Getting Rid of Saves?

[I'll probably write about the problems I have, and the problems I don't have, with WOTC at another time and in another post. Suffice it to say that I'd like to buy OD&D. Let's leave it at that for the purposes of this post.]

I don't usually spend much time monitoring WOTC. Usually any new developments that are of any interest to me at all are more than covered by more than one of the wonderful Old School blogs whose RSS feeds I subscribe to. (Yes, if you've been wondering why I'm not following you on blogger, it may be just because I don't follow anyone on blogger because I use an RSS reader instead.)

Lately, though, with a few posts predicting or offering advice for 5E, I thought it might be time to swing by WOTC's D&D website and see what's up. Mike Mearls column called "Legends and Lore" is interesting, mostly because I don't know of any other place on the web where someone who knows D&D history quite well messes with D&D rules from a New School perspective while still engaging older editions respectfully and seriously. That last bit, by the way, is particularly impressive to me; Mike Mearls appears to genuinely respect older editions.

Anyway, in the latest "Legends and Lore" article, Mearls argues for a streamlining of things in D&D by getting rid of separate saving throw values. He figures that we've already got six attributes, so why should we have separate save values when we can just roll attribute checks for the appropriate attribute when we need to roll a saving throw?

Wow! Mike Mearls is advocating attribute checks?! Cool! That's one of my favorite tricks I learned from the Swords and Wizardry Quickstart! Isn't that something that's really, really Old School? Please correct me if I'm wrong on this, those of you who are more familiar with 3.x and 4E than I am, but attribute checks are really rare, if ever used at all, in later editions, right? I don't remember a single attribute check in the 3.5 campaign I played in.

So Mearls' suggestion, from my perspective, has these two things going for it. One is that it is rules-lite - "minimalist" in his terminology. I'm always a fan of simplified game systems both because they are easier to quickly learn and use with players that for whatever reason don't take the time to learn the rules themselves and because simple systems are easier to homebrew, and I just enjoy homebrewing for the sake of homebrewing. On the face of it, Mearls' idea seems like an elegant way to approach saves, and even seems like it could be considered an Old School thing to do.

There are a few hitches with that, though.

One is that Mearls' suggestion is about streamlining the system down to a "core mechanic." In contrast, one of the tenants of Old School game design, from what I can tell, is modularity, that is, making sure that there is no one "core" mechanic, but that instead games are conglomerations of different mechanics that deal with resolving different things. So, for example, dealing with traps is done differently than fighting is done differently than turning undead, and so forth. This makes it so that if I don't like the way fighting is dealt with in the default rules set, I can devise my own system, pull out the combat "module" in the default rules and graft in my own system, all without worrying about messing up the non-existent core mechanic. What Mearls is suggesting, though, is to give D&D a core mechanic, revolving around the six attributes. That's interesting, and it might still be possible to preserve some modularity because it might be possible to have lots of different ways to use the six attributes to resolve any one activity, but, for my tastes anyway, I see this as getting much too close to making it so that any tinkering I do will mess up a beautifully streamlined core mechanic.

The other way that Mearls' suggestion isn't Old School is that it doesn't address the idea that saves aren't all about the performance of the character, but are instead super-abstract reductions of all that can go in the character's favor when something bad happens, ranging from, yes, some quick thinking on the character's part to some sort of fumbling on the trap designer/attacker/whatever's part to some sort of quirky biological resistance to a poison to even divine intervention. Making saving throws dependent on attribute scores would make it so that anything external to the character has no effect on the outcome of the saving throw.

I'm not an expert on saving throws. I've only ever used Swords & Wizardry's single save system, as well as 3.5's triple save system when I first started playing, though I'm growing more and more interested in switching over to the original 5 saving throw categories. I do get the impression that this last point, about saving throws including influences external to the character himself, is an important one.

Thoughts?