AC: is it worthless?

[SynissterSyster]

{{scrubbed}}

[Malachei]

I once handedly won a lvl 3 arena match against a Wilder using Psionic Shot and Empower Power with Wild Surge to deal ridiculous damage in a single shot. I won it by waiting until he got just about ready to alpha-strike me into oblivion, and then taking total defensive (+6 AC, standard action, readied) while dropping prone (+4 AC vs ranged, free action, part of the readied action), behind cover (+4 AC). Halfway through manifesting the power, instead of aiming for 14 touch AC, he was looking at 28 touch AC... and it was too late for him to abort.

I think that was my proudest single action of my gaming career.

A wonderful example of how nice AC can be if it stacks. My experiences are similar. Readying moves is also nice, if the result breaks line of effect. That's why I love pillared halls. And then you don't need high AC anymore, as well

[sonofzeal]

Please close this thread. Why? Opening post mentions someone whom is banned (which I believe is against TOS). Said person also cannot be included in the debate to defend their position making a side no heard and there for the debate can be slanted to one opinion.

Also another post mentions another forum. People from that forum may or may not be on this forum so the post is either baiting or the rules should say no other forum should be mentioned as the people from the other forum may not be here to defend their position.

Lastly this has been beaten into the ground. All will have their own opinions on the subject and will follow "what is their experience". Thank you.

I've removed the header you referred to. I don't know off the top of my head if it's against TOS or not, but it wasn't necessary for the discussion.

Also, the whole point of the thread is that I haven't heard much debate on the subject. Certainly not nearly as much as on Monk, or Truenamers, or "PF vs 3.5", or whatever. I don't see any other open threads on this issue, and I can't remember seeing any, ever, that cited data and tried to come up with a cohesive argument one way or the other. I'm trying to do that here, and so are some of the people with the opposing stance. People seem to be approaching this with intellectual honestly and openmindedness so far, breaking down numbers rather than just repeating rhetoric. Just because you don't want to engage in the discussion is no reason to close the thread. You're welcome to post somewhere else if you'd rather.

[Doug Lampert]

Mithral Fullplate instead of Breastplate (+3_
Defending Armor Spikes (+1 - +5)
Defending Shield Spikes (+1 - +5)
Dusty Rose Ioun Stone (+1) <- for 5k gp, this is cheaper than upgrading a Ring of Protection from +1 to +2.
Skin of the Defender (+4; stacks with Ammy of NA since it's not an enhancement bonus)

That should be more than enough to keep you golden for quite some time. And that's without dipping outside the SRD. Seriously, I think the lack of education on how to manage AC is behind most of the criticism it gets.

Almost every bonus type in the game can improve AC. There are a LOT of bonus types in D&D.

Enhancement alone can improve it 4 different times (all stacking), enhancement to Dex, enhancement to natural armor, enhancement to armor, enhancement to shield.

Whenever someone starts talking about +5 armor being the best way to improve AC you KNOW they're either near epic, epic, or they've missed the point. You improve AC by stacking a dozen +1 to +3 modifiers.

[Doug Lampert]

How exactly did you count these? Consider for example a Crusader (no ranged weapon prof) that fails his reflex save vs. Entangle and then gets shot to death by some archers he can't reach or retaliate against. Would that count as death by damage or death by failed reflex save?

Asking because in the level range we're talking about (5-10) many spells don't kill, but incapacitate the enemy to such extent as to make actually killing them trivial.

These are PC deaths, unless the crusader has wandered off alone or there is a TPK the enemy DOESN'T finish him off, because the rest of the party is killing them.

[ILM]

Mithral Fullplate instead of Breastplate (+3_
Defending Armor Spikes (+1 - +5)
Defending Shield Spikes (+1 - +5)
Dusty Rose Ioun Stone (+1) <- for 5k gp, this is cheaper than upgrading a Ring of Protection from +1 to +2.
Skin of the Defender (+4; stacks with Ammy of NA since it's not an enhancement bonus)

My vision is colored by my games. I don't play with psionics, and I dislike the concept of a fixed spike glued to your helmet somehow making you much less likely to get hit. I know, I know, magic. I can tolerate a defending shield spike but an armor spike is a bit much for my suspension of disbelief. Stacking both would get you a book to the face in my game. That's just my personal position though, and I can't disagree that it's perfectly RAW (or balanced, for that matter).

Arguing that you just need to use a shield at lower levels, however, is a bit disingenuous considering that everyone and their mother here keeps saying sword and board is the biggest crime you could ever commit in a D&D game after playing a straight fighter. If you do sword and board, your AC may come on top but you're stuck with crap damage while your friendly chargebarian obliterates whatever he hits. It's also impractical for all the casters and archers and what-not, who still occasionally get attacked but need their two hands.

Swiftest correctly points out you should bet on party casters to buff you and get them an offering of pearls of power to compensate, which is cheaper than increasing enchantments. You correctly point out that there are a bunch of ways to get your AC higher than I listed (assuming your DM lets you get whatever equipment you want when you need it, which isn't necessarily a given - neither is the fact he'll stick to WBL, actually). Still, the weaknesses of AC remain:
- you may have more options but you're still sinking craptons of your WBL into it
- however high your AC, spells ruin your day big time and there's increasingly more of them flying around as ECL increases
- if you focus too much on AC and too little on offense, enemies can just step around you and go for the squishies.

Note that I'm not saying you should settle for AC 10 and waltz in naked and stupid. At lower levels, AC is a necessity and should represent a hefty part of your investment in your character. However, I feel that as the game progresses and more options present themselves to the opposition - options that have next to nothing to do with AC - investing more than 150k just so an average CR20 monster doesn't auto-hit you may not be the value for money out there. Plus, cost of opportunity: every ring of protection you get is a ring of freedom of movement or evasion or counterspells that you'll have to skip.

P.S: Full plate, breastplate, whatever. You get an extra 3 armor bonus, and lose 2 from the lower max dex. Big whoop.

[Piggy Knowles]

As an aside, here is a link to a spreadsheet I put together. This lists the highest attack bonus (versus AC only, I still need to do one for touch attacks) of every monster in the SRD, organized by CR, with the lowest, highest, average and median attack bonuses for each CR.

I'd like to add more data points to it, and start incorporating things like touch attacks, etc., but I use it for reference whenever I'm concerned about a character's armor class.

As a general rule, if I'm building an AC-focused character, I want to make sure that my AC is going up by an average of 2-3 points at every ECL in order to stay competitive. It's certainly doable, even in core. A Runesmith/Abjurant Champion wizard I'm building for a game now should be significantly above the curve for AC. But it can be a pain in the butt, and a lot of times I prefer to have characters with only a moderate investment in AC and other ways of avoiding attacks or damage.

[Ravens_cry]

It is interesting to note that the common defences, AC, Miss Chance, and DR, all have different kinaesthetic game feels.
AC is just a miss, either because you hit the armour or they dodged out of the way, but a basic miss. It has a nice solid feel.
DR is that part where the hero stabs the monster, only for it to be a dull thunk rather than a slash.
At high levels it can have an intimidating feel, where the monster is like 'Sure, hit me, it won't do any good."
Miss Chance, ooh, Miss Chance is tricksy. Miss Chance snatches defeat from the jaws of victory.
Miss Chance is where you hit . . but don't.
From a DM perspective, Miss Chance is good, in my opinion, for trickster monsters, creatures who outsmart and outplay the PC.

[JadePhoenix]

Recently my party fought a guy with high AC, miss chance and regeneration. What a pain!

[Ravens_cry]

Recently my party fought a guy with high AC, miss chance and regeneration. What a pain!

Wow, that does sound annoying.
"Even if you hit me, you might not hit me. And even if you actually hit me, I'll just heal."
That fight must have been brutal.

[JadePhoenix]

Wow, that does sound annoying.
"Even if you hit me, you might not hit me. And even if you actually hit me, I'll just heal."
That fight must have been brutal.

The worst part is that he fled.

[Rejusu]

I don't think AC is worthless certainly. But I've always been on the opinion that D&D promotes the idea that the best defence is a good offence. Rather than trying not to get hit it's generally better not to give your enemy the chance (or too many chances) to hit you. Not to mention that with so many defences in the game (touch, flat-footed, and regular AC plus will, reflex, and fortitude saves not to mention things like strength checks for trip/grapple/etc) it's very difficult to guard against all the possible attacks that might be thrown your way while still maintaining good offensive capabilities.

After all you can have as much AC as you like, but if you can't kill your enemy quickly then he's going to get a lot of chances to overcome your defences. On the other end of the scale offence is a bit more universal. Yes the enemy might be able to land a few blows on you quite easily, but when you can turn him into paste in a few rounds it's not going to matter that much.

[LordBlades]

These are PC deaths, unless the crusader has wandered off alone or there is a TPK the enemy DOESN'T finish him off, because the rest of the party is killing them.

Unless the enemy sees some benefit in finishing off somebody while he's down. I can think of quite a few scenarios where it would make sense:

-PCs are storming into something (castle/dungeon etc.) and the archers are in a hard to reach place (wall, ledge etc.). It makes more tactical sense to cause some permanent damage by shooting a guy full of arrows before retreating to the next line of defense rather than moderately wound everybody.

-The Warblade is the best target because the rest of the party is engaged in melee with the team archer's frontliners.

-Mindless/very dumb creatures that have picked the Warblade as target before he gets entangled and can't recognize it would be smarter to switch

[Morph Bark]

A chain shirt, dastana, chahar-aina and +5 animated heavy shield each with +5 defending armor spikes gets your AC ridiculously high, but the cost will be high as well, at 337000 gp for the enhancement bonuses and some trivial money for the gear they enhance. AC will be 44, but the four +5s from the defending armor spikes will also apply to touch and flat-footed AC. That is excluding an amulet of natural armor (+5 for 50000 gp, does not apply to touch), ring of protection (+5 for 50000 gp, applies to all) and Dex bonus. With a cloak of displacement, you can bet on being hit very little, but you'll have many other troubles to worry yourself with.

If we take certain classes, we can discard a lot of this, due to those classes being very capable of buffing AC, often with bonuses that would otherwise overlap with the items above.

[Doug Lampert]

P.S: Full plate, breastplate, whatever. You get an extra 3 armor bonus, and lose 2 from the lower max dex. Big whoop.

You only gave +3 for dex. I can use all +3 from dex in mithal full plate.

He losses NOTHING unless you want to revise your prior estimates up.

[Doug Lampert]

Unless the enemy sees some benefit in finishing off somebody while he's down. I can think of quite a few scenarios where it would make sense:

-PCs are storming into something (castle/dungeon etc.) and the archers are in a hard to reach place (wall, ledge etc.). It makes more tactical sense to cause some permanent damage by shooting a guy full of arrows before retreating to the next line of defense rather than moderately wound everybody.

-The Warblade is the best target because the rest of the party is engaged in melee with the team archer's frontliners.

-Mindless/very dumb creatures that have picked the Warblade as target before he gets entangled and can't recognize it would be smarter to switch

In ALL of which cases he'd still be killed if not webbed! The example was a crusader being shot up while webbed. How is he in any more danger in any of your scenarios because of the web?

These are all straight damage kills where AC would help.

[FMArthur]

AC is never bad, but when frailer characters simply cannot take the 1-in-20 risk to begin with, I just find it more prudent to invest in attack avoidance through other means for them. Those with d4-d6 hit dice or low Con need to form a careful defensive strategy because you should plan for the character to last indefinitely. A 1-in-20 chance to get decommissioned doesn't stand up very well to repeated trials, and encounters for an adventuring career provide repeated trials absolutely.

AC is a focal point of investment for heavy melee characters who plan to get hit, grappled and knocked around and continue the average fight afterward. And since it's relatively inexpensive to assemble lots of small AC bonuses, I recommend doing so on frail characters as well. When it gets more expensive, though, it's not as useful for them compared to other things that prevent the chance for harm or help you survive when something gets past your AC, because it is inevitable without a more thoroughly reliable survival mechanism.

That means Shadow Cloaks (more than one for the day even), Rings of Spell-Battle, Rings of Greater Counterspells, Deathward and Deflection armor/shield enhancements, items of Constitution, various swappable armor/shield crystals, lots of variations on invisibility/nondetection effects, etc. There is a ridiculous number of items, enhancements and spells that can help you ensure survival from modes of attacks to a great degree rather than improve your chances, and IMO they are more valuable than the equivalent amount of gold in AC after you've exhausted the inexpensive means for it.

There is no such thing as a universal set of gear for all characters because you just can't afford everything in normal circumstances. You have to carefully pick and choose defenses based on what your character will be exposed to and what risks they can take with them; armor class is not above this consideration.

[Snowbluff]

Uh... AC isn't worthless. Try playing a game with not armor or any AC and see how far your get.

[Amphetryon]

In ALL of which cases he'd still be killed if not webbed! The example was a crusader being shot up while webbed. How is he in any more danger in any of your scenarios because of the web?

These are all straight damage kills where AC would help.

Note, however, that if the target has been webbed or otherwise stuck in heavy BFC, he's almost certainly being hit against his Flatfooted AC, which is almost always a lower target number.

[LordBlades]

Note, however, that if the target has been webbed or otherwise stuck in heavy BFC, he's almost certainly being hit against his Flatfooted AC, which is almost always a lower target number.

The web was just an example...replace with whatever you feel like. The idea is that most low-to-mid level BFC is largely non-lethal, and I wanted to know if Saph counts enemies mopped-up by damage but severely impaired by various BFC effects as 'damage kills'.

[ILM]

You only gave +3 for dex. I can use all +3 from dex in mithal full plate.

He losses NOTHING unless you want to revise your prior estimates up.

I gave +3 for dex at ECL 5, where your 3,000 gp will be a bit on the short side to pay for 4,000 gp mithril medium armor, let alone 9,000 heavy. I gave +5 at ECL 10 and up.

The same breastplate is now mithril, +3 with a +1 ring and +1 amulet. Your AC is 10 +8 (+3 breastplate) +1 (ring) +1 (amulet) +5 (dex, yay mithril) +2 (you got another class ability, go you!) = 27. Already a tiny bit behind the curve.

[moritheil]

AC is worthwhile if your DM permits it to be worthwhile. If you're fighting ghosts, mages, or wraithstriking trolls all the time, then obviously you aren't going to get your money's worth from AC.

As I said in the other thread (before it got derailed and shut down) you can get an AC over 40 at level 10 if you're willing to spend about half your resources on it. This then increases quite well, even while decreasing the amounts you spend (spending about 17% of your resources on AC at high levels.) A good character builder will no doubt find these numbers feeble, but I submit this partial optimization as a general idea of what is possible if the DM permits you to optimize AC.

Anyhow, you want 40AC at level 13? It's doable before level 10.

1: Be an outsider. This can be attained via Neraph, Otherworldly, Aasimar, or whatever floats your boat.
2: get Alter Self 4/day at caster level 5+ (whether by being a caster, or getting your DM to sell you an item, doesn't matter. Being a caster is better because you will have longer duration.)
3: Either you, or someone in your party, needs to have enough Knowledge (planes) to make a check that reveals the Dwarven Ancestor, with 18 natural armor. You can take this form at level 5. You do lose 1 AC from being Large, but if you are the tank, being Large is a good thing. Do note that at 10 minutes/level, you can easily fight two combats off of a single Alter Self.

...

Supposing you are level 10, you might plausibly have the following AC bonuses:
8 armor + 4 shield + 19 natural +1 deflection + 3 dex + 1 dodge + 5 combat expertise -1/size = AC 40.

You might want to get a lesser metamagic extend rod, an amulet of natural armor, and obviously you'll want the best armor you can buy that doesn't interfere with your buff casting - likely a mithral fullplate or enchanted mithral breastplate - but this is all doable well within budget. You have 49,000 gp; a +4 mithral breastplate is a little over 20k gp, amulet is 1k, ring of protection is 1k. Extend rod is 3k. I've used 25k out of 49k at level 10, that's half the budget on defenses for a character made to tank in 4+ encounters a day. Take the rest of the money and buy save gear and a nice weapon. If you're an Abjurant Champion your shield spell will be automatically extended and quickened, so all you need the extend rod for is the Alter Self.

All of this is off the top of my head. If you took this apart and spent a few hours examining resource allocation you could probably squeeze out another 5 AC, easily. 40 AC is really trivial to get if you permit optimization.

A CR 13 Iron Golem has +23 to attack
A CR 16 Hound Hero Archon has +25 to attack
A CR 17 Frost Giant Jarl has +30 to attack
A CR 17 Marilith has a bunch of attacks at +25.
A CR 20 Balor has +33 to attack.

EL 10: AC 40 earlier in this thread, done with 25k wealth spent on AC (50% of resources at 10, < 25% at 13.) This build actually includes AbChamp levels and does not factor in their bonus into the Shield spell, so the AC is really a little higher than 40. Also, I mistakenly wrote the +4 mithral breastplate as 8AC when it's really 5+4=9AC. So it's actually 41 AC + whatever you get from AbChamp depending on exactly how you build it.

EL 16: 260k wealth, only needs ~5 more AC, trivially easy as all you do is progress the +1 items in my EL 10 build. (Make ring of protection +3, amulet of natural armor +3. Also note that +5 AC comes from Ab Champ levels, which you will have finished.) New result: 50 AC. Difference in GP: 18k *2 - 4k = +32k GP; total 25+32 = 57k GP. 57/260 = 22% of your resources spent on AC.

EL 17: 340k wealth, needs 50AC. We already have 50AC at EL 16. 57/340=16.7% of wealth spent on it AC.

EL 20: I don't really believe a balor's most dangerous attack is physical, as any reasonable DM will probably swap a feat for Quickened SLA:Greater Dispel, and hit you with that + Implosion or something in the same round.

Still, AC 53 needed? We can do better. Progress the ring of protection and amulet to +5. Swap in Greater Mage Armor +11AC which is 2 better than your old armor, as you probably want to use your Armor +s for bonuses like Soulfire or Heavy Fort or something. That's 6 more AC for AC 56. 2*(50k-18k)=+64k, 57+64 = 121/760 = 16% of wealth spent on AC.

This is far from optimized for 20. You'll want to go through splatbooks to actually optimize it. For example, if you want to use Greater Luminous Armor from BoED it's AC13 and a -4 to attack you. This is effectively 6 AC better than using Greater Mage Armor, so you basically have an AC of 62. EDIT: And you can get it by level 10, so you could go back and bump up all the other numbers by 6 if you wanted to.

To compare, a great red wyrm has a base attack of +40, so without the DM optimizing it a bit, it could only hit you on a critical. (Unlike the balor, it wants to attack you physically. It can't compete with your group for caster levels and you're probably all immune to fire.) EDIT: It was pointed out that dragons will have a lot of bonuses to pile on to their base attack. So no, you may not be totally unhittable. But the point still stands that you can get AC to a useful level.

I'm sure this isn't optimized at all because I recall someone hitting 108+ AC at 20 (shapechange was involved.) But you get the idea. AC is there for the taking if you want to build for it.

Dragons are currently being ignored because they are more highly optimization-sensitive than other monsters, take way more effort to calculate for, and are generally considered unrepresentative of their CR anyway.

Not a problem right out of the box, because you can get your AC well over what you need to resist their attack bonuses. But they are arcane casters, so they can wraithstrike, make you save-or-lose, and do other things, if the DM wants them to.

Personally, I think the numbers here are pretty conclusive. It should be easy for most characters who care about it to get AC into the effective range at lower levels. At lvl 10 there's actually something of a peak, where AB has scaled faster than AC likely has, but even then a committed character can manage. But after that, AB hardly seems to scale at all, and the vast wealth of the higher levels can easily catch your AC back up without leaving you short in other areas.

With a little optimization you can get AC at all levels to be untouchable by nearly all challenges of your ECL. You are legitimately using up your feats, spells, gold, and other resources to do this, so I don't see it as any worse than other focused characters (consider rogue types with the Dark template for Hide in Plain Sight and around +40 to hide checks by level 8.) The way I showed above uses a trick, but there are other tricks available even if this one isn't. (The Dwarf Ancestor is unusually good, but even without it Alter Self can still give you 12-15 natural AC to work with unless the DM bans a great many Outsiders or houserules the spell itself.) The bottom line is, as with most things, it's up to your DM and whether or not he or she will let this fly in the campaign.

[moritheil]

Recently my party fought a guy with high AC, miss chance and regeneration. What a pain!

I built a recurring vampire boss, totally by RAW, to have an AC of 40 at CR 6. Even when the party was level 8-10, they still couldn't take him down without a lot of tactics and preparation (lesser globe of invulnerability to negate his cleric buffs, etc.) After this epic battle in a desecrated temple where the melee were mostly KOed, half the party had negative levels to varying degrees, and the mages and artificer burned through a lot of resources to get him . . . he simply turned gaseous and fled. They did achieve a win for the campaign, though.

[moritheil]

Either you invest a lot in AC, or you invest next to nothing into it. At higher levels, there seems to be no middle ground and armor is only for the enhancements. If monsters buff up or you play against NPCs with PC class levels, then you need a much higher AC for it to be effective.

Yes, this is because AC is an all or nothing proposition: either an attack hits, or it misses. It's inherent in the system. Also, something important you touch on is that it depends on what you're up against; I firmly believe that it's up to the DM to be permissive of things that players are trying to do. If your DM is OK with you not being hit in 90% of combats, because that's your character build, then great; if not, and your high AC is his cue to send in armies of ghosts and Psychic Warriors, then you probably don't want to put your resources into optimizing AC.

IIRC, there was a formula that monsters generally should have an AB of CR+13, so that an AC of CR+23 or higher is okay and having an AC of CR+33 is unnecessary, unless you absolutely want to dodge True Strike attacks.

And if the DM is willing to send True Striking monsters, you can bet the Wraithstriking monsters aren't far behind. AC working well effectively depends on DM permission.

[navar100]

There is potential for some DMs to make AC worthless on purpose regardless of RAW numbers. They reflexively coil against the idea that it's ok for monsters/bad guys to miss, at least on the first attack iterative. If the bad guy misses because the DM rolled a 5, fine, but to miss when the DM rolled a 17 makes particular DMs shudder. Combats are to be "challenging". If there's no threat of death, what's the point? That sort of thinking. These DMs need "auto-hits" on some roll and above, say Natural 15 and higher.

Such DM need to learn it's ok for a PC to have an AC just that good. Such a character is supposed to be hardly hittable for the non-BBEG. The player knows the BBEG will hit him. That's expected regardless of his AC. The player wants the high AC so that the mooks and lieutenants miss a lot. That's a feature, not a bug, these DMs need to learn.

[FMArthur]

It seems unsporting from the perspective of DMs who don't target other defenses enough. Variety in targetted defenses is just one of the things to keep in mind when building dungeons and encounters. It's not even about making massive AC investment less awesome so much as giving accrued benefits for other sensible defense methods their value. The lady with a cranked Will save should get to see it in use, you know?

Pure mundane monsters and opponents all day every day is the easiest DMing, and it screws up values of defenses.

[Ravens_cry]

The worst part is that he fled.

Well, that sounds like an all around unsatisfying experience I must say.

[Doug Lampert]

Well, that sounds like an all around unsatisfying experience I must say.

Eh? How can victory be sweet if it's handed to you? An intelligent foe that wants to live should ALWAYS consider things like fleeing, negotiating, surrendering, or have backup and escape plans.

You need that for versimilitude if nothing else. People mostly don't want to die. Those who become undead abominations to live are even less likely to want to die.

I've had quite a few villains escape 3-4 times, and the PCs are usually overjoyed when they encounter him the 4-5th time, and casually curb-stomp him prior to any escape because they've leveled far faster than he has :) (and likely they dimensionally anchored him first thing just in case).

PCs often don't FEEL awesum because their foes are tougher as they level, encountering the same thing repreatedly is one way to show how much they've improved.

In any case, figuring ways to track and pursue is part of the game.

[Ravens_cry]

No, you should earn your victory, but the combination of tactics with a kinaesthetic with high potential frustration value (miss chance and regeneration) along with the guy running away, combines for a feeling that time was just wasted, that nothing was accomplished, and when players feel they accomplished nothing, it can be very unsatisfying.
It's the same reason why you don't throw overwhelming enemies at PC, only for NPC to bail them out.
Now, there is times when such feelings are in fact desirable to elicit, but you should make sure that is the key you want to hit.

[Emperor Tippy]

That's the side effect of not being good enough, paranoid enough, lucky enough.

You are adventurers; you survive (if the DM is playing the world and enemies remotely honestly) by being exceptionally skilled, incredibly paranoid, a firm believer in careful planning (with multiple contingency plans), and luckier than the next guy. Sure, the form of your plans and options changes with your level but the kind of detail and planning that is required doesn't.

There is a reason, IC, that surviving to level 5 makes you a legend and surviving to level 20 makes you a demigod (effectively). An equal CR encounter is supposed to have a reasonable chance of killing you (I believe 1 in 5 is the official number but I can't remember where from). You are supposed to face 13 such encounters per level.

You will not survive unless you do everything possible to slant things in your favor and, IC, cheat shamelessly.

At level 1-5 you should be scouting constantly with your scout (Rogue, Factotum, familiar, psicrystal, etc.) and shouldn't enter any fight that you haven't already planned for and laid the ground work for. At later levels more exotic methods are usually necessary but the same principal applies.

In the end, if you let your enemy act in combat then you have messed up spectacularly. If your enemy is surviving the surprise round then you are depending on pure luck to survive. And that makes AC a near total waste; spend the money you would spend on it on ways to ensure that you can get that surprise and kill your enemies in the opening move.

---
Low OP or High OP is irrelevant in the end, play-style, gaming philosophy, and approach to the game are all far more relevant.

Relatively few people find it fun and interesting to actually role play their characters realistically (largely because they don't find that kind of paranoia planning fun) but if you do, a lot of the game makes a whole lot more sense.

Think about it, any level 1 PC is skilled and capable enough to walk into any city and get a job living a life of luxury. That means that anyone who goes adventuring has 1) a sufficient motive to risk their very probable death to achieve or 2) is flat out insane. And that's at level 1. Anyone adventuring past level 10 or so has a goal that they are willing to risk the obliteration of their soul to achieve, has raised paranoid planning to an art form, has the luck of the gods, is incredibly skilled, and is at least mildly insane.

Adventuring is a job you simply don't survive performing unless you are the best.