The save chance issue is debatable as this thread shows. For limits on casting, there may only be limited slots and the concentration mechanic, but their overuse of that one mechanic hurts a lot of the benefits reaped from loosening the other restrictions.
At first, targets will fail a bad save something like 60-70% of the time (low level non-proficient saves are as bad as -2), while proficient targets might save 50-60% of the time. As you level up, proficient saves proximately keep pace (maybe falling behind a little when casters stats hit 18 and 20), but non-proficient saves fall further behind.
Officially, they may be the most flexible ever, but I've seen some variant or other that basically accomplished the prepare spells but cast them spontaneously in just about every home game of 3rd edition and PF I've ever played, so for me, that isn't that much of a gain, but rather a formal acknowledgement that the older official version was usually ignored or modified.
I find your assertion that a rule was 'usually' ignored or modified in the RAW-uber-alles 3.x era... implausible. One of the goals of 5e was to get back to the pre-3.x tradition of the DM feeling free to use extensive variants, in contrast 3e RAW-based player 'entitlement.' So, yes, the Tier 1 casters in 5e are even more versatile than they were in 3.5, which is saying something.
And the use of the concentration mechanic doesn't ease pressure on spell slots, it just changes how the pressure is applied, as an example early in the thread of being faced with the decision of which concentration spell to cast shows quite clearly, as does the point that people felt compelled to save their actual slots for truly meaningful moments,
Sounds like fewer spell slots being used. Also, sounds like D&D.
Mechanically, wizards are still doing fine overall, and very few people are challenging that, but what really hurt more than anything was the nerfs to the individual spells that did things like force an abjurer to take sleep if they want to be half way functional in combat even if they have absolutely no in game reason to support taking that spell over an abjuration spell.
Sleep has always been /the/ 1st-level combat spell. In classic D&D, you hoped your magic-user started with Sleep (it was random), because without it you might as well just throw darts. It's nothing to do with 'nerfing' other spells, it's just another way 5e evokes the classic game.
My concerns are variety. If I specialize in something, I want to do it well enough to be useable in combat. If I'm an enchanter, I want to be able to take a mind over in combat. I don't need to do multiple like 3E. I would like to have a very high chance of maintaining a single domination spell. That doesn't appear possible in this edition.
With good reason, there are a lot of /very/ powerful individual monsters in the DMG. In the playtest, a common encounter design was multiple lesser monsters and one more powerful boss - dependably being able to dominate the latter would own the encounter. In the DMG guidelines, a lone monster is the most straightforward way to design a challenging encounter.
What you are asking for is excessive, game-breaking power. 5e doesn't say you can't have it, but you will have to be 'clever' or game the DM a bit to get there. It doesn't give it to you by default, in RAW, the way 3.x did, but it doesn't actively try to block you from ruining the game experience for everyone else like 4e did, either.
It is that way for a lot of spells. There seems to be only one sure path to useful spells: no save spells attack roll spells or short-duration spells like a round or two for effectiveness or spells that don't easily break from a common event during the course of gameplay. That limits your spell list substantially.
If you want a 'sure thing,' yes, you'd be limiting your choices. If you're willing to take a chance - you have more. Use those riskier choices cleverly and you'll get some very good results, some of the time. That's part of what made casters fun in classic D&D and 5e has come through with a similar feel.
