So I ask again what are the magical dependencies that martial characters need to be able to have a reasonable chance at succeeding at reasonable challenges?
This I think also ties into the issue of not actually giving an adventuring party enough to do concurrently. Its not so much that martials need some magical ability to be useful, and more that theres seldom anything
to do that isn't either easily resolved by a turn-off-this-mechanic button or is directly meant to be a magical problem.
The simultaneous devaluation of substantive mechanical exploration and distribution of absurdly cheap access to turn-off-mechanics buttons play a big part into why that is.
Martials have need to be able to resolve problems on their own when the Mage has their own problems that they can't divert attention away from.
I believe it was this topic where I talked about my mountain top encounter with a Ancient Red Dragon-lead cabal of Arch Mages leading an army of thousands of Orcs to besiege a city, where the point is for the players to break the siege and defeat the bads without letting the city fall to ruin.
Part of why that encounter works, by my estimation, is that it successfully diverts the attention of the party. The party mages are the only ones who can really tackle the rival Arch Mages, and can't waste turns slinging spells at the Dragon or the Orcs, and the rest have to both beat back the Orc army and fend off the Dragon long enough until at least one of these elements collapses dead, and
then they can justify focusing fire to win the day; but even then, the couple of times Ive run this the party usually still ends up split between the Dragon and the Orcs. (The Arch Mages understanably have always ended up the first to fall).
At a much smaller and less grandiose scale, the same general tenet applies. Attentions can be divided not just by way of multiple problems but even spatially. The Dragon in that example typically keeps 2-3 players roaming through the battlefield as they chase it, preventing any easy means of them jumping in on the other two targets, both of which keep their respective players relatively locked in place separate from one another.
In the same vien, the classic heist is another example of how this can be done, and theres a dearth of environmental puzzles that can serve as inspiration too. Even the "infiltrate the Lords Feast" sort of quest. Theres a lot of options here, is what Im saying.
And when you design adventures like this, your Martials
don't have to compete with Mages. They can just do things on their own in their own way, because the situation at hand can't just be turned off by a button.
Thus, you can design Martials any way you want. It fundamentally doesn't matter if they can match Mages or not, because, by way of how adventures are designed, they will always be able to contribute regardless of what the Mages do short of them neglecting their own proverbial turf to deliberately intrude on the Martials, at which point we're looking at a bad player.
The only real way to dispute the validity of this approach is by pointing at Simulacrum, but that spell, like Wish, is just bad game design masquerading as some integral part of the game; obviously broken things that can't easily be fixed without deleting them outright are outliers in the system, and shouldn't be the basis of any of these arguments.