D&D 5E Little rules changes that still trip you up

[I'm A Banana]

I found the problem.

The Assassin has already declared the attack.

The attack is happening.

You can't go back in time and make the attack not happen. Further, the DM shouldn't be saying which NPCs won initiative. The DM should be narrating what the NPCs do.

The NPC when they lose initiative does nothing. The NPC win they win initiative does nothing. So they are indistinguishable. There is nothing for the DM to describe because the NPC isn't doing anything.

The only difference is that if the NPC wins initiative they might have a possible reaction to an attack. If that happens then the DM can say, 'okay Assassin, you have made your attack but your foe throws up the Shield spell (or whatever) at the last second.'

Yeah, this is not clear at all that it is RAI and it's definitely not RAW. There's no other place in the game where you don't get to decide your action on your turn. You don't even have to react to triggers for readied actions if you don't want to. Your adjudication is a solution, but it's one that significantly hoses an assassin player, makes an ambush a much less appealing tactic, and goes way beyond what the rules spell out ("forcing" a player to take an action that maybe they don't want to take anymore).

The alternative would be to roll initiative after the attack resolves, but that doesn't sound like it's RAI even though it'd be RAW, and it might make the assassin TOO powerful because they get two assassinate attempts.

A third option might be to use the Ready action out of combat ("I ready an action to snipe the guard as they come around the corner")...but that is basically the same as rolling initiative after the attack resolves, and potentially allows the assassin two assassinate attempts, still.

This is all part of why surprise is befuddling as frig and one of the major reasons that this is one of the rules changes that still trip me up.

 

[thalmin]

The surprised character who wins initiative is not acting before the surprising attacker. If he is surprised he doesn't act during the first round. But he can react. Something the ninja did in drawing or shooting the bow, throwing the star, swinging the sword, whatever caught the attention of the alert victim, who is able to react.

This works both ways. After all, a lot of the time the PCs are the ones who are surprised.

 

[Flamestrike]

Charmed by itself doesn't do much (though if you're a spellcaster with a good Cha, it's useful!), but a lot of other effects pass through the "charm barrier" first.

I really like this personally. Rather than charmed being 'auto social skill win' it simply makes the target more succeptable to the party face, or lets you act as the party face for a short period of time.

Invisibility is the same. Its useless unless you have a good Dex + Stealth and/ or cunning action. You still need to take the hide action for it to work. If you dumped Dex and dont have Stealth proficiency, youre always gonna be better off casting it on the party rogue.

Both spells arent auto invalidate the class thats spent a lot of resources getting good at it any more. They just create favorable conditions that make it easier.

Things like this contribute to taking the 'Tier 1, I can do everyones job better than them' tag off casters. They might be able to blow a precious resource and concentration to do the job nearly as good as the party face or rogue, but they dont invalidate that class by doing so. The path of least resistance is to use that resource to assist the PC thats already good at the task at hand, which is much better direction from a teamwork based game.

 

[Flamestrike]

The wonkiness there is that the assassin hasn't actually done anything for the victim to react to when the victim stops being "surprised." There's no attack yet.

Youre falling victim to a sequencing fallacy.

The assasin launches his attack. This is what triggers the initiative roll (an opposed dex check to see if the target can react in time to that attack).

In your example, its resolved via simple narration.

Assasin player: 'OK DM, I'll shoot the bugbear from hiding!'
DM: 'Cool mate. You draw back your bowstring, and the bows frame shudders with the slight creak of straining wood as you line up the foul creatures heart in your sights. The Bugbear is surprised. Roll initiative'.

From this point it doesnt matter what the PC and/or the Bugbear roll. That bugbear is copping an arrow regardless of a higher or lower initiative result (he gets higher and he cant act or take turns) - the initiative check is only there to see if the Bugbear can react to it in time to take a reaction (if any) to the attack, and to sequence the combat.

Should the Assasin roll higher in initiative, he actually gets two turns before the Bugbear can act (including the ability to assasinate him on turn one!).

For a surprised critter, the initiative check on round one only really determines if he can act just in the nick of time to take a reaction to whatever surprised him (a shield spell, or a Monk deflecting arrows, or a battlemaster riposting or parrying the attack).

If the attacker rolls higher, he not only removes the ability of the target to take reactions at all on round one, but also gets two whole turns in a row (which in 5E, is a death sentence for most targets).

 

[I'm A Banana]

Youre falling victim to a sequencing fallacy.

The assasin launches his attack. This is what triggers the initiative roll (an opposed dex check to see if the target can react in time to that attack).

In your example, its resolved via simple narration.

Assasin player: 'OK DM, I'll shoot the bugbear from hiding!'
DM: 'Cool mate. You draw back your bowstring, and the bows frame shudders with the slight creak of straining wood as you line up the foul creatures heart in your sights. The Bugbear is surprised. Roll initiative'.

From this point it doesnt matter what the PC and/or the Bugbear roll. That bugbear is copping an arrow regardless of a higher or lower initiative result (he gets higher and he cant act or take turns) - the initiative check is only there to see if the Bugbear can react to it in time to take a reaction (if any) to the attack, and to sequence the combat.

Should the Assasin roll higher in initiative, he actually gets two turns before the Bugbear can act (including the ability to assasinate him on turn one!).

For a surprised critter, the initiative check on round one only really determines if he can act just in the nick of time to take a reaction to whatever surprised him (a shield spell, or a Monk deflecting arrows, or a battlemaster riposting or parrying the attack).

If the attacker rolls higher, he not only removes the ability of the target to take reactions at all on round one, but also gets two whole turns in a row (which in 5E, is a death sentence for most targets).

That's not a fallacy, that's just the rules in the book. There's nothing I can see that locks the attacker into attacking in the RAW, and it's not at all clear that that's the RAI. The use of initiative would, in fact, suggest that the attacker's turn happens just like any other turn in combat: you decide what you do on your turn, and if you decide you don't want to attack on your turn - because the enemy is no longer surprised because they went first - that would be no different from deciding you didn't want to act on your turn because the target you were fighting got charmed and is no longer a threat.

The surprised character who wins initiative is not acting before the surprising attacker. If he is surprised he doesn't act during the first round. But he can react. Something the ninja did in drawing or shooting the bow, throwing the star, swinging the sword, whatever caught the attention of the alert victim, who is able to react.

This is also a fair ruling, but I don't think it's RAW, and it's not clear that it's RAI. The ninja assassin was never detected - he's still hidden, and if he makes no attack, he remains hidden, and there's nothing to alert the guard to his presence.

Which is enough to say that Surprise Trips People Up.

Stealth, too, while I'm at it! I've got a ruling that I think works well enough, but it is far from clear that I'm doing it RAI or RAW. When conducting prolonged stealth, how long does one stealth check last for? I've basically ruled that it applies up to your speed (assuming you aren't seen in the meantime), but I can see cases for half speed, double speed, or "however long I really feel like as a DM," too!

 

[Flamestrike]

Order of Operations:
1 - Assassin: "I shoot my bow at the unaware guard!"
2 - DM: "Okay, the guard is surprised. Roll initiative."
3 - Init is rolled. Guard wins.
4 - DM: "Okay, the guard is not surprised anymore because of reasons. Your turn."
5 - Assassin: "Okay, I don't shoot the guard." / "...I ready an action to shoot the guard when she IS surprised."
6 - DM: "Uhh...okay, no combat happens, and the guard continues on its patrol, still unaware of you. What do you do?"
7 - Assassin: "I shoot my bow at the unaware guard!"
8 - ...etc?

No, the player declared an action at 1.

Do you let players declare actions and then retrospectivly decide not to do so, after the outcome of a roll is not favorable to them?

Its no different to:

1. Assasin: 'OK DM - Ill try and wrestle the Guard into the spiked pit!'
2 - DM: "Okay, Roll opposed Athletics."
3 - Athletics is rolled. Guard wins.
4 - DM: "Okay, the guard wins, pushing you into the pit."
5 - Assassin: "Okay, I don't wrestle the guard."

If a player declares he's going to shoot someone, he does so. He doesnt get to change his mind simply because the DM called for an opposed Dex (initiative) check to see if his surprised target can react in time.

Once those dice hit the table, you're rolling to resolve an action the player has declared. All bets are now off, and I certainly dont allow 'takings back' at this stage!

 

[Flamestrike]

There's nothing I can see that locks the attacker into attacking in the RAW,

Didnt your player just declare an attack? He's fired his arrow or whatever. Hes told you thats what his character is doing, which is why you the DM have moved to rolling dice rolling to resolve the action (Dex check followed by an attack roll).

We've moved beyond declare action, and into action resolution. He's gone too far to 'go back' now!

Arent you only rolling that Dex (initiative) check to see if the target can react in time to that attack (that has just been declared)?

Maybe it works differently at your table, but if a player declares they want to do X, and then roll dice to resolve that action (and the Dex initiative check is part of that resolution here remember) then I dont allow 'takings back'.

 

[I'm A Banana]

No, the player declared an action at 1.

That's a fine ruling, but in the rules, the player doesn't declare their action until their turn. There's no "declare an action" step in the RAW before Surprise is determined by the DM. There's no "trigger" for surprise that must happen regardless of the initiative roll.

Didnt your player just declare an attack? He's fired his arrow or whatever. Hes told you thats what his character is doing, which is why you the DM have moved to rolling dice rolling to resolve the action (Dex check followed by an attack roll).

If that's the case, then initiative should be rolled after that attack is resolved (potentially letting the target be surprised for two attacks), not before with the attacker being required to do a certain action. Which, again, isn't anything the rules ever say happens, though I wouldn't dispute a DM threading the needle like that in their own campaign because, again, Surprise is a "little rule change that still trips me up."

 

[ad_hoc]

Your adjudication is a solution, but it's one that significantly hoses an assassin player, makes an ambush a much less appealing tactic, and goes way beyond what the rules spell out ("forcing" a player to take an action that maybe they don't want to take anymore).

It does none of those things. The auto-crit on the Assassinate ability is a minor benefit that Assassins get. I think the only reason it is there is because they probably already have advantage on an attack if they have surprised their opponent so they need something else if they also win initiative in that circumstance.

It's important to note that all aspects of the Assassinate ability work only if they win initiative. This is deliberate.

Ambushes are still incredibly powerful. Enemies get 1 less full turn than if they hadn't been surprised. It's just that if they win initiative they might be able to use a reaction if they have one. Houseruling to create a surprise round like 3.x is limiting. It is going back to the regimented style of move actions and the like.

There is no forcing the player to do anything. The player should have no idea that the opponent won or lost initiative if the opponent doesn't doesn't do anything.

I'm not sure how else to explain it so I will leave it there. Yes, the game works differently than it did in 3.x. Yes, this is on purpose. Yes, it is okay not to like it. No, following the rules of 5e is not a punishment to players.

 

[Flamestrike]

That's a fine ruling, but in the rules, the player doesn't declare their action until their turn.

There were no turns when the player declared his action. You're in narrative time.

If he declares he is doing X, and then dice are rolled to resolve that action, its happening. Im not sure that needs a rule.

If during a social encounter, a player says 'I'll hurl my sword at the King looking to impale him on his throne!' (after the shocked silence at the table) you say 'Ummm OK' roll initiative.

Lets see if the guards (or other players, or the king) can react in time to thwart this madman as he goes for his sword with murder in his eyes.

There's no "declare an action" step in the RAW before Surprise is determined by the DM.

You're not in 'combat' prior this point so there are no turns. Something has happened to trigger the switch from narrative time (when turn sequencing doesnt matter) to combat time (an abstraction, with turn sequencing for ease of play).

If it really bothers you, I suggest letting the player roll his attack first, and then before you compare it to the critters AC or make a save or whatever, roll initiative to see if the target is surprised or not.

It would go:

1: Player initiates combat by declaring an attack.
2: Player rolls his attack roll, casts his spell or whatever. None of it is resolved yet.
3: DM and Player now make opposed dex checks to see if the target can react/ dodge the attack/ is caught totally unawares by the attack/ counterspell in time (and to sequence the rest of the combat).
4: With that information now at hand, resolve 2.

Same outcome, different sequencing of rolls. I guess I just trust my players that when they say 'Im going to draw back my bow and shoot the Bugbear' once I accept that this is what they are doing and dice start getting tossed around to resolve this action (a Dex check to see if the Bugbear can react in time) there is no 'takings back' if those dice are unfavorable to them.

Its never an issue for me as I roll initiative in secret behind the screen. The players know what they got, but not the monsters result (although after a round its pretty obvious).

In this situation (player declares an attack initiating combat), the player and I roll initiative, and I note the results. Then (regardless of the result) I look at the player and say, 'roll your attack'.

Once resolved, the player finds out then if the creature is surprised or not.

There's no "trigger" for surprise that must happen regardless of the initiative roll.

If my players are wandering down a forest trail, and are ambushed by a bunch of goblins, surprising three of them (one, Seans Ranger - makes his perception check), then I narrate it thus:

'Alright; as you walk down the trail, your thoughts turn to your recent adventures, and warm beds that await you at town. Suddenly the air is filled with a whistling sound and a hail of black fletched arrows fills the sky! You're all surprised... aside from you Sean - your Ranger spots several goblins off to the north side of the trail, about 40' from where the party is. Roll initiative.'

Even if Sean goes first and fireballs the goblins into ash (he multiclassed Wizard) before the Goblins attacks are resolved, nothing changes. You just narrate it as the arrows falling harmlessly around the party, and get on with the game.

Remember, combat is simultaneous (despite the cyclical nature of turns its resolved in). If a player declares he's doing action X, then action X is happening.

The opposed Dex check made by all combatants is simply checking reaction speeds to this stimulus. All initiative is is simply an opposed Dex check to see if the surprised critters can maybe react in time to get a reaction up to that stimulus, and to see if they cop one whole rounds worth of attacks before they can do anything, or two.