11
\$\begingroup\$

Many times, I've seen the case made for a specific prepared caster class being superior to its spontaneous equivalent (e.g. Wizard vs Sorcerer, Cleric vs Favored Soul, anything vs its Uneathered Arcana spontaneous caster optional rule version) and said case is usually based on the spontaneous caster happening to be one level's worth of spell progression behind the prepared version. However, in my experience, the wider community considers prepared casting to be always superior to spontaneous casting. Why is this? The specific cases alone aren't proof of a general rule and all else being equal, spontaneous casting ought to be superior. What argument, if any, is so strong that it knocks down all of the spontaneous casters?

\$\endgroup\$
14
\$\begingroup\$

This could be a very difficult question, if things were not so lop-sided, but they are. For example, one very simple illustration: compare a sorcerer’s spells known to a wizard’s spells per day.

Aside from cantrips, the numbers are extremely similar. For most spell levels, when the sorcerer first accesses that level, they know 1 spell. Likewise, a wizard gets 1 spell per day of a new spell level. The sorcerer’s spells known for a given spell level scale, for most, to 4—which is the same number that a wizard’s spells per day scales to. For 1st- and 2nd-level spells, the sorcerer eventually learns 5, but that’s more than counterbalanced by 6th-level and higher spells stopping at 3. But still, “around 4.”

What does this mean? While choosing his spells for the day can be challenging for the wizard, the sorcerer has the vastly more difficult challenge of choosing his spells for the rest of his life. If the wizard gets it wrong—or, more realistically, if his needs change—he can trivially make different choices the next day. The sorcerer gets nine total chances to redo spell selections over the course of 20 levels.

If the sorcerer’s spells known were significantly better, this case would be harder to make. Depending on how things work out, we might still say preparing is better—after all, having exactly the perfect spell for the job within a day is a fabulous feature—but it would be debatable. But as it is, the sorcerer ends up with the exact same limitations on “prepared” spells that the wizard has—and has to make those preparations for a vastly larger time-scale and with near-zero opportunity to adjust. So in this environment, it’s not even arguable; it’s just fact. Any freedom that comes from spontaneity is ruined by locking in the spells known before that.

And in case there was any doubt, remember that wizards aren’t actually locked at 4 spells per day in a given spell level. They’re probably a specialist, and they definitely have high Intelligence, so they actually have more spell slots than the sorcerer has spells known.

This has been one example: wizard vs. sorcerer. And we didn’t even cover everything about that comparison—we skipped the glaring issue of the spell level delay, didn’t even get into issues with spontaneous metamagic, and so on. But even with just this one comparison, even focusing on just this one issue, we can already make conclusions. We can safely generalize this: the sorcerer is the archetypal spontaneous spellcaster, and the direct spontaneous analogue of the wizard. Other spontaneous spellcasters who reach 9th-level spells follow the sorcerer’s example almost exactly, and any who don’t reach 9th-level spells have concomitantly fewer spells. In other words, if you’re talking about another example, you’re probably only making things worse for the spontaneous side.

The question does raise two cases that are better off than the sorcerer: Unearthed Arcana’s spontaneous variant clerics and druids. First, it has to be stated upfront, since the question is “why is X considered..?”—by and large, people don’t consider these classes. They’re obscure variants and rarely used, in my experience, and general statements about 3.5e won’t be made with them in mind. That said, they aren’t substantially better in this regard. They’re vastly better than the sorcerer in general, it’s true, because there is no spell-level delay. But as far as their number of spells known is concerned? It’s literally the sorcerer’s table with a “0” added one level before the sorcerer (and these classes) get their “1” for each spell level. That “0” is filled in with the relevant summon nature’s ally spell for the druid, and with the two domain spells of that level for the cleric. Which means the druid or cleric knows 1 or 2 more spells than the sorcerer—but they cannot choose those spells freely.

For the druid, the summon nature’s ally aren’t really all that good, so that’s not really much of an improvement (it is, for example, very unlikely that a druid would choose to learn all nine of them if they weren’t free).

The cleric is in a better place—there are some really good domains out there, with really good spell lists, so that can be some solid spells that the cleric knows. So OK, the spontaneous cleric has more spells known than the wizard can prepare... before we consider bonus spells, including from specializing (and spontaneous clerics lose the bonus domain spell slot). Most wizards are going to specialize, and have a ton of Int, though, so really it still looks about the same. And the spontaneous cleric is still locked in, still “preparing” for their entire life rather than a single day.

| improve this answer | |
\$\endgroup\$
  • \$\begingroup\$ Do you have a citation for that last sentence? \$\endgroup\$ – J. Mini Aug 6 at 12:40
  • \$\begingroup\$ Great answer, but I noticed you didn't even bring up spontaneous casters getting access to higher level spells a level later, which will also have a notable impact over the course of campaign. \$\endgroup\$ – Oblivious Sage Aug 6 at 12:44
  • \$\begingroup\$ @J.Mini Just look at their spells known tables? For full-casters, the table is almost-always a copy of the sorcerer’s (e.g. favored soul). For anything less than a full-caster, there is concurrently fewer spells known—hence worse off. I maintain that sorcerer is “best” here—rather than tied with all those who copied its table—by virtue of the sor/wiz spell list. Which might not be entirely clear in this answer, but doesn’t seem worth going into this tangent to explain to me. \$\endgroup\$ – KRyan Aug 6 at 12:44
  • \$\begingroup\$ @ObliviousSage My reading of the question was that it was struck by the word “always” and wondering how true that was when the spell levels matched, e.g. on even levels. Otherwise, yes, every other consideration pales in comparison to that delay; that’s massive. \$\endgroup\$ – KRyan Aug 6 at 12:44
  • 1
    \$\begingroup\$ @J.Mini I see that you are also considering Unearthed Arcana spontaneous variant classes, which I didn’t really consider. Those are, of course, exceptions. I suppose I’ll have to consider those here, but they don’t change much. \$\endgroup\$ – KRyan Aug 6 at 12:56

Not the answer you're looking for? Browse other questions tagged or ask your own question.