[quote=@The Nexerus]First of all, I'd like to affirm--before anyone has even had the chance to post it--that the standpoint that a roleplay isn't really a roleplay if it has mechanics is idiotic. Anyone who has ever said anything to that effect is too stupid a moron to warrant taking anything that they ever said, ever, even half seriously. There is nothing more innately roleplay-ey about freeform roleplays versus structured roleplays. It's as hair-brained an idea as accusing limericks and haikus of not being [i]true[/i] poems.[/quote]

The validity of game mechanics as part of a roleplay experience is a subjective construct which an individual can choose to accept or reject. THere is nothing innately wrong with considering roleplays with game mechanics outside the scope of what said individual considers a valid roleplay. The standpoint that said individual considering such roleplays unfit is wrong is idiotic. Anyone who has ever claimed anything to that effect is too stupid a moron to warrant taking anything that they ever said, ever, even half seriously.
[i](This is what your comments read like. There is a line between wit and derision and this reads much closer to derision.)[/i]

To answer your question [@GraySkye], it partially depends on your target audience. Different groups of people enjoy different quantities of game mechanics. I am more of a purist but there's plenty of people who are happy going full D&D dice-and-stats-for-everything style. In that regard its a case of "if you build it, they will come." The other major consideration, people already touched on. Playability. Overcomplicating things is a great way to narrow down your audience to zero.

My counsel would be that, if in the course of writing a normal post for the roleplay in question, you have to refer back to more than one additional reference page (or a singular reference page more than 4 or so times) .... or if you have to spend more than 10-15 minutes tops working on math for stats, you may want to evaluate the possibility of cutting back. Those are some tests I suggest to check how burdensome the systems are. Burden will very quickly turn to burnout and that will ultimately kill a project.