On a failed save, the target obeys the Flee command by

[Spending] its turn moving away from you by the fastest available means.

Three questions:

  1. Does that mean they take the Dash action?
  2. If not, do they take the Disengage action?
  3. In case the answer to 2 is no, do they provoke Opportunity Attack?

I'm leaning towards 'yes' to 1 ('fleeing' seems to suggest 'dashing'); no to 2 ('fleeing' doesn't suggest 'careful retreat'); and 'yes' to the 3rd (since it's not 'forced' movement, at least as per PHB page 195).

  1. If the Dash action is the fastest means of fleeing available to them, then yes, they will use it. For a counterexample, a monster with the ability to teleport (like a yugoloth) would probably do that instead.

  2. If the Disengage action is the fastest means of fleeing available to them, then yes, they will use it. (I really can't think of a way to make this true, though.)

  3. A creature under the effect of a Command to flee uses its own movement, so yes, it will provoke opportunity attacks.

    You also don’t provoke an opportunity attack when you teleport or when someone or something moves you without using your movement, action, or reaction (PHB p.195).

| improve this answer | |
  • \$\begingroup\$ on 2.: I'm not that into 5e, but in other systems' disengage equivalents equal it to one "unit" (squares, meters per second, foot per action, whatever speed is in your system) of moving away. so any monster with a 1 "unit" movement speed could still disengage as it gains no fleeing advantage of not using it. \$\endgroup\$ – Patrick J. S. Dec 19 '15 at 5:23
  • 1
    \$\begingroup\$ For 2. Goblins - get a bonus disengage or hide so can disengage and dash \$\endgroup\$ – Dale M Dec 19 '15 at 7:28
  • 1
    \$\begingroup\$ @PatrickJS it doesn't work like that in 5e. Disengage doesn't move you, but your movement doesn't provoke opportunity attacks for the rest of your turn. \$\endgroup\$ – xanderh Dec 19 '15 at 19:38
  • 1
    \$\begingroup\$ Sage advice backs this up: sageadvice.eu/2016/08/02/… \$\endgroup\$ – Shaun Oct 26 '18 at 7:05

The combat effectiveness of the Command spell is always a matter of how the GM interprets:

The spell has no effect [...] if your command is directly harmful to it.

If you explicitly include this proviso to the text of the Flee option, it becomes "The target spends its turn moving away from you by the fastest available means which does not cause it direct harm". If the target has a credible expectation of being stabbed, I'd rule that they Disengage rather than Dash. So:

  1. The target Dashes if they don't believe they'll be exposed to direct harm in doing so, such as opportunity attacks.
  2. The target Disengages if they do believe they'll be exposed to opportunity attacks.
  3. The target provokes opportunity attacks as usual if they did not choose to Disengage. (Which suggests they were mistaken about who was armed and dangerous.)
| improve this answer | |
  • \$\begingroup\$ @adclark09 Do you think playing "Russian roulette" wouldn't count as a directly harmful act, since it's only potentially fatal? I'd say an action can be potentially AND directly harmful; they're not mutually exclusive. \$\endgroup\$ – Sebkha Dec 20 '15 at 15:48
  • 5
    \$\begingroup\$ I don't think this is correct, given that one of the provided commands is Grovel, which grants advantage to melee attacks (which, at low levels, is almost like granting an additional attack, sans the damage). However, at higher levels, granting advantage to 4 attacks is more deadly than merely granting a single attack. \$\endgroup\$ – Khashir Dec 20 '15 at 19:39
  • \$\begingroup\$ Crawford's ruling contradicts this: "Command: Flee isn't directly harmful; OA may or may not happen. "Jump in the lava" & "Stab yourself"—harmful." \$\endgroup\$ – V2Blast Dec 24 '18 at 2:21

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.