Stupid Idea #1: Opportunity Rolls

[FreakyCheeseMan]

Working on ideas for a game system I'm building from the ground up - here's one of my stupider more insane less conventional ideas, which I wanted to run by everyone. Here goes.

Proposed Rules

At the start of their turn, players roll a single d20; this number represents a (very) abstract notion of the options available to them at the moment based on the positioning and defences of enemy combatants, fluxes in the local magical field, etc. We'll call this number their "Opportunity Number," until I think of something better.

Every action will likewise have a number associated it - we'll call this an "Opportunity Cost." You can only take an action if the Opportunity Costs divides your current opportunity number. Whenever you take an action, you divide your current Opportunity Number by that cost. (So, if you have an Opportunity Number of 12, and take an action with an Opportunity Cost of 3, then your Opportunity Number for the rest of the round becomes 4.)

Your most basic cut-and-thrust actions - basic attacks, simple offensive spells, standard blocks and parries - would have an Opportunity Cost of 1 - you can always do them, and doing them does not limit your capacity to try other things with the rest of your turns.

Somewhat more complex actions - trip attempts, bashing someone with your shield, dispelling magic, feinting, etc - will have a somewhat higher opportunity cost, in the 2-5 range. More powerful options - throws and disarms in combat, AoE spells, etc - will have a higher cost still, at around 6-10.

Your most powerful options - instant death spells or attacks, single actions that have the potential to alter the course of the entire battle - will generally have a cost greater than 10, so they can only be performed on specific rolls.

Depending on how I feel further down the line, there may be more or less wiggle room - some character options might let people re-roll or add/subtract a single point from their opportunity number, or you may be able to take actions you don't qualify for, at a significant penalty.

Justification - Realism
The options available to a person in combat rely on a lot of (sometimes complex) details of the situation: their placement, where their opponent's attention is focused, the current distribution of weight and momentum, etc. If you look at real fights (martial arts, in this case) they're very reactionary - most people don't have a single technique they're trying to use over and over, but rather, are working with what's available to them.

Justification - Gameplay
I'm hoping that the proposed system will allow for more varied and interesting combat, encourage broader, less focused builds, and allow options to be available that would simply be too powerful, if they were reliably available.

Part of the inspiration for this came from the plight of the poor 3.5 fighter; he basically is relegated to always taking the same action (attack), regardless of the situation. When better options are introduced to him (tripping), it doesn't really improve things - it changes from "You will always be attacking, and it is foolish to invest character resources in anything else" to "You will always be tripping, and it is foolish to invest character resources in anything else."

Additionally, I feel like chance is a great way to create variety in combat - it keeps things away from that stagnant place of "Everyone knows what their best strategies are, so fights always flow in the exact same way." However, having the chance mechanism in 3.5 - when you take an action you roll a die, and that tells you if it worked or not - is rather limited, and somewhat frustrating (when you're routinely taking actions that fail.)

I also feel like this gives me the freedom, as a designer, to introduce cool and powerful options without worrying too much about them breaking the game - spells or manoeuvres that would determine every fight (even if they were only usable once per encounter) might not be so insane, if they only have a 5% of being available in a given round - so, even if I do screw up and give the Wizard a "Win" button, the game can still be fun: he has to answer the challenge of what to do when it isn't available, and the fighter has a chance to contribute/survive alongside him.

[Amechra]

And rolling prime numbers?

[Grod_The_Giant]

It's an interesting idea. I kind of like it. A d20 seems way too swingy for something like this, though. 2d10 or 3d6 would probably work better-- something a with more of a bell curve thing.

[Vadskye]

It's an interesting idea. I kind of like it. A d20 seems way too swingy for something like this, though. 2d10 or 3d6 would probably work better-- something a with more of a bell curve thing.

A d20 is exactly what should be used for this. That's what the division mechanic is for: the lower the number, the more likely it is that a character will be able to take the action. It's fairly intuitive in that sense.

That isn't to say that I am necessarily a fan of the idea, though. People vary their techniques in real life to adapt to different situations. This system makes people vary their actions randomly; you may see 20 goblins charging in a tightly packed circle, but for some reason your character forgot how to throw a fireball for a round. That's irritating. It's even more irritating (and unintuitive) when he "remembers" next round - after the goblins are already engaged in melee with the party and are harder to target exclusively.

A more robust way to solve the problem you present is to make situations in D&D more different, such that tripping is the right option in some (but not all) situations. In other words, tripping should never be more effective at just dealing damage than attacking. However, it should serve some specific purpose, such as stopping people from escaping or making people easier to hit in a group setting, that it can specifically fulfill.

[FreakyCheeseMan]

And rolling prime numbers?

There are eight prime numbers between one and twenty; pretty much every character would have at least one ability associated with the lower value ones, (2, 3 and 5), leaving the remaining five (7, 11, 13, 17 and 19) more questionable - likely, those would cover "Special" attacks, if you had them. If you didn't have them... it'd be like you rolled a 1.

Mind, rolling a "1" doesn't mean you're useless - the options you most rely on would pretty much fall under there.


That isn't to say that I am necessarily a fan of the idea, though. People vary their techniques in real life to adapt to different situations. This system makes people vary their actions randomly; you may see 20 goblins charging in a tightly packed circle, but for some reason your character forgot how to throw a fireball for a round. That's irritating. It's even more irritating (and unintuitive) when he "remembers" next round - after the goblins are already engaged in melee with the party and are harder to target exclusively.

This is definitely a concern; if I go with this, I'll have to be careful, especially with spells. There'd also be something where, if you could take a minute to focus, you could prep up anything - maybe, during a surprise round, you could pick your Opportunity Number.

One thing - hopefully, ideally, there will be enough different options that (usually) you'll have several associated with any number, or at least with the lower primes. Also, there will be a lot of options with an opportunity cost of 1 - so, you'll still be making choices, you'll just have a different landscape of choices available to you.


A more robust way to solve the problem you present is to make situations in D&D more different, such that tripping is the right option in some (but not all) situations. In other words, tripping should never be more effective at just dealing damage than attacking. However, it should serve some specific purpose, such as stopping people from escaping or making people easier to hit in a group setting, that it can specifically fulfill.

That would be ideal, but much more difficult to design for - and, it runs into the min/max issue where players feel like investing resources in "Secondary" options just sabotages their ability to fill out their primary role. Really, though, a lot of this is coming from a recognition on my part that I'm a newbie designer - trying to make a system work naturally and fluidly like that is easier said than done.

Oh, and here's an example of one of the specific ways I'd like to use this:
The system I'm making uses "Action Points" that renew at the end of every turn, and describe both what a character can do during their turn, and some off-turn actions they can take. Mostly, these "Off Turn" actions take the form of "Active Defences", where you can expend action points to get a better chance at resisting certain attacks.

One of the most basic Active Defences is Blocking, which has a very, very strong chance of preventing an attack from getting through. (And wounds are nasty - players won't want to take a hit if they can help it.) This, hopefully, leads to my personal favourite bit of emergent behaviour - a player can get put on the defensive, where they have to keep spending their actions to block over and over, without ever being able to make an attack of their own.

I have a few ways that either side could break out of this - either the defensive player having something like a shield bash that lets them get back on the offensive, or the offensive player having something like a feint or an overrun where they break through their opponent's defences entirely.

However, I don't want the existence of those options to completely negate that dynamic - that's where the Opprotunity system would come in. One player would put the other on the defensive, and the other would keep blocking over and over - all that they could do with Cost:1 actions. The options that would let them break this cycle would have a higher cost, however, so there'd be a little bit of luck involved in who got such an option first.

[Vadskye]

That would be ideal, but much more difficult to design for - and, it runs into the min/max issue where players feel like investing resources in "Secondary" options just sabotages their ability to fill out their primary role. Really, though, a lot of this is coming from a recognition on my part that I'm a newbie designer - trying to make a system work naturally and fluidly like that is easier said than done.
It is definitely difficult to design for. But why invest time and energy into developing something that will always be less immersive and intuitive than a well designed system, even if you do get it working? I'd rather spend the time focusing on something where the ideal is... ideal.
Then again, I've been writing my system since last July. So I admit that there's definitely a tradeoff there.


Oh, and here's an example of one of the specific ways I'd like to use this:
The system I'm making uses "Action Points" that renew at the end of every turn, and describe both what a character can do during their turn, and some off-turn actions they can take. Mostly, these "Off Turn" actions take the form of "Active Defences", where you can expend action points to get a better chance at resisting certain attacks.

One of the most basic Active Defences is Blocking, which has a very, very strong chance of preventing an attack from getting through. (And wounds are nasty - players won't want to take a hit if they can help it.) This, hopefully, leads to my personal favourite bit of emergent behaviour - a player can get put on the defensive, where they have to keep spending their actions to block over and over, without ever being able to make an attack of their own.

I have a few ways that either side could break out of this - either the defensive player having something like a shield bash that lets them get back on the offensive, or the offensive player having something like a feint or an overrun where they break through their opponent's defences entirely.

However, I don't want the existence of those options to completely negate that dynamic - that's where the Opprotunity system would come in. One player would put the other on the defensive, and the other would keep blocking over and over - all that they could do with Cost:1 actions. The options that would let them break this cycle would have a higher cost, however, so there'd be a little bit of luck involved in who got such an option first.

Your attempt to limit and redefine options definitely makes more sense in the context of action points. Or more precisely, it makes sense if there are specific unusual options you know characters will want to perform, and you don't want them to always be able to do it. Though this brings to mind a different way of solving that problem: why not say that succeeding by enough on an attack roll / defense yields some special effect? If you succeed by enough on the attack roll, you can get a feint/overrun/etc. effect that can override the action point. On the other hand, a defender spending an action point who succeeds by enough (or perhaps this triggers if the attacker fails by enough) automatically gets to shield bash or otherwise reverse the dynamic. That ties the solution into the existing system in a more immersive way.

[FreakyCheeseMan]

It is definitely difficult to design for. But why invest time and energy into developing something that will always be less immersive and intuitive than a well designed system, even if you do get it working? I'd rather spend the time focusing on something where the ideal is... ideal.
Then again, I've been writing my system since last July. So I admit that there's definitely a tradeoff there.

For me, it's less about time, and more about unpredictability - I don't know what players are going to think of as far as game-breaking options go, and really, without a *lot* of playtesting (which is difficult to make happen), I don't know what options will turn out to be under/over powered; some of it I can just run the math on (Figuring out average damage with a certain set of options, etc,) but the rest is kinda foggy; I feel like this system gives me a cheap and dirty pre-emptive, cause so long as I don't make any of the truly OP options have a cost of one, even if they do turn out to be game-breakingly powerful, they won't actually render it unplayable.




Your attempt to limit and redefine options definitely makes more sense in the context of action points. Or more precisely, it makes sense if there are specific unusual options you know characters will want to perform, and you don't want them to always be able to do it. Though this brings to mind a different way of solving that problem: why not say that succeeding by enough on an attack roll / defense yields some special effect? If you succeed by enough on the attack roll, you can get a feint/overrun/etc. effect that can override the action point. On the other hand, a defender spending an action point who succeeds by enough (or perhaps this triggers if the attacker fails by enough) automatically gets to shield bash or otherwise reverse the dynamic. That ties the solution into the existing system in a more immersive way.

Yeah, it occurred to me as I was writing that that that would probably be a better way to handle it - I may do that in general, in fact, and say that a certain number of successes on certain actions opens up extra choices.

[Vadskye]

so long as I don't make any of the truly OP options have a cost of one, even if they do turn out to be game-breakingly powerful, they won't actually render it unplayable.
It would help keep an accidentally broken system playable by limiting access to overpowered abilities. But as long as you know that you're in beta testing anyway, I think your players should expect that you will step in to tweak broken mechanics if they come up. And giving players the option to do whatever they want each round will make it that much easier to notice and change overpowered (or underpowered) options. Your idea has some merit, though.


Yeah, it occurred to me as I was writing that that that would probably be a better way to handle it - I may do that in general, in fact, and say that a certain number of successes on certain actions opens up extra choices.

Threshold systems have merit. I like the idea that I can do so well on my attack that I get to throw in a free disarm, trip, or some other cool add-on to the action. Just be careful - one of the things that makes D&D run faster than other systems is that you don't have to add up the full math on every single roll. As long as you can eyeball whether the roll is over or under the DC, it's good enough - and this makes it run smoother. If you're going to add extra options at certain thresholds, I'd recommend an "If you beat the DC by 10 or more..." threshold; 10 is a really easy number to do math with.

[FreakyCheeseMan]

Threshold systems have merit. I like the idea that I can do so well on my attack that I get to throw in a free disarm, trip, or some other cool add-on to the action. Just be careful - one of the things that makes D&D run faster than other systems is that you don't have to add up the full math on every single roll. As long as you can eyeball whether the roll is over or under the DC, it's good enough - and this makes it run smoother. If you're going to add extra options at certain thresholds, I'd recommend an "If you beat the DC by 10 or more..." threshold; 10 is a really easy number to do math with.

I think that actually works a little better, under my rather ridiculous system for determining chance outcomes.

My Rather Ridiculous System of Determining Chance Outcomes:
All chance outcomes have the same system: players flip a coin (or roll a die and go high/low, whatever) until they get a result different than the one they started with - however long the streak went, they get that many successes or failures. So

HT = 1 Success
TTH = 2 Failures
HHHHHHHT = 7 Successes

and then modifiers (which are few and far between) are added or subtracted directly. (1 Failure + 1 = 1 Success - there is no "Neutral.") There are no damage rolls - spells and weapons determine damage based on successes.

I've been going back and forth about whether or not I like this system - the downside is pretty much just that it uses a lot of die rolling. The upsides are that it only requires a single die, maintains a steady, gradiated probability curve for all actions, and, as even a single point of difference is significant, means I can keep the math to a bare minimum.

[Vadskye]

Well, that's a unique system. It is indeed simple - but dang if that isn't a lot of die rolling. Rolling one die at a time is a pain, and you're rolling a minimum of two separate dice (or two coins, etc.). As long as you're experimenting with super simplified roll mechanics, I'd be happier if you could compress the number of rolls - in a lot of situations, D&D would actually have faster rolling mechanics than your current system.

But anyway! I ran some math on your system using Anydice.com to see what the numbers look like. The default chance of success is 50%. Let's say that we want extraordinary success to happen about 10% as often as you succeed. According to my calculations, you get at least 5 successes 6.25% of the time. That's good enough for me. So extraordinary success could be a +4 threshold over regular success.

[FreakyCheeseMan]

Hmm... I could keep the probability model but reduce the die rolls by switching to a d8.

First roll would be
1: 3 failures + roll again
2: 2 failures
3-4 : 1 failure
5-6: 1 success
7: 2 successes
8: 3 successes + roll again

Subsequent rolls would be the same, but would just treat everything on one side of the die as "stop rolling and keep what you have so far." If you then put an order on them, rolling 2d8s at once could ensure that almost every action was resolved with one roll. Die rolling would actually be reduced from 3.5, once you factor in damage rolls.

...I think I did that math right. I might switch to a D20 with a similar distribution - the curve would be less elegant, but be mostly similar while requiring less die rolling.

...if only there were a D16.