What effects would allowing same-class multiclassing have on interclass balance?

Question

In Pathfinder, taking multiple copies of a class is controversial from a rules-based perspective. Most of the groups I have played with allow same-class multiclassing in 3.5 but not Pathfinder. I don't see much reason from a balance perspective, however, to treat the systems differently in this regard. What changes in Pathfinder when same-class multiclassing is allowed, and in particular what balance issues become more prominent? I'm particularly interested in the interplay between the following class options:

• Fighter
• Wizard
• Cleric
• Rogue

as well as the consequences of being able to have separate, parallel levels in multiple archetypes of the same class.

Answer 1

First of all, 3.5 and Pathfinder are really in the same place here as far as the rules are concerned—this was never allowed. I have never heard of it before as a houserule either, but that’s neither here nor there.

Secondly, there definitely could be weirdness in both systems with such a houserule. Repeatedly talking the first level of wizard to sell the starter spellbook (depending on the interpretation of that rule, anyway), for example. More cleric domains. Both ranger combat styles. Whatever. None of these is necessarily “too good to allow” but they all do run contrary to the concepts behind those classes.

But thirdly, and most relevantly when comparing Pathfinder to 3.5, where you had no trouble with this rule, Pathfinder punishes multiclassing harshly in almost all cases. While there may be a (very) few cases where such a thing is advantageous, in most cases it’s a very bad idea to actually use this rule even if you allow it. From that perspective, I would be leery of it as a GM—either the player has some abuse in mind, or, honestly more likely, the player is about to walk into a trap. Pathfinder has more than enough traps as it is, I don’t want to add any.

The other Pathfinder-specific issue is archetypes. Archetypes are all-or-none, and many of them take advantage of that by balancing superior trades at some levels with inferior trades at others. Multiclassing already makes it possible to skip balancing factors (which doesn’t really matter mostly because of the problems with multiclassing), but with this it is getting very close to having your cake and eating it too. I do not know that this would necessarily cause problems but it is something that would worry me about attempting this.

Any player who asked me for this would instead be asked what they are trying to accomplish, so we could work out together an appropriate solution—which might be the multiclassing you describe, or very similar to it, but also might end up being something else entirely, and in any event would not necessarily open up such multiclassing to everybody in general.

Answer 2

Probably weaker, but possibly stronger.

---

Let's look at the classes you mentioned specifically, and take them to level 5. Since this system is probably only useful for the purpose of archetypes and other class-specific choices (like domains), I've added a note for archetypes as well:

• Fighter 1 (x5) - 7 total feats, +10 base Fort save, +0 base Ref and Will saves. 7 feats is a lot, but your fighter is more limited than normal 5th-level fighters (doesn't qualify for Weapon Specialization or anything that requires bravery). Very high Fort save but worthless other saves, so there's some unbalance there. Overall, this probably breaks even.

  • Archetypes - Most fighter archetypes don't kick in until level 2, so same as above. The few that do trade in their 1st-level feat arguably get something better in return, and giving up armor proficiency does nothing here (since the fighter will have four other copies of it), so the clever fighter might wind up stronger.

• Wizard 1 (x5) - Five sets of level 1 school powers, lots of 1st-level spells, +0 BAB, +10 base Will save, +0 base Fort and Ref saves. This one is complicated, and will require some GM adjudication. Can the wizard choose the same bonded item five times, granting one item that can spontaneously cast five spells? How would this interact with the Magical Knack trait? Even with the most generous interpretations, though, the wizard is severely underpowered compared to normal 5th-level wizards. No fireball or haste or dispel magic, and many level 1 school powers aren't very useful at level 5 (and forget using any of them that require an attack roll). Very high Will save but worthless other saves. Overall, this is weaker.

  • Archetypes - Most wizard archetypes give up the arcane bond and/or specialty school, but won't gain anything in return that will make up for the power differences mentioned above. Still weaker.

• Cleric 1 (x5) - Between five and ten sets of level 1 domain powers depending on GM (a deity only grants so many domains), five sets of channel energy 1d6, lots of 1st-level spells, +0 BAB, +10 base Fort and Will saves, +0 Ref save. Game-breaking Will save (since Wisdom is probably the cleric's highest stat) but worthless other saves. Similar problem to the wizard; no lesser restoration or animate dead (if you're into that) or remove curse, weak domain powers-- and since most domain powers require a melee touch attack, they're unusable anyway. Doesn't qualify for anything that requires channel energy 2d6 or higher. Overall, this is weaker.

  • Archetypes - Virtually all cleric archetypes give up at least one domain; some also give up channel energy. With the already-limited number of domains granted by each deity, the clever cleric might be able to squeeze out some extra power here, but won't be able to make up the difference. Still weaker.

• Rogue 1 (x5) - Sneak attack +5d6, five redundant copes of trapfinding, five redundant copies of Weapon Finesse (if Unchained), BAB +0, +10 base Ref save, +0 base Fort and Will saves. Getting sneak attacks to land will be difficult, but when they do... wow. The damage output of a 9th-level rogue at level 5. The clever Unchained rogue player will retrain four of his Weapon Finesses, giving him more feats to work with than a 5th-level fighter. The Core rogue trades out evasion, uncanny dodge, trap sense, and two rogue talents for an extra +2d6 sneak attack; the Unchained rogue also trades out Dex-to-damage for one weapon, which definitely hurts (and a skill unlock, which hurts less). Overall, the Core rogue probably breaks even; the Unchained rogue is definitely weaker.

  • Archetypes - By far the most traded-out class feature for rogue archetypes is trapfinding. Very few give up anything else at level 1, so the breakdown above will likely be unchanged. Since multiple copies of trapfinding don't stack anyway, this is all to the good and the rogue can easily add some extra power this way. The Core rogue probably winds up stronger; the Unchained rogue might break even.

• Worth mentioning: Barbarian 1 (x5) - Lots of rage rounds, +10 ft movement speed (based on the wording of Fast Movement they probably wouldn't stack, but debatable), +10 base Fort save, +0 base Ref and Will saves. Full BAB means the barbarian hits just as hard as anyone else, but take a look at that rage. With a 16 Con, the barbarian is looking at 35 rounds of rage per day-- more than twice the amount a 5th-level barbarian has to work with. So the trade is two rage powers, trap sense, and uncanny dodge/improved uncanny dodge for all-day rage. Pretty good trade, if you ask me. Overall, this is probably stronger.

  • Archetypes - Some barbarian archetypes trade out fast movement at level 1, which, if the GM rules they don't stack, is only for the better. The clever barbarian can piece together archetypes that leave rage untouched but give up fast movement or armor proficiency (which is a non-trade; see Fighter) and become even stronger.

---

Conclusion: Virtually all spellcasting classes will wind up weaker this way than simply advancing like normal, so the player will probably be walking into a trap. Martial classes might wind up about the same but could be even stronger than their normal-advancement counterparts. Even then, though, there will be some unbalance issues with their saves-- unless they're a Core monk (which will have great saves but be mostly useless at anything else) they'll have one (or two) very high save and two (or one) very low saves. Overall, since Pathfinder wasn't designed with this option in mind, allowing it is best avoided.

Answer 3

Most answers only consider taking the first level multiple times but when allowed to multi-same-class you are not limited to 1st level.

Fighter: Compare a level 11 base fighter with a 3/5/3 base fighter/weapon master/ brawler fighter

Base fighter: 6 Feats, Bravery+3, Armor training 3, weapon training 2; saves: 7,3,3.

Multiclass: 7 Feats, bravery+2, armor training 1, weapon training 1, Weapon Guard, Reliable Strike, Close Control, Close Combatant; saves: 10,3,3

=> 1 more feat, 1 less bonus from bravery, only armor training 1, weapon training and close combatant should stack (when selecting a close weapon) for +2 hit +4 damage compared to +2/+2 with one group and +1/+1 with a second, he gets weapon guard, Reliable Strike and close control in addition and has a +3 better fort save. For me a net win by some margin.

This example not even used archetypes that trade away armor prof, class skills and the like.

For the wizard the "biggest" issue I'd see is going wizard 1/wizard x and selecting a bonded object and a familiar.

Answer 4

If we're looking at the capabilities of the character itself, then the face value capabilities of the same-class multiclasses that you requested are as follows:

Fighter

At level one, a Fighter gets a bonus feat, and nothing more. If the character were to keep taking a level one Fighter each level, then over the course of the development the character would on average gain 1.5 feats per level instead of 1 and would have full BAB, but no bonus to non Fortitude saves and any other feature of the Fighter. Seeing as Fighters already have a great many feats, having even more at the cost of normal Fighter advancement skills seems woefully underpowered.

Wizard

As mentioned earlier, Wizards would in theory get more of their tangibles from selecting the same class. I can't speak to the potential power of having multiple bonded items or familiars as I am not a Wizard player, but the early game money from cycling the spellbooks and the vast number of low level spells per day seems overshadowed by the drawback of not getting higher levels spells quicker. Not to mention, as only one Wizard class could be leveled at a time, in the long run only one familiar would be particularly useful. The other would fall off very hard.

Cleric

While a player may be able to select multiple domains, they would only reap the benefits of how much they leveled each Cleric due to the powers obtained through progression. On top of this, Channeled Energy does not stack between classes, so even if a player wanted to enhance both Clerics, or however many he has, his Channeled Energy power would never surpass that dictated by his highest Cleric level.

Rogue

The only benefit to this would be extra sneak attack damage early. Unless one really cares about having 2d6 sneak attack at level 2 instead of 3, this seems completely worthless because rogue talents don't come into play until level 2.

In general, it seems like a strategy like this is quite weak due to the fact that any early benefits are neutered in the long term because there's no progression on them. Perhaps there's more abuse potential if the campaign is short and the characters don't advance greatly, but that might have to be play-tested. I think archetypes would also be affected the same way as the normal classes due to the progression issue.

If I were a GM, I would probably prohibit this style of leveling not because of the abuse, but because I can't see a way that a player wouldn't be setting themselves up for a disappointment.