D&D 3E/3.5 The Red Avenger

[Celebrim]

I agree with pretty much all you're saying, though maybe I tackle the issues in a different way. The introduction of PrCs within the DMG was good, and it should have stayed that way. PrCs are a DM tool, and a potential option for them to give to the players/table should the situation arrive.

The way Monte Cook pitched PrCs was almost like way Gygax did with artifacts. Cook style PrCs would be almost campaign level secrets that were uncovered in play and granted as options to players through play as part of joining these secret societies. And that's a cool idea I can get behind, albiet not one that I think was ever really realized during 3.X D&D's run.

I think the biggest issue with PrCs (and feats) is the "culture" that developed around them, they were presented as player options in the splats not as DM options.

This is in a nutshell probably the biggest problem with 3.X generally. The brand managers of 3.X D&D were more focused on short term profits than they were on protecting the integrity of the game. This came out in decisions like every splat book had to have both player options and DM material, and in the very aggressive production schedule that saw books being printed vastly faster than the material in them could be play tested.

Feats have their own issues...

Feats were I thought the real genius of 3e, in that feats allowed you to build a class in a way that was much less broken than other 'do it yourself' class options were. D&D kept with its core strength of class based design, but customizable feats and skill expenditures allowed you to have some of the benefits of point buy based chargen as well.

The biggest trouble with feats is that martials weren't given enough feats (and skills) to compete in coolness with spellcasters who got lots and lots of spells as they leveled up, and that feats were much much harder to write well than was generally recognized. The short terseness of a feat made it seem like they were easy to design, but in fact I'd say feat design was probably the hardest thing to do in 3e. This resulted in as many badly written feats as there were badly written PrCs - indeed most 3rd party material for 3e D&D was focused heavily on badly written feats and PrCs.

What handy things do you do to help fix key issues for your games?

I have extensive house rules, but next to banning PrCs the one that has probably helped my game the most was removing the spell level from the DC to resist a spell. That is to say, in my game a 1st level spell, 3rd level spell, and 9th level spell all have the same DC to resist. This goes a very long ways toward all on its own balancing martials against spellcasters. "Save or Suck" gets a lot harder to pull of consistently, and as a result spellcasters are more reliant on being support classes in my game, as often their best move is to buff a martial class rather than debuff bad guys. They can still be decent artillery, but artillery was never what made 3e spell casters broken.

 

[Bestbake]

That's super interesting to hear, Celebrim. An idea like that is something I've been mulling over too. I like the idea of teamwork and spellcasters appreciating a warrior be present from a mechanical sense and not just by being a meat-shield/distraction. It hasn't been a problem in games yet, though I think laying a good foundation to be the right thing to do.

I really like how Trailblazer, by Bad Axe Games, developed the "spine"- breaking down the stats of 3.5 so we can make educated guesses at where numbers should be. The saving throw portion of the spine shows some pretty abysmal numbers, especially considering how D&D up until 3.X handled saving throws.
An option I had considered was increasing the rate that good and poor saving throws improve. Rates I wanted to try were +2/3 for good saves and +1/2 for poor saves (with some type of 1st level bonus, like how good saves start at +2). A fix like this would also alleviate any issues that come from the DCs of non-spell sources. Had you thought of using something like this to fix the spell DC problems?
I'm yet to trial this though, after a hiatus I've returned to 3.X intent to lean much more heavily towards 3.0 because I like its monster design more. I'd like to see (or create) a 3.0 spine, because the numbers are different enough from a quick glance between the two. I know to take the idea of the spine with a large pinch of salt because of the challenge rating clause can be out of place, but still, a rough idea is better than none........ though in typing all this out, there is no harm in implementing better save progressions and trialing it.

 

[Bestbake]

The Halfling Outrider, also from Sword and Fist, was a bit bizarre too...

And the damned table doesn't show a BAB progression! Hahaha oops

It is very odd, and until you pointed it it out I thought it to be pretty much the same class as the 3.5 version.
It's interesting that they gave it the Deflect Attack ability, it's already so similar to the mounted combat feat. It's actually interesting how similar it is to the Devoted Defender I the same book. I guess the designers wanted to explore the defensive end of combat. The duelist is also in the book, though I think that PrC first showed up in a dragon mag.

 

[Voadam]

I used a bunch of house rules in my d20 games.

Here is a list of House Rules I used in Freeport game I ran here in 2007.

Some were for my general preference in game style (multiclassing and prerequisite hacks), some were to get characters more appropriate for the specific D&D pirate game theme (a number of the AC ones to encourage lightly armored swashbuckling types and make them effective without heavy armor tanks).