Reflections about multiclassing


Pathfinder First Edition General Discussion


Hi guys,

After seeing some builds on the board I was wondering about how you do the maths when multiclassing. I am just speaking about BAB and saves here, the rest is obvious.

I see two methods when multiclassing, one where you just add whatever you find in the table at the appropriate level for each classes, one where you split the tables in components (full BAB, 3/4 BAB and 1/2 BAB; good saves and poor saves) then you reassemble them in each categories before summing them.

First method: you just add the table lines

Example method 1:
Example: Bbn 1, Clr 2, Drd 2, Wit 1, Wiz 1

We have:

  • Bbn 1 (BAB +1, Fort +2, Ref +0, Will +0)
  • Clr 2 (BAB +1, Fort +3, Ref +0, Will +3)
  • Drd 2 (BAB +1, Fort +3, Ref +0, Will +3)
  • Wit 1 (BAB +0, Fort +0, Ref +0, Will +2)
  • Wiz 1 (BAB +0, Fort +0, Ref +0, Will +2)

So BAB +3, Fort 8, Ref +0, Will +10.

With this you can have great saves, but if you have a poor one he will stay that way.

Second method: you add like this BAB = (N*1+M*3/4+P*1/2), Save = (X*good+Y*poor)

Example method 2:
Example: Bbn 1, Clr 2, Drd 2, Wit 1, Wiz 1

We have:

  • Bbn 1 (1*BAB, Fort G, Ref P, Will P)
  • Clr 2 (3/4*BAB, Fort G, Ref P, Will G)
  • Drd 2 (3/4*BAB, Fort G, Ref P, Will G)
  • Wit 1 (1/2*BAB, Fort P, Ref P, Will G)
  • Wiz 1 (1/2*BAB, Fort P, Ref P, Will G)

Where G = good save (2+lvl/2 rounded down), P = poor save (lvl/3 rounded down).

Which gives us: BBA = (1*1+4*3/4+2*1/2), Fort = (5*G+2*P), Ref = (7*P), Will = (6*G+1*P).

So BAB +5, Fort +4, Ref +2, Will +6.

With this you have a better BAB, better poor saves, and no great saves. Moreover base saves will always be between +6 and +12.

I personnally use the later, mostly because my first D&D DM used it and I grew to appreciate it. I am notably fond of the saves framing.

I was wondering about what you actually use? And if you can (gentlemanly) argue about your point of view I would like to listen.

P.S.: please don't use RAW, house rules or anything like that. I just want a debate about pro and con, not about legality.


Con: You have to take care of fractions. Which you will have to do anyway, if using some of the alternate Favored class bonuses.
Pro: About everything else.
Result: For simplicity's sake, things will probably stay the same.

BTW: For the saves, I'd recommend another approach.

Suggested Save Progression wrote:
Save = 1/3 (sum of bad classes) + 1/2 (sum of good classes), with an additional + 2 if any of your classes had good saves.


If there was any doubt as to the 'official' method, it is the first one. I generally prefer the second, but multi-classing is often deeply unsatisfying in Pathfinder.


By the books, it's Method 1. Method 2 is a commonly-used houserule though.

And ninja'd by the evil ex-prez.


Evil Lincoln wrote:
If there was any doubt as to the 'official' method, it is the first one. I generally prefer the second, but multi-classing is often deeply unsatisfying in Pathfinder.

I was aware of that fact, just wanted to know people opinions about them.

For the part about unsatisfactionI believe it greatly depend on the party, the campaign and your vision of your character. But it will probably not be optimal. Using gestalt is probably better.


I don't know anyone who would use the second, fractional, method. It wouldn't fly with any of the judges I know, nor would I allow it myself.

My opinion is that players are trying to squeeze every last possible benefit from the rules. It's an escalation, comparted to the levels of optimization that we normally use, and therefore umwelcome.

I'd much rather run and play games somewhere in the middle where the rules expect you to be. When you start to leave the sweet spot, you have to make more adjustments, and then more adjustments, and so on...


Is this for when you create a multi class at a set level or when you start at 1st and add new classes as you gain xp ?


rkraus2 wrote:
My opinion is that players are trying to squeeze every last possible benefit from the rules. It's an escalation, comparted to the levels of optimization that we normally use, and therefore umwelcome.

The base saves will always be between +6 and +12 at level 20, I don't see it as humongous optimization more like increased survivability because poor saves will never be <(lvl/3 rounded down). BAB will be better but not drastically.

The Saltmarsh 6 wrote:
Is this for when you create a multi class at a set level or when you start at 1st and add new classes as you gain xp ?

I use the fractional method for both. It takes a little more time when leveling but not as much as choosing feat or spells (my players don't plan much).


I use a different system will put up an example when I finish work but end result is fairly similar to your 2nd method

Liberty's Edge

Wow, I've never seen or heard of anyone using the second method or anything like it.

everyone I know or have ever played with uses the correct method as outlined in the rules (method 1 in your example)

I actually don't see any problem at all with the way saves and BAB work out using the official method ...


Did I let the impression that I have a problem with the official (tables) method? If so let me rephrase it: I don't have any problem with the official method, I was just asking about which method you use and why.


Evil Lincoln wrote:
If there was any doubt as to the 'official' method, it is the first one. I generally prefer the second, but multi-classing is often deeply unsatisfying in Pathfinder.

Agreed. That said, for amusement I'd like to share that I'm playing a RAW-method fourth level character that's proved rather amusingly fun and effective.

She's a Cleric of Callistria 1 / Barbarian 1 / Inquisitor 1 / Psychic Warrior 1. Total BAB of +1. Saves are darned nice, she's got lots of low-level spells and powers, tonnes of hit points, and has a lot of flexibility in what she does each round. Having true strike three times a day makes her combat maneuvers really reliable, and she uses her scorpion whip to great effect. Tripping, disarming, stealing... you name it... when it needs to get done she does it. Being able to do an expanded raging true strike-augmented disarm attempt at reach while stacking in a judgement is... astounding.

But in general multiclassing isn't nearly as effective. In my opinion the problem is the relative lack of ways to maintain progression of abilities. The above character gets four rage rounds a day and only six channel energy uses. Ideally there would be a feat that would let her use her hit dice for some progression. Extra Rage doesn't really cut it and the amount healed by her channels won't ever increase even if she took Extra Channel. There should be more multiclass feats like Boon Companion.

Liberty's Edge

Cinabre wrote:
Did I let the impression that I have a problem with the official (tables) method? If so let me rephrase it: I don't have any problem with the official method, I was just asking about which method you use and why.

No problem!

I've just never heard of anyone even coming up with an alternate way of doing it ... I guess I was more curious why anyone would want to come up with a different way - the official way seems perfectly logical and balanced to me.


The fraction method is less punishing when taking a second poor/medium BAB (or poor save) class, since you actually get to +1 BAB at 2nd level (which is when you'd get it if you didn't multiclass) as opposed to at 3rd level.

Plus, it's somewhat less powerful when taking multiple good save classes, since you're not stacking the +2s at first level of each class.

So, those are the reasons some prefer to use it. Personally, I see the merit of both systems.

Edit: I'm sure it existed before then, but the fraction system was published in 3.5's Unearthed Arcana, as an alternative rule.

Grand Lodge

Evil Lincoln wrote:
If there was any doubt as to the 'official' method, it is the first one. I generally prefer the second, but multi-classing is often deeply unsatisfying in Pathfinder.

In 3.5 multi-classing was the bomb, but conversely staying single class in any class was the loser. (In certain classes like the Cleric or the Sorcerer it was epic fail)

It's a zero sum game. Either multi-classing has significant tradeoffs, or single classing sucks. It's clear what Paizo has chosen in this balancing act.

Liberty's Edge

Out of curiosity, why was multiclassing in 3.5 the bomb? It's been years since I've played 3.5 (I don't even have my books any more) but wasn't it pretty much the same as it is in Pathfinder? Am I forgetting something?


Primarily because many classes didn't get any additional bonuses for staying single-classed. Sorcerers and clerics were the worst offenders (neither gained any new class features after level 1). So, there was no reason to not multiclass or, more accurately, to take a prestige class.

Grand Lodge

And there were prestige classes that were extremely munchkin material.

Abjurant Champion I'm looking at you!


Marc Radle wrote:
Out of curiosity, why was multiclassing in 3.5 the bomb? It's been years since I've played 3.5 (I don't even have my books any more) but wasn't it pretty much the same as it is in Pathfinder? Am I forgetting something?

One: you stacked everything.

Cleric 1/Ranger 1 give what? +4 Fort, +2 Reflex, +2 Will, +1 BAB. Not to mention skills.

two: classes were front loaded.

Three: 3.5 had better Prcs, PF has barely any in main books.
But it is otherwise exactly the same as PF.

Community / Forums / Pathfinder / Pathfinder First Edition / General Discussion / Reflections about multiclassing All Messageboards

Want to post a reply? Sign in.
Recent threads in General Discussion