Initiative Through the Ages

I’ve played with lots of different initiative systems – individual rolls, group rolls, based on a stat, drawing cards from a deck…  Seems like turn order is a thing that game designers just love to tinker with.  Personally, I think the simpler the better.  I use a single group roll at the start of combat and then just go around the table clockwise until it’s over.  It’s fast, it’s light, everyone knows when it’s their turn – easy.  I was going to write a post espousing this method when the thought struck me – I wonder how it has changed in D&D over the years?  Time for a Through the Ages post!

Let’s start with OD&D.  Nope, can’t do that, it has nothing to say on the topic.  Please correct me if I’m wrong, but I seriously can’t find a single comment on turn order during combat or any instance of the word initiative.  Well, OD&D does say I have to have a copy of Chainmail for combat, so let’s start there:


  1. Both opponents roll a die; the side with the higher score has the choice of electing to move first (Move) or last (Counter-move).
  2. The side that has first move moves its figures and makes any split-moves and missile fire, taking any pass-through fire possible at the same time.
  3. The side that has last move now moves its figures and makes any split-moves and missile fire, taking any pass-through fire possible at the same time.
  4. Artillery fire is taken.
  5. Missile fire is taken.
  6. Melees are resolved.
  7. Steps 1 through 6 are repeated throughout the remainder of the game.

For the sake of brevity I’m not including the second alternative, titled “the simultaneous movement system” which outlines a mechanism of writing down orders for units and then executing as a second pass.  All this sounds pretty reasonable for a war game.  A couple things of interest to note for later comparison – the winner of the die roll gets to choose to go first or last, which makes sense given that situationally the advantage may lie in either direction.  Movement and actions are separate – initiative determines who moves first, but then missile fire and melees seem to be simultaneous actions.  It feels like there may be order of operations questions there – does a unit still get to attack if it is killed that turn?  Anyway, lest we get embroiled in war game analysis, let’s move on and see how this gets reflected into D&D once D&D starts to have its own combat rules.

Holmes Edition:

When two figures are brought into position 10 scale feet (or less) apart they may engage in melee. The character with the highest dexterity strikes first. If the Dungeon Master does not know the dexterity of an attacking monster he rolls it on the spot. Subject to the limitation of heavy weapons the two figures exchange blows in turn until the melee is resolved. If dexterities are within 1 or 2 points of each other, a 6-sided die is rolled for each opponent, and the higher score gains initiative — first blow.

Fascinating, all turn order is dictated by dexterity.   I love the qualifier about “if the Dungeon Master does not know the dexterity of an attacking monster he rolls it on the spot.”  I mean, that’s basically true for every single monster.  I love the idea of rolling dex for every monster – this orc is fast, but I rolled a 3 for the dragon, so I guess it’s just a slow, clumsy dragon.  Dice only come into play to break ties, and this text may be the first place we see the term “initiative”.

Ultimately it looks like Holmes made his own stuff up here.  His system does not feel rooted in Chainmail / OD&D to me.  He also creates the precedence for having dexterity impact initiative – what seems to be a contentious choice that we’ll see come and go throughout later editions.


INITIATIVE: To determine initiative, each side rolls 1d6 (the DM rolls for the monsters).  The side with the higher roll may move first and attack first in combat for that round.  If both sides roll the same number, the DM may either have both sides roll again, or may consider movement and combat for both sides to occur at the same time (known as simultaneous combat).

The side “with the initiative” has the first choice of actions.  Members of that side may choose to fight, run, throw a spell, take defensive positions and wait to see what the other side does, start talking, or do anything else that the players or DM can imagine.

Now there’s additional text here about an optional method of having each individual roll instead of per-group, but I’m not including it here.  Additionally though, it’s important to note the “combat sequence” laid out later in the text:

Combat Sequence

A.  Each side rolls for initiative (1d6).

B.  The side that wins the initiative may act first (if simultaneous all actions are performed by each side at the same time):

  1. Morale checks, if needed (page B27).
  2. Movement per round, meleed opponents may only move defensively (spell casters may not move and cast spells).
  3. Missile fire combat:
    1. choose targets
    2. roll 1d20 to hit; adjust result by Dexterity adjustment, rang, cover, and magic
    3. DM rolls damage
  4. Magic spells (roll saving throws, as needed: 1d20).
  5. Melee or hand-to-hand combat:
    1. choose (or be attacked by) opponents
    2. roll 1d20 to hit; adjust results by Strength adjustment and magic weapons
    3. DM rolls damage; adjust result by Strength adjustment and magic weapons.

C.  The side with the next highest initiative acts second, and so on using the order given above, until all sides have completed melee.

D.  The DM handles any surrenders, retreats, etc. as they occur.

Wow, so much to unpack here.  I did not realize that this is where the “no cast and move” rule was hidden.  Also fascinating that the DM always rolls damage, I don’t think I ever noticed that – did anyone play this way?  I imagine not knowing how much damage you’re doing really changes the feel of combat.  Note too that the first part of the text mentions that the winner of initiative “may move first and attack first,” and I wonder if that use of “may” is meant to imply choice as it was in chainmail?  

The combat sequence here really reflects the text from chainmail, but then item C seems to countermand the simultaneous nature of actions after movement.  I feel like I’ve stared at this text many times thinking I was supposed to interweave actions until noticing item C. 

The concept of simultaneous combat is also something that I’ve struggled with through the years.  It seems rather neat but is so complicated and so little actual guidance is given on how to make it work.  With a d6 for initiative, ties are pretty common, and nothing is quite as infuriating as multiple ties and re-rolls.  Recently I’ve been using a simple “ties go to the players” rule.  Sure, it gives them an advantage, but hey, they’re the players, why not?  It at least makes things much easier and faster.

OK AD&D, what’ve you got?

Surprise gives initiative to the non- or less-surprised party. It is otherwise determined when an encounter occurs and at the start of each combat round. It indicates which of the two parties will act/react. Again, a d6 is rolled, and the scores for the two parties are compared. (It is recommended that such initiative rolls be made openly unless there is some reason to hide that of the encountered monster party – such as special bonuses which would be unknown to the player characters involved.) The higher of the two rolls is said to possess the Initiative for that melee round. (While it is not accurate to roll one die for all individuals comprising each party, it is a convenient and necessary expedient. Separate rolls could be made for each member of two small groups, for instance, but what happens to this simple, brief determination if one party consists of 9 characters and 6 henchmen and the other of 7 giants and 19 dire wolves, let us say?) Possession of initiative allows the individuals to take action or reaction as desired according to the foregoing list of alternatives, and as detailed hereafter. The dexterity or speed of individuals or weapons is not considered in the 1 minute round except as hereafter noted.

I’m limiting myself here to just the rules around determine who has initiative – what to do with that info gets really complicated in AD&D.  AD&D starts adding in weapon speeds, casting times, and all other manner of complexity that frankly I’ve never really been able to deal with.  The example combat has all this interesting stuff going on with spells being interrupted left and right, but the mechanics for dealing with that feel like they’d take forever to calculate and would slow combat to a crawl.

And yet, what I find really interesting here in this text is a little nod to expediency around group vs. individual combat.  Group initiative is described as a “convenient and necessary expedient”, and then there’s a hilarious aside about a complex group containing multiple different types of enemies, players, and henchmen.  Ultimately I think the point is a fantastic one, but where does this sentiment go when we introduce concepts like casting time and spell interruption? 

Other interesting notes – we have here a complete dismissal of dexterity modifiers to initiative, and also the concept of the winner of initiative getting to choose to go first or not.  The former seems odd without taking into consideration the historical context (sorry Holmes), and the latter is something I feel like I’ve never noticed nor seen used in play.

OK, 2nd edition, can you clean up this mess?

The initiative roll determines who acts first in any given combat round. Initiative is not set, but changes from round to round (combat being an uncertain thing, at best). A character never knows for certain if he will get to act before another.
Initiative is normally determined with a single roll for each side in a conflict. This tells whether all the members of the group get to act before or after those of the other side(s).
There are also two optional methods that can be used to determine initiative. Each of these optional methods breaks the group action down into individual initiatives. However, the general method of determine initiative remains the same in all cases.

Standard Initiative Procedure
To determine the initiative order for a round of combat, roll 1d10 for each side in the battle. Normally, this means the DM rolls for the monsters (or NPCS), while one of the players rolls for the PC party. Low roll wins initiative. If more than two sides are involved in combat, the remaining sides act in ascending order of initiative.
If both (or all) sides roll the same number for initiative, everything happens simultaneously — all attack rolls, damage, spells, and other actions are completed before any results are applied. It is possible for a wizard to be slain by goblins who collapse from his sleep spell at the end of the round.

Nope, 2nd edition just piles it on.  In addition to standard group initiative we have two optional methods.  Also not included here is a decently sized table of standard initiative modifiers.  Interesting that we’ve moved from a d6 to a d10 – possibly because we wanted a finer grain of modifiers, or possible to make simultaneous combat more rare?  Simultaneous combat is still here though, and this is perhaps the best description I’ve read on how exactly it’s supposed to work.  Still more effort than it’s worth in my opinion.

D&D 3rd Edition:

Initiative Check
At the start of a battle, each character makes a single initiative check. An initiative check is a Dexterity check. If two or more characters have the same initiative check result, the characters that are tied act in order of Dexterity (highest first). If there is still a tie, roll dice to break the tie.

The Combat Round
Each round represents about 6 seconds in the game world. Anything a person could reasonably do in 6 seconds, a character can do in 1 round.

When a character’s turn comes up in the initiative sequence that character performs his entire round’s worth of actions.
There are no simultaneous actions. All effects of a character’s action fully resolve before the next character acts. A character cannot split an action to allow another character to act between portions.

Effects that last a certain number of rounds end just before the same initiative count that they began on.

I have to say, despite my feeling like 3rd edition begins the “board-game-ification” of D&D, this I think is a really good refinement.  Individual initiative takes over, but there’s no such thing as a tie, no such thing as simultaneous action, no interruption of spells, just do all your stuff and the next player goes.  Also, welcome back dexterity influence.

OK, so where does all this leave us?  Honestly, I can’t say any single edition ever came up with a good solid answer.  I think ultimately I have to hand it to later editions for just doing away with interruptions and simultaneous actions.  My preferred method is basically that, but with a group roll instead of individual rolls.  Going around the table is just so fast, and best of all, predictable.  I think a lot of delay happens in combat when players can’t tell when their turn is coming up. They delay making any tactical choices until notified it’s their turn, so then as each turn comes up we have to wait for the player to make decisions and then execute them.  When you go around the table you know your turn is next as the player to your right is executing their action, and you should be ready with your actions by the time that player completes their turn.

10 thoughts on “Initiative Through the Ages

  1. Personally, I adore the “Elective Action Order” system:

    Basically, pick any means you like to determine who goes first (players choose; highest attribute; predetermined by order around the table; etc.), and then after that person finishes their turn, they pick who goes next. Only after everyone, good guys and bad guys alike, have had a turn does the next round begin. And so on. I like this system for two reasons. First, it adds a layer of player agency where one wouldn’t normally find it. And second, the tactical decision-making turns out to be exciting — the players have to weight the benefits of lining all their turns up in a row to whomp on the bad guys against the risk that the bad guys will then get to do the same to them. Super fun.

    1. That is a really interesting system. I may run this past my DemonWars group. Given how much that game emphasizes team tactics, it may be a really good fit.

  2. I’ve looked into these early initiative systems a bit over the years, driven by trying to figure out where Holmes’ system came from. OD&D Vol 1 does have one clue, buried in the description of Dexterity: “It will indicate the character’s missile ability and speed with actions such as firing first, getting off a spell, etc”. This appears to be Holmes’ textual basis for using Dex for initiative, as he included a version of this sentence in Basic and logically extended it to melee weapons: “Characters with high dexterity can get off the first arrow, throw the first spell or draw a weapon and strike the first blow” (pg 5). Gygax explained initiative in the D&D FAQ in Strategic Review #2 (Summer 1975), where it is “simply a matter of rolling two dice (assuming that is the number of combatants) with the higher score gaining first attack that round. Dice scores are adjusted for dexterity and so on”. With it is an example where he gives a +1 for high Dex. I tend to think that Holmes didn’t see this. Eldritch Wizardry (1976) has a very complicated initiative system (possibly written by Tim Kask) where Dex modifies only spells & missiles. Holmes was probably also influenced by the Warlock variant D&D rules (1975), which uses Dex for order of spells, and possibly by Metamorphosis Alpha (1976), which also uses Dex for all order in combat. Regarding Gygax, there’s also a section buried in the back of the PHB (1978) that mentions using Dex for important combats, but obviously he changed his mind by the time of the DMG.

    1. Wow, really excellent research there. Thanks for sharing! Overall I’d say it feels like an area of the rules that was likely highly debated. I’d love to have been a fly on the wall for some of those early discussions.

      I also find the origin of terminology fascinating. OD&D does not yet use the term “roleplaying game”, and similarly I could not find the word “initiative” in the text anywhere, though I wouldn’t be surprised if war games were using the latter term by this time. Unfortunately my copy of Chainmail has not been OCR’ed, so searching for specific words is a difficult manual process.

  3. I’m trying something like written orders. In theory it’s good. But it’s clunky at the table.

    You pick one of five categories of actions – Move (a full move, not partial), Missile, Magic, Melee, Misc (drink a potion, try to unlock a door in combat, whatever)

    Then roll initiative by side.

    Then when it comes to your turn you are limited by the kind of action you picked. But you could cast any spell at any target, for instance.

    So choose action, then roll initiative

    1. That is really interesting. I would think the draw for written orders would be evaluating actions based on type. Eg, after everyone has selected what they’re doing execute all movement, then all missile fire, then all spells, and finally all melee combat. I feel like that would approach the original war game roots closer, and might approximate the “simultaneous actions” that B/X and AD&D give so little guidance on. Though that said, I suppose you’d still have cases where you need to know which spell goes off first, or who takes missile damage first, etc.

  4. Cool stuff! A few added notes:
    – Swords & Spells has a Turn Sequence [p. 3] that some assert constitutes a newly included initiative system for OD&D. To my eye it’s effectively the same as that first system from Chainmail. (I expected Zenopus to beat me to that.)
    – Chainmail has the original restriction of wizards’ movement: “In order to cast and maintain any spell, a Wizard must be both stationary and undisturbed by attack upon his person.” [p. 32]
    – Another motivation for 2E’s d10 is that 1E AD&D connotes in several places the initiative die-roll to action on a particular segment within the round. Since there are officially 10 segments to the round, someone might have thought the d10 felt more complete for that.
    – 3E had other mechanics for interrupting actions and spells, e.g., attacks of opportunity, readied actions, and counterspells. That stuff alone pretty much turned off Isabelle from the 3E system.

  5. For years (and through several editions now) I’ve uses a number of variations of no initiative. I’ve been able to refine it via 5e quite a bit, but it basically comes down to actions occurring simultaneously or in a logical order.

    In most cases it’s really not important who strikes first. Where it does matter, you can make an initiative check between the relevant combatants at that time. Otherwise we go with what makes sense: the orc looses an arrow before you can close the 30 feet and attack with your sword.

    In practice it means I generally have everybody jump in with what they are doing. If something that somebody else does, or even something as we’re resolving the round, changes that, then they can alter their action with whatever seems within reason. If you’re charging 40 feet to attack the orc on the right, and the wizard drops it with a lightning bolt, you can attack the orc on the left.
    I tend to resolve attacks/counter attacks together. For example, the fighter is attacking the Irvin the left, so I resolve that orc’s attack at the same time. If multiple people are attacking the same creature, then just figure out the resolution order.

    We have a short list that details an order of actions, shooting a loaded crossbow is faster than loosing an (non-drawn) arrow, which is faster than a light/finesse weapon, which is faster than a two-handed sword or a pole arm being used as a slashing weapon, but a thrust with a pole arm is faster than a shorter weapon, etc. they are basically guidelines (along with Dex) to help determine what happens first to avoid having to make an opposed initiative check.

    It’s super simple, very fast, and allows a more fluid and narrative resolution of combat.

    1. That’s pretty fascinating, I’ve never heard of anyone using a style like this. Out of curiosity, how many players are you generally playing with? The chaotic nature of this sounds like it would work better with a smaller group. Do you think this scales to a group of say, 10 players? As I try to imagine it, I can’t help but assume it would be difficult to follow as various players talk over each other trying to jump in with their actions, and some players totally miss out because they weren’t assertive or loud enough.

Leave a Reply

This site uses Akismet to reduce spam. Learn how your comment data is processed.