Gender Distribution In OCs

Did you know that you’re bad at estimating odds? It’s true!

One way you’re bad at estimating odds is that you overestimate uncommon values. I could do fun kinds of goofy tricks with examples like words or dogs on leashes and break out the Dan Gilbert playbook, but for now, trust me when I tell you that your brain is very bad at reasonably estimating odds. That means things like a chance for things to happen or a distribution of things. That’s why there’s an actual puzzle to things like ‘guess how many beans are in the jar.’

Anyway, we’re not here to talk about cognitive biases and functional memory, we’re here to talk about my OCs.

I roleplay a lot! It’s one of my favourite things to do. I have a lot of yearning for story spaces and I love my friends but I and my friends can’t just interrupt our days to go punch a dragon, so instead, we roleplay and we play in our own little fictional worlds with their own rules. I have a huge pile of characters over the years, and since I have so many, I do what any sensible nerd would do and put them in a spreadsheet.

After all, spreadsheets are cool. 😎

With the shutdown of City of Heroes (and that’s a complicated story now), I took a look at this spreadsheet and did a big crunch, realising that one column in my piles of data I had was pretty consistent. I mostly just made males. I had some nongendered characters, who were robots, and they were mostly referred to as males. At the shutdown of City of Heroes back in 2012, I had a cast of characters that numbered over a hundred characters, and that list was 95% male.

I thought about this but it wans’t until a few years later that I considered this spreadsheet and how I wanted to change that stat. I make characters to play in spaces, and I want those spaces to have cool characters in them, and I want there to be cool women characters in those spaces, and so… why not make more women? Why not write women, especially in the low-key low-impact way of roleplay spaces where I’m primarily playing pretend with friends and not creating a greater coherent text that is going to get reviewed by editors. That ensued a period of revamping some characters that were unused as girls, and making more women characters. In some cases, I even rolled a dice to just check ‘hey, is this character cis?’ because it was a question I’d literally never considered when I’d made them.

Then I thought, after two years of this that maybe I’d gone too far. Maybe my population of characters was massively female dominated and now was just an insight into the Horny of my mind, and I was seized with a pre-emptive embarassment. To the spreadsheet I went, to fill in fields and give myself that balm that is data.

Turns out that this fear that I was wiping out my male character population had resulted in a whopping 33% of the characters now being women. I thought I’d gone too far and I hadn’t even gone half way.

One example from a study of women in meetings shows how we handle stats badly. Dudes polled will often believe women have overwhelmed a group when they represent 33% of it; they will believe women talk 20% more than men when they spoke 20% less. And as with me, I thought I’d gone overboard when I hadn’t even finished addressing my massive bias.

We are bad at recognising stats. We are bad at estimating odds. We are ferociously bad at recognising our own biases. It’s worth checking yourself, and considering just what you’re doing.