Yeah...I guess it could be explained many ways. I didn't really mean to make it as complicated as it sounds after reading it again. I'll try to explain it again, and I'll follow up with something that hpsolo said that I don't quite agree with (don't worry, its not all bad ^^)
SE likes to repeat formulas for alot of things. While magic accuracy hasn't been determined, I personally mirror it to how the formulas for hit rate work. You can get alot of the information from wiki see how Skill, Eva Skill, ACC, EVA, DEX and AGI work together to come up with a single result.
The basis of my assumptions consider the "Hit Rate" for a spell, and I like to call it MACC%. This value, like hit rate, gives a single value that determines the chance of a spell landing. This is completely different than MACC.
We know that this MACC% is influenced directly
by many things. Elemental Staves (+10%, +15%), Weather (+10%,+25%), Day of the Week (+10%), are a few that directly influence MACC% (also consider reverse effects!). We also know that base MACC% is calculated by things such as Magic Skills, Stats, and Levels.
This is where MACC and MEVA come into play. MACC also has direct modifiers (Alkalurops with its mighty +20!). Most of MACC however, is calculated by some formula using Skill and stats (chr). MEVA has only recently been known and has direct modifiers too - with roots starting from DNC's "Stutter Step" as well as the new RDM WS "Death Blossom". It is assumed that MEVA is also calculated in some way, with some of the better known pieces of the formula being based on stats (chr).
Once you have your MACC and MEVA levels, there is a bit of a calculated 'face off' of sorts, to determine your base MACC%. After this calulation is complete, a level difference/correction formula is applied, and you arrive at your final MACC% rate.
Hopefully this is explained a bit better. Unfortunately, this is how I have theroized how this works, and I could be completely wrong. But I do think this fills in alot of gaps to some questions like:
Why does completely gimp person with HQ staves beats someone with none (the flat MACC% bonus), or...
Why a well geared person can't seem to land spells any better than the next player on end-game events(the level correction), and others...
Which brings me full circle to...
If the former really is indeed how the staves work (I'm inclined to believe it myself), this is often the reason why elemental staves continue to be the best accuracy weapon.
A bit of a disclamer though - I know what you were implying, and you are absolutely correct. End-game wise, fighting higher level mobs, and general game-play where MACC% really matters, staves will probably always win - hands down.
The part I do want you to think about is this -
If your level 75, and walk up to a lvl 1 bunny and you have a choice of +20 MACC or a flat +15% MACC - which is really
+20 MACC at level 1 is an incredible amount, and I'm betting the bonus would probably outweigh the MACC% bonus. Of course, this is only theretical due to the fact that, who really cares? It's a level 1 mob! (The real answer - level correction wins. Even if you had severe negative stats, the level correction formula would give you a tremendous bonus)
However, by looking at this extreme, and swapping roles, you can see how some incredibly high level mobs, such as Kirin, will always be a pain in the ass!