Informed decisions and dice rolls.

[Talakeal]

Why is it that whenever an RPG gives you the ability to modify or reroll a dice it almost invariably does it from a position of ignorance?

For example, in WoD you can spend a point of willpower to receive a bonus on a roll, but you must declare doing so before any dice are rolled. Furthermore, in Pathfinder the improved save feats allow you to reroll saving throws and Eberron Action points allow you to add to rolls, but only before you learn the DC of the test and whether or not you succeeded.

Do people genuinely prefer to play guessing games as some sort of thrill rush akin to gambling? Or is it because the authors think these abilities would be overpowered if they allowed you to make an informed decision?

I personally find always firing blind to be annoying, especially when you already have the randomness of the dice roll in place, and think the strategic level of play would be much more interesting if they didn't include the guessing game element.

My personal favorite way to handle it is the Lord of the Rings skirmish game (not really an RPG, but close) which allows you to spend might to modify rolls and fate to avoid wounds after learning the result of the dice, and if your initial expenditure doesn't prove to be enough you can then go add more. But even it has a weird guessing game in that if two characters on opposed sides want to modify the same roll it has a weird little bluffing game where you simultaneously write down what you want to spend (and you can retroactively decide to spend less than you initial said, or even nothing) and then reveal them.

Any ideas why they are so prevalant? How do you feel about them?

[Vitruviansquid]

Do people genuinely prefer to play guessing games as some sort of thrill rush akin to gambling?

First of all, yes.

Second, let's imagine you're rolling 1d6 and trying to hit a target number of 4. If you have to allocate points to add before knowing the result, you get to decide between adding nothing, adding 1, adding 2, and adding 3 to your result.

If you get to see the result after allocating points, you can add nothing or add enough to hit your target number for HALF of the time. The other half of the time, you don't have a choice because you've already hit the target number of 4.

[jaydubs]

Do people genuinely prefer to play guessing games as some sort of thrill rush akin to gambling? Or is it because the authors think these abilities would be overpowered if they allowed you to make an informed decision?

Yes and maybe. In general, dice are introduced to add a certain degree of randomness and thrill to otherwise dry systems. It's weird that humans are wired to be excited by random risk, but we are.

Statistically, making a player choose whether or not to use a reroll ability before knowing the result makes it less powerful. Whether or not that's why the writers generally make it that way, who knows? Another explanation is that it makes turns go faster.

Player rolls.
"Did I make it?"
DM checks.
"No."
Player thinks about it.
"I'm rerolling. Did I make it?"
DM checks.
"Yes/no."

VS

Player rolls, maybe adds ability.
"Did I make it?"
DM checks.
"Yes/no."

[Milodiah]

I think it's because they're intended for use in especially important situations, like swinging at the BBEG or trying to save another PC from dying. If you could modify it after you know the result, people would be more inclined to bump up barely-failed rolls in more mundane situations, simply because they can. Besides, if you can make the decision afterwards, why would you ever make the decision before? You're essentially just grabbing the die and then moving it so the number's one higher, it's far better to actually have to think about whether or not you want to take the risk before you roll, as opposed to there being no risk since you can just apply it at leisure with no negative consequences or potential to "waste" it.

[Talakeal]

Yes and maybe. In general, dice are introduced to add a certain degree of randomness and thrill to otherwise dry systems. It's weird that humans are wired to be excited by random risk, but we are.

Statistically, making a player choose whether or not to use a reroll ability before knowing the result makes it less powerful. Whether or not that's why the writers generally make it that way, who knows? Another explanation is that it makes turns go faster.

Player rolls.
"Did I make it?"
DM checks.
"No."
Player thinks about it.
"I'm rerolling. Did I make it?"
DM checks.
"Yes/no."

VS

Player rolls, maybe adds ability.
"Did I make it?"
DM checks.
"Yes/no."

I don't know about you, but I would spend a lot longer deciding on whether or not to use my 1 per day reroll vs. a save or lose spell if I didn't know the result. If I do it is a know brainer, if I don't I am going to be running all sorts of math in my head and be very hesitant and indecisive about the issue.

I think it's because they're intended for use in especially important situations, like swinging at the BBEG or trying to save another PC from dying. If you could modify it after you know the result, people would be more inclined to bump up barely-failed rolls in more mundane situations, simply because they can.

I don't know, if it is a limited resource (most abilities like this a once a day or in the case of action points something like 5 a level) that isn't a bad thing. The player knows that if they use it they won't have it around later. To me that should be enough of a gamble to make people only use it on special things.




Also, running the math on D&D Action Points: You start the game with 5 and get one every 2 levels. That is hardly something that is trivial to use. They add 1d6 to a d20 roll, which means that there is only a 17.5% chance they will actually turn a success into a failure. When you only get 15 over the course of your entire adventuring career that's a pretty crappy non ability.

[Tengu_temp]

Many games which use an opposed rolls mechanic let you make an informed decision when you decide to reroll or add modifiers when you're the defender - and not when you're the attacker because the attacker rolls before the defender. Mutants and Masterminds, Fate, Legends of the Wulin - those are just a few examples.

[Sartharina]

Also, running the math on D&D Action Points: You start the game with 5 and get one every 2 levels. That is hardly something that is trivial to use. They add 1d6 to a d20 roll, which means that there is only a 17.5% chance they will actually turn a success into a failure. When you only get 15 over the course of your entire adventuring career that's a pretty crappy non ability.

You don't get 5+1+0+1+0 etc... you get 5+6+6+7+7+8+8... etc, but lose any 'unspent' action points.

[Vitruviansquid]

I don't know about you, but I would spend a lot longer deciding on whether or not to use my 1 per day reroll vs. a save or lose spell if I didn't know the result. If I do it is a know brainer, if I don't I am going to be running all sorts of math in my head and be very hesitant and indecisive about the issue.

Most games I've seen, excluding DnD 3.5, don't like to boil a fight or a campaign down to "save or lose," but if that was to happen to me, I'd take 0 time thinking about modifying that roll, because expending any kind of resource is better than straight up losing.

[Talakeal]

You don't get 5+1+0+1+0 etc... you get 5+6+6+7+7+8+8... etc, but lose any 'unspent' action points.

Sorry, its been forever since I played it. Even so, even if you use them every time an important roll comes up they are unlikely to make a difference more than one a level (and even that is being generous).

Most games I've seen, excluding DnD 3.5, don't like to boil a fight or a campaign down to "save or lose," but if that was to happen to me, I'd take 0 time thinking about modifying that roll, because expending any kind of resource is better than straight up losing.

Exactly, IF you know the result.

If you don't know the result you need to think carefully because the reroll might be unnecessary or, worse, cause a previous success to turn into a failure.

[Vitruviansquid]

I don't know if we're talking about the same thing here, because as I perceive it, you're telling me you're more annoyed by this situation

DM: The enemy wizard casts "Dude to Toad" at you, do you want to use your points to add to your dodge roll?
Player: How many points should I spend to avoid it?
DM: You're gonna have to give it your best guess.

than by this situation:

DM: The enemy wizards casts... a spell on you.
Player: What spell?
DM: I can't tell you.
Player: Well, what does it look like?
DM: You're gonna have to guess. But it could do barely conventional damage... or instantly kill you.

[Talakeal]

I don't know if we're talking about the same thing here, because as I perceive it, you're telling me you're more annoyed by this situation



than by this situation:

No, I wasn't talking about unknown spells.

I was talking about "Wizard casts Finger of Death on you. Roll a fortitude save or die."

I roll something low, say a 7, which may or may not fail. I have improved great fortitude, which gives you one reroll a day. The feat specifically says you can't use it if you know the save DC. Therefore, I don't know if the 7 fails. I might have already succeeded, in which case I am wasting my reroll for nothing. Worse than nothing, because I could change my 7 which succeeds to a 1-6 which fails, meaning that my own feat killed me.

[Vitruviansquid]

I don't play DnD 3.5 or 5e, but it seems like this feat is just terribly designed. If they wanted you to be better against an important fortitude roll, it would be perfectly fine to just give you a reroll on fortitude rolls without forcing the DM to stop and run things differently than he did before the player had that feat. What happens if someone casts the same spell on you twice? Can you not use the feat if you knew the target number from the last time someone cast it on you?

Most games with rules that give you a resource to expend for rerolls or roll bonuses aren't that badly written.

[Talakeal]

I don't play DnD 3.5 or 5e, but it seems like this feat is just terribly designed. If they wanted you to be better against an important fortitude roll, it would be perfectly fine to just give you a reroll on fortitude rolls without forcing the DM to stop and run things differently than he did before the player had that feat. What happens if someone casts the same spell on you twice? Can you not use the feat if you knew the target number from the last time someone cast it on you?

Most games with rules that give you a resource to expend for rerolls or roll bonuses aren't that badly written.

Going by pure RAW the feat does nothing if you already know the DC. I think the intent is that the DM doesn't have to tell you whether or not you succeeded before you commit to using the reroll, but that is still terrible imo as it means the feat can often be wasted or worse actively hard you, which is both a terrible feat and a psychological punch in the crotch.

[jaydubs]

I don't know about you, but I would spend a lot longer deciding on whether or not to use my 1 per day reroll vs. a save or lose spell if I didn't know the result. If I do it is a know brainer, if I don't I am going to be running all sorts of math in my head and be very hesitant and indecisive about the issue.

I suppose that's true. But that's also what decision timers are for. A DM can say you have X number of seconds, and start counting down. And if you haven't decided, you don't use it. Players tend to be much more comfortable with decision timers (which penalize people for taking too much time to choose) than turn timers (which also penalize people who are less familiar with the rules).

I get where you're coming from though. I certainly feel more comfortable if I can completely control the situation, and run math from an informed position. But that wouldn't necessarily be more fun. There's a reason most popular games introduce randomness, rather than being no-chance. Aka, backgammon is often more exciting than a game of chess.

[Talakeal]

I suppose that's true. But that's also what decision timers are for. A DM can say you have X number of seconds, and start counting down. And if you haven't decided, you don't use it. Players tend to be much more comfortable with decision timers (which penalize people for taking too much time to choose) than turn timers (which also penalize people who are less familiar with the rules).

I get where you're coming from though. I certainly feel more comfortable if I can completely control the situation, and run math from an informed position. But that wouldn't necessarily be more fun. There's a reason most popular games introduce randomness, rather than being no-chance. Aka, backgammon is often more exciting than a game of chess.

I agree, but it gets to the point where excitement turns into frustration. I don't think anyone feels good about the game after the dealer hits 21 for example.

To use the improved great fortitude example, it is a reroll, not an automatic success. So for example it should be (imo)

You fail a crucial roll.
You get to decide if it is crucial enough to spend your 1/day ability. (This gives you player choice).
You roll the dice again. (This is where randomness and excitement come in).

Vs. How it is:

You roll poorly on a crucial roll.
You try and decide if it is crucial enough to spend your 1/day ability on. (Here you get player choice.
You try and judge whether or not your roll was enough to succeed in the first place (A secondary player choice which slows down the game and involves a lot of metagame assumptions).
You roll and hope that your new roll is better (this is where randomness and excitement come in).
You potentially learn that you chose poorly and you actually succeeded the first time. You either wasted your ability, or worse, hurt yourself. (This will, imo, always be far more frustrating than fun).

[Milodiah]

I actually feel opposite ways on the two, then. Adding a modifier should be done in ignorance, as I said before. But rerolling should be an option more or less, and it really kind of is anyway...I mean, the player sees the number on the die, it's not like the DM makes them close their eyes before they decide to reroll or not. If it came up a 4 on a d20, 4's pretty bad, and if it's an important thing that you're pretty sure a 4 can't cut it (and at least in most games you have a basic idea of what the target is, whether or not the DM tells you what it is right now) then you should probably re-roll. The way I see it, the only thing that's objectively worse than a 4 is a nat 1, since if a 4's gonna fail then a 3 or 2 is gonna fail with the same consequences anyway.


And just to be technical, of course we don't enjoy the house winning when the dealer hits 21...but we keep gambling, don't we? We don't storm off from the table, it's (probably) not like the dealer is cheating or anything. Same thing with a DM. We don't enjoy the monster rolling a nat 20, but we don't just cash out either (and unlike gambling, there's not an incentive to stop playing entirely from a run of bad luck).

[Talakeal]

And just to be technical, of course we don't enjoy the house winning when the dealer hits 21...but we keep gambling, don't we? We don't storm off from the table, it's (probably) not like the dealer is cheating or anything. Same thing with a DM. We don't enjoy the monster rolling a nat 20, but we don't just cash out either (and unlike gambling, there's not an incentive to stop playing entirely from a run of bad luck).

I have seen a number of people get up and go to a different table when they feel that a dealer is "too hot," and people certainly don't tip dealers when they are losing.

But gambling is about winning money, RPGs should be about having fun. The DM doesn't make a profit when the players get frustrated and lose, all that happens is the game gets less fun for everyone.

[Gavran]

4E makes use of a lot of "reroll and keep the better roll", which I think makes a lot more sense. But it has "add X to the triggering roll / targeted defense" too which I agree, feels kinda bad to use when it still does nothing.

Actually I think I read somewhere that reroll + keep best equated to like, +5 or something on a d20. I don't know the math behind it but it sounded plausible. It's interesting to think that you could use either system (rerolls or add modifiers) to get the same effect, because that way you could phase out the one that's less satisfying.

[Talakeal]

Actually I think I read somewhere that reroll + keep best equated to like, +5 or something on a d20. I don't know the math behind it but it sounded plausible. It's interesting to think that you could use either system (rerolls or add modifiers) to get the same effect, because that way you could phase out the one that's less satisfying.

It is a bell curve. Roll twice take the best is between a +1 and a +6 depending on what the target number is and how you round.

The problem is that rerolls are seldom take the best, they are almost always final roll stands.

[Sartharina]

... Deal? Or No Deal?

[NichG]

I'd say neither case is inherently better or worse, but they have different roles when it comes to overall design.

If you ask the player to allocate resources before knowing the result of a random process, you're tying their performance to a particular test of their awareness of probability and/or context. E.g. in the Finger of Death case, you're testing whether the player is roughly aware of what kinds of DCs might be out there for Finger of Death. This kind of test is actually pretty common in D&D: whenever there are multiple ways to approach a situation to resolve it but which have different statistics associated with them, the game is testing the player's ability to understand the probabilities and also estimate difficulties. For example, choosing to cast a spell that targets Fort or Reflex saves against a particular group of enemies. So abilities that work this way are another angle on presenting that kind of challenge.

On the other hand, if you ask the player to allocate resources after knowing the result of the random process then it tends to sort the mechanic into two particular branches of design space. One branch is that the resource is sort of like hitpoints - there's no way that the character isn't going to reroll that failed Save vs Death or Save vs Complete Shutdown, so the reroll is a secondary hitpoint track that works specifically with regards to that kind of ability. The fact that the player is 'choosing' to use it is basically an illusion in that case, because the situations where it will and won't get used are pretty clear-cut. This is useful for presenting something narratively as extremely dangerous and sustaining that impression, while making it less dangerous in practice in play. E.g. if you had 3 points you could spend to auto-dodge any attack that hit you, but any attack that hit you kills you instantly, then the resulting impression is that e.g. a knife can still be a dangerous threat to a powerful character, but it's the lack of ability of the person wielding the knife to overcome their skill that blunts the danger and not that 'the knife doesn't do enough damage to get through their hitpoints in one shot'. So that can be a useful trick.

The other branch of the design space is that you restrict the resource to being used for things which are not no-brainers. E.g. here's a pool of points that you can use to force an enemy to re-roll a save against one of your abilities for 3 points, let yourself re-roll an attack roll for 2 points, or re-roll your damage roll for 1 point. In this case the challenge/tactically interesting element is determining the best choice to make in the situation, and the important thing is to make sure there are a number of possible uses which are each sufficiently equivalent in value that their relative valuation becomes dependent on the details of the scenario. This can be done by linking incomparable things - e.g. '1/day you can either reroll a failed saving throw or gain an extra full round action'

[Nagash]

I have seen a number of people get up and go to a different table when they feel that a dealer is "too hot," and people certainly don't tip dealers when they are losing.

But gambling is about winning money, RPGs should be about having fun. The DM doesn't make a profit when the players get frustrated and lose, all that happens is the game gets less fun for everyone.

As a las vegas native for the smart people gambling is about having fun too.

The only sure way to lose all your money is to sit down and try to win. Thats why we have a saying "dont play it if you cant lose it".

You sit down to enjoy the highs and lows, the rush and the free drinks.

For anyone else who comes here... one more bit of local wisdom, "play slow, drink fast". Even if your losing the free drinks can make it a break even night if your smart.

Oh and yes, lots of us DO tip the dealer after losing. Those people basically live on tips, the cards are not their fault.

[Grim Portent]

I'm rather fond of how the 40k games handled fate points.

You can spend them (refreshing each session) to add +10 to a roll before you roll it or spend them to reroll after you've rolled. You can also burn them (permanently losing one) to avoid dying when killed.

[aspi]

Why is it that whenever an RPG gives you the ability to modify or reroll a dice it almost invariably does it from a position of ignorance?

Do people genuinely prefer to play guessing games as some sort of thrill rush akin to gambling?

You phrase it rather dramatically but basically I think that's the reason, yes. In their core, dice-based RPGs have a strong element of risk, chance and what you make from it. Imagine a system where you can spend points of some sort to improve a roll. If you're allowed to spend them before the roll, you have to weight probability against cost and either spend a lot for an ensures success or very little and take a risk of failure. If you were allowed to spend points after the roll, you could just spend enough points to never fail until you run out of points to spend.

That would remove the element of luck to a large degree and replace it by bookkeeping. That doesn't really sound all that fun.

As for the entire concept of dying due to one failed roll: I personally find that to be a rather bad concept in any case, and there are better was of dealing with this sort of problem than removing the element of risk alltogether.

[nedz]

If you can modify the result after the fact then the game becomes one of resource management rather than risk — the dice are just a monte carlo simulation, based on your modifiers, of how many resource chips you have to spend. There is nothing wrong with this, but it is a different game: Settlers of Catan versus Risk.

[kyoryu]

Many games which use an opposed rolls mechanic let you make an informed decision when you decide to reroll or add modifiers when you're the defender - and not when you're the attacker because the attacker rolls before the defender. Mutants and Masterminds, Fate, Legends of the Wulin - those are just a few examples.

That's actually untrue for Fate. In Fate, either side can spend Fate points after the initial roll is resolved. This can lead to Fate Point "bidding", which is a deliberate feature of the system as the rolling of aspects into the resolution leads to a kind of mini-story.

Fate is definitely a situation counter to Talakeal's post - all Fate Point expenditure is done after the dice are rolled.

If you can modify the result after the fact then the game becomes one of resource management rather than risk — the dice are just a monte carlo simulation, based on your modifiers, of how many resource chips you have to spend. There is nothing wrong with this, but it is a different game: Settlers of Catan versus Risk.

It certainly adds a resource-expenditure aspect to the game, but there is still an element of risk. Where that balance falls can vary from game to game.

Only if you remove dice entirely do you remove the element of risk entirely. If I spend my last Fate Point to mitigate a result, for instance, I'm banking on there not being a *worse* result down the road.

[Beleriphon]

No, I wasn't talking about unknown spells.

I was talking about "Wizard casts Finger of Death on you. Roll a fortitude save or die."

I roll something low, say a 7, which may or may not fail. I have improved great fortitude, which gives you one reroll a day. The feat specifically says you can't use it if you know the save DC. Therefore, I don't know if the 7 fails. I might have already succeeded, in which case I am wasting my reroll for nothing. Worse than nothing, because I could change my 7 which succeeds to a 1-6 which fails, meaning that my own feat killed me.

This is why I'm fan of Mutants and Masterminds 2E/3E method of rerolling. Blow a hero point and you get to reroll, but if the die roll is less than 11 it counts as 11, that way you don't waste points by making your roll worse.

[Amphetryon]

You phrase it rather dramatically but basically I think that's the reason, yes. In their core, dice-based RPGs have a strong element of risk, chance and what you make from it. Imagine a system where you can spend points of some sort to improve a roll. If you're allowed to spend them before the roll, you have to weight probability against cost and either spend a lot for an ensures success or very little and take a risk of failure. If you were allowed to spend points after the roll, you could just spend enough points to never fail until you run out of points to spend.

That would remove the element of luck to a large degree and replace it by bookkeeping. That doesn't really sound all that fun.

As for the entire concept of dying due to one failed roll: I personally find that to be a rather bad concept in any case, and there are better was of dealing with this sort of problem than removing the element of risk alltogether.

All of this. If the risk is entirely removed from the encounter, then there is no need to play out the encounter; the GM can merely say 'you win' and move on to the next thing in the story. The dice add an element of randomness and potential for failure, without which 'success' can feel quite hollow.

[Talakeal]

All of this. If the risk is entirely removed from the encounter, then there is no need to play out the encounter; the GM can merely say 'you win' and move on to the next thing in the story. The dice add an element of randomness and potential for failure, without which 'success' can feel quite hollow.

What sort of "encounter" is played out using a single dice roll?

Also, if the players have some sort of ability that actually makes success guaranteed rather than merely giving a reroll or a bonus it would likely be an extremely limited resource, in which case the certainty of using it up is a much bigger threat than the risk of a failed dice roll in almost all cases (and if it isn't, something like "You are hit by a poison trap. Roll a fort save or die" the encounter isn't going to be fun in any case, merely frustrating).

[Amphetryon]

What sort of "encounter" is played out using a single dice roll?

A Disintegrate spell cast at a party member, to use an example from earlier in the thread, is decided by a single die roll in most D&D permutations that I can recall.

A vast majority of 3.X's Enchantment spells can essentially turn entirely on a single die-roll.

Last time I looked (it's been a while, and I believe at least 2 editions have come and gone), WHFRPG combats were often fundamentally decided in a single roll, or combat round; many people, both on the forums and off-site, play 3.X encounters (combats especially) in such a way that a single roll - even just the Initiative roll - either wins the day for them, or has things going very badly for them.

Many 'stealth-based' obstacles are overcome - or not - based on a single stealth roll of some variety (opposed or otherwise), in several different systems.

Almost every edition of D&D's version of Trapfinding is reasonably expressed as an encounter that is played out using one Disable Device check; I have played other systems that handle traps similarly.

Pendragon's combat system is such that I do not ever recall both combatants getting off a successful attack before the combat was either over, or all over but the shouting.

I could probably go on.

Also, if the players have some sort of ability that actually makes success guaranteed rather than merely giving a reroll or a bonus it would likely be an extremely limited resource, in which case the certainty of using it up is a much bigger threat than the risk of a failed dice roll in almost all cases (and if it isn't, something like "You are hit by a poison trap. Roll a fort save or die" the encounter isn't going to be fun in any case, merely frustrating).

Note that the initial assumption in this quote is not mentioned explicitly, or even hinted at where I can see it, in your opening post, and reads from here as a shifting of the goalposts. Note, also, that the assumption in your parenthetical - that a Save or Die encounter is not going to be fun 'in any case' - may be true from your perspective, but is not empirically true.