This article was originally written in May 2024, but has been lightly rewritten and edited.
Short article this time around!
The more I’ve worked on various projects, from Adamiir to the Road Back Home, the more I’ve thought about randomization. I figured I’d share what I’ve been ideating on in a quick little article.
Moreover, share how to implement randomization into your games and why your choice matters.
For the purposes of the article, let’s categorize randomization tools into the following:
Fixed-Outcome Variability: Results correlate to a specific outcome.
Examples include playing cards, drawing unique tiles from a bag, drawing tarot cards, etc.
Granular-Numeric Variability: Results correlate to shared or mixed outcomes.
The prime examples are dice in their various forms (in some cases adding / subtracting numbers or dice from results).
If we agree that randomization largely falls into these two categories, my biggest question when creating a game is what tool makes the most sense for the job.
Like choosing a type of saw or screwdriver for a home improvement job (for those in the know, you’ll understand why that metaphor is on my mind 🙃), you can get away with using the wrong tool for the job.
But it will be painful and nowhere near as intuitive as using the correct one. So, when it comes to roleplaying games, how do you know which tool to use? It might be easier than you think.
Here are the two most important decision points you need to make:
1. Is the result of randomization finite or repeatable?
2. Is the result of randomization truly random or somewhat weighted?
Let’s use an example of when you’d want to use each of the categories listed above:
Ragi is deep in the dungeon with his companions, Serala and Mertan.
They come to a heavy, stone door that needs to be lifted to continue. Ragi tells the others he’s going to try to lift the door.
As a GM, you have a few options:
All things being equal, you could simply not use randomization at all. Narratively, you can say that Ragi takes time to open the door (consequence or conversation: how much time does it take?) or that monsters on the other side start to attack as Ragi is holding the door open (choice: hold the door open or let it fall?).
If a difficulty is assigned to the door, you could use dice to meet/beat said target difficulty. This has an element of randomness, but is usually tempered by the character's abilities or features. If Ragi is a strong fighter, they may have a very low chance of failing this task, prompting you as the GM to determine what that small chance of failure is caused by.
If there are outside factors at play and the door itself is merely an obstacle, you could use cards to determine those outside factors. Because you don’t want the same outside factors rolled twice (imagine rolling “Door is Stuck” twice on a random table), cards allow for finite results, especially when you remove said card from the deck after drawing it.
There are, of course, tons of augmentations you can do to any of the above applications. You can mix and match any die rolls, card pulls, drawing tiles, etc. But ultimately, the tool you use does have a long-term effect on play, not just through what the players interact with but also through the amount of brainpower you need to use actively in play.
Thanks for reading!

This article is brought to you by the following paid subscribers who make this newsletter possible:
Azzlegog
Colin
DSPaul
Michael Phillips
Mori
Space Pirate
To Nate as well as WBD or anyone else tackling the topic of randomization, I will never stop to invite to read the great work of Torben available at the link here below!
Enjoy and may the fun be always at your table!
http://hjemmesider.diku.dk/~torbenm/Troll/RPGdice.pdf