How much damage does the Shield spell soak at each level?

[Daphne]

On average, how much damage do you reduce by casting it? In your experience, how many hits does it turn into a miss?

We can start by looking at the average damage of monsters by CR, as per the DMG table:


LVL
DMG


1st
11.5


2nd
17.5


3rd
23.5


This is probably not very accurate though.

[stoutstien]

with DMG chart doesn't factor in is the number of attacks that damage is broken down into. The good news that AC is pretty linear. +1 AC reduce your chance of being hit by 5% until you hit the soft cap where only a critical hit lands.

[Daphne]

with DMG chart doesn't factor in is the number of attacks that damage is broken down into.

Indeed, most monsters have around 2 or 3 attacks per turn I'd say.

[stoutstien]

I forgot to bring up the other end of the range. it's possible that your base AC is low enough that shield won't really help at all.
Let's set your standard wizard who gets caught without mage armor is attacked by a fire giant. There's a good chance that even with shield they are going to get hammered.

[Quietus]

In practice, soft floors and ceilings are irrelevant. You aren't using Shield when it won't stop an attack, and you aren't using Shield when they're already missing. You're using Shield specifically when it will make a difference. If you want to make estimates of how much damage it will absorb, then I would start by looking at the average damage done by a monster in one attack at a given level, and assume that Shield will stop exactly that one attack, and no others.

[jas61292]

In practice, soft floors and ceilings are irrelevant. You aren't using Shield when it won't stop an attack, and you aren't using Shield when they're already missing. You're using Shield specifically when it will make a difference. If you want to make estimates of how much damage it will absorb, then I would start by looking at the average damage done by a monster in one attack at a given level, and assume that Shield will stop exactly that one attack, and no others.

Technically, that is a playstyle thing. I know it is common for people to play that the DM tells them the exact number of the attack roll, and then they choose to use Shield only if it will work. But I also know that there are people who play such that the DM simply tells them that the attack hits and not the number rolled, and so they have to guess whether or not Shield will help. Neither is more or less RAW.

If calculating things, it would probably be good to do separate calculations for games where you know if it will work and games where you do not. Obviously it is a better spell in the former case, but it is a strong spell regardless, and the information is still useful in the latter.

[stoutstien]

In practice, soft floors and ceilings are irrelevant. You aren't using Shield when it won't stop an attack, and you aren't using Shield when they're already missing. You're using Shield specifically when it will make a difference. If you want to make estimates of how much damage it will absorb, then I would start by looking at the average damage done by a monster in one attack at a given level, and assume that Shield will stop exactly that one attack, and no others.

The estimated damage reduction is 25% per attack including the triggering one.

[clash]

I mean even if you dont know if shield will help it is easy to calculate. Just take the monsters attack bonus at a specific cr say at cr 5 they have an average of +7. Take your own ac. Call it 15 for a wizard with 14 dex and mage armor. That means on average they hit on 8 or higher which is 12 numbers. You can make it miss on 5 of those. So on average you block a single attack 42% of the time. Assume you always use it on a hit, take the creatures average damage per hit say 20 and multiply it by 42% for an average of 8 damage blocked. I mean I am just making these numbers up, but that's how you would go about it not knowing what they rolled when they hit. Obviously if your ac was higher, say 20 then they need 13 or higher to hit which is only 8 numbers you are causing 5 of those to miss reducing 63% of damage coming your way. So the higher your ac is the more effective shields becomes when you use it, but the less often you need to use it.

[Pex]

Couldn't say what the raw numbers are, but I can remember when it literally saved my character's life. That's what really matters.

[Misterwhisper]

Couldn't say what the raw numbers are, but I can remember when it literally saved my character's life. That's what really matters.

Yep, and I am sure it has happened many many times.

The spell is absolutely amazing.

It is a great spell at level one or twenty
It is a great spell on a full caster, a half, or a third.
It is even a reasonable pick for magic initiate.

[LudicSavant]

The estimated damage reduction is 25% per attack including the triggering one.

That would only be true if A) your chance to be hit was 100% before applying Shield and B) you had to use Shield before rather than after the roll. Shield is actually a lot better than that.


with DMG chart doesn't factor in is the number of attacks that damage is broken down into. The good news that AC is pretty linear. +1 AC reduce your chance of being hit by 5% until you hit the soft cap where only a critical hit lands.

It should be noted that the returns on AC are not linear. For example, if you go from a 10% chance to be hit by something to a 5% chance to be hit by it, you just doubled your effective durability (+100%). By contrast if you go from 100% chance to be hit to 95% chance to be hit, you just got +5% to your effective durability. Mind, this gets altered a little bit by crits, but this should be enough to establish the shape of the curve. It'll look something like this:

https://i.postimg.cc/MTvcMjvC/Arcane-Trickster5save-HP.png

Also, Shield is worth a lot more than just a 1 round +5 AC buff, because of the "you can use it after seeing the roll" aspect, which means it's almost guaranteed to negate the first hit (if it wouldn't negate the first hit, you don't use it).

So Shield's value is basically the full damage of the first hit, plus the difference in DPR between what all future hits in the round would do to you with vs without Shield. In other words, Shield is really good.

[stoutstien]

That would only be true if A) your chance to be hit was 100% before applying Shield and B) you had to use Shield before rather than after the roll. Shield is actually a lot better than that.



It should be noted that the returns on AC are not linear. For example, if you go from a 10% chance to be hit by something to a 5% chance to be hit by it, you just doubled your effective durability (+100%). By contrast if you go from 100% chance to be hit to 95% chance to be hit, you just got +5% to your effective durability. Mind, this gets altered a little bit by crits, but this should be enough to establish the shape of the curve. It'll look something like this:

https://i.postimg.cc/MTvcMjvC/Arcane-Trickster5save-HP.png

Also, Shield is worth a lot more than just a 1 round +5 AC buff, because of the "you can use it after seeing the roll" aspect, which means it's almost guaranteed to negate the first hit (if it wouldn't negate the first hit, you don't use it).

So Shield's value is basically the full damage of the first hit, plus the difference in DPR between what all future hits in the round would do to you with vs without Shield. In other words, Shield is really good.

The seeing portion is not universal. The shield states that you can activate it when you know you've been hit but know where does it say that you actually know the value of the roll.
Tables rolling in the open or not have a pretty big impact on stuff like shield or protection fighting style.

I don't roll in the open but I do tell the players if it's a Crit before they do decide to use whatever feature.

I should have been more clear that decreasing your chance of being hit is not a direct correlation to the amount of survivability you gain for it. It's rarely going to be a case where someone AC is low enough that shield is not a huge boost in survivability.

[LudicSavant]

The seeing portion is not universal. The shield states that you can activate it when you know you've been hit but know where does it say that you actually know the value of the roll.
Tables rolling in the open or not have a pretty big impact on stuff like shield or protection fighting style. Yeah, it does lose some value if your DM hides all their rolls for whatever reason, but even then it's mathematically better than applying the AC boost prior to the hit being confirmed.

[stoutstien]

Sure, it loses some value if it's a hidden roll, but even then it's mathematically much better than applying the AC boost prior to the hit being confirmed.

Always. It's the defensive equivalent of divine smite. It's hard to waste it. Find familiar is the only first level spell that even close to keeping same value as shield does for the same slot.

[JellyPooga]

It's probably worth noting the difference in value of the slot being used at different levels. A 1st level spell slot at 1st lvl is a much greater investment than it is at 3rd, 5th or higher when the value of that 1st level is commensurably lower. This determines, to an extent, the frequency of use. If you're out of slots, whether because you've spent them already or are reserving them for offence or utility, then you won't be as willing to blow the slot on defence.

So while intuitively you might think Shield is at its best at lower levels, it probably actually stops more damage at higher levels than at low. Conversely, it probably saves your life more often at low levels.