Sunday, October 19, 2025

Fixing 5e Initiative Part 1: The Problems

I don't like the way initiative works in 5e. Really, I don't like the way initiative has worked in D&D from 3rd edition onward. Really really, I don't like the way initiative works in any edition of D&D. This time, I'm going to try to fix it for 5e, though.

You can't fix a problem until you know what the problem is. Here's my diagnosis of the problems with 5e's initiative system:

Rolling initiative is disengaging. Just as something very exciting is about to happen, everything stops as the DM and players roll dice, add modifiers, and then take notes of all the results. Once that's done, you begin cycling through characters and monsters one at a time, deciding on each creature's turn what it's going to do, and then resolving it. This is disengaging in two ways. First, rolling and recording initiative itself breaks everyone out of the game's imaginative space for a few minutes to deal with something purely mechanical. Second, there's not much for players to do when it's not their turn. They could be planning their next turn, but conditions might have changed a lot by the time that comes around. Or they could see what's new on Instagram.

Rolled initiative results are mainly randomization independent of player choices at the table. The main influences a player can have on initiative results are decisions made about ability scores and class features, which provide modifiers to a d20 roll. In almost all cases, the roll itself is the most significant determinant of initiative. As a rule of thumb, though, mere randomness is almost as bad for player agency as pure determinism. The ideal is player-modulated risk or uncertainty; there is a chance element, but player decisions in play can influence the probabilities.

Here's why none of the initiative variants in the 2014 DMG solve the problem.

Initiative Scores. Instead of rolling initiative, use passive Dexterity checks. 

This approach reduces disengagement, because the DM can have all the relevant initiative scores written down ahead of time. It eliminates randomization as well. Unfortunately, it replaces randomization with an absolutely deterministic mechanic that is still entirely separate from decisions players make in play. It is a little better than the default system, but it doesn't give players enough control over when their characters act.

Side Initiative. No modifiers apply to initiative. Instead, each side rolls 1d20, high roll goes first, and the two sides alternate thereafter. Creatures on each take turns in whatever order they choose.

Like using initiative scores, this approach reduces the time spent dealing with initiative. It also lets players decide their place in the initiative order round to round — relative to their own side. Those are good things.

However, this system amounts to flipping a coin for which side gets first action in combat. No decisions anyone makes can help or hurt their initiative. So, features like Jack of All Trades, simply having a high Dexterity score, or cat's grace lose some of their benefit. Relative order within a side rarely matters as much as going before or after the opposition, so this approach still leaves the most important part of initiative up to chance in a way player's can't influence.

Speed Factor. Every round, everyone declares their actions and rolls initiative. The roll is modified by penalties for slow actions, bonuses for fast ones, and the size of a creature.

This approach makes initiative depend on player decisions about what they will do in combat. That's good. That's really good. It injects an element of risk or uncertainty into the system that was otherwise a matter mainly of chance.

The main problem with this system is that it replaces the once-per-combat ceremony of rolling and recording initiative with a once-per-round ceremony of declaring actions, rolling initiative, and recording everyone's results.

Next up: Some other possibilities.

No comments:

Post a Comment

Animals

Many of the monsters in B/X are animals. While there are differences among them, their are enough similarities that it’s worth having a gene...