Inspired by a recent conversation about initiative and turn order, I felt compelled to record the video above to compare how different systems handle this. Synopsis below.
This post is not meant to be an exhaustive list, and I’m sure I’m missing some approaches. I’ll also present a potentially controversial “hot take” at the end of the video.
Most, but not all, of the turn order approaches can be split into the following dimensions:
- Random or fixed
- Determined once for the entire encounter, or once per turn
- Individual or side based (where the latter typically means that players go in any order they want; but they may all go before or all after the enemies)
Approaches
Let’s jump into the different options for handling turn order:
1) Randomly determine individual turn order for the entire encounter
Examples: D&D 5e/4e/3e, Pathfinder, DCC
This is a common mechanic that most of you are probably familiar with.
When combat begins, each player rolls a die to determine their place in the initiative order. Usually you add a stat to the result. E.g. in D&D 5e, you add your character’s Initiative stat, which in turn is based on their DEX, with a few other modifiers.
It works essentially the same in D&D 4e, 3.5, as well as in systems derived from D&D, such as Dungeon Crawl Classics or Pathfinder - although the details may vary slightly. E.g. Pathfinder 2e uses the Perception attribute instead of Dexterity, and in some cases even specific skills, such as Stealth. This kind of makes sense to me, as might be more about how observant the character is than their physical dexterity, for example.
When multiple characters end up with the same number, you typically resolve this via a roll-off. I would argue that this is utterly unimportant at least for ties between PCs, but some tables insist on resolving these ties.
Now you have to remember the order somehow, which can be surprisingly tedious! You might write it down on a piece of paper, or use some sort of tokens, index cards, or the digital equivalent.
Overall, I find that this approach tends to kill all momentum. The players suddenly face a dramatic challenge, but instead of keeping the excitement high and immediately jumping into actions, the call to “roll for initiative” makes things slow to a crawl until these logistics are resolved.
As a slight variation, some Free League games, such as Forbidden Lands or Vaesen, use numbered cards to determine initiative order. At the beginning of the combat, each player draws a card, and the GM draws a card for each NPC/monster. Actions are resolved in order of increasing numbers. This sounds slightly more streamlined, but I haven’t played enough Free League games to comment on how well this works in practice.
2) Randomly determine individual turn order each round
Example: Savage Worlds
Savage Worlds is one of the prominent systems that do this, in this case using cards rather than dice as randomizer, which kind of fits well with the feel of that system.
Counting down card values works pretty well, so you don’t have to separately write anything down. When playing in person, GMs can see each player’s card in front of them, which might be helpful, too. They added some other aspects to make it more fun, e.g. Jokers give the player a bonus for the round.
Troika also has an interesting spin on this: The GM draws tokens from a bag that represent all the characters. That player (or GM) resolves the turn, then the GM draws another token. Repeat until everyone went that round, or until the special “end of the round” token is drawn. The latter immediately ends the round and proceeds to resolving end of round activities (e.g. magic effects, bleeding out) without the remaining players getting a turn.
I haven’t played this myself, but while it sounds quick enough, it seems to remove player agency. I would be annoyed if I got unlucky and had to sit through multiple combat rounds where I didn’t get to act. Especially if I’m in a dangerous situation and unable to save my ass.
Overall, fewer systems use this approach, because it adds additional overhead. I think it can be fun to mix things up, but overall I prefer simpler options.
3) Randomly determine side based turn order for the entire encounter
Example: Mork Borg, Pirate Borg, Sharp Swords & Sinister Spells
In Mork Borg, Pirate Borg, and presumably other Mork Borg derived games, the GM rolls a d6 at the start of combat. On a 1-3, enemies go first, while on a 4-6 the PCs go first.
(Note that there’s also an optional individual initiative rule, where each player rolls d6 + Agility.)
I consider Sharp Swords & Sinister Spells a variation of this concept: Characters normally act in order of Hit Dice (i.e. level for player characters). Ties are resolved by having PCs make an Agility check to see if they go before or after their enemy.
So we effectively split the PCs into two groups: One group that goes before enemies, and one that goes after. Depending on the level of the PCs compared to the enemies, all players might be in the same group or not (e.g. a level 2 party fighting a bunch of HD 1 goblins would always go first).
4) Randomly determine side based turn order each round
Example: D&D B/X, Old School Essentials
In D&D Basic/Expert (B/X) or its modern reimagining Old School Essentials (OSE), each side rolls a d6 to determine who goes first. If tied, everything happens at the same time.
But the whole thing feels a bit messy. While it’s nice that we don’t worry about individual characters, B/X then uses an archaic order to resolve actions within each turn: Each character moves, then we resolve missile attacks, followed by spells, and finally melee. Why not simply resolve one person at a time?
I could see this making sense if PC and NPC actions were interleaved, but each side is completely resolved on its own before the other side goes through their sequence of actions (unless a tie was rolled).
That’s more of a critique of the specific mechanics, though. Overall, I think side based initiative is a very reasonable solution, and we’ll see some variations later that handle this in a more modern and streamlined fashion.
5) Fixed individual turn order
Examples: Symbaroum, Fate Core, Fallout 2d20, Call of Cthulhu, WFRP 1e
As you can see from the long list of examples (and I could have listed several others), this approach is very common. Turn order is usually based on a fixed attribute value. For example:
- Symbaroum
- Based on Quick, with ties broken based on Vigilant
- Players only only roll a d20 if they are still tied after that
- Fate Core
- For physical conflicts, in order of Notice skill. Ties are broken by Athletics, then Physique
- For mental conflicts, in order of Empathy, Rapport, then Will
- Although there’s a suggestion for GMs to have all their NPCs go on the same turn order, to keep things simple
- Many groups also use Popcorn initiative instead (see below)
- Fallout 2d20
- In order of the character’s Initiative stat, with the GM deciding order in case of ties
- For characters, Initiative based on Agility and Perception attributes, but can be modified via perks; creatures use different stats
- Call of Cthulhu
- Based on DEX
- WFRP 1e
- Had a fixed Initiative stat as well
Purely from a streamlining perspective, fixed individual turn order might seem like a nice option. But it’s actually one of my least favorite approaches. I find it annoying to have the same player go first every time. It’s predictable, boring, and disconnected from the narrative.Everyone should be able to act first sometimes, if it makes sense in the fiction.
6) Fixed side based turn order
Example: EZD6
EZD6 uses a very simple turn order approach: Unless ambushed, all PCs go first, followed by all NPCs.
It flows amazingly well. There’s no awkward pause when combat starts - we immediately jump into the action. Since this is side based, we can go in the order that narratively makes sense. E.g. if a warrior heaves open a portcullis in a dungeon and this attracts a bunch of kobolds, chances are that the warrior takes the first action against them - but if appropriate, someone with a ranged weapon might try to take a shot at them first.
This particular system means that the GM’s monsters and NPCs always go last. I’m perfectly fine with that; it just means that I might need to add an extra opponent or two to keep the challenge interesting for the PCs, as they will likely take down a couple of foes before I have a chance to act.
7) Alternating player and GM actions
Examples: Warlock!, Warpstar!
Warlock! and its Sci-Fi cousin Warpstar! use a nice middle ground: One side begins. Which one is often obvious from the narrative. Otherwise, you roll a d6 to determine which side begins. I usually don’t bother and let the PCs begin unless they’re surprised.
Then you alternate between player characters and the GM’s NPCs/monsters. To be precise: You alternate actions, not characters. PCs only have 1 action, but some monsters may have 2 or even 3 actions. In that case, one PC goes, the monster takes 1 action, another PC goes, the monster takes its second action, and so on.
This leads to a good flow, similar to the EZD6 fixed side based turn order, but a bit more interesting, as the PCs can be challenged before all their turns are completed.
This system really sings at the table, especially with Warlock!’s unique melee mechanic: All melee attacks are opposed roll, where the loser (which can be either the attacker or defender) takes damage. Often, once one character engages an enemy, the enemy will immediately retaliate. This means that depending on the rolls, they might each win one of the two rolls, or one side might win both of them, leading to potentially a lot of damage, which in turn creates a lot of excitement.
8) Fast & Slow turns
Example: Shadow of the Demon Lord (SotDL)
This system is pretty unique, and I haven’t seen this in any games besides Shadow of the Demon Lord.
Each player decides whether their character takes a Fast Turn or a Slow Turn. A fast Turn means they get to act before the enemy, but they can only attack or move, not both. A slow Turn means they go later, but they can both move and attack.
So the turn order becomes:
- Players Fast Turns
- GM’s Fast Turns
- Players Slow Turns
- GM’s Slow Turns
This is amazingly fluid at the table, while still allowing for some interesting tactical choices.
9) Popcorn initiative
Examples: Marvel Heroic Roleplaying, Cortex Prime (if using Action Order), Fate Condensed
While often used, some people really dislike the term “Popcorn Initiative”. An alternative might be “Elective Action Order”.
This mechanic was first introduced in the Cortex Plus based Marvel Heroic Roleplaying game, and carried over into Cortex Prime. Note that in Cortex Prime, it’s just one of several ways turn order works. Others are more free form. Action Order is used only in Cortex games where turn order matters. This was a popular variation in Fate Core as well, and is the official turn order mechanism in the more recent Fate Condensed book.
The GM chooses a PC to begin (though I often let them choose). That player picks who goes next, which can be a PC or NPC. And so on. The last player in the round decides who goes first next round, which can be themselves. So there’s an incentive for players to choose an NPC, to prevent the GM from going first with all their NPCs the next round and potentially demolishing the PCs.
This is one of my favorite ways to handle turn order. It keeps a very logical flow, driven by the narrative, and allows players to set up interesting combos.
It might not resonate with all groups, though, as nominating an enemy character to go next could take some getting used to.
10) Everyone goes at the same time
Example: Mothership
In Mothership, players describe how they react. Once they commit, the GM resolves everyone’s actions at once.
I Haven’t had a chance to play it myself yet, so not sure how well it works in practice.
Note that Mothership also has an optional rule for everyone to roll a Speed Check at the beginning of the encounter to determine whether the player goes before or after the GM.
11) Randomly determine individual action count/order each round
Examples: 7th Sea
7th Sea resolves Action Sequences this way: Each round, players announce their character’s approach. The GM tells them what the consequences and opportunities are. All players then roll at the same time and count their so-called Raises. These raises essentially act as Action Points, allowing players to take one action for each raise. Raises also determine turn order.
The player with the highest number of raises goes first, takes an action, then the player with the next highest number of raises (which might still be themselves) goes, and so on. So depending on their roll, players may be able to take different number of actions each round. If they roll well, they can use these to set up advantages for the other players, for example.
This approach seems interesting, and I know some players that really like this mechanic. I only had one experience with 7th sea at a con, and that was pretty terrible, so I’m a bit tainted from that session. :) Overall, it’s more complex than my sweet spot.
12) Variable length actions instead of discrete turns
Examples: Hackmaster, Aces and Eights
In Hackmaster, all characters roll for initiative first, which determines when they start acting in any given encounter. The GM then counts up second by second. When players are up, they announce what action they take. That action might take anywhere from 0 to multiple seconds, and therefore determine how soon they get to go again. The GM needs to keep track of what second we’re at.
I believe Aces and Eights works similarly.
It seems like a lot of bookkeeping and opportunity to lose track, but I haven’t played it, so can’t speak to how well it works at the table.
Conclusion
Summary
So that’s my brief overview of initiative systems. And again: This was not meant to be an exhaustive list; I’m sure I’ve missed some approaches.
My point was mainly to illustrate that there are many different ways to handle turn order in RPGs, and that they range in both complexity, as well as what I might term permissiveness: The most permissive being EZD6, where all player characters always go first, followed by all NPCs.
You might have already picked up on my bias in some of these descriptions. It’s no secret that I’m generally not a fan of crunch, and gravitate towards lightweight systems these days.
Controversial Opinion
So here’s my controversial opinion: Collectively (as a hobby), we waste way too much time and effort on determining turn order. I’m not convinced that it really contributes much to our enjoyment at the table.
My sense is that this is largely an artifact of the wargaming origins of D&D. In wargames or board games in general, this stuff is important. However, in roleplaying games, I don’t think it is.
I realize this touches on another bias of mine that’s important to call out: I don’t care much about the tactical side of RPGs. I do enjoy combat (as long as it doesn’t dominate the session), but just as a way to create fun and exciting action scenes as part of the larger story. I love exploration, interacting with NPCs, overcoming challenging non-combat situations, or perhaps even the occasional puzzle, just as much as combat. In particular, I really don’t like how combat often feels like an entirely separate mini game, disconnected from the rest of the game. If I wanted a truly tactical game, I’d probably play a miniatures board game instead.
Back to my opinion on turn order and initiative: In games with loose turn order, I never felt that something was missing. Whether that’s using Popcorn Initiative, Warlock!’s alternating actions, or EZD6’s “players always go first” approach. Instead, not having to worry about establishing and maintaining turn order allows both the GM and players to focus on more important (and more fun!) things.
Consider this:
- Encountering an enemy and immediately describing how I charge and attack them is exciting and fun.
- Encountering an enemy and halting the game to do a bunch of initiative dice rolls, then writing down initiative order, is tedious and breaks all momentum, immersion, and excitement.
While the order of players vs. NPCs might matter, order definitely doesn’t matter between PCs (unless it’s a player-vs-player game). Roll-offs between tied PCs just feel silly. So at the minimum, some sort of side based initiative seems like a reasonable compromise.
Aside from the tedium of establishing random turn order, there are other reasons why I don’t like turn order that’s fixed - whether just for that combat or even permanently: In my experience, this can lead to players zoning out, because they know it’s not going to be their turn for a while. Instead, when turn order is up to the players, they need to be more engaged to decide when to go. And this might lead to some surprisingly interesting tactics, too.
Say one player casts a spell that temporarily blinds an enemy (or simply throws sand in their eyes). This would be the perfect time for an assassin type character to go next and exploit the enemy’s weakness by backstabbing them. By allowing the players to decide who goes when, we unlock much more interesting ways for them to collaborate and build up on each other’s actions, rather than each player taking their turn in isolation, disconnected from the rest of the action.
Some games have ways for players to change the turn order. E.g. 5e allows players to ready an action that triggers based on a specific event (e.g. “as soon as the enemy moves within range, I shoot my bow”). But the way this interacts with the established turn order, as well as movement, always feels a bit clunky and adds more complexity and tracking. Instead, using a more permissive turn order system allows players to go in whatever order makes sense narratively or tactically, and avoids all this complexity.
My favorite options
That said, here are my favorite approaches:
- Popcorn Initiative (aka Elective Action Order), as used in Fate Condensed and some Cortex Prime games
- It’s simple, quick, permissive, balanced between players and GM, and flows well from the narrative.
- Warlock!’s alternating player and GM actions
- Likewise: Simple, quick, permissive, and perhaps even more balanced.
- But this requires slightly more bookkeeping, particularly for monsters with multiple actions (although I usually still do this in my head without issues)
- EZD6’s fixed side based order, where players go first, followed by the GM
- Even simpler, with even less to track
- But the balance is tipped towards the players here, so you just might need to compensate a bit by increasing enemy numbers or strength
- SotDL’s Fast & Slow Turns
- Allows for more elaborate tactics, in case your players enjoy a bit more crunch, while still keeping the action flowing, without much bookkeeping
Depending on which game you play right now, why not give one of these options a shot? Especially Popcorn Initiative is a pretty general approach that makes no assumptions about the game, so should be easy to swap into many games.
I do realize that not all these approaches work with all games, though. E.g. 5e makes a lot of assumptions about turn order being consistent from round to round. Especially spell casters may find their spells ineffective or inconsistent if this changes, because this often determines how long enemies are affected by a spell.
Side based initiative, whether fixed as in EZD6 or random as in Mork Borg, could be a good compromise here, as it’s mostly the turn order of the players relative to the enemies that matters, and this approach would still be consistent each round at that level.
Also note that not all games, or perhaps not all scenes within a given game, require a well defined turn order at all. Instead, sometimes it’s fine to simply follow the narrative or the flow of the action, moving the spotlight from one character to the next as it makes sense. I will however acknowledge that this does require some experience and discipline with spotlight management, and it can be more difficult if you have one or more very dominant players at the table.
Still, I’d encourage you to experiment with this approach for a scene or two. Depending on which direction the action takes, you can always zoom in and switch to one of the more rigid turn order approaches at that time.