Ilbranteloth
Explorer
OK, I've been running my 5e games for several years without initiative now. However, Mike Mearle's tweet about initiative made me look at it again. The path it led me down was a similar initiative system using his system of multiple die sizes.
The reason I like the approach is that it automatically addresses speed factor without math. But it also allows you to sometimes get the shot in first if you roll well. For example, this is what I settled on:
So you roll initiative using the die indicated, and that sets the order. The nice thing about "lower is faster" for initiative is that it actually sets the order. #1 goes first, then #2, etc. I don't have an issue with ties, but if needed you can either compare Initiative modifiers or roll an opposed initiative check. You roll initiative every round.
Going back, I'm reminded that this is fairly similar to how AD&D, and even more so AD&D 2e worked. If you opted to use the Speed factor rule, those numbers were added to your initiative check (Dexterity modifiers were not). In 2e, spells often had casting times of a number - Casting time: 1, Casting time: 6, etc. and that was also added to your initiative.
In AD&D the round was split into segments, and casting times were in segments. If the casting time was longer than the number of segments you had left in a round, the spell was completed in the next round.
So in my new initiative system I had settled on 10 segments per round. If you want to change your action in the middle of the round, no problem, just roll the initiative die again. Since that could put you over 10, it would carry over into the next round. A segment in this system is 1/2 second, so we're not talking a lot of time.
Ready uses your reaction like the normal ready rule.
Undeclared just means you don't roll the initiative die until you declare an action.
--
OK, if you're still with me, I realized that the biggest problem I have with the way 5e turns work is that you take your move, action, and bonus action all at once. While everybody else is doing nothing. In theory the round is 6 seconds long, and everything "is happening together" but the way the game is designed, that's just not true. You can move 30 feet, attack, and if you're a rogue use a bonus action to move another 30 feet before anybody else can do anything. It really bugs me.
But if movement is separated from your action - everybody can move at any time during the round, then it solves the problem. Since the round is 10 segments, you move up to 1/10 (3 feet) of your move each segment. Somebody pointed out the issue this causes with people using a grid (I don't) so you could shoehorn it into 5 feet every 2 segments, and switch it to 12 segments.
One big advantage to this system - people can run away. If you start 30 feet away from your target and want to close for melee, they can actually attempt to run away. In which case you can chase them. One modification I made is that you can use your action and/or your reaction to dash. Of course, a rogue can also use their bonus action.
With the normal system, if you're 30 feet away from somebody, and you win initiative, there is nothing they can do to get away from you until after your attack.
--
So thinking further, what if there aren't any rounds? At least not in the usual sense.
What if there are just actions, with a variable amount of time for an action? In AD&D, if the speed factor of the weapon you were using was less than 1/2 of your opponents, then you got an extra attack. And of course there was the old "3 attacks every 2 rounds" multiple attack progression that also modified the usual wall between the rounds.
In this case, a combat starts with segment (or call it round if you want) 1. Everybody rolls their initiative die, and they end up on 1, 3, 4, 6, 7. They use their actions, and then instead of another round, as soon as you complete one action, you declare the next and roll initiative again. So the guy with initiative 1 attacks, and then intends to attack again on his next turn. So he rolls a d6 and gets a 4. So his next attack is segment #5. So his second attack comes before the slowest attack (and ties the second slowest). Is that really a problem? Note that creatures who get multiple attacks still get multiple attacks each time they use their action. So I don't really think so. But let's play it out a bit longer. Here are some random rolls:
Character A (d4): 1, 5, 9, 12
Character B (d6): 3, 6, 9, 13
Character C (d8): 4, 10, 11, 14
So the first two characters had three actions before the third character got his second action, but then he got his third action before they got their fourth. It's quite possible for somebody to have a lucky streak and get a lot of actions, and for somebody to have a really unlucky streak. But with no modifiers it's just the luck of the die.
What about things like Lair Actions? They can have a variable or fixed amount of time. Spell durations? If we're basing it off of a 10 segment round, then anything that lasts until the start of your next turn is 9 segments, and the end of your next turn is 10. Start of somebody else's turn? Doesn't change - because it's based off of them and affecting them for a certain number of turns rather than time. Spells that last a minute...last a minute. We just need to define the amount of time a "round" takes. Perhaps each slice of time is a second and we're just counting time?
We do have to address things like reactions. As Mike Mearles points out, bonus actions are often attached to actions. Some are independent (like Bardic Inspiration). We could just assign a fixed amount of time - 10 seconds between them minimum.
But really, the easy answer is that you only get one bonus action and one reaction between your actions.
I'm excited about the prospect, time to go do some testing...
Edit: This formatted very poorly. Try this instead: https://docs.google.com/document/d/12xPT-45PCBvaAJ3FaKI2Tkz2UmMGzoFcCzG0i7BfOss/edit
The reason I like the approach is that it automatically addresses speed factor without math. But it also allows you to sometimes get the shot in first if you roll well. For example, this is what I settled on:
Action | Initiative Die |
Attack, finesse or light weapon | d4 |
Attack | d6 |
Attack, heavy or 2-handed weapon | d8 |
Grapple/Grapple Maneuver | d4 |
Ready | Special |
Spellcasting, | d4 + level |
Undeclared | Special |
So you roll initiative using the die indicated, and that sets the order. The nice thing about "lower is faster" for initiative is that it actually sets the order. #1 goes first, then #2, etc. I don't have an issue with ties, but if needed you can either compare Initiative modifiers or roll an opposed initiative check. You roll initiative every round.
Going back, I'm reminded that this is fairly similar to how AD&D, and even more so AD&D 2e worked. If you opted to use the Speed factor rule, those numbers were added to your initiative check (Dexterity modifiers were not). In 2e, spells often had casting times of a number - Casting time: 1, Casting time: 6, etc. and that was also added to your initiative.
In AD&D the round was split into segments, and casting times were in segments. If the casting time was longer than the number of segments you had left in a round, the spell was completed in the next round.
So in my new initiative system I had settled on 10 segments per round. If you want to change your action in the middle of the round, no problem, just roll the initiative die again. Since that could put you over 10, it would carry over into the next round. A segment in this system is 1/2 second, so we're not talking a lot of time.
Ready uses your reaction like the normal ready rule.
Undeclared just means you don't roll the initiative die until you declare an action.
--
OK, if you're still with me, I realized that the biggest problem I have with the way 5e turns work is that you take your move, action, and bonus action all at once. While everybody else is doing nothing. In theory the round is 6 seconds long, and everything "is happening together" but the way the game is designed, that's just not true. You can move 30 feet, attack, and if you're a rogue use a bonus action to move another 30 feet before anybody else can do anything. It really bugs me.
But if movement is separated from your action - everybody can move at any time during the round, then it solves the problem. Since the round is 10 segments, you move up to 1/10 (3 feet) of your move each segment. Somebody pointed out the issue this causes with people using a grid (I don't) so you could shoehorn it into 5 feet every 2 segments, and switch it to 12 segments.
One big advantage to this system - people can run away. If you start 30 feet away from your target and want to close for melee, they can actually attempt to run away. In which case you can chase them. One modification I made is that you can use your action and/or your reaction to dash. Of course, a rogue can also use their bonus action.
With the normal system, if you're 30 feet away from somebody, and you win initiative, there is nothing they can do to get away from you until after your attack.
--
So thinking further, what if there aren't any rounds? At least not in the usual sense.
What if there are just actions, with a variable amount of time for an action? In AD&D, if the speed factor of the weapon you were using was less than 1/2 of your opponents, then you got an extra attack. And of course there was the old "3 attacks every 2 rounds" multiple attack progression that also modified the usual wall between the rounds.
In this case, a combat starts with segment (or call it round if you want) 1. Everybody rolls their initiative die, and they end up on 1, 3, 4, 6, 7. They use their actions, and then instead of another round, as soon as you complete one action, you declare the next and roll initiative again. So the guy with initiative 1 attacks, and then intends to attack again on his next turn. So he rolls a d6 and gets a 4. So his next attack is segment #5. So his second attack comes before the slowest attack (and ties the second slowest). Is that really a problem? Note that creatures who get multiple attacks still get multiple attacks each time they use their action. So I don't really think so. But let's play it out a bit longer. Here are some random rolls:
Character A (d4): 1, 5, 9, 12
Character B (d6): 3, 6, 9, 13
Character C (d8): 4, 10, 11, 14
So the first two characters had three actions before the third character got his second action, but then he got his third action before they got their fourth. It's quite possible for somebody to have a lucky streak and get a lot of actions, and for somebody to have a really unlucky streak. But with no modifiers it's just the luck of the die.
What about things like Lair Actions? They can have a variable or fixed amount of time. Spell durations? If we're basing it off of a 10 segment round, then anything that lasts until the start of your next turn is 9 segments, and the end of your next turn is 10. Start of somebody else's turn? Doesn't change - because it's based off of them and affecting them for a certain number of turns rather than time. Spells that last a minute...last a minute. We just need to define the amount of time a "round" takes. Perhaps each slice of time is a second and we're just counting time?
We do have to address things like reactions. As Mike Mearles points out, bonus actions are often attached to actions. Some are independent (like Bardic Inspiration). We could just assign a fixed amount of time - 10 seconds between them minimum.
But really, the easy answer is that you only get one bonus action and one reaction between your actions.
I'm excited about the prospect, time to go do some testing...
Edit: This formatted very poorly. Try this instead: https://docs.google.com/document/d/12xPT-45PCBvaAJ3FaKI2Tkz2UmMGzoFcCzG0i7BfOss/edit
Last edited: