For many years, I secretly harbored the knowledge that I HAD rolled a character with straight 18s for a 1e D&D game. And that bothered me, because I know the odds against doing that were… huge. Beyond any reasonable chance of it happening. And yet, I remembered it happening.

But it hadn’t.

During one of my recent moves, I found the character in question — Buskirk, an elven rogue/fighter/magic-user (multiclassing and demi-humans were weird in 1e). And I made two startling discoveries.

First, he didn’t have all 18s. It was three 18s, and three 16s. Awesome, but not the same.

Second, those ability scores were rolled using a method from the 1e Unearthed Arcana, rolling 9d6 for your primary ability score *and keeping the best 3), then 8d6, then 7d6, and so on, down to 3d6. I know this because I wrote down the results of all the dice on the back of the character sheet when I first made the character. I haven’t done the math on this, but it obviously produces higher ability scores.

So, memory vs reality. I can see why I remembered it the way I did… but my instinct that there must be something wrong with such a spectacularly unlikely result (astronomical in scale) was accurate.

Memory has been shown to be spectacularly unreliable, and yet many of us cling to it as our most trusted information source.

(Do you enjoy the content on this blog? Why not become a patron, and support the creation of more free material!)