I was going to write a much longer post this week, but alas, I've been busy working on a new project and so I'll just discuss archetype lengths this time.
Recently, I had a very good conversation about archetype lengths with a few fellow freelancers. Without going into details, I maintained that Paizo archetypes are, on average, about 450 words long, though of course individual archetypes are often much longer or shorter than that. When I first made my claim, I didn't actually check every archetype in existence, the estimate was just based on my designer's hunch (yes, you should have one if you're a designer).
My point was that if the average length of archetypes in any given product is much longer than that, the authors should ask themselves why they are so long. Would they be able to write equally good archetypes -- or even better ones -- if they exercised more restraint? My hypothesis is that most archetypes that look conspicuously long would indeed benefit from having a few hundred words shaved -- or hacked -- off.
Let's have a look at a few things that have led me to think this way.
RPG Superstar: I designed my first archetype in 2011 when I was an alternate in RPG Superstar. In both 2011 and 2013, the round 2 challenge involved designing an archetype using just 450 words. After I started freelancing I noticed how the word counts used in RPGSS actually make sense in the world of freelancing. Bestiary entries (600 or 700 words) roughly correspond to one page in a printed product.
Archetypes and other small entries like feats and magic items are different in that you're generally not assigned just 450 or 300 words. You're more likely to have a number of pages assigned to you, and you should make the content you're designing fit within a word count determined by the number of pages, the amount or art, and the type of content.
Actual Paizo products: While I still have limited experience in writing archetypes, I've certainly read and used a lot of archetypes! I didn't do any analysis prior to writing this article, but to back up my claims (or to disprove them!), let's have a look at Paizo's latest big book with archetypes, the Advanced Class Guide.
The book has 60 pages worth of archetypes and related content (discoveries, rage powers, a cavalier order, dares, a magus arcana, and a combat style). The related content takes up approximately 4 pages altogether, so we're looking at 56 pages of archetypes. There also some intro text for each class, but it's generally quite short, so let's assume the intros for the 29 classes only take up 1 page in total (I'm being generous). We've got 55 pages left.
Now, let's count the archetypes: 3 + 2 + 6 + 7 + 2 + 2 + 8 + 4 + 7 + 2 + 3 + 2 + 3 + 2 + 2 + 6 + 2 + 5 + 1 + 2 + 3 + 1 + 1 + 7 + 10 + 1 = 94
94 archetypes, 55 pages. That's 1.7 archetypes per page or 3.4 per 2-page spread. How many words is a page or a two-page spread? Based on my personal experience and some estimates I've seen before I even started freelancing, I'd say a 2-page spread is usually around 1,400 words, but the actual number depends on the amount of art and type of content. If we assume each 2-page spread is 1,400 words, we'll get 409 words per archetype. Not very far from my estimate of 450 words. (Editor's note: An earlier version of the article stated that the word count for a 2-page spread can be as high as 2,000 words, but I later learned that in practice, that's doesn't happen a lot.)
If someone has the time, feel free to count the actual lengths of the ACG archetypes. I'm quite sure the average lies somewhere between 400 and 500 words.
Archetypes are all about small but significant changes: The point of an archetype is not to change everything but to change just enough. It makes sense to replace any ability that doesn't make sense, such as wilderness-based abilities on an aquatic archetype. However, abilities that don't detract from the concept of the archetype should not be replaced just because you can, only if you really need to. Each change should be weighed carefully.
Besides, replacing a small number of abilities
means that the archetype is more "modular" than a very complex archetype. It is
more likely that you'll be able to combine it with another archetype,
which increases the archetype's utility.
I think a very good example of an archetype that involves small but significant changes is the infiltrator (inquisitor archetype). The changes are indeed small (limited to just a few abilities and the new abilities are described in less than 200 words) but the flavor and role of the inquisitor changes considerably.
Someone might argue that if you limit yourself, you're giving the reader/player less, but I'd argue that you're in fact giving the reader more. If you are able to deliver 8 archetypes with an allotment of 3,000 words, you are giving the reader more than an author who only managed to squeeze 3 archetypes into that word count.
In a PDF-only product that isn't so much of an issue because you're not limited to a specific word count. Of course, every extra word still costs your publisher extra cents. Also, I'd argue that restraining yourself in terms of word count leads to improved quality and more user-friendly content (less to read, less to remember). It's also easier for the layout designer to make good-looking page layouts if the combined word count of the archetypes matches the expected word count for a page or two-page spread. (E.g. each page or two-page spread begins with a heading, not a paragraph of text.)
And lastly, as I've mentioned in my monster articles, you should make any abilities you design simple but elegant. (Monsters, magic items, and archetypes aren't all that different.) If a single ability needs 200-300 words to describe, you're at risk of creating overly complex mechanics.
Sometimes, albeit rarely, you do need the words, which brings us to...
Exceptions exist: There are always exceptions to just about any rule, and archetypes are no exception (pun intended). One of my favorite archetypes is the zen archer which is nearly 800 words long. Some ideas need more words to be expressed clearly. The zen archer is a well-established monk archetype in both real life and fiction, but its focus is very different from that of the "vanilla" monk in Pathfinder who doesn't even have a bow proficiency. Thus, many abilities needed to be changed for it to make sense.
But as mentioned, the zen archer is an exception to the general rule and it would be a mistake to think that you can always justify a huge word count because each of your archetypes is a special exception. If you are assigned to write 4 pages worth of archetypes and the assumption is 1.7 archetypes per page, you're going to have to make your other archetypes much shorter if one of them takes up a whole page.