Does shield retroactively change the result of an attack? [duplicate]

Question

Shield spell Casting time line says (PHB, p. 275):

1 reaction, which you take when you are hit by an attack or targeted by the magic missile spell.

(bold highlight by me)

PHB, p. 194, 1st paragraph under "Attack Rolls" say:

To make an attack roll, roll a d20 and add the appropriate modifiers. If the total of the roll plus modifiers equals or exceeds the target's Armor Class (AC), the attack hits.

So, disregarding the magic missile spell case, the trigger for the Shield is the fact of hit, i.e. attack roll was made and it was determined that the target was hit. What would be the point of casting the Shield spell at this point if attack is already resolved?

Spell effect description says:

An invisible barrier of magical force appears and protects you. Until the start of your next turn, you have a +5 bonus to AC, including against the triggering attack, and you take no damage from magic missile.

(again, bold highlight is mine)

The description seem to imply that it retroactively changes the outcome of the attack roll, otherwise the bolded text does not make any sense; but on the other hand it does not spell it out explicitly and it specifically has different wording for being hit by an attack vs. being targeted by magic missile (if the intention was to retroactively make attack miss, the wording could have used "targeted" for both cases; and PHB, p.194 explicitly mentions picking a target for attack, so targeting is a well-defined step).

Is there something that supports retroactive outcome change?

Answer 1

It does retroactively affect the attack.

One of the main benefits of the shield spell is that, when using it to defend against an attack, you know whether or not the attack would've hit you first. If it was phrased 'when you are targeted by an attack' rather than 'when you are hit by an attack', that would imply you wouldn't know the outcome and could potentially waste the spell against an attack that was already missing you.

In general, reactions occur after their triggers unless an explicit exception is made to have it 'interrupt' their trigger, which shield does (as quoted/emphasized in the question). To quote from @Rubiksmoose's excellent answer to 'Do reactions interrupt their triggers or not?':

Xanathar's Guide to Everything outlines the general rule for the timing of reactions:

If you’re unsure when a reaction occurs in relation to its trigger, here’s the rule: the reaction happens after its trigger completes, unless the description of the reaction explicitly says otherwise.

So, if the timing of the spell/ability you are using does not specify a timing then the reaction will come after whatever triggered it.

This is essentially just rephrasing what was already in the DMG since its release:

Various spells and features give a creature more reaction options, and sometimes the timing of a reaction can be difficult to adjudicate. Use this rule of thumb: follow whatever timing is specified in the reaction’s description. For example, the opportunity attack and the shield spell are clear about the fact that they can interrupt their triggers. If a reaction has no timing specified, or the timing is unclear, the reaction occurs after its trigger finishes, as in the Ready action. (D&DB)

Note the specific mention of shield interrupting its trigger in that quote from the DMG, for confirmation of that functionality.

Answer 2

It triggers after the attack but before damage is applied.

As you highlighted, the bonus applies to the triggering attack itself and so must be retroactive.