Worlds of Design: “Old School” in RPGs and other Games – Part 1 Failure and Story

Alzrius said:

On a related note (I forgot to mention this before), remember how Matt Mercer issued a public apology after a character on Critical Role permanently died during the course of the campaign?

Click to expand...

I read the rest of his statement at the link - I get a hint of sarcasm in the full 'apology.' As in, 'if you found this ONE, SINGULAR MOMENT..." I think I commented to someone at the time, "The old-schooler in me wants to say, 'suck it up, it happens,' but at the same time I can get how when you're really invested in them, you absolutely hate when a beloved character dies." And it was a moment I hope that gamers who are new by way of Critical Role did learn from....Because based on some of the comments made by various posters at Twitch and Youtube and Reddit, some absolutely did not, basically trying to bully him into "bringing Molly back."

I think I was never that invested in a particular character, even my name sake, I was into the game. Roll up a new PC and get payback on that dungeon...and DM. Then again I'm deeply hacky and slashy.

Flexor the Mighty! said:

I think I was never that invested in a particular character, even my name sake, I was into the game. Roll up a new PC and get payback on that dungeon...and DM. Then again I'm deeply hacky and slashy.

Click to expand...

I think it was more about the setting (and exploring it) than about the PC originally. If you enjoyed the world / game, you wanted to get back into it. I think the difference between OS and NS is the focus on setting vs. (player) character. Currently, I think most games fall into the center somewhere; you have to enjoy the setting / game world / story (and game system) and you have some level of investment in the PC. If you are heavily invested in the setting you are more accepting of PC death. If the main reason you're playing is character development, you are less accepting of PC death. My 2 cents, and, of course, all imho.

I find that the lethality of older editions is often vastly overstated. Yes, characters could die, and it could be fairly easy to kill them off....a first level Magic User could cast Magic Missile once per day and then started throwing darts at monsters, after all.

Those first few levels were dangerous. Then you got into Raise Dead territory, and a huge amount of risk vanished from the game.

Today’s D&D is largely the same. Although PCs are overall more durable, they’re still mostly vulnerable for the first few levels, and then the Raise Dead spells start. I think the game is slightly less deadly mostly as a side effect of making it less boring at lower levels. I never once played a Magic User in 1E....it just didn’t seem fun for the first couple of levels. The modern iterations have taken that into consideration. As a result, the game is more fun at lower levels. And also a bit less deadly.

I’d also argue that losing a character that you didn’t really grow attached to because the game was so lethal isn't really something I’d consider risky. If you simply crumpled up the sheet and wrote up a new one, then who cares if you failed or not?

But so far, it’s all about D&D. What about other games?

Blades in the Dark springs to my mind. It’s a storytelling game, but there is great risk to the characters, and they fail pretty often. They also suffer pretty strong consequences, which are (ultimately) unavoidable. Eventually, your character will accumulate enough Trauma that they simply cannot keep adventuring. They have to retire.

The game gives players great influence over if/when their characters are harmed, and also how much. That’s probably something that most “old school” GMs (as described in the article) would not be okay with. But the game is that dangerous to the characters that even when a player has mechanics that allow them to resist or avoid harm, eventually they’ll die or be forced out of action.

So in that sense, maybe what Lew is really trying to get at is the “player focused” aspect of his old school games, versus the “character focused” aspect of many modern games? Because I really don’t think that lethality is as big a deal as stated, and chance of failure I don’t see as a factor at all.

The conflation between 'danger' and 'failure' in the article is, I suspect, no accident at all.

In OS games, danger leads to a risk of failure and failure leads to a risk of danger - they're very often interwoven. There's also no brakes or training wheels put on either one: failure can be just that - a complete dead stop where the only option for the PCs is to completely abandon what's being done and go do something else e.g. an entirely different adventure or mission; and danger can and does include TPK and much more often includes loss of a significant portion of the party - along with magic items, levels, major possessions, and so forth.

In NS games, danger is danger and failure is failure, and while one can still lead to the other it's not as clear-cut as with OS. And, in general, the brakes are on: the (IMO horrible!) concept of fail-forward means actual failure - as in a hard 'no' - is quite rare, and the usual options become 'succeed' or 'succeed with complications'; while danger - both in types and severity - has been very much mitigated. There's no more level loss, little if any magic item loss or destruction, and PC death is both rarer and easier to recover from. TPKs are also much less common, with the one bizarre exception of 4e D&D where in touch-and-go situations it seemed parties either all survived or all died.

So, where the article references either failure or danger it could just as easily be referring to both at once.

For my own part, as both DM and player I don't mind the concept of hard-no failure; if for no other reason than it's realistic: not everything you try will succeed. And if it's the scale of failure of a sort that means the party I play in has to find another adventure, so be it; and if this happens when I'm DM - well, that's why I've always got another adventure or two in reserve.

As for OS-level danger - as long as everyone is vaguely equal in their willingness to engage with it and take some risks and occasionally just be reckless, I'm fine with it. But I've come to loathe adventuring with cowards.

hawkeyefan said:

I find that the lethality of older editions is often vastly overstated. Yes, characters could die, and it could be fairly easy to kill them off....a first level Magic User could cast Magic Missile once per day and then started throwing darts at monsters, after all.

Those first few levels were dangerous. Then you got into Raise Dead territory, and a huge amount of risk vanished from the game.

Click to expand...

Not all the risk vanished, though: you still had to make that resurrection survival roll and even if you did you still came back down a point of constitution - and note that both of these drawbacks to death have disappeared from the game over the years, further supporting the point the article is trying to make. Raise Dead cost a lot too, and Resurrection even more.

And Raise in the field took having a 9th-level Cleric in the party, which is pretty late-game for 1e.

Today’s D&D is largely the same. Although PCs are overall more durable, they’re still mostly vulnerable for the first few levels, and then the Raise Dead spells start.

Click to expand...

At automatic chance of success and at much lower cost.

Blades in the Dark springs to my mind. It’s a storytelling game, but there is great risk to the characters, and they fail pretty often. They also suffer pretty strong consequences, which are (ultimately) unavoidable. Eventually, your character will accumulate enough Trauma that they simply cannot keep adventuring. They have to retire.

Click to expand...

So, Trauma here replaces Sanity in CoC?

The game gives players great influence over if/when their characters are harmed, and also how much. That’s probably something that most “old school” GMs (as described in the article) would not be okay with.

Click to expand...

Damn right we wouldn't! If I ever get to the point of having to say "Joe, can I please have permission to kill off your character?" then I'm doing it wrong on a host of levels right from square one - and Joe's probably in the wrong game, too.

But the game is that dangerous to the characters that even when a player has mechanics that allow them to resist or avoid harm, eventually they’ll die or be forced out of action.

So in that sense, maybe what Lew is really trying to get at is the “player focused” aspect of his old school games, versus the “character focused” aspect of many modern games? Because I really don’t think that lethality is as big a deal as stated

Click to expand...

Yet as you yourself note, there's much more to danger than simple death. Level loss is gone. Magic item destruction is gone, or close enough. Limb loss is either gone or close to gone. Most save-or-dies are gone, along with many save-or-hosed; and those save-or-hosed that remain have been greatly mitigated in duration and-or severity.

There's just no denying that dangers in most NS games are less than in OS games.

and chance of failure I don’t see as a factor at all.

Click to expand...

Chance of failure isn't a factor - 50-50 odds, for example, are the same in any system.

But what's become different is what a failure represents, and how the roll is interpreted.

What it represents:

In OS, a failure usually means 'no, you can't do it', whatever it was you were trying to do. You don't climb the wall. You don't find the secret door. You don't talk your way past the guards. You don't find the princess before her kidnappers kill her off. (as a side note, this is what 'failure' means - the opposite of success)

In NS, a failure quite often in fact means success (and is thus the wrong term to use) but with a complication. You climb the wall but there's a guard at the top waiting for you. You find the secret door because a monster comes out of it. You talk your way past the guards but one is suspicious and runs to inform her boss. You find the princess but she's been horribly disfigured. (note: none of these are actual failures!)

How the roll is interpreted:

In OS, in cases where success or failure is not necessarily a binary condition, many DMs would interpret rolls such that a narrowly-made success might end up with what we'd now call a fail-forward. For example, if you're trying to climb a wall and need to roll 11 to make it, on 10 or less you fail outright but on 11 or 12 - a narrow success - a DM might say you made it but there's a problem - you were too noisy, or you took longer than expected, or whatever. Failure was never mitigated, but success sometimes was.

In NS it's the reverse: success is never mitigated, but failure often is.

Interesting debate.

For me, the divide between new and classic design isn't even about lethality, though I recognize that angle, from the standpoint of D&D.

No, the main thing for me, is how much narrative power players get outside the immediate actions of their characters? How much the rules are about "telling a story", or presenting a world?

In that sense, "new school" games actually degrade my immersion. I prefer the game's world working on it's own inner logic and consistency, rather than the whims of the players. Also, if I made the wrong choices for my character, it suffers, or dies, the plan misfires, etc. I actually get to have more interest in the character like that, because I feel I'm playing a living person in a living world, not an artificial character in a story.

Falling forward is interesting. I recognize the benefits of giving incentive to the players to actually play out their character's flaws, something that often get neglected in classic-style rpgs. However, it could go into places I don't prefer, making the characters charicatures of themselves, or players getting the idea of not succeeding being "better".

My feeling on Old School gaming is that it was not really "difficult" it was more "Nintendo difficult". If you step on the wrong place then you die. If you roll too low then you die.

It was more of a test of the Player then the Character.

Derren said:

And every challenge which arose through failure can also be failed forward so that in the end real failure is not possible.

Click to expand...

Only if you completely ignore the part where I said "the GM reworks failures that would stop a game cold".

During non-story-stopping failures, it's just good old normal failure. You fail. That's for playing, you don't even get a consolation prize.

For something that stops the game cold, there is still a heavy penalty for failure, it's just that the failure hits the characters in another way. Maybe it takes too long and the princess IS sacrificed to the summon the demon, and now the story is dealing that that. Still a failure, just one that will move a story forward in an interesting way.

So let's try this again. Failure STILL EQUALS failure. Always. Just that if it would end up being a boring failure that stops play, it's turned into an interesting failure that continues. That failure might even be worse, it's not coddling. It's just got to be interesting.

Derren said:

Not to mention that if the failure results in a combat encounter its very often a bonus as many systems like D&D give out XP and loot for combat.

Click to expand...

True, many systems still use the old school paradigm of combat for XP system. Though it's an interesting take on risk that your default position that a combat encounter only ends up in XP and loot. That may be where your idea of real failure may not be possible is coming from - DMs who are coddling. And that happens in old school and new school games.

Lanefan said:

Not all the risk vanished, though: you still had to make that resurrection survival roll and even if you did you still came back down a point of constitution - and note that both of these drawbacks to death have disappeared from the game over the years, further supporting the point the article is trying to make. Raise Dead cost a lot too, and Resurrection even more.

And Raise in the field took having a 9th-level Cleric in the party, which is pretty late-game for 1e.

At automatic chance of success and at much lower cost.

Click to expand...

I agree that AD&D is a deadlier game. I just think how much so tends to be overstated.

And I think what’s often overlooked is investment in character. As I mentioned in my post, and which you chose to not quote...if you don’t invest in your character, then who cares if he dies? I had a buddy who had Drexel the Fighter, who met a grisly fate. Then there was Drexel II, Drexel III, and so on. Eventually they were replaced with Lexerd the Fighter.

The game doesn’t really end, right? Is it really all that different for one player to make three consecutive characters that all die and another to play one character that has a couple of close calls?

And ultimately a lot of this comes down to optional rules and what’s implemented, right? A lot of the things you mention are options. I mean, critical hits weren’t an official thing until 3E right? And as much as they’re fun, they’re far more detrimental to PCs than to monsters. 5E has a ton of optional rules to make it more deadly.

So I don’t think it’s a matter of the system. At least not with D&D.

Lanefan said:

So, Trauma here replaces Sanity in CoC?

Click to expand...

Not really, no. There’s a mechanic called Stress that each PC has. It’s a rrsource that allows them to resist harm or push themselves and their abilities. However, if they accumulate 8 Stress, then they’re out of play, knocked out or senseless or whatever. When this happens, the PC gains a Trauma, which is a permanent effect. Once you have four traumas, the PC’s career is over.

Lanefan said:

Damn right we wouldn't! If I ever get to the point of having to say "Joe, can I please have permission to kill off your character?" then I'm doing it wrong on a host of levels right from square one - and Joe's probably in the wrong game, too.

Click to expand...

I wouldn’t say so, no. Therr’s more than one way to play, especially once you consider other games beyond D&D. If the game is designed to give the players some narrative power (like Stress in Blades in the Dark) then you’d certainly not be doing it wrong.

Lanefan said:

Yet as you yourself note, there's much more to danger than simple death. Level loss is gone. Magic item destruction is gone, or close enough. Limb loss is either gone or close to gone. Most save-or-dies are gone, along with many save-or-hosed; and those save-or-hosed that remain have been greatly mitigated in duration and-or severity.

There's just no denying that dangers in most NS games are less than in OS games.

Click to expand...

Yes, there is. I am denying it.

How can we quantify and compare something like 1 point of Constitution to a more narrative element like failing to get justice for a PC’s murdered brother? Which is “worse”? You seem to be focused entirely on mechanics....failing a skill check, losing a level, and so on. But what about the story elements where a PC can fail? Or how they can be harmed in ways that don’t adjust their character sheet?

Again, limiting the discussion to editions of D&D then I would say that the older the edition, the deadlier the game. The size of the delta is debatable; the game has always included ways to come back from the dead, and optional rules to make it harder or easier, so i think many folks site it as sooo much deadlier when it’s really only moderately so. But that’s a matter of opinion.

But since we’re not solely talking D&D but instead seem to be talking about “classic” versus “modern” game design, I think the premise is simply wrong.

Lanefan said:

Chance of failure isn't a factor - 50-50 odds, for example, are the same in any system.

But what's become different is what a failure represents, and how the roll is interpreted.

What it represents:

In OS, a failure usually means 'no, you can't do it', whatever it was you were trying to do. You don't climb the wall. You don't find the secret door. You don't talk your way past the guards. You don't find the princess before her kidnappers kill her off. (as a side note, this is what 'failure' means - the opposite of success)

In NS, a failure quite often in fact means success (and is thus the wrong term to use) but with a complication. You climb the wall but there's a guard at the top waiting for you. You find the secret door because a monster comes out of it. You talk your way past the guards but one is suspicious and runs to inform her boss. You find the princess but she's been horribly disfigured. (note: none of these are actual failures!)

How the roll is interpreted:

In OS, in cases where success or failure is not necessarily a binary condition, many DMs would interpret rolls such that a narrowly-made success might end up with what we'd now call a fail-forward. For example, if you're trying to climb a wall and need to roll 11 to make it, on 10 or less you fail outright but on 11 or 12 - a narrow success - a DM might say you made it but there's a problem - you were too noisy, or you took longer than expected, or whatever. Failure was never mitigated, but success sometimes was.

In NS it's the reverse: success is never mitigated, but failure often is.

Click to expand...

I’ll have to ask you to be specific; what game are you describing?

Again, to site Blades in the Dark, players roll D6s when they attempt an Action. On a 1-3, they fail. Often horribly so. On a 4-5 they succeed but with a setback or complication. On a 6 they fully succeed.

The better the PC is at something, or the more willing they are to push themselves and take some Stress, the more dice they roll. So their chances to succeed go up. But there is still plenty of chance for failure.

The other narrative games I’m familiar with like Dungeon World or City of Mist also allow for failure. Roll 2D6 and add a bonus, 6 or lower and you fail, 7-9 and you succeed with a complication, 10 or more and you fully succeed.

I’m not sure what game you’re siting that simply doesn’t allow for failure; can you be specific?

Also, to use one of your examples....”you climb the wall, but there’s a guard”; you describe this as not a failure. Why not? What’s the character actually trying to do? Simply climb a wall? Or climb a wall so they can infiltrate a location unnoticed? Probably the latter, in which case, a guard being there means they failed, or that they can possibly fail. So again, I don’t think that there is no failure in new school games simply because they function differently.