thought to cast True Strke, and to hear that the whole game is designed around it just boggles my mind.
I ran some calculations to check just how bad it is if a player does not adhere to these "Generalist Spellcaster" assumptions. A level 9 wizard who does not max out their intelligence, and who targets a level 12 Lich with a Will-based Incapacitation spell, will have only a 5% chance that their spell does anything at all.
This level of optimization is typical of my players, and it's clear to me that the game is unplayable for casters in its current state without an unreasonable level of optimization. I want to change the game's rules to eliminate the need to play a generalist spellcaster.
When selecting a character, my players almost exclusively want to play a specialist caster, not a generalist. Typically they pick a theme, such as "water" or "fire", or "illusions", and select spells around that theme, with a small handful of go-to spells that they use almost every fight. I want all these spells to be useful, regardless of the opponent's relative defenses.
How can I modify the rules to eliminate the "generalist spellcaster" paradigm from the game, and remove the need to target weak defenses without making spellcasters too weak or too powerful?
Best Answer
You cannot remove it without extensive redesign
I always had the feeling that PF2 was targed at people who found DnD5 too simplistic. You get way more feats, positioning now matters, learning about your enemy (Recall Knowledge) helps here, and so on. These are not only some part of the game, these were the design goals.
Wanting to remove these is like asking how you can remove the Force and Light Sabers from the Star Wars RPG. It would likely be much easier just to play another game.
Flexibility was never wrong
I want to mention that picking varied spells has always been a recommended best practice, at least since ADnD:
I do not know what game you were playing in the last 15 years, but it is subtantially different from what I played. Yes, PF2 takes tactical planning to new heights, but that is why I prefer it to DnD5.
Needing to think is not a bug, but a feature
If the player does not like to think during an encounter, he picked the wrong character, and should play a martial instead.
If the player does not like to think during character creation/level up, he picked the wrong game.