A friend of mine has mentioned he is thinking of trialing rolling initiative at the start of every round in combat, vs 1 roll at the start of combat for his table.

This sounds like a really bad idea to me for a few reasons, for example.

Any spell that lasts "until the end of your next round" potentially becomes far weaker/stronger depending on your initiative rolls (if you rolled low and then high the duration halfs, if you roll high then low it doubles)

Players death saving throws and healing in general but clerics etc.

But these are all anecdotal.

So is there a statistical way to show how re rolling initiative every round of combat would impact the DnD system?

Just to clarify my friend uses a VTT so the issues of manually rolling and working out initiative order are resolved (to his mind the only issue with rolling every round)

  • 1
    \$\begingroup\$ Exalted 3 did shifting initiative based on each previous round, and it is rather troublesome to track. \$\endgroup\$
    – Trish
    Jun 4, 2021 at 23:09
  • 2
  • 1
    \$\begingroup\$ Note that there is a variant rule called Speed Factor initiative (see DMG Chapter 9, "Initiative Variants") that does involve rolling for initiative every round (but also some other differences as well). You might find some relevant analysis and advice in questions about using Speed Factor initiative. \$\endgroup\$ Jun 5, 2021 at 0:14
  • 2
    \$\begingroup\$ I didn't downvote, but the question seems too broad. D&D is a game of stochastic processes which have bounds on possible outcomes-- there necessarily is a way to demonstrate the impact of changes using statistical techniques. Answerers will need to look at every possible element of every class, feat, spell, ability, and action, and then do some nontrivial math for each of them, and those results will themselves have probability weights (ordering of turns will matter a lot, which is also random). And D&D isn't too tightly balanced in the first place, so it seems like overkill to me. \$\endgroup\$
    – Upper_Case
    Jun 5, 2021 at 0:57
  • 3
    \$\begingroup\$ What does "statistically" mean in this context? What's the real question? Are you interested in how much frequently you may get two turns in a row, or the probability that the enemies have their turn always after the party, for example? \$\endgroup\$
    – Eddymage
    Jun 5, 2021 at 7:40

3 Answers 3


"It's the (action) economy, stupid!"

In analyzing D&D 5e, we often talk about the "action economy" -- in short, the team that gets to take more actions has a huge advantage in the fight.

This is why "boss monsters" are difficult to balance; a big scary monster that gets only one action to the party's four or five is just at a big disadvantage, even if it hits like a freight train. Fifth Edition features purpose-built "legendary" creatures that get to take extra actions outside their own turn specifically to adjust the action economy back towards a more balanced state, making them a more reasonable threat in a one-on-many fight.

Going the other way, if you dig into the encounter difficulty calculations, the DMG has you multiplying the difficulty of a group of enemies based on the sheer number of enemies. For example, if there are 7 or more creatures in the group, you triple the difficulty of the encounter -- the fact that the enemy team just gets to act a lot more often than the heroes makes them vastly more dangerous than their strict CR would suggest. (This is why "crowd control" spells are important; area damage and delaying effects that can break a fight into multiple waves have an outsized effect on large groups of weak enemies.)

While rerolling initiative every round can make "until your next turn" effects unpredictable, the real problem is the way it impacts the action economy. It makes fights virtually impossible to balance.

In the best case, the players may get a lot of turns in a row as they act late in one round and early in the next, allowing them to take out a lot more enemies before the enemies get to hit back and swinging the action economy in the party's favor.

But the nasty one is when it goes the other way; if the enemy just happens to get two turns in a row before the party can respond, they may very well take out one or more player characters without any apparent way to stop them or retreat, which isn't fun, and getting multiple actions in a row can essentially break the tactical movement system by denying the PCs a chance to respond to enemy maneuvers. And of course, once a player character goes down, the action economy turns against the party -- they've lost 20% or more of their damage output, which can turn into a "death spiral" where the disadvantage leads to the party taking more damage than expected, which leads to further knockouts, which leads to more damage, and so on.

Is that going to happen in every fight, or even all that often? No, it won't. The randomness of initiative rolls makes it likely to spread out pretty evenly, in general. But all it takes is one spate of unfortunate initiative rolls to unexpectedly ramp up the difficulty of a fight, or completely ruin a encounter that was meant to be tough, and it can really suck the fun out of a game session. Keep in mind that most fights in D&D only last a handful of rounds, so two initiative rolls that interact badly can mean almost the entire fight is impacted.

Now, there is a time when this happens in D&D as written: when the party is surprised. When one group is surprised, the other side effectively gets two consecutive turns. And what does the DMG have to say about that? Well, on page 85, there's a list of "situational drawbacks" that can raise the encounter's difficulty by a full step (from Hard to Deadly, for example). The very first one on the list? "The whole party is surprised, and the enemy isn't."

To put that another way, depending on how initiative shakes out in the first few rounds, a Hard encounter could end up anywhere from Medium to Deadly, which is a pretty wild change in difficulty!

Worse, a bad action economy is not as obvious as other similar situations. If the players just can't roll above a 5 today, everyone can see what's going on -- nobody can be that upset about what is clearly just the Curse of the Dice Gods. It's frustrating, but nobody's fault. But a funky series of initiative rolls is a lot less visible, which may make the players feel like they're being mistreated. It's not immediately obvious why they're dropping like flies, so it's likely to feel more like the DM has inexplicably hit the party with an incredibly difficult fight, or is somehow cheating, or just "being mean" in some vague way.

It's just not a good change to make in my opinion. The difficulties are not insurmountable, and a mature group of gamers can deal with it (though every gaming group seems to overestimate how mature they actually are), but it seems like adding a significant time cost at the table that comes with a lot of risk and no real reward. What's the actual benefit of randomizing initiative? The game becomes more random, but that's not an inherent benefit. I'd personally much rather get to my turn faster and keep the game rolling along than stop every ten minutes to redo the initiative slots.

  • 1
    \$\begingroup\$ I feel like this could be further strengthened by linking it to Surprise Rounds, which do more or less what good or bad luck does here and the DMG is pretty clear that it swings the difficulty by a lot. \$\endgroup\$
    – Erik
    Jun 5, 2021 at 6:29
  • \$\begingroup\$ That's true. I added a paragraph to discuss surprise. \$\endgroup\$ Jun 5, 2021 at 20:10

Some effects (like the monk's Stunning Strike) last until your next turn. Rerolling initiative each round would make effects like this much more erratic, giving a level 5 monk a pretty good chance to stun an enemy for two rounds in a row if they just rolled a high initiative value.

With rerolled initiative, it's possible for a player with Sentinel to make an opportunity attack and still have that enemy escape, since they could have a second turn before the player's next turn.

Overall, I wouldn't recommend it, but feel free to try it out :)

  • 1
    \$\begingroup\$ The OP is asking for statistical explanations of the results of rolling initiative each round. This answer currently gives only anecdotal evidence (one of which is already mentioned in the question itself). Including statistical evidence would improve the answer greatly, as, in its current state, it doesn't actually answer the question asked \$\endgroup\$ Jun 5, 2021 at 5:14
  • \$\begingroup\$ @Medix2 Which is a different version of "do my homework for me" at a number of SEs that get shut down hard. \$\endgroup\$ Jun 5, 2021 at 16:19

For a game like 5e that is already very slow the most notable impact would be that combat would drag on even longer than it already does. Any other effect would be unnoticeable and/or average out over time. Look at Deadlands for an example of a game that does it right.

  • 2
    \$\begingroup\$ This is a reasonable answer, but backing up your assertion that everything else won't matter much would improve it. You also might want to give a little more detail about how Deadlands gets it right. As is, I can’t really judge your answer one way or the other, since I don’t know if you’re right or not and your answer doesn’t give me enough info to find out. I won’t downvote for that (though some might), but I’m not going to upvote, either. So to get more upvotes, you probably want to flesh this out a little more. \$\endgroup\$
    – KRyan
    Jun 5, 2021 at 1:25
  • \$\begingroup\$ What @KRyan said. If how Deadlands does this is a good format, showing how that works makes for a good frame challenge type answer. \$\endgroup\$ Jun 5, 2021 at 16:20

Not the answer you're looking for? Browse other questions tagged .