**
When optimizing a character, it is key to identify what you will be spending most of your time doing.
**
In this case, it seems like you want to make a melee character who fights in melee. The problem is, as a Magus, you will be combining spell use with melee attacks, and relying on spells to do your damage or to attach Saving Throw based effects. However, your melee attacks still need to hit to apply these spells. So we need to optimize two things - your to hit and your spell effects.
Why not rely more on melee damage? As a non-Fighter and non-Barbarian and non-Rogue, your melee damage is low. You can't easily use PA to increase your damage as your to-hit is key. Ergo, relying on pure melee damage is a bad idea with this build.
To-Hit
To Hit is made of three parts -
- Stat added to to-hit
- spell bonuses and gear bonuses to hit
- feat bonuses to hit
Spell and gear bonuses are simple - you select spells (such as Haste and Cat's Grace) that give you bonuses to hit and try to have them cast before combat goes up. Additionally, you buy or steal or barter or trade or select items (gloves of dexterity, magical swords, banners, duelist's gauntlets, sashes etc) that increase your to-hit. It's usually fairly binary, and choices between say, a great save or die to apply via sword hits or a spell that increases the chance of hitting, can be decided upon by comparing the great spell to your pre-existing spells, thinking about how often you hit currently, etc.
Stat added to hit is also fairly simple. You want more of the stat that adds itself to your to-hit, within reason, again, by comparing it to other things you might get with the same money.
Feat bonuses are slightly more complex, as some feat chains offer debuffs which can affect to-hit, and computing the result is non-trivial. However, as you aren't a Fighter, you can't afford enough feats to make the Improved Trip line worthwhile (Expertise, Trip 1, Trip 2, Fury's Fall), as you aren't a rogue the Shatter Defenses line isn't worth it, and most of the others aren't great shakes. Weapon Focus is an okay choice as it pays off immediately, doesn't require anything else, and is easy to use with your weapon choice of 'always scimitars'. The Two-Weapon Fighting chain is mandatory. That leaves you with very few feats, so the only real 'bang for buck' feat choice is Dazing Assault. Daze is a great status effect and worth applying.
Spell Effects
Here we start to run into problems. In the proposed build, you have 4 lost caster levels. In a partial casting class like the Magus, that's effectively suicide. Unless there are some vastly powerful low level spells in the Magus list that you are going to be able to rely on, and you're taking the levels in those classes later - which there is no sign that you are, as they appear to be low level dex and BAB boosting classes.
Unless you are going to dual-wield scimitars and take power attack/twf chain etc, I recommend you only take one level instead of four. Taking a second level later might be workable, but 4 levels of delayed casting as a Magus will suck.
If you have 3.5e backwards compatibility, things like the Abjurant Champion prestige class might help you out in getting a better BAB while maintaining your casting progression (although you still lose out on arcane pool and arcana - the only worthwhile things about advancing pure magus).
Now, as for advancing your spell effects, we have two main things to worry about - save DC and damage.
Save DC is increased by int and various feats, none of them amazingly good. Spell Specialization is a good choice, as is spell Focus in a school you have multiple good options in (like Conjuration or Transmutation).
Damage is increased by CL and metamagic feats. The Gifted Adept and Metamagic Master traits are key to increasing the power of a spell you'll use a lot, such as Shocking Grasp. Otherwise Spell Focus, Mage's Tattoo (Varisian Tattoo), Spell Specialization, can all increase your CL. At lower levels, an Empowered Shocking Grasp for 5d6 x 1.5 damage is crazy at level 4 or whatever. At higher levels, an Intensified Empowered Shocking Grasp out of a 2nd level spell slot for 10d6 x 1.5 damage is great in addition to a full attack.
Taking a level in Crossblooded Sorcerer (Draconic/Orc, or anything else that boosts damage) can greatly increase the damage of touch spells you use also.
Once you have all that handled, damage, status effects, how you are applying them, then you can think about ratios of int to dex.
And in this case, mathematically speaking, you want 2 more points of Int than Dex at any one time, but both are important.
You are overthinking this especially since it's backstory, not a real situation in play.
A 17th level wizard has a +10 Will save base. (He loses a level from the rez, but it's still +10.) That's a 25% chance to fail the DC 15 Will save, assuming he's not super wise (most wizards that embrace lichdom aren't). So... He failed it! Done and done.
You're depicting a past event, so why does it matter? How does it change things if his chance to fail is 25%, or 50%, or 95%? It's never zero or 100% (since 1 always fails and 20 always succeeds) so you are trying to push precision into something that doesn't need it.
If you're looking for background color, then sure he "cast various spells to lower his Wisdom/Will save/resistances/whatnot," though it's mostly impossible RAW to land them while he's dead (readied action or quickened before slapping on the helm, perhaps). Or slapped another cursed item on him to that effect (vary something like a robe of powerlessness to hit Wisdom instead of Int). But in the end, it really doesn't matter what the real number is - you are as the GM just saying "he failed," whether he had an effective +20 Will save or a -10 Will save.
It's easy to get stuck in a rabbit hole as a GM - you have a lot of more important things to prep, a lot of things that will make real different to your players' enjoyment of the game, than this kind of obsessive detail. You are effectively making your game worse by choosing to spend your time in the hole. Pull yourself out, write the backplot, and move on.
Best Answer
It sounds like the foe is always going to succeed on all his saving throws. That means developing another method of
Getting a Target to Fail a Willpower Saving Throw
The ioun stone (flawed mulberry pentacle) (8,700 gp; 0 lbs.) causes its possessor to take a -2 penalty to Willpower saving throws. According to the glossary penalties
In other words, ask the DM if the penalties from multiple mulberry pentacles stack (maybe couched as, "Were my character to have several of these spinning around his head, would he suffer the penalty multiple times?" if the DM-player relationship is particularly adversarial--see below). Then determine if creatures can be involuntary possessors of ioun stones: "When a character first acquires a[n ioun] stone, she must hold it and then release it, whereupon it takes up a circling orbit 1d3 feet from her head," says the description, but later the text says that ioun "stones only float when sent spinning around the head of an intelligent (Int 3+) creature." Emphasis mine. Experiment on the cohort. If penalties stack and involuntary possession is possible, the spell invisibility [illus] can affect objects; a level 17 alchemist should be able to manufacture such ioun stones and render them invisible, while a level 20 ninja may be able to succeed on a sufficiently high enough Sleight of Hand skill check to send spinning around a sword saint's head a handful of invisible mulberry pentacles without the victim noticing.
That's all I have that don't also require another saving throw or conspirator (like MrLemon's witch). If the first method doesn't work, the ninja's still facing a level 20 character who likely has multiple methods of rerolling his saving throws (e.g. a golf bag full of luck blades). Thus, even with a limited wish and a wish involved, success is far from guaranteed.
That's because failing a saving throw ends characters' careers. It should be difficult--if not impossible--to make a character fail a saving throw outright by any means. We readers--at this point in the question's evolution--don't know what spell the cohort's casting from that scroll, but a failed saving throw could lead to the success of spells like dominate monster [ench], imprisonment [abjur], magic jar [necro], or trap the soul [conj]. Neither the DM nor you want those kinds of effects possible sans a saving throw, or even with a saving throw yet obscenely reduced.
The Foe's Probably Invulnerable Anyway
I'm speculating here, but if the DM's already told you the foe always acts in the surprise round, is never flat-footed, and can't be flanked, then--while that combination of abilities is possible--it sounds like the DM's subtly telling you No, and he wants you to find the foe's Secret Hidden WeaknessTM instead of confronting the foe directly. That shouldn't stop your attempts to confront the foe directly, but expect such confrontations--no matter how convoluted your plan--to fail until your ninja's found the MacGuffin or mastered the secret technique or whatever. It might be worthwhile questing for that rather than a method to make the foe fail a saving throw.