Failing Forward

[nomotog]

I love the idea of fail forward systems. They help the story flow forward. The annoying bit is that your can't always match up logical disadvantages with the action. You often have to make up your own on the spot and that can feel a little random if handled poorly. Like I recall one zombie game that had a system were the environment could get more or less deadly based on your actions. Like if you made a lot of noise the danger level would go up and that could attract more zombies to you. It made sense you know zombies are drawn to noise, but high danger level could also do things like summon bad weather or make your missions take place at night. Trying to logic out that didn't make as much sense, but the trade off was you had a lot more verity in danger results. (I am still conflicted on if it's a fair trade.)

 

[Maxperson]

Well, then don't do it, and avoid systems that have 'success, but at a cost' in their mechanics.

Nah. No need to avoid systems like that. It's an easy change to make. If I avoided RPGs that have systems that I don't like, I'd never be able to play an RPG.

 

[grendel111111]

Ah. I'd instead say that Cha is a valid thing to roll to find a guide who could help you get to the top, not that Cha was appropriate for a climbing check. Remember that there are systems that allow you to do odd pairings of skill and stat (so, you might use INT + Climbing when asked what you know about the technical aspects of climbing). If you are making a Persuasion roll, you aren't using Climbing skill, so it isn't a Climbing check, you see.

Unless, of course, you're playing in a game in which spirits of the land/mountain may be invoked, in which case Cha may well be right for a climbing check

That is true, but if the climbing part is not in anyway involved (No matter what you roll you will get to the top) Is climbing skill in anyway needed? Shouldn't you roll on a skill that matches what the failure might be?

 

[grendel111111]

I don't have time for my own personal commentary or to address specific facets of specific posts that I would like. However, here is a link to a blog post on Dungeon World's failure that is apropos. I'm going to include some of the relevant bits below.

(for some reason this part did not copy:
I read an article about emergency treatment that totally changed how I look at failure from a narrative point of view. Serious complications are unavoidable in a hospital situation, but some hospitals keep more patients alive despite this. The key to their success is in how they quickly rescue patents when things go wrong. As the author put it the only failure is a failure to rescue.


And this works well in Dungeon world or similar games. It matches the degree of focus that they are trying to achieve with their system. If you want to try and mimic similar in D and D you can, but the more you do the better off you are just using dungeon world because it is a better match.

I voted for I like it but only in certain situations because many of the Fail forward results don't follow.

We are partly limited by the degree of focus that the game puts on each action.
In the above hospital situation they are talking about medical complications..... the surgeon encountered something unexpected that hadn't shown up on a scan, a cut was made too deeply, etc. All of those things are included in the 1 medicine check in D and D. And they are all related to your medical skill. Other wise you would need to do the operation step by step. (Oh you cut an vein, roll medical to clamp it. His heart stopped, roll Str (medical) to keep his heart pumping while you finish the operation). Instead it is all put into 1 roll for the operation.

So if the DM decides that it is essential the operation is successful for the "story" so why roll for it?
(For me) The fail forward should be something medical if you are rolling medicine. Maybe a sponge was left in the patient, so they later develop a fever, etc.

But if you decide that on a failed medicine roll, then the patients family turns up and demands that the operation stop for religious reasons there is no link between the skill and the result. The family would turn up independent of the doctor skill. They would turn up if a good doctor was working on the patient or if an OK doctor was working on the patient. The doctors skills in no way affect the chances of the family turning up. So either it is so interesting that it happens anyway, or assign a random chance of it happening independent of what is happening in the operation.

 

[grendel111111]

I love the idea of fail forward systems. They help the story flow forward. The annoying bit is that your can't always match up logical disadvantages with the action. You often have to make up your own on the spot and that can feel a little random if handled poorly. Like I recall one zombie game that had a system were the environment could get more or less deadly based on your actions. Like if you made a lot of noise the danger level would go up and that could attract more zombies to you. It made sense you know zombies are drawn to noise, but high danger level could also do things like summon bad weather or make your missions take place at night. Trying to logic out that didn't make as much sense, but the trade off was you had a lot more verity in danger results. (I am still conflicted on if it's a fair trade.)

I agree with this. Tying unrelated events to a skill makes it too disjointed for me.
In picking a lock a fail resulting in broken tools or alerting the rooms occupants is a logical follow on from failing to pick a lock (you used too much force, dropped you tools, etc.) but a failure that let you "open the door, but it starts raining" (somehow if you were better at picking locks you could have controlled the weather) is a step to far for me.

 

[Janx]

I agree with this. Tying unrelated events to a skill makes it too disjointed for me.
In picking a lock a fail resulting in broken tools or alerting the rooms occupants is a logical follow on from failing to pick a lock (you used too much force, dropped you tools, etc.) but a failure that let you "open the door, but it starts raining" (somehow if you were better at picking locks you could have controlled the weather) is a step to far for me.

That's probably a GMing issue. Initially, I cannot imagine deciding that because you failed the lock pick roll, it starts raining and the door opens. I would suspect my first choices of what the failure would mean, would relate to the act of lock picking and your equipment, body, etc. Perhaps the pick breaks, jamming your finger for 1 HP. Which causes you to yelp, which a monster on the other side hears and opens the door to investigate (thereby solving the locked door problem, but destroying the element of surprise for the PC.

 

[nomotog]

I agree with this. Tying unrelated events to a skill makes it too disjointed for me.
In picking a lock a fail resulting in broken tools or alerting the rooms occupants is a logical follow on from failing to pick a lock (you used too much force, dropped you tools, etc.) but a failure that let you "open the door, but it starts raining" (somehow if you were better at picking locks you could have controlled the weather) is a step to far for me.

I am not 100% sure about either version. I mean if you think about it, then you can kind of justify the logic. "It took you longer then you expected and the rain started." (You could also foreshadow something like that letting the players know a storm is coming and that they need to move fast to avoid it.) It also gives you a lot more wrenches you can throw. If all you can do is brake tools and alert guards, then that is going to get old faster then you think. Rain is a more interesting wrinkle then a broken lock-pick.

 

[grendel111111]

I am not 100% sure about either version. I mean if you think about it, then you can kind of justify the logic. "It took you longer then you expected and the rain started." (You could also foreshadow something like that letting the players know a storm is coming and that they need to move fast to avoid it.) It also gives you a lot more wrenches you can throw. If all you can do is brake tools and alert guards, then that is going to get old faster then you think. Rain is a more interesting wrinkle then a broken lock-pick.

But it rains on the skilled and unskilled alike. If it is going to rain then your lack of skill isn't going to cause it to happen. then it becomes to different things unrelated to you picking the lock (that has been decided already that yes you do). Does it rain? not in any way affected by your lock picking (but maybe due to time).
So I could see the rain approaching and you feel drops on your face, make a lock picking roll, If you make the roll you get through the door just before the down pour begins and have slightly damp clothes. Fail and you are caught in the down pour and are drenched (torches go out, supplies are ruined, maybe a light acid effect). The difference then is did you get inside before the rain hit (the rain is still falling irrespective of if you are inside or outside). Rather than if you make your roll you get through the door and it isn't raining but if you fail you get through the door but it starts raining.
I know it is subtle and I suspect it is partly dependent on if you see the narrative being the most important force in the game or having a consistent world being more important.

 

[billd91]

Rain is a more interesting wrinkle then a broken lock-pick.

Depends on your definition of interesting. Personally, I don't consider a vexed, WTF feeling to be interesting.

 

[nomotog]

Depends on your definition of interesting. Personally, I don't consider a vexed, WTF feeling to be interesting.

Well I mean rain introduces a more diverse set of changes. If your picking a lock and brake your pick, well you can't pick other locks. Meanwhile rain dose a lot. You have all kinds of penalties in it, some bonuses attached to it.

Though the general idea wasn't bad lockpicks cause rain, it more that the failure penalty doesn't have to be instant and directly related to the action. That opens up a lot of options to the GM to use.

Actually you can make things more nebulous. Like if we jump back to the zombie game. Making noise didn't summon zombies. It raised the danger level. It didn't have any concrete and direct results. You knew was going to lead to something bad, but you didn't know how it would come around to bite you. It was like a chaos theory or karma thing.

Now this thought is kind of unrelated to the idea of direct connecting. With fail forward, you want to take care to make sure that things move forward. You need to add choices not just take them away. (One of the reason broken tools doesn't work as well.) Jumping back to the zombie game. I toyed with a scavenging system were you always found what you were looking for. (If it was possible.) If you failed, you had to roll an extra 1d6 and that added buts to what you found. Like you found what you wanted, but someone else already looted it, you found what you wanted but it's inside a safe, you found what you needed, but it's not exactly what you wanted (like you were looking for a gun and found a spear gun.) That last one was extra common because it fed into my crazy crafting idea.