Binary Success vs Multiple Levels of Success

[zakael19]

It seems only skills are multiple levels of success. Combat you hit or miss and saves are for nothing or half damage mostly. The exploration and social pillars get opposed by the DM wanting to push information to the players vs the players not getting the clues or looking in the right places, so the DM needs to get it to them.

Instead of you fail and the players need to figure out something else if we use the binary pass/fail system. When talking to a NPC it is easy to see that he could be partially swayed or mostly swayed or totally in your pocket. But the PCs cannot be swayed unless they want to.

I guess the question here is a) is combat not a pass/fail conflict drawn out over a long period with a whole lot of highly-mathed rolls akin to the blog's idea about skills and b) are we sometimes solving a conflict that might be a skill roll in some circumstances but a full on initiative into combat in others (eg: you fail the social challenge, COMBAT TIME!).

Instead of adding more rolls per se, I find what Errant does really interesting. It merges in Position and Effect from Forged in the Dark play to a D20 based adjudication system. On top of that, it has straightforward calculations for the GM to determine the Difficulty Value of an check ("something of which the result is uncertain and failure has consequences"), and if the DV is 0 or lower once all adjustments have been made - the character automatically succeeds. Position and Effect tell you "for this action, on a failure the outcome will be Eh, not great / about as bad as you'd expect / oh so much worse" and on a success the outcome will be "not quite as much as you wanted / about what you were hoping / even better then expected."

 

[Staffan]

I think both methods have their place. I like having multiple possible success levels (at least basic/critical success) on a single roll, for when you're just doing a single thing. But I think collapsing a larger effort to a series of rolls can also be useful in many cases – particularly if those rolls can be spread out over multiple characters in order to engage multiple players. But a thing to consider in those cases is that a "failure" shouldn't be a showstopper, but more along the lines of "you didn't do so good at that particular aspect".

Troubleshooters, for example, use what they call Challenges frequently. They're basically "roll for three to five different skills to see how well you did overall." For example, you might be setting out on a wilderness expedition to a forgotten city in the jungle. The Director might call for a challenge for Credit Rating to finance the expedition, Charm or Contacts to get local help, Survival to manage the wilderness, and Endurance to deal with the actual walking. A failure on e.g. Credit Rating doesn't mean "Sorry, you couldn't afford it", but it might mean "You had to settle for some gear that wasn't as good as you would have wanted" which then leads to a less than ideal result when summed up. IIRC, on a four-skill challenge you need 3 successes for a clean success, while 2 would be a success-at-cost (you got there but it took longer so your rival got there first and has a head start) and 4 would be success-with-upside (you got there with plenty of time to spare and managed to find something cool on the way as well).

In another scenario you might want to get into a locked room to look for clues. That might be a three-skill challenge: Security to figure out the guard patrols, Prestidigitation to pick the lock, and Search to look for clues. A failed Prestidigitation check doesn't mean you don't get the lock up, but it could mean it takes a little longer so you don't have the time you had hoped for to search. Two successful checks would mean you find the clues, three would find some additional clue or something else useful, and one might mean you find the clue but the guard shows up just as you were leaving.

Basically, multi-roll checks are good for "montages", but not for doing several things at the same time.

 

[Laurefindel]

On the idea of multiple tiers of results...
I think extra scales of success or failure are extra steps in just doing what PBTA has been showing people how to do all along.

So you need a 10 to succeed at a Persuade check?

1 = fumble, bad things happen
5= you didnt do it, but things are not ruined for ya
8= you kinda did it, but only in a limited way
10 = you did it, but only in a basic way
15= you did it and you get a little extra
20= you succeeded in the best way possible and get bonus results that are good too

like... sheesh.... what a overly extra tedious way of just doing =
10+ you did it as best as can be done within the fiction
9- you did it with some complications
1 the GM gets to decide based on the theme and plot, and here is a quick ref guide.

The extra granularity of the first set in fact adds nothing. You can fold 5, 8, together for the same effect based on the context of the roll. you can fold 10 and 15 together because they are the same thing. 20 can be reserved for character special abilities or just folded onto 10 because at the end of the day, you still just "did it".... move on to the next fun bit already...

And if you want, you can use a PBTA style 'themed suggested results set' for success with complications if you need a little creative juice. You got success with complications to Persuade the guy = ok, well they go along with it but ... pick one

• you have to pay them extra
• you owe them a favor
• it makes someone else an enemy of you
• it adds danger to the scene
• etc etc...

That is just as quick as any binary roll, and it helps get the mind thinking of what the risks were in the first place.
And it gives the GM what they really want deep down = challenges and drama for the players to engage with.
Even better it was the roll that made the result = the GM isnt the enemy to outthink, they are the person who helps bring these results to like in big and cool way


.................
On the article....
The author of the article is laughably wrong on this = "Mixed successes provide a framework for an assumed situation, which might not fit what’s actually going on"
its utter nonsensical gibberish. assumed situation where mixed success does not fit?

Here is the real nitty gritty kitty litter truth = this is really exposing bad GM habits.... ones where the GM has no clue what the stakes are, so its just easy for them to say "you fail, nothing happens". or "I don't really know if I want you to succeed here, so uh... here is a DC, if you make it then fine, you do it..." = yuck!

If the assumed situation has some chance of failure, then failure can be quite a few things... like... a lot of possibilities...

And a "success" is absolutely part of a failure state. (i.e. yes, you got the reactor working, its online and everything has power, but its overloading, so in 1 hour it will blow up)

Even in "fail" too... (i.e. You didn't get the reactor online your hack failed, but since the power is out the enemy can't open the maglock doors, so you are safe... for now...)

Mixed success always makes sense. And if it doesn't, its a bad GM making someone roll when they should not have been rolling at all. That, is the death of fun and a waste of time.



.......................

On Pass/Fail and binary results...
Are trash.
Ok, that's too harsh. But hear me out. "I roll to hit the ork, I miss. nothing happens." its binary. You hit or you don't hit. Nothing interesting happened when I missed (even if the interesting thing as the ork getting a one-up on me).

This also makes gaming worse for success. "I roll to persuade the king to kill himself." I succeed, he does. its binary. There is no reason for this, there is no build up. In a single roll we have given up what would have been a interesting dialog of why, and to what end, and what emotions or reasons were involved.

Binary pass fail is junk because it requires a GM to be a gatekeeper of fun. They, the person who is the GM, have to say "No you can't just persuade them to to that". That is a person telling another person what they can try in a game and what they cant. And its not always agreed upon, there are a thousand posts on this very forum over this exact thing...





................

IMHO, it is proven day after day, everyone (deep down in truth) wants to play PBTA, its what everyone really enjoys... we are all tired of pass fail and extra layers of nothing, and we all want rolls to be interesting, even failed ones. And PBTA serves as a good test of bad habits and exposes unfun rulings that too many people have become used to suffering through.

There are so so so many blogs, posts, podcasts, even critical role talks = of people saying how to make failed rolls more interesting, narrate things happening, getting players to not spend hours arguing over how to out think the GM, players getting bad rolls and it removing them from play, GMs not handling missed perception checks leading to failed plots, tedious combat of 45 minutes of waiting for your turn to miss... the list goes on and is really obvious


Funny thing is... these gripes, the heartaches... all for pass fail systems (most of which are really wargames in drag). None of these things are breaking the backs of PBTA games, no such wealth of players who fully embrace and play mostly only PBTA talk of of untenable roll results or plays having to do nothing from bad rolls, PBTA isn't missing out on plots or suffering through a GMs guessing game of clue locations...

................

I used to not get PBTA. I thought it was "training wheels for roleplay" or some such. I thought the moves were limiting to players and the GM had no clue what to run or do... I was wrong.

I was trying to force my bad habits and my overly controlling GM habits into a game that solved all that and more.

Now, when I run PBTA, as a GM I work less, and get more fun out of my plots, my players don't see me as a villain to fight, and as a player I get to say the thing I wanted to do, do it, and see how much fun snowballs from the successes and the complications both. Players embrace both harm and risk more, and that lets me add to the stakes

Now now, you’re trading bad GM habits for terrible GM practices. You don’t like pass/fail systems, I get that, but you’re laying it pretty thick there. PbtA is a good system and obviously it worked for you, but it is far from the panacea you think it is.

 

[billd91]

What do you think? Do you prefer binary success or multiple levels of success? Why? Do you agree with the author's idea that you can make binary success more interesting by making checks more complex?

That depends on what you’re doing. Some things work fine with pass/fail, some with varying degrees of success. Brennan Lee Mulligan has made some excellent uses of degrees of success when judging skills in D&D (for example, DC 10 gets you something, DC 15 gets you something more, DC 20 gets you everything). Knowledge type checks have had them since 3e. And combat pretty much always has had degrees of success since the damage roll has always essentially been a method for it.

 

[aramis erak]

...............

IMHO, it is proven day after day, everyone (deep down in truth) wants to play PBTA, its what everyone really enjoys...

................

I strongly disagree with the quoth bit.

Most of the games I like have multiple success levels, but not all have a fumble/botch/crit_fail.

MegaTraveller has 4 standard levels of success/fail, Exceptional Success, Simple Success, Simple Failure, Exceptional Failure... and Exceptional failure is probably also a fumble, but there's a separate roll for fumble severity, which has 5 levels: none, superficial, minor, major, and destroyed.

White Wolf is a count successes system, so is most of Free League's product line.

The two lowest I've run as primaries are D&D and Pendragon; Pendragon has 4 success levels in simple rolls (CS, S, F, CF), and 6 in opposed resolution (CS, CT, S, PS, F, CF). D&D has an option for critical results; I used them in 3.X and 5-14 (DMG page 242), and success at cost is allowed for any fail (ibid.) As is allowing an increased failure at failed by 5 or more. (ibid.)
Right or wrong, I also used the D&D outcomes range in Stargate (p 134 shows 4 levels, but the rules on checks don't, but combat does.) and Monarchies of Mau... (MoM p 97: Triumph, Success, Failure, Botch - 4 level).

I'll note TFT has multiple levels as well - at least in combat. Note that it also modifies target numbers, not the die roll. Combat rolls of 3 are triple damage, of 4 are double, 5 to controlling adjusted attribute are single, higher to 16 are fail, 17 is drop weapon, 18 is break weapon. It's my current. Non combat has 4 levels, but the dice rolls vary by difficulty, as difficulty is in dice rolled.

I find 3 levels of success at least one too few. Hell, the one game I've run that's built from a PBTA core concept has 5 outcomes... Sentinel Comics.

I've known people who genuinely don't like anything but boolean.
They're not the kind who stay at my table.

 

[Andvari]

I prefer a single resolution roll. I don't necessarily have a preference for multiple degrees vs binary, although I do mind multiple degrees if they have clearly defined outcomes specific to each action amongst a large variety of defined actions, as it results in having far more rules to memorize. Instead, I'd prefer each degree has a more general rule that can be broadly applied to most any action. For example, success, success with a cost (perhaps leaving it to the GM and player to negotiate the type of cost), and failure.

 

[Crimson Longinus]

That depends on what you’re doing. Some things work fine with pass/fail, some with varying degrees of success. Brennan Lee Mulligan has made some excellent uses of degrees of success when judging skills in D&D (for example, DC 10 gets you something, DC 15 gets you something more, DC 20 gets you everything). Knowledge type checks have had them since 3e. And combat pretty much always has had degrees of success since the damage roll has always essentially been a method for it.

Yep, this is what I do as well. I like degree of success. However, it is not needed for everything and some things can be simple pass/fail. I like a system that can easily handle both.

 

[Breaking Star Games]

I really like multiple levels focused on generating complications because I think it maximizes player agency. The largest difference between rolls come up with complications vs binary is that by having player action create complications, you focus on that PbtA Play to Find Out rather than the traditional prepped obstacles/adventure.

To spell that out, when I run a PbtA game, my prep is very fluid to use as the players lead the story and really, it's those 7-9 results and misses that shape what kind of problems they deal with. Whereas when you have binary success/failure, you don't generate these new obstacles, it's just as a matter of how many resources you spend based on the luck of your rolls. Instead, you prep these obstacles ahead of the game in a more linear order like a published adventure or a dungeon. So, the prep guides the play, rather than player action. Obviously between sessions, you can take into account player action for your next prep, so it's hardly a railroad, but it's nowhere near as fluid as player action directly leading to what next they deal with, which I find diverts play in very exciting ways.

 

[Ruin Explorer]

A friend of mine shared this blog post with me and I though it would make for an interesting topic here.

I don't really buy the core of the argument that all you have to do to make binary results interesting is to add more rolls. That makes things tedious. In the drive, shoot and resist psychic attack example, the order of those things will matter.

Note that I am not saying that "partial successes" are easy to come up with on the fly. It can be a real pain depending on the task -- but that is why most PbtA and FitD games put some of that work on the players, too.

What works for me with games like D&D is instead of setting DCs for binary results, is I know the DC scale (8-12-16-20-24) well enough to interpret the results of a check on the fly. The specific situation will determine whether that 8 means "success at a cost" or "limited success" or what.

What do you think? Do you prefer binary success or multiple levels of success? Why? Do you agree with the author's idea that you can make binary success more interesting by making checks more complex?

I'm definitely in the "multiple levels of success" camp, but two I things I really don't like are:

1) DMs who try to treat binary success games as "multiple levels of success", particularly when it's not like, a house rule, they're just freestyling it. Like, no, making the DC but only by a small amount should not mean I basically fail, or have so many consequences I'm screwed, this is D&D, that's now how it works, stop that! Bad DM! Either make it explicit and reliable how it works (and apply it to rolls from NPCs), or treat binary as binary and and multiple levels as multiple levels. I would literally rather have pure binary than this weird kind of DM fiat I've seen a few times.

2) Having to do math to work out multiple levels of success. Just find another way to make the dice work if that's what you're thinking! So PtbA and so on are thus fine, but I loathe "for each X points beyond the DC your success level increases one rank"-type stuff.

I haven't seen a lot of difficulty with coming up with "successes with complications" - I think that's generally the better approach than "partial successes", because in my experience most DMs end up making "partial successes" into just "slightly creative failures". Whereas success with complications for whatever reason seems to be more clear - you succeeded, period - but other stuff went wrong.

Adding more rolls to a binary game is a very bad idea, I agree there. Because it's binary, it's essentially pass/fail, and even if you say "Welll, only one roll should really be pass/fail, the rest add nuance!" or something, the reality is most DMs are going to treat failures as failures, and are thus going to just make your success worse - because in most cases a success is already as good as you're going to get - critical successes which aren't silly just don't apply to a lot of situations.

 

[payn]

I prefer the multiple levels of outcomes to the roll a bunch until you succeed method. This is why things like PbtA and Mongoose Traveller have built it into the system for both ease of GMing and understanding of the player. It's also why things like 4E skill challenges nor PF2s spell tier of success (ultimately it just feels like two stage binary) didnt work for me. In the first instance it feels intentionally built into the system, in the second it feels like a bolt on that is just an exercise in convolution.