Is it Unbalancing to Not Require an Attack for Shield Master?

Question

The Shield Master feat states:

If you take the Attack action on your turn, you can use a bonus action to try to shove a creature within 5 feet of you with your shield.

I've encountered several nonsensical situations at the table where it was helpful for a Shield Master player to just shove someone and not attack them (perhaps dashing away or drinking a potion was a better idea). Indeed, it's creating occasional instances where a character shoves to get themselves out of danger and then needs to move right back into danger to comply with the attack requirement.

I'm considering a house rule to not require the Attack action when using this feat and if it created imbalances. Have others tried this and if so, what problems did you see (if any).

The issue I have is that if you don't have the feat, you have to use your Action to shove somebody. Given that the player has expended a feat, it seems they should get some versatility with it. My vision would be they could shove, then Dash away to safety; shove, then cast a ranged spell or make a ranged attack without penalties; shove, then take the Dodge action.

As the feat is usually taken by a tank, it'd be nice for them to be able to create breathing space without having to put themselves back into danger.

Answer 1

There is a false premise in your question. You say "Given that the player has expended a feat, it seems they should get some versatility with it." - they do: they get the ability to make one more shove than anyone else.

Given that it is not guaranteed that your first (or any) attempt at shoving a creature is guaranteed to work(unless you have a net +19 on the contest), having another chance is a huge advantage. If your chance of succeeding with your shove is \$p\$, your chance of succeeding with 2 shoves is \$2p-p^2\$, with 3 it is \$3p-3p^2+p^3\$ and with 4 it is \$4p-6p^2+4p^3-p^4\$, and so on. For an evenly matched pair, the chance of succeeding with 1 is \$0.47\$, 2 is \$0.724\$, 3 is \$0.853\$ and with 4 is \$0.925\$. I would argue that this is some measure of "versatility".

In addition, they can "make a ranged attack without penalties" (assuming that they are now not within 5 feet of an enemy) - any remaining attacks from the attack action are available to make ranged attacks (or melee attacks, or grapples, or more shoves etc.).

Thematically, the designers appear to be saying with this feat - when you do this aggressive thing (attack action), you can do that other aggressive thing (shove someone). This is different from combining an aggressive action with a non-aggressive action like a Dash or Search.

That said, will allowing it to be combined with dash break your game? Probably not. It will be advantageous in a few encounters like the one you describe or where the player is out of movement range (but within double) as it will allow them to close and shove. Given that the PCs are expected to win every encounter - who cares about such a minor advantage?

Answer 2

The action economy strictly prevents exchanging Actions for Bonus Actions (and vice versa). You need special features to do so (for example, Cunning Action).

Consider the example of the grappled character underwater. Normally, escaping a grapple requires an Action (trying to break the grapple, forcibly moving the grappler, killing the grappler, etc). Allowing a shove attempt as a Bonus Action (freeing up the Action for Dash) means you are allowing this Action to made as a Bonus Action.

This is unbalancing; it is your call as to whether it is "too" unbalancing. It did require the character to choose a Feat, which is a high cost, so I don't think its all that unbalancing. On the other hand, the feat also gives them useful other stuff.