Explaining Hit Points - An essay by Justin Alexander


I found this to be a great way to open up the mind to how HP’s can be perceived in our beloved hobby. Found at: https://www.thealexandrian.net/creations/misc/explaining-hit-points.html


*An Essay by Justin Alexander

Hit points have been around for more than 30 years now. (Longer if you count their antecedents in wargaming.) So you might think that, by now, people would have a pretty firm grasp on how these mechanics worked, what they represented, and what it means to lose hit points.

You might think that. But you would be horribly mistaken.

Basically I’m writing this little mini-essay because I’m tired of engaging in the same painful debate every two months. I want to be able to simply point people to this essay and say, “Look, this is the way it works.” (This won’t actually have much effect with people who prefer to be mired in fallacies and foolishness, but it will at least delay the day on which I will inevitably succumb to a lethal case of carpal tunnel syndrome.)

So, first I’m going to repudiate the two primary fallacies which lead the innocent astray when it comes to understanding hit points. And then I’m going to pull my Big Reveal and explain the beautiful abstraction which lies at the heart of the hit point system.

(It should be noted that this essay specifically deals with the type of inflationary hit point system found in D&D. The term “hit points” may also be used for very different damage tracking systems, to which this essay probably won’t apply at all.)



The first fallacy goes like this:

  1. Dupre has 100 hp.

  2. A goblin with an axe has hit him 10 times and done 78 points of damage.

  3. Clearly, the goblin has hit Dupre in the face 10 times and Dupre is still alive! That’s ridiculous!

The fallacy lies in the illogical leap from point 2 to point 3. A moment’s consideration clearly reveals that there is absolutely no reason to assume that, every single time the goblin landed a solid blow, it meant that the goblin planted his axe straight between Dupre’s eyes.

For example, imagine that you’re talking to someone in real life and they said, “Did you know that Bill was actually shot three times during a mugging a few years back?” Given that Bill is still alive, would you immediately assume that, during the mugging, Bill was dropped to his knees, a gun held to his head execution-style, and the trigger pulled three times?

Of course not. You would assume that Bill was probably hit in the legs or the arms. If he was hit in the chest, you’d assume that he only survived because he got prompt medical attention. And if one of the bullets actually did take him in the head, you’d know that it was a medical miracle that he was still alive.

Similarly, there’s no reason to assume that Dupre was hit in the face ten times with an axe. In fact, quite the opposite is true: There is every reason to assume that he wasn’t hit in the face ten times with an axe.



The second fallacy is most often committed because, after escaping the trap of the first fallacy, people go to the other extreme:

  1. Dupre could not have been hit 10 times in the face with axe and survived.

  2. Therefore, Dupre was never hit with the axe.

  3. Dupre won’t be hit by the axe until a blow causes his hit points to drop below 0. When that happens, the goblin will have finally hit him with the axe.

The fallacy here lies in the leap from point 1 to point 2.

Let’s go back to the example of Bill’s mugging. If you friend said to you, “Did you know that Bill was actually shot three times during a mugging a few years back?” Would you assume that he meant, “The guy mugging bill shot his gun three times, but never actually hit Bill.”?

Of course not.

In terms of D&D, the nature of this fallacy is more explicitly revealed when you look at something like poison. If the orc’s axe is coated with poison and we embrace this fallacy, then Dupre has been exposed to the poison 10 times despite the fact he’s never actually been hit by the axe .

And you can reproach this fallacy from multiple directions: If the hit point loss from blows that don’t “really” hit is because Dupre is wearing himself out from dodging, why is dodging a +1 flaming sword more exhausting then dodging a +1 sword ?

And, of course, you also have the oddity that, apparently, dodging a blow from a sword can be even more deadly to you than being hit by the sword.


The trick to understanding the hit point system is understand that a hit point is not equal to a hit point. In D&D, 1 hit point of damage always represents a physical wound. However, the severity of the wound represented varies depending on how many hit points the victim has.

For a character with 1 hp, that 1 hp of damage represents a serious wound – a punctured lung, a broken leg, or something of that ilk.

For a character with 10 hp, that 1 hp of damage represents a meaningful wound – a deep but or a broken rib.

For a character with 100 hp, that 1 hp of damage represents an essentially inconsequential wound – a scratch, a bruise, or the like.

The reason any particular character has fewer of more hit points (and, thus, varying the severity of any given wound they receive) is abstracted. For some characters its skill; others luck; others physical toughness; others divine grace; others magical protection; and so forth. For most PCs, it’s some combination of all these things.

This is a beautiful abstraction because it allows for quick, simple, and entertaining gameplay. One could certainly design a system with variances in skill, luck, toughness, divine favor, magical protection, and the like were all separately modeled. Many such systems exist. But none of them are as simple, easy, or fun or as hit points have proven to be.


No. The system is an abstraction, and that brings with it both advantages and disadvantages. The advantages are simplicity and completeness. The disadvantage is that you can’t reliably pull concrete information back out of the abstraction – and, if you try, you’ll eventually find corner cases where seeming absurdities crop up.

For example, it’s theoretically possible for a sufficiently weak character to deal no more than 1 hp of damage per attack. Such a character could, theoretically, land 100 blows on a character with 100 hp before taking him down. Add poison to the scenario, and you’ve now locked yourself down to a scenario where the weak character is, apparently, whittling his opponent to death.

Such corner cases are statistical oddities, but they’re going to reliably crop up in any system which doesn’t require several additional orders of complexity.

Which leaves the only significant and intractable problem with the hit point abstraction: The cure spells. Despite the fact that the number of hit points required to represent a wound with a particular severity varies depending on the character’s total hit points, a cure spell heals a flat number of hit points. Thus, a cure light wounds spell used on a 1st level fighter will heal grievous wounds. When the same spell is used on a 10th level fighter, on the other hand, it can’t handle more than a scratch.

This is a legacy issue which has been retained for reasons of game balance. But if you want to fix this, simply have cure spells work more like natural healing: Multiply the number of hit points cured by the creature’s HD.

Hit points aren’t a one-size-fits-all solution to tracking wounds in roleplaying games. There are lots of reasons why you might want a more concrete representation of actual wounds or a realistic modeling of incapacitation.

But hit points are often attacked for the most erroneous of reasons. And, as I said up front, I don’t necessarily expect this little essay to make any sort of huge dent in that tidal wave of ignorance and faulty logic. But it might help me keep my blood pressure down.


Isn’t that the issue with increasing hp so many times?
How does a person go from 10 hp to 50? When a goblin does hit Dupre square between the eyes, which can and will happen, what should happen?


Exactly. In my mind… it should be a kill. But, unless your players are told up front that this is a lethal game (the way Dungeon Professor apparently runs his games), then what does one do? Do you… fudge things somewhat? Hurt them but not to death? Come up with some lame reason they don’t die after all? It’s tough when you want some level of realism but there is no real way to duplicate realism in dice form. I don’t think it’s been actually figured out yet.

This is one of the key reasons I stick around here… waiting, hoping that our all mighty deliverer from the past systems brings us forward into the final one true system to rule them all. Brandish… we await your final masterpiece. You are the Beethoven of RPG system developing.


In my opinion the single biggest issue is that in many games there are simply too many damn hit points. Conversely, I think ICRPG does very well with the 10 HP setup, where taking damage actually means something. I also think that narration plays a big part and it is something that can’t be quantified in a ruleset (which is the beauty of it). Here’s how I would handle the goblin axe to the forehead thing:

Goblin swings an axe to the forehead of a helpless target: The target is dead, end of story.

Goblin swings an axe at a targets head: First question: Does he hit? (hit roll) Second question: If he hit, how well did he hit? (damage roll). Narration fills in the graphics based on the damage roll. (yes, the goblin swings at the target with the intent of hitting him in the face, the damage roll indicates how well this was accomplished). Just because a goblin swings at a targets head doesn’t mean he actually hit the target in the desired location, this is combat, it’s chaos.

Here’s another layer (For ICRPG that handles this situation well IMO): The goblin swings an axe at the target with the specific intent of focusing on the targets head. GM: “That is a called shot, the TN is now HARD, if the goblin succeeds he does ULTIMATE damage instead”. This gives the goblin a chance to take out his opponent in one fell swing but because it is a called shot it will be more difficult to perform.

I don’t mean to sound abrupt but don’t hold your breath for a final one true system to rule them all by any designer at any point in the future, I think that is an unrealistic expectation (and impossible). What we need is to enjoy what we have in front of us, not overthink things too much, and appreciate the imperfect perfections.


So, by setting a target of say… 17 for a called shot at the head which is up around 5’ in the air and Goblin is a short dude… he rolls a 17 which could indicate he JUST hit and thus an ear got cut off. Or maybe Goblin dude rolled a 20. Crit hit. But, lets say the GM doesn’t use crits for monsters… thats fair. Okay, so the 20 would be a solid hit… death save? Dead? Dying? If Goblin dude rolled a 25 due to stats allowing to roll over 20 in your system… then for sure dead in my opinion.


Amen to that mi amigo.


Just curious, what system do you have in mind when you are giving this example?

Ok, I’m going to sort out your example from the ICRPG lens (or at least how I run ICRPG which could be different from someone else).

The goblin makes a called shot, the room target is 14 and a called shot is HARD so the TN would be 17. The 5’ height and short goblin is irrelevant mechanically but narratively adds flavor (since the room TN and HARD TN are already set). The goblin rolls a 17, indicating a successful hit. ULTIMATE EFFORT is rolled and does 5 damage, reducing the target to 5 HP, it’s a substantial hit!

If the same attack is made except a 20 is rolled that indicates a critical success, scoring an EXTRA ULTIMATE EFFORT roll! In this case a 5 and 12 are rolled for damage, so 17 HP of damage is done, a lethal hit! The target drops unconscious and the death timer starts on his next turn (a 1D4 to indicate how many rounds until the target kicks the bucket for good).

This is just my theory but I think the death save/death timer mechanic is put into games to soften the potential end to player characters, it’s a way of keeping a level of happiness at the table. I also think it gives characters the chance for a dramatic re-entry into the story against all odds (I picture how things go all the time in super hero movies, the hero popping up in that critical moment to fight the baddie). Death saves can feel a little unrealistic and maybe too forgiving but I think they are ultimately a good thing to have.

I do have to say that if I were running a game I would not use called shots against characters, unless there was some sort of special ability like it for an end boss or something. I think generally running called shots against characters is too aggressive and antagonistic and would not be good for the morale at the table.


Well said. I’m thinking of any system really but your answer with the ICRPG death save mechanic is a good idea to add to any system in my opinion. I have.

Balancing the reality off the narrative aspect such as short vs tall being handled simply by the TN changing is good. It needs to be verbalized I think so the table gets that is part of the TN and I personally feel that more narrative is good to keep players engaged as opposed to just spouting out numbers and TN’s all the time. Yawn.

Great answer mate!


I think the core take away of Mr. Alexander is correct, but he’s in desperate need of an editor. Yeah, HP is an abstraction, since HP has no diegetic explanation; they’re like experience points, completely mechanical, representing something similar to reality, but not really functioning like reality in any way. I think calling different ways of running HP a “fallacy” is a bit rich. As far as I’m concerned HP could make a hit sometimes mean a real hit, and sometimes means a near miss - and there’s no “wrong” way to do it. It’s a functional system, and too much digital ink has been spilled trying to connect the mechanics to the fiction. The best explanation I heard was describing HP as dramatic tension - and I think that works. Injury in RPGs work most of the time like they do in action movies - largely cosmetic, and for dramatic purposes, but mileage may vary for any explanation.,


HP is an abstraction, that is correct.

Alexander’s fallacies are also correct, except the second one has an oversight. You can clearly argue that a flaming sword is harder to dodge than a regular sword, because it is more dangerous, so that argument falls flat.

I see HP as a rough gauge of stamina (luck included) and nothing more. If an attack hits, I assume that the blow successfully landed on the target, but the severity, and sometimes the location of the blow is determined by the damage dealt. This is very simple for ICRPG like systems, where HP is low. For D&D and other games that have HP bloat, the severity is determined by the rough ratio of the damage dealt to the target’s HP maximum.

Sometimes a hit lands but it just scratches armor (low damage rolled), but you spend the effort to dodge it anyway. That is why your HP gets lower; your stamina is lower, you start to get tired. Eventually, you will be tired to even to lift your sword = your HP is zero. You are defeated.

This is more or less in line with this dude’s reasoning, but it includes all corner cases he mentioned. Under this abstraction, a goblin can do 1 HP damage each time and hit the target a 100 times, which will eventually “kill” it. Explaining this is very easy: Target absorbed the hits 99 times with success, but eventually his ankle broke, got a heart attack, just gave up, ran out of breath and any other misfortune we can imagine.

Real life works like this. You lift the same weights every day, but some days it feels significantly easy, some days it feels very hard, and sometimes you get injured. It means that you got dropped to 0 HP that day. :smile:


If HP is stamina, wouldn’t attacking also spend HP?

I don’t think HP has a diegetic component - it’s purely mechanical, born out of convenience and ease of use - and I think in many ways that’s a strength, because it doesn’t have to be consistent, but can fill whatever role the fiction demands.


That’s what I mean.

Sure, if you want a boring game. We conveniently overlook that portion.