Personally, I have no intention of ever utilizing the following rules, but it goes into the Game Design theory I was talking about earlier:
Daily powers, and encounter powers, are used partially on a Use Them Or Lose Them basis. This can be contrasted to consumable items and powers, which are not Use Them Or Lose Them. My own (FULLY ANECDOTAL) experience has been that players tend to be a lot more willing to use the former than they are the latter-though, in a total pinch, because they have to use consumable items to win, they will, as no one wants a TPK. The exception to this is when consumable items are so cheap-as is the case with low level reagents at mid to high paragon and beyond (prior to the rarity rules)-that the real expected cost of the consumable is so low that it might as well be something akin to an At-Will.
As I said, I don't have a problem with this mechanic-it encourages players to use every part of the character build, so to speak, but if you happen to like the idea of there always been a trade-off between using a power or not, at the end of an Extended Rest, consider rewarding characters who have unspent daily resources. The reward should be proportional to the number of encounters faced over the course of the day, the power (probably simply related to level) of the power, and any other factors concerned. This has the result of making burning off daily powers a little less desirable because of the associated opportunity cost. It does, however, put martial Essentials classes (barring the Assassin) in a bit of a pickle, especially considering how controversial they are to begin with.
That being said, it's not that it's complicated-a daily power that goes unspent is a one-time resource that wasn't used and as thus is wasted. Consider giving a "bonus" consumable power of comparable level, with a rarity based on the difficulty of that adventuring day. Obviously, common consumables are little more than worth 1/5th their level as bonus treasure, but uncommon and rare consumables are a major benefit.
But be prepared for the last fight of any given adventure to be a lot more difficult-your characters may start hoarding consumables, but whether or not they actually use them depends on their response to Use It Or Lose It free resources.
Showing posts with label Game Theory. Show all posts
Showing posts with label Game Theory. Show all posts
Friday, April 8, 2011
When is an Encounter Power not an Encounter Power?
In my upcoming Dungeons and Dragons game, Sands, Serpents, and Scales, I'm using a number of variant rules regarding daily power usage and healing surges. The reason being that with the style of narrative I'm using, it's more difficult to create Extended Rest time periods that conform to the dramatic narrative, which tends to run to the unhappy situation where every even remotely difficult encounter can be rendered irrelevant by dumping a whole bunch of daily attack powers. Whatever-that's not what's at issue here.
What's at issue here is that, reading the Vampire Preview, I find that the mechanics of the Vampire Class are actually remarkably similar to the mechanics that I'm utilizing-very few "daily" healing surges, but the ability to regain (or rather gain) healing surges upon successful attacks. In the utilized system (which can be found at http://sandserpentsscales.wikidot.com/chapter-based-gameplay ; I have the wiki set to open due to the complications of having my players sign up, so please don't abuse the wiki), characters get a certain number of healing surges per encounter. So, what does this have to do with the topic?
Well, in addition to making healing basically impossible when you have 10 encounters per day (or however many), many of whom are more than EL (yes-you level up more than once per "day"), characters are able to use daily powers less often than normal. To allow characters to "Nova" a little bit more often, you'll notice the Finishing Move rules. Once per encounter, a character can use a daily attack power at the cost of healing surges.
Surges per encounter are use them or lose them. That means that, if you're not damaged at the end of an encounter, or if you're damaged only such that one surge (along with the bard's Song of Rest capability-good choice, By the Way) and you have a pool of two surges left over, that's a wasted resource-a resource that, had you used them differently, perhaps you could have prevented someone else from having spent resources.
In a standard campaign, barring the knowledge that certain enemies are more powerful while bloodied (and as thus should be engineered to spend as little time being bloodied as possible), it's natural to unleash one's encounter powers up front, to be sure that you're doing "as much as possible." After all, if you hang on to your strongest attacks, there's a heavy chance you'll have basically wasted them. Daily powers are the same way, except that the calculations are based on whether or not the power will be available this encounter, or available next encounter-so the "Use it or lose it" sense doesn't trigger until the last encounter of an adventure.
Of course, a Killer DM (http://tvtropes.org/pmwiki/pmwiki.php/Main/KillerGameMaster) may be a master at setting up "False" final battles for an adventure, perhaps throwing a Purple Worm encounter on the way back home. This distorts how much information a player has for making decisions about when to use powers or not, but that's more an issue of Transparency As It Relates To Game Theory (of which I'm sure there are many books on the subject that I am only semi-qualified to even comment on).
But the interesting point is that we're looking at a binary state of Use or Don't Use as it pertains to being Wasted or Not Wasted. When you have something like the new Vampire Class that can utilize healing surges to power other abilities, suddenly, spending a healing surge isn't just an issue of whether you heal or not-it's the opportunity cost of those options you no longer have. Likewise, with the aforementioned Finishing Move rules, the information about how much those surges costs you changes over the course of a battle.
That's the beauty about having multiple Use it or Lose It resources that can interact together in different ways. Instead of dropping Fireball (a 2 surge Finisher) in the first round because, if you don't, you lose the ability to use a finishing move for that encounter, you wait to see just how valuable two surges are going to be to you over the course of the encounter. So if, as the fight is winding down to a close, you notice "Holy crap, I'm at full health and have two encounter surges left and haven't used my finishing move," you know that the opportunity cost of using Fireball goes from "Healing myself to full or not" to "Jack Squat."
They're called Finishing Moves for a reason.
What's at issue here is that, reading the Vampire Preview, I find that the mechanics of the Vampire Class are actually remarkably similar to the mechanics that I'm utilizing-very few "daily" healing surges, but the ability to regain (or rather gain) healing surges upon successful attacks. In the utilized system (which can be found at http://sandserpentsscales.wikidot.com/chapter-based-gameplay ; I have the wiki set to open due to the complications of having my players sign up, so please don't abuse the wiki), characters get a certain number of healing surges per encounter. So, what does this have to do with the topic?
Well, in addition to making healing basically impossible when you have 10 encounters per day (or however many), many of whom are more than EL (yes-you level up more than once per "day"), characters are able to use daily powers less often than normal. To allow characters to "Nova" a little bit more often, you'll notice the Finishing Move rules. Once per encounter, a character can use a daily attack power at the cost of healing surges.
Surges per encounter are use them or lose them. That means that, if you're not damaged at the end of an encounter, or if you're damaged only such that one surge (along with the bard's Song of Rest capability-good choice, By the Way) and you have a pool of two surges left over, that's a wasted resource-a resource that, had you used them differently, perhaps you could have prevented someone else from having spent resources.
In a standard campaign, barring the knowledge that certain enemies are more powerful while bloodied (and as thus should be engineered to spend as little time being bloodied as possible), it's natural to unleash one's encounter powers up front, to be sure that you're doing "as much as possible." After all, if you hang on to your strongest attacks, there's a heavy chance you'll have basically wasted them. Daily powers are the same way, except that the calculations are based on whether or not the power will be available this encounter, or available next encounter-so the "Use it or lose it" sense doesn't trigger until the last encounter of an adventure.
Of course, a Killer DM (http://tvtropes.org/pmwiki/pmwiki.php/Main/KillerGameMaster) may be a master at setting up "False" final battles for an adventure, perhaps throwing a Purple Worm encounter on the way back home. This distorts how much information a player has for making decisions about when to use powers or not, but that's more an issue of Transparency As It Relates To Game Theory (of which I'm sure there are many books on the subject that I am only semi-qualified to even comment on).
But the interesting point is that we're looking at a binary state of Use or Don't Use as it pertains to being Wasted or Not Wasted. When you have something like the new Vampire Class that can utilize healing surges to power other abilities, suddenly, spending a healing surge isn't just an issue of whether you heal or not-it's the opportunity cost of those options you no longer have. Likewise, with the aforementioned Finishing Move rules, the information about how much those surges costs you changes over the course of a battle.
That's the beauty about having multiple Use it or Lose It resources that can interact together in different ways. Instead of dropping Fireball (a 2 surge Finisher) in the first round because, if you don't, you lose the ability to use a finishing move for that encounter, you wait to see just how valuable two surges are going to be to you over the course of the encounter. So if, as the fight is winding down to a close, you notice "Holy crap, I'm at full health and have two encounter surges left and haven't used my finishing move," you know that the opportunity cost of using Fireball goes from "Healing myself to full or not" to "Jack Squat."
They're called Finishing Moves for a reason.
Sunday, April 3, 2011
Solos: Survival of the Solitary
I've been deliberately putting this topic off to last because I'm not sure what to say about it.
It's not that solo monsters being more subject to the laws of Critical Existence Failure (http://tvtropes.org/pmwiki/pmwiki.php/Main/CriticalExistenceFailure) is somehow wrong-it's just that it's different, and that difference means that some of the common conceptions of how combat works are out of touch.
I have to confess, much of my theory on the topic of solo monsters is based on an arbitrary number someone told me that encounters last for five rounds. And that's about it. It's a completely arbitrary statistic that I don't have the numbers to back up, and yet it's going to found the basis of this essay.
Now, the reality is that no two parties are identical (unless, well, they're identical) and how any given group deals with combat encounters varies. A group with three wizards all tossing Icy Rays everywhere may have monsters drop three at a time but dealing little overall damage. But a group with two strikers is likely to take out one monster per round for the first few rounds while they burn through encounter powers. So, even though there's five monsters in a standard encounter, the longevity and effectiveness of those five monsters isn't a constant.
If we just assume that a solo monster has five standard monster level attacks, that's great-ignoring other action economy problems-but we run into the nasty problem that it's not indicative of how a standard encounter functions*; it's like if everyone was attacking a different creature and not focusing fire. This means that, in a five round combat, you'd expect to see the equivalence of twenty five creature rounds.
Compare this to an ordinary encounter, where there's five creatures in the first round, four in the second, three in the third, two in the forth, and then one straggler in the fifth. That's only fifteen creature rounds by comparison, so your solo monster is 66% more effective over the lifetime of an encounter!
On the other hand, perhaps not. Controllers can't use multi-target damage, meaning that the net damage per round being dealt by the party is reduced. This can partially be balanced by the fact that there is no chance for lost damage in the form of overkill. On the other hand, if you're playing with an Essentials Assassin, there's also no chance for the Death Attack feature to trigger! And finally, despite being worth five creatures, ever since Monster Manual 2, solo monsters are only considered to have 4x HP in order to avoid having fights drag on too long!
The fact of the matter is that, I don't have the hard data to come to a decisive conclusion here. If we assume striker overkill balances out lost controller damage, then the fact that a solo monster only has 80% HP of a standard encounter leads us to the conclusion that we can multiple those 25 creature rounds by 80%. That's still 20 creature rounds, or 33% effectiveness over a standard encounter. So, what's the solution?
If we assume that because it only has 80% HP the solo monster only requires four rounds to defeat, we can expect that the first round it will be unbloodied, and sometime around the middle or end of the second round it will be bloodied. Working from this, we can use a 3 Creature Round x2, plus 4 creature round x2 model; in short, if the creature's effectiveness increases by about 33% while bloodied, but its effectiveness pre-bloodied is only three creatures rather than four, we're somewhere around the proper level we're aiming for. But again, this is very ad hoc, and making a lot of unsupported assumptions. For instance, most multi-target attack type monsters can't effectively attack the entire party, but a creature with sufficiently large close burst powers can.
Also, again, I think a lot of this has to go back to the Game Theory of Solo Monster design. If the players know that solo monsters are more powerful bloodied than not bloodied, they will want a greater percentage of the fight to occur with the monster not being bloodied. Upon first instinct, you may think that doesn't make sense-after all, in addition to our anticipated values above being "Bloodied by the end of the second round," half of the creature's HP is half of it's HP, right?
Not really-the fact of the matter is that party damage is wildly different between at-will output, encounter nova output, and daily nova output. If players know that the creature is more deadly while bloodied, then players can reserve powers for after the creature is bloodied. Of course, if the solo monster is aware that because of this PCs are taking it easy, the solo monster can use that information as well, so it gets complicated. But in reality, it's been my experience that most PCs don't think this way-they like to burn through encounters.
And to a large extent, that's wise. Not every encounter ends with every encounter resource having been spent, and for creatures with regeneration, every round you procrastinate is a round of additional HP you have to deal with.
Other alternatives for dealing with this kind of playstyle are having a creature deal low damage but have extremely high defenses-typically in conjunction with regeneration-for the first half of the battle. Each round is relatively non-damaging to the party, but still a cost in resources; procrastination only works in the monster's favor. Once the monster becomes bloodied, it becomes truly effective, but the party may have to expend its novas in order to get it that far, or risk being bogged down. The major downside of this gameplay style is that it's time consuming, and has a risk of alienating players, regardless of whether it makes for dynamic combat.
It's not that solo monsters being more subject to the laws of Critical Existence Failure (http://tvtropes.org/pmwiki/pmwiki.php/Main/CriticalExistenceFailure) is somehow wrong-it's just that it's different, and that difference means that some of the common conceptions of how combat works are out of touch.
I have to confess, much of my theory on the topic of solo monsters is based on an arbitrary number someone told me that encounters last for five rounds. And that's about it. It's a completely arbitrary statistic that I don't have the numbers to back up, and yet it's going to found the basis of this essay.
Now, the reality is that no two parties are identical (unless, well, they're identical) and how any given group deals with combat encounters varies. A group with three wizards all tossing Icy Rays everywhere may have monsters drop three at a time but dealing little overall damage. But a group with two strikers is likely to take out one monster per round for the first few rounds while they burn through encounter powers. So, even though there's five monsters in a standard encounter, the longevity and effectiveness of those five monsters isn't a constant.
If we just assume that a solo monster has five standard monster level attacks, that's great-ignoring other action economy problems-but we run into the nasty problem that it's not indicative of how a standard encounter functions*; it's like if everyone was attacking a different creature and not focusing fire. This means that, in a five round combat, you'd expect to see the equivalence of twenty five creature rounds.
Compare this to an ordinary encounter, where there's five creatures in the first round, four in the second, three in the third, two in the forth, and then one straggler in the fifth. That's only fifteen creature rounds by comparison, so your solo monster is 66% more effective over the lifetime of an encounter!
On the other hand, perhaps not. Controllers can't use multi-target damage, meaning that the net damage per round being dealt by the party is reduced. This can partially be balanced by the fact that there is no chance for lost damage in the form of overkill. On the other hand, if you're playing with an Essentials Assassin, there's also no chance for the Death Attack feature to trigger! And finally, despite being worth five creatures, ever since Monster Manual 2, solo monsters are only considered to have 4x HP in order to avoid having fights drag on too long!
The fact of the matter is that, I don't have the hard data to come to a decisive conclusion here. If we assume striker overkill balances out lost controller damage, then the fact that a solo monster only has 80% HP of a standard encounter leads us to the conclusion that we can multiple those 25 creature rounds by 80%. That's still 20 creature rounds, or 33% effectiveness over a standard encounter. So, what's the solution?
If we assume that because it only has 80% HP the solo monster only requires four rounds to defeat, we can expect that the first round it will be unbloodied, and sometime around the middle or end of the second round it will be bloodied. Working from this, we can use a 3 Creature Round x2, plus 4 creature round x2 model; in short, if the creature's effectiveness increases by about 33% while bloodied, but its effectiveness pre-bloodied is only three creatures rather than four, we're somewhere around the proper level we're aiming for. But again, this is very ad hoc, and making a lot of unsupported assumptions. For instance, most multi-target attack type monsters can't effectively attack the entire party, but a creature with sufficiently large close burst powers can.
Also, again, I think a lot of this has to go back to the Game Theory of Solo Monster design. If the players know that solo monsters are more powerful bloodied than not bloodied, they will want a greater percentage of the fight to occur with the monster not being bloodied. Upon first instinct, you may think that doesn't make sense-after all, in addition to our anticipated values above being "Bloodied by the end of the second round," half of the creature's HP is half of it's HP, right?
Not really-the fact of the matter is that party damage is wildly different between at-will output, encounter nova output, and daily nova output. If players know that the creature is more deadly while bloodied, then players can reserve powers for after the creature is bloodied. Of course, if the solo monster is aware that because of this PCs are taking it easy, the solo monster can use that information as well, so it gets complicated. But in reality, it's been my experience that most PCs don't think this way-they like to burn through encounters.
And to a large extent, that's wise. Not every encounter ends with every encounter resource having been spent, and for creatures with regeneration, every round you procrastinate is a round of additional HP you have to deal with.
Other alternatives for dealing with this kind of playstyle are having a creature deal low damage but have extremely high defenses-typically in conjunction with regeneration-for the first half of the battle. Each round is relatively non-damaging to the party, but still a cost in resources; procrastination only works in the monster's favor. Once the monster becomes bloodied, it becomes truly effective, but the party may have to expend its novas in order to get it that far, or risk being bogged down. The major downside of this gameplay style is that it's time consuming, and has a risk of alienating players, regardless of whether it makes for dynamic combat.
Friday, April 1, 2011
Solos (Followup): Game Theory of Cleanser Rules
As an economist by training (if not by trade) I have a strange fascination with Game Theory. In my previous entry, I mentioned that one potential "solution" (if it can be called that) for dealing with status effects on solo monsters is the usage of what I called "Cleanser" rules, which effectively render one creature as being a different creature as it pertains to things like status effects.
The concept was introduced not for a specific gameplay reaction per say, but rather to better simulate a situation where a solo monster functioned like five individual standard enemies-that is, what applies to one enemy doesn't apply to all five. However, looking at Cleanser rules using Game Theory actually leads to some interesting thoughts.
First, let's assume that the players (and by extension PCs) know that the monster is able to use Cleanser at least once. Thus, players are aware of the fact that after using their powers, there is some chance that they will simply be negated. In this sense, the player and monster get into a relationship where, if the party lands too many status effects at any given time, the monster will remove them all. But if the party delays applying multiple status effects, the monster-in wanting to negate as many status effects as possible-will "Let it pass." However, in ensuring this state where the party uses only one status effect at a time, the party is themselves giving up a valuable resource-the ability to land status effects earlier in a fight.
The resulting dynamic is that players are more hesitant to drop multiple status effects-be they end of encounter, save ends, or even end of next turn-all at once as it applies to limited resources. However, how hesitant players are depends on the monster's own tactics as well, as if players realize that one power won't be suppressed, perhaps two won't be, or three won't be, and so on. Ultimately, we are left with a Price is Right-esque mechanism, whereby players will test the waters until they "bust" and end up negating their hard work-
-or will they? If the party as a whole drops one status effect at a time, it is unlikely that the monster will utilize its cleanser. However, as the fight nears its conclusion-and the monster recognizes what a precarious position it is in-the monster will be less hesitant to expend its cleanser. As thus, players recognize the difference in behavior of monster cleanser usage, and adjust their tactics accordingly.
Of course, this assumes a simple finite usage of cleanser. If the monster treats Cleanser as a rechargable power, then it has an extremely high incentive to use its cleanser when it recharges, as in addition to not removing status effects for the round in question, the creature is effectively "missing" potential cleansers. How missing these cleansers actually affects what status effects it will suffer over the course of the battle, of course, goes back to how the party and the monster play off each other-that is, whether the party "Forces" a usage of cleanser or not. After all, even if it's a rechargable, what monster would risk being stunned for a round in order to negate a -1 penalty to attack? But in making that decision, it is thus accepting that -1 penalty to attack.
Of course, as with any game theory, this only really applies if both sides know what the other side has to gain. A solo monster doesn't know that the enemy party has either one or five stuns ready. And the players don't necessarily know the mechanism by which the monster can use cleanser. The DM, on the other hand, knows both, leaving the DM in a monstrously (pardon the pun) superior position. To make the game "fair" the DM can overcompensate-by letting the monster make deliberately wrong decisions. Or perhaps the DM uses this advanced knowledge to the monster's advantage.
A more cunning DM might even make usage of skill checks as bonus actions in order to determine "what sort of information" a given creature is going to have access to. Can the party bluff the monster into thinking they're holding onto their strongest abilities until after it's dropped its guard? Perhaps an arcana check would reveal to the party the circumstances of the monster's cleanser. If both parties are on equal footing, then game theory can commence-
-however, if someone is misinformed about what the other has to gain, you can generally bet that they'll be at a disadvantage every time.
The concept was introduced not for a specific gameplay reaction per say, but rather to better simulate a situation where a solo monster functioned like five individual standard enemies-that is, what applies to one enemy doesn't apply to all five. However, looking at Cleanser rules using Game Theory actually leads to some interesting thoughts.
First, let's assume that the players (and by extension PCs) know that the monster is able to use Cleanser at least once. Thus, players are aware of the fact that after using their powers, there is some chance that they will simply be negated. In this sense, the player and monster get into a relationship where, if the party lands too many status effects at any given time, the monster will remove them all. But if the party delays applying multiple status effects, the monster-in wanting to negate as many status effects as possible-will "Let it pass." However, in ensuring this state where the party uses only one status effect at a time, the party is themselves giving up a valuable resource-the ability to land status effects earlier in a fight.
The resulting dynamic is that players are more hesitant to drop multiple status effects-be they end of encounter, save ends, or even end of next turn-all at once as it applies to limited resources. However, how hesitant players are depends on the monster's own tactics as well, as if players realize that one power won't be suppressed, perhaps two won't be, or three won't be, and so on. Ultimately, we are left with a Price is Right-esque mechanism, whereby players will test the waters until they "bust" and end up negating their hard work-
-or will they? If the party as a whole drops one status effect at a time, it is unlikely that the monster will utilize its cleanser. However, as the fight nears its conclusion-and the monster recognizes what a precarious position it is in-the monster will be less hesitant to expend its cleanser. As thus, players recognize the difference in behavior of monster cleanser usage, and adjust their tactics accordingly.
Of course, this assumes a simple finite usage of cleanser. If the monster treats Cleanser as a rechargable power, then it has an extremely high incentive to use its cleanser when it recharges, as in addition to not removing status effects for the round in question, the creature is effectively "missing" potential cleansers. How missing these cleansers actually affects what status effects it will suffer over the course of the battle, of course, goes back to how the party and the monster play off each other-that is, whether the party "Forces" a usage of cleanser or not. After all, even if it's a rechargable, what monster would risk being stunned for a round in order to negate a -1 penalty to attack? But in making that decision, it is thus accepting that -1 penalty to attack.
Of course, as with any game theory, this only really applies if both sides know what the other side has to gain. A solo monster doesn't know that the enemy party has either one or five stuns ready. And the players don't necessarily know the mechanism by which the monster can use cleanser. The DM, on the other hand, knows both, leaving the DM in a monstrously (pardon the pun) superior position. To make the game "fair" the DM can overcompensate-by letting the monster make deliberately wrong decisions. Or perhaps the DM uses this advanced knowledge to the monster's advantage.
A more cunning DM might even make usage of skill checks as bonus actions in order to determine "what sort of information" a given creature is going to have access to. Can the party bluff the monster into thinking they're holding onto their strongest abilities until after it's dropped its guard? Perhaps an arcana check would reveal to the party the circumstances of the monster's cleanser. If both parties are on equal footing, then game theory can commence-
-however, if someone is misinformed about what the other has to gain, you can generally bet that they'll be at a disadvantage every time.
Subscribe to:
Posts (Atom)