Psychochild's Blog

A developer's musings on game development and writing.

11 May, 2016

Design Challenge: doing away with hit points

I’m still feeling a bit under the weather, so let’s do something a bit different. A weekend design challenge in the middle of the week.

So, here’s the challenge: design a combat system that isn’t the current standard “beat on a target until their bag of hit points is depleted.” I’ll post a few ideas, but get ready to share yours!

I had an idea a while ago of measuring the flow of combat. Actual combat tends to be less about beating an opponent into a pulp and more about wearing them down until you can go for an incapacitating blow. And, such a blow wasn’t always immediately lethal; sometimes people fatally wounded in battle would linger on for long afterwards. As medical technology got better, previously fatal wounds could be recovered from. In a world with magical healing, this would likely also be possible.

In the description of hit points in early D&D books, they were described as an overall measurement of combat ability. A “hit” didn’t necessarily mean that you were wounded, but that your combat effectiveness dropped by some amount. Perhaps you were put off balance, or had to strain yourself to get out of the way of a blow and were slower to dodge a potentially fatal blow. This is why fighters had more hit points, as they had more training to get the hell out of the way, not because they could take a sword perforating their spleen better than the wizard. But, people tend to think of hit points as overall health anyway, despite any more nuanced abstraction.

So, my idea was to measure two thing about a combat: the stamina/fatigue for each character, and the advantage one had. The idea was that a combatant would wait for the right time to strike, spending stamina to try to get an advantage in combat. Once you had enough advantage, you could go for that incapacitating blow. The idea was that each pair of combatants would have their own advantage measure between them, meaning that one person fighting three targets had to deal with 3 different advantage. Fighting 3 opponents became a lot harder unless they were pushovers!

I did a little mockup of the combat system. In practice, it wasn’t quite as fun as I was hoping. With evenly match opponents, advantage would go back and forth too easily, leading to a stalemate. Realistic, perhaps, but not terribly fun. It also lead to a feedback loop, where the loser would tend to keep losing when they were disadvantaged. The last problem was displaying the advantage measurement between pairs of combatants, which got really confusing if you had, say, a 3v3 fight where there would be 9 different measurements between all the combatants. The big advantage of hit points is all that information in a simple bar. Although, recalling the system I’m tempted to go back and revisit the system.

Another system, kinda related, would be to have statuses instead of hit points. The goal of a fight would be to inflict statuses on people which would need to be countered. So, for example, you might try to inflict a strain on an opponent (perhaps on a specific body part) which would give a negative to specific actions. The defender would have to counter it properly; for example, perhaps massive attacks cause strains when they connect with shields, so the right answer is to dodge a massive attack rather than block it.

This would be interesting as statuses could limit the player’s options. A strain might mean you can’t do heavy attacks, meaning that you couldn’t strain your opponent back very easily. The disadvantage for a status system, again, is that this becomes a lot harder to display the current state of combat in a simple way. And, if statuses have negatives, you have the feedback problem where the first person to score a status might have a mighty advantage over an opponent.

So, what do you think? Could you come up with a system that replaces hit points? Preferably one that isn’t too hard to display and doesn’t give an advantage to someone already winning.

Special note: I’ve disabled the plugin that let you preview your comment. Something seems to be playing poorly with comments, and I’m trying to narrow it down. It seems longer comments are more likely to get rejected, so I guess you should keep your comments short. If your comment does get discarded, email me the text and I’ll add it manually. Thanks for your patience!

« Previous Post:


  1. I have a question: what exactly is the purpose or goal behind doing away with hit points? Even if we bring down the number of hit points to “1″ each, and implement a whole system of interlocked defences and offences before scoring that one victorious hit, is that still not having hit points ultimately?

    The phrase “Big bag of hitpoints” seems to indicate we want to reduce the time-to-kill or the boringness inherent in just repeating a pattern of skills that will subtract a random number from the hit point reservoir over time until it goes down to zero. That might be done with combat systems adding a layer of action or strategy on top of a hitpoint system (and reducing the number of hitpoints per mob from in the absurd millions to something substantially lower, but harder to score hits on.)

    Comment by Jeromai — 11 May, 2016 @ 10:39 PM

  2. Afterthought: A spin-off from this might be giving each bodypart or wear location its own set of hitpoints, per se. Successfully reducing that part to zero creates some kind of status effect, in essence forming a kind of cc that does not necessarily mean every npc or pc -must- be killed. Perhaps disabling them in some way is sufficient, removing their weapons takes away their ability to fight back, or leg shots affect their mobility, etc.

    Comment by Jeromai — 11 May, 2016 @ 10:45 PM

  3. A possible approach is the insta kill. If you do not succeed to block/parry/avoid you are disabled. This is used in ‘realistic’ shooter ( Arma ) and some fun one (Quake with mod). Tryong to remember the name oc the game but a swordplay simulator also used this approach.
    Two co dition for this to work :
    - making a hit shall notbe easy and need some skill
    - blocking or evading shall be quite easy
    The length of the combat could adjusted with the average probability to hit. Being better than the enemy means that you are able to increase your own probability or doing less mistep.

    This works well with action combat, but it could be more of a challenge to make it work with dice based game.

    Comment by Ettesiun — 11 May, 2016 @ 10:47 PM

  4. Hit points are easy, you’re right about that. But there are other systems also.

    Some have no hit points and instead track wounds, each of which comes with a disadvantage (usually a modifier).

    I kind of like that, except my biggest gripe with hit point based systems is lost here as well, namely that losing the use of an arm during one handed combat, while bad, isn’t as incapacitating as losing the use of a leg.

    I’m more in favour of tracking a degree of woundedness for major body parts. That might be a few numbers to track, but only when a hit produces damage. The advantage is that modifiers are easily derived from this.

    In fact, status effects are also easily derived from this, e.g. if the head gets a wound, you’re stunned, etc.

    Of course that means you have to have some way of soaking up ineffective attacks, if wounds only track actual hits, and stamina makes sense here.

    Trouble is, losing stamina should also by rights incur penalties, and then you have to track penalties from stamina and all body part wounds. That can get messy.

    (Bleeding, incidentally, would be based on stamina.)

    That makes stamina kind of a bag of HP again, except you can regenerate some of it very fast.

    While I think I the above should work well in terms of mapping real life into game mechanics, it’s too complex a system for me to love :(

    The much simpler alternative would be to have a single wound track and ignore different body parts altogether, and an intermediary would be to do the same, but track modifiers for body parts so you get the distinction of different wounds again with fewer numbers to deal with.

    With regards to advantages, a bunch of thoughts.

    First, I like how Rolemaster deals with some of this. You have a single modifier for your “combat effectiveness”, which you can split up how you like for defensive and offensive purposes. That effectively lets you deal with single or multiple opponents, go for an all out attack, etc. Therefore penalties to this modifier are a disadvantage to you, and therefore an advantage to all of your opponents (in their sum).

    So combat should be about trying to put the opponent at a disadvantage rather than about gaining an advantage. That’s easily, if verbosely tracked with the above system.

    Any extra penalties from e.g. unbalancing an opponent should IMHO be achieved by successful attacks specific for that purpose. That’s a lot closer to how grappling systems tend to work. If the attack succeeds, the opponent loses only a little stamina from the failed defence, perhaps, but gains a penalty for a status effect.

    That means you can bash at a person wildly to wear them down, or you can wear them down until a strike to a soft spot becomes worthwhile, or you can unbalance or grapple them to put them at enough of a disadvantage that you have them under your control.

    I really do like that. I just don’t like tracking wounds on all relevant body parts, however “realistic” it may be, and lean more towards the intermediary solution above for my preference.

    Comment by unwesen — 11 May, 2016 @ 10:48 PM

  5. The Harn roleplaying system didn’t use hit points, but it did use endurance. It’s published, so I guess there’s no harm in describing it.

    Skill/attack rolls used a percentile system, like in RuneQuest. Except that if your roll was divisible by 5, you got a “critical” result, either success or failure, depending on whether you rolled under or over your skill (under succeeds).

    Endurance subtracted from your skill numbers (High skill number means you’re better, just to be clear).

    On any opposed action, like an attack, both parties rolled the skill they were using, such as “sword skill” to attack and “shield skill” or “dodge” to defend. The rows of the matrix were “critical success”, “marginal success”, “marginal failure” and “critical failure”. Likewise the defender’s columns for the defense skill roll. I don’t recall the exact contents of the cells of the matrix, but I do recall that if you got a critical success on attack versus a marginal failure, you got very powerful results, such as a “wound” that gave you further penalties, or denied the use of a limb. There was probably another chart referenced, which might have also had hit locations. You could lose endurance to the point where you couldn’t really do anything (though there were “rest” actions”), but it would never kill you or drop you. The bad results in the matrix are what would kill you.

    Comment by Toldain — 11 May, 2016 @ 11:35 PM

  6. I’m with Jeromai. I think this is semantics (although it has to be said that I’m a semiologist at heart so I believe everything is semantics). If you are positing a situation where two (or more) opponents in a video game are required to simulate combat and reach a resolution, no matter what criteria you set or mechanics you use you are ultimately always going to end up at the exact same point: winner/loser. How you measure that is a naming convention not a substantive change to outcome.

    Of course, there’s the purely subjective concept of “fun” to be considered. Fun is a real buzz word in MMO development these days, with huge projects being cancelled left and right because they fail to meet some unrevealed definition or metric of “funness”. Whatever alternate set of signs you implement to indicate progress towards the identical ultimate victory condition, the only concern in a video game should be whether it persuades the player to keep doing it or to stop and go find something else to do instead. Removing hit points from the equation seems to be a solution looking for a problem to me.

    Personally I’d like a system that hid ALL of the information. I really, really enjoy the way in Black Desert that the HP bar of creatures doesn’t change at all until suddenly they drop dead. That’s perfect for my tastes. Unfortunately, your character does eventually “learn” by experience and opens up the knowledge of actual HP damage per type of character. I would have kept that invisible permanently and left the player to learn the names and appearances (actually I wouldn’t display mob names either so it would just be appearances) which to fight and which to leave alone.

    Comment by bhagpuss — 12 May, 2016 @ 1:39 AM

  7. I have argued this time and time again. The tech is good enough now that if you slow down combat a touch you can have actual blocks and parries. If a sword goes through you, it does. Combat in games would be better if it is more like it is in movies. dodges, parries, etc. etc. When your dagger goes through the orcs head, he dies. It can even be done with multiple enemies just fine (think combat in Batman: Arkham but with swords and shields).

    The build up and anticipation of the misses and blocks would be better than “swing my sword through is torso 50x to win”. When you finally do hit and kill it would be far more satisfying. In my opinion of course (qualifier needed. Some like the A-D-D gameplay elements)

    Comment by Isey — 12 May, 2016 @ 6:26 AM

  8. Something that always bothers me is how prevalent the idea of suicidal enemy forces is within games. I don’t know how many people out of a unit need to be shot before that unit starts devoting some serious thought to retreat, but I imagine it should be less than 100%.

    For starters, I would take something similar to what Red Orchestra 2 has in regards to its damage. As I recall, you can take minor or major wounds to body parts, which can be lethal or not depending and some can be patched up with the player only moderately inconvenienced. I would expand this to including incapacitating wounds that don’t kill, such as losing the use of a leg and being able to drag yourself around but not stand up anymore, and then introduce the ability for players to assist others in movement. Then, incentivize survival over killing, while still rewarding murder because after all this IS a game we’re designing. Including a sort of joint squad- or platoon-level “moral meter” that directly impacted player capabilities would round this system out.

    It’s not hit points because individuals taking damage do so on a more analogue scale, and morale would go up and down depending on what happens–getting hits or routing enemies would raise morale, someone getting killed would lower permanently vs morale loss from an injury lessening as it gets patched up, etc.

    Comment by Drew — 12 May, 2016 @ 11:51 AM

  9. At a high level, I’ve been thinking this comes down to continuous versus discrete — or to put it another way, economics versus achievements.

    What I mean is that adding or subtracting points from a pool is basically an economic operation. The number goes up and down in small enough units to feel like a continuous process.

    The most distinctive alternative to this is representing combat info as discrete objects. In this form, change happens by modifying unique individual pieces of a whole thing — left pauldron destroyed; right leg disabled; etc. Rather than progressively whittling down a Thing through multiple economic operations, each Thing is eliminated as a clear achievement.

    There can be overlap between these modes of representing combat status and effects. But I suspect that what feel like the most elegant systems minimize this overlap.

    Comment by Bart Stewart — 12 May, 2016 @ 11:59 AM

  10. Some great insight here. Thanks for the comments, all! :)

    Jeromai wrote:
    …what exactly is the purpose or goal behind doing away with hit points?

    A fun little exercise to think about. It’s good to take a deeper look at why hit points exist, what role they play, and if we can design a better system. As I said, hit points do have a lot of advantages in that they convey information quickly and are fairly well understood by most people.

    I’m not calling for them to be abolished. I used the term “big bag of hit points” as that’s how some people refer to monsters in games.

    …giving each bodypart or wear location its own set of hitpoints…

    I was working on a game that was trying to do that. The big problem is representing the information easily. Assuming four limbs, a head, and a torso, that’s 6 bars to measure instead of one. You also need some good design to make the successful strategy something besides “attack weakest body part that will cause death as hard and as fast as possible.”

    Ettesiun wrote:
    Trying to remember the name of the game but a swordplay simulator also used this approach.

    I think you’re talking about Bushido Blade. The problem this had is that it was easy for a master to slaughter a novice by doing in for an aggressive hit. The advantage of hit points is that while the novice might still lose, they at least feel like they have some chance rather than being killed immediately.

    Isey wrote:
    The tech is good enough now that if you slow down combat a touch you can have actual blocks and parries.

    The problem is that people perceive slow combat as “boring”, even if it’s not. I like slower combat with more actions and reactions, but I’ve seen a lot of other people complain about such pacing. But, in MMOs you still have to worry about latency effects and out-of-sync issues, though.

    Comment by Psychochild — 13 May, 2016 @ 7:20 PM

  11. There was some discussion over on Google+, copying some of the comments from there.

    Ian Borchardt (Reverance Pavane) wrote:

    (a) I quite like my Chainmail inspired combat system for D&D. It improves the growth potential for fighters considerably even though it does mean that it is particularly fraught (especially if you are low level and a class not trained for combat). [A high level wizard is still pretty safe (at least initially), but it is a lot easier for them to get mobbed in melee and killed.]

    (b) The French editions of Nephilim did an interesting trick with the POW vs POW table for damage resolution which provided a status arther than hit point loss. However this was reverted to the normal hit point system when Chaosium published the English version of the 3rd edition.

    (c) Statuses work very well but generally only if the combat system is detailed enough to make use of them. ironclaw does excellent work with it’s use of damage statuses. For example being Injured means any subsequent attack is more likely to do serious harm, and it has Stunned and Reeling as well, which inhibit tactical responses.

    (d) One alternative might be to keep hit points constant as a threshold but increase the “damage” (actually it’s the potential to be seriously wounded) of subsequent attacks. Of course that’s an extra step in the resolution procedure to ask what the damage bonus will be (in terms of steps up the damage dice chain). So a kobold that might do d4 to you now does a d20 damage. Do you want to take the risk that he rolls over your hit point threshold. [Remember this is constant - it does not abalate.]


    Also something to beware of is that most humans are pretty bad at evaluating risk, so any system that involves probability will get outrage “when their luck runs out” (even if luck has nothing to do with it).

    The good thing about hit points is that it is easy to work out when stuff becomes more dangerous.

    [I've noticed a tendency in my Chainmail games for players to get too overconfident and end up getting mobbed once they gain a few levels and have a measure of "invulnerability."]


    And one of my favourite systems, although it is totally unplayable, is the one in first edition Albedo (the T&I/Chessex version). characters generally acquired fatigue when they did something. When the accumulated fatigue crossed certain thresholds, you received penalties to actions, and eventually there was a chance of the character falling unconscious of otherwise dropping out.

    When a character was injured, the severity and location of the wound indicated how much fatigue the character immediately gained from being hit, the continual fatigue penalty for having an untreated wound, and an additional fatigue penalty if you tried to use the affected body part.

    It did a good job of mimicking the general results of trauma, especially continuing trauma. The take extra fatigue/damage from using the part is particularly important in this regard.


    Generally though I consider only a couple of injury states in D&D (and most other games).

    (a) Uninjured – minor cuts and bruises and even wounds. This is what hit point damage in my D&D game generally represents – shock, fatigue and stress – it heals extremely fast with rest (but may also be reduced if you are fatigued or hungry or drugged for example – or increased with the right sort of drugs. Basically you are trading extra exertion to avoid being hurt.

    (b) Wounded – this is the classic walking wounded status in the military. You have a definite injury but can deal with it yourself and can usually walk out there (with assistance). Doing anything strenuous is going to hurt and may cause complications.

    (c) Seriously wounded – You may be conscious but you aren’t doing anything without help. You need to be medivaced out of there – you are not going to make it out by yourself. Most imnjuries of this type require a save vs crippling.

    (d) Mortally wounded. Utterly out of it. Save vs death to avoid actually dying if someone can render first aid. [You can even recover from a mortal wound after failing the saving throw.]

    (e) Dead – Instant death. The trauma is too severe to

    Recovery from injury will take a number of weeks equal to the damage of a weapon, for example. Less an effect of armour. So a normal dagger blow will invalid you for 1d4 weeks.

    Critical hits tend to cause actual injuries.

    Comment by Psychochild — 13 May, 2016 @ 7:23 PM

  12. On Google+, Paul Gestwicki wrote:

    Check out Feng Shui 2. It has some of the most elegant action-oriented combat systems I have seen in a tabletop game. There are tiers of enemies: mooks go down with one hit and exist primarily for the joy of eliminating them; featured foes have the equivalent of hit points and are treated similar to players; bosses have a lot of hit points and don’t always go down when reduced to zero. I am especially fond of the mooks system since they really capture the feel of a Hong Kong action flick.

    Comment by Psychochild — 13 May, 2016 @ 7:25 PM

  13. I address bhagpuss’s and Jeromai’s issues about if this is just semantics in a follow-up post on Why it’s not all semantics because game design matters.

    Comment by Psychochild — 14 May, 2016 @ 6:16 PM

  14. I’m curious how you do your mock ups? Do you make a working model program, or create a pen and paper version to play around with?

    I love brain storms when it comes to mechanics. To replace something as heavily used as hp, it’ll require a system that’s simple but elegant enough to work well with other established systems, less you have to redesign them (like you mention about healing in your semantics post)

    This idea of wounds sounds like it would make combat feel more real, vs seeing numbers float above an enemy till you equal it to a magic amount and they are pronounced dead. Going down this path, I’d say would need a stamina meter, something akin to Dark Souls. It’ll get easily used up, but just as easily comes back. As you make mistakes, wounds will affect this meter, shrinking it to where you do less, or making it regain slower, making you wait longer to get off that big combo. A healer can undo some of that damage during combat, but may leave them open to a strong combo from the enemy, so the other player will need to have the mobs full attention first. Preferably while it’s in the middle of a combo on said “tank” player.

    In this deviated version of your system, I would say you only have 1 health point. When you are open to a death blow and are hit, then you are defeated. To deplete an opponents stamina, you attack at them. Attacks that land take out a chunk of their stamina, attacks that are blocked take out tinier slices of stamina. When you attack an opponent and get their stamina to zero and hit them, that’ll open them to a death blow. If you have enough stamina, or can regain enough before the player comes out of a death blow, you can take a swing and defeat them.

    Comment by Tyrannodorkus — 18 May, 2016 @ 10:48 AM

  15. It seems to me that if you want to get away from hp, you have to introduce the very real possibility of instant kills. This is true whether hp are supposed to represent how close to death you are, or morale (in LOTRO), or unconsciousness, or simply being at the mercy of your opponent(s). Basically it represents the idea that no matter what you do, you will die/fail/be out of action until revived by an external power (rez, respawn, etc). If you aim to avoid the “big bag of hp” problem, and make different attacks do different things like a lot of the suggestions above with status effects and so on, then I think you need to make it possible for very skilled – or very lucky – strikes to immediately take an opponent out of the fight. From what we’ve seen in player communities in the past, insta-kill is entirely unwelcome when it can happen to you. That is effectively the main problem anti-gankers have with open PvP systems, it feels like they have no chance and are totally at the mercy of the gankers. Or, maybe I am wrong and the objection to insta-kill is the PvP element, maybe players would be more open to it in a PvE environment.

    One idea that has come to me while writing this, is taking a leaf from the old Wing Commander and X-Wing books. You had 4 zones – front, back, left, right – that were covered by shields, and you could adjust the strength of each zone independently on the fly. Attacks would deplete shield energy, and then would start inflicting ‘permanent’ damage on the ship. So, weapons would be knocked out, engines could be crippled, and so on. In a character-based MMO, that could represent your active defence, being the combination of blocking, parrying, and evasion. As you take attacks, your defences begin to show strain, and you can shore them up by exposing yourself to attacks from other directions. Once you start taking direct hits, you are sustaining permanent damage that has to be treated with magic or medical attention, and you receive penalties until you do. Movement penalties, defence penalties, some abilities might be locked out, e.g. you might not be able to use a bow any more.

    I guess my main brick wall when it comes to this ‘alternative to hp’ problem, is including magic systems. Healing and support are doable, but how do you reconcile being enveloped in a poison cloud, or set on fire, or doused in acid, or struck by lightning, in a system that is designed around physical wounds? I mean, this is also a potential problem from natural causes too (being caught in a burning building, for example) and it feels like there needs to be a separate system to deal with that? I don’t know.

    Comment by Dahakha — 23 May, 2016 @ 2:34 AM

Leave a comment

I value your comment and think the discussions are the best part of this blog. However, there's this scourge called comment spam, so I choose to moderate comments rather than giving filthy spammers any advantage.

If this is your first comment, it will be held for moderation and therefore will not show up immediately. I will approve your comment when I can, usually within a day. Comments should eventually be approved if not spam. If your comment doesn't show up and it wasn't spam, send me an email as the spam catchers might have caught it by accident.

Line and paragraph breaks automatic, HTML allowed: <a href="" title=""> <abbr title=""> <acronym title=""> <b> <blockquote cite=""> <cite> <code> <del datetime=""> <em> <i> <q cite=""> <strike> <strong>

Email Subscription

Get posts by email:

Recent Comments


Search the Blog


December 2019
« Aug    



Standard Disclaimer

I speak only for myself, not for any company.

My Book


Around the Internet

Game and Online Developers

Game News Sites

Game Ranters and Discussion

Help for Businesses

Other Fun Stuff

Quiet (aka Dead) Sites

Posts Copyright Brian Green, aka Psychochild. Comments belong to their authors.

Support me and my work on