Do you ever set up your PCs to fail?

[Trasilor]

In most story arcs, 'Team Opposition' need to get a win. Stories in which the good guys always triumph over evil, tend to become boring or ends up feeling like the DM is using the 'kid gloves'.

So how do other DMs set up the PCs to fail? Or, set it up in such a way in which the bad guy wins? (without the whole "the PCs were working for the BBEG the whole time" trope)

Look for stories / suggestions that I can use.

I have a large party - casual players, not very optimized, but will surprise me with creative problem solving - level 5.

[Rhyltran]

In most story arcs, 'Team Opposition' need to get a win. Stories in which the good guys always triumph over evil, tend to become boring or ends up feeling like the DM is using the 'kid gloves'.

So how do other DMs set up the PCs to fail? Or, set it up in such a way in which the bad guy wins? (without the whole "the PCs were working for the BBEG the whole time" trope)

Look for stories / suggestions that I can use.

I have a large party - casual players, not very optimized, but will surprise me with creative problem solving - level 5.

Option 1: There's several ways. Have the pcs believe something but it turns into a trap. The BBEG tricks them into positioning themselves where they are surrounded and completely outgunned.

The counter problem? Your PC's may think they can win and may try to fight their way out when it's just not possible resulting in their deaths.

Option 2: You have them encounter the BBEG too early and he might be too powerful. Maybe his champions. The group ends up defeated but the enemies might have motivations/orders not to kill them yet.

This scenario can also go badly because no matter how powerful the enemy is as long as it has a stat block the players have a chance to come up with a way to kill it.

Option 3: The players find out where the enemies are hiding but find when they get there the place is mostly vacant. Upon fighting their way to the BBEG they realize only one of his lieutenants is left behind and that the BBEG had taken his real threats to go deal with the kingdom/whatever you're protecting. Resulting in the death of some favorable npcs or the ransacking of some important town.

This doesn't require a party's defeat but does let them feel it was a loss. It doesn't make it entirely pointless either because the death of the lieutenant could be a bittersweet victory. The problem with this method is the players may feel cheated if they did a proper investigation and felt stuff was purposely left out or they may not even take the bait.

Every method comes with it's own flaws and problems but that's true of anything but if handled carefully it can be done. We kind of frustrated our DM in our last campaign. Even fights that we weren't meant to win we won. This wasn't the DM pulling punches (he's wiped parties before based on their choices. He believes the death of the heroes can be a good story just as their victory) but we really crushed his threats. Despite this there were several very memorable encounters. Sometimes the heroes winning the whole way doesn't end up boring and it can be fun. Sometimes unless you say "Rocks fall everyone dies/loses" the party WILL come up with ways to win.

[Mendicant]

I don't generally set my players up to fail. Protagonists that never lose can be boring as a story, but it's different as a game. A lot of the useful things protagonist failure does for you, like broadcasting how dangerous the opponent is, or contributing to a character arc, are accomplished better in other ways when there are multiple authors and no audience. For instance, if they don't know already, my players will learn how dangerous that goon is the first time it power attacks somebody for a whole bunch of HP. I don't need to have them lose in order to create a moment of dramatic tension.

Now, that doesn't mean that enemies can't win sometimes. If you give them resources and play them smart, they'll beat the PCs occasionally even with all the party's advantages. When that happens, try to avoid a tpk but do create some kind of failure consequence that leads to something interesting, like an escape or a tense negotiation, or a new quest.

Second, you can guarantee the bad guys win sometimes through the medium of other NPCs. Just because the players win all or most of their fights doesn't mean their allies do. Take a bit of time doing worldbuilding at the table and show them a set of NPCs with a particularly fearsome reputation--a knightly order or a college of wizards or the evil empire, whattever. This org could be a background for one character, or maybe a semi-regular quest giver or even an antagonist. When you want to introduce the new baddy, have the players come across a wiped out team from that group, or maybe visit the headquarters looking for advice or assistance and find nothing but a blasted shell.

One way to play with "failure" is to have the PCs fight their way out of an overwhelming attack on a plot important site. This only works with players who are willing to retreat, but if you have that sort of group, it can be a fun way to make a bad guy seem powerful and dangerous while also making the PCs hate his guts.

[Crake]

In most story arcs, 'Team Opposition' need to get a win. Stories in which the good guys always triumph over evil, tend to become boring or ends up feeling like the DM is using the 'kid gloves'.

So how do other DMs set up the PCs to fail? Or, set it up in such a way in which the bad guy wins? (without the whole "the PCs were working for the BBEG the whole time" trope)

Look for stories / suggestions that I can use.

I have a large party - casual players, not very optimized, but will surprise me with creative problem solving - level 5.

I once had a player that followed a particular demon lord (a demon lord of seduction called Nephthys) who was quite out of favour with the rest of hell (custom setting, there's just 1 hell, no 9 hells, abyss, gray wastes etc). He hated his father, and ran away from home a long time ago, but it just so happened that he was back in the same city his father lived in. He found out his father was a city guard, so he gathered all his fathers friends and murdered them all in front of him, then, as he was laughing at his father struggling to escape his bindings, his father managed to actually break free and stab him with a cold iron dagger (the player was partway along transforming into a demon and had DR/cold iron, so that was a bit of a shock). He then started chanting, a ritual, I cant remember the exact words anymore, i might have them written down somewhere, but something like "by the blood of the four, and the blood of the one, may the decieved before me become undone!". The four being his four friends, the one being the player, and then he promptly stabbed his own chest, and out from the wound came two arms clawing themselves out of his body, shortly followed by the demon lord Pithius who had banished Nephthys to the dredges of hell.

Needless to say, the player was terrified, and fled with his proverbial tail between his legs. He later found out that his parents had been worshippers of Pithius, but his mother had rebelled when she found out her son would be condemned to hell for their worship of Pithius, and, with no other option available, dedicated her son's soul to Nephthys, both as a spit in Pithius' face, and because she knew Nephthys, lacking any followers whatsoever, would hopefully treat her child well (which she certainly did). His father imprisoned his mother for her betrayal, and continued his worship to Pithius, until that day, where he sacrificed himself, using his own blood, and the blood of his 4 unwitting friends and his long lost son to summon Pithius into this world. It was even more of a terrible loss for the worshipper of Nephthys, because they had just established a chain of cults disguised as brothels, and the summoning had taken place in the basement of one of those brothels, so not only did Pithius burn down the place, but it also forever ruined it's reputation, causing them to have to rebuild from scratch under a different name.

[Kurald Galain]

So how do other DMs set up the PCs to fail?
I don't. The PCs do that for me.

But a nice way of pulling this is having two evil NPCs that are fighting, and suggest the PCs to pick sides. No matter which they pick, they'll (likely) be victorious over one of the evils while strengthening the other.

[Geddy2112]

I never set up PC's to fail, but I routinely challenge them and set up difficult challenges and situations that cause setbacks, etc. I know they are very likely to overcome them and eventually come out ahead, but the plot needs some rising action now and again.
The players always have a choice on how to deal with these things, but these challenges are set up to have multiple solutions with pros and cons to each solution(including the ones I don't think of and my clever party figures out).



This doesn't require a party's defeat but does let them feel it was a loss.

I really like this mentality. PC's don't have to be mortally wounded, expend finite resources or epically fail to have a loss or setback. Maybe the boss fight got ugly and some of the treasure was destroyed? Maybe the actions here had consequences down the road with future NPC's. Maybe some NPC's died, or favor changed, or because of some PC action power has shifted to a new ruling family. If your PC's are invested in the game, they can lose things that are not on their character sheet and should mourn those losses more than a broken wand or ripped bag of holding.

you can guarantee the bad guys win sometimes through the medium of other NPCs. Just because the players win all or most of their fights doesn't mean their allies do. Take a bit of time doing worldbuilding at the table and show them a set of NPCs with a particularly fearsome reputation--a knightly order or a college of wizards or the evil empire, whattever. This org could be a background for one character, or maybe a semi-regular quest giver or even an antagonist. When you want to introduce the new baddy, have the players come across a wiped out team from that group, or maybe visit the headquarters looking for advice or assistance and find nothing but a blasted shell.

Your PC's are not the only people in the world. I always have multiple factions at play in any setting I make, and many of them are not opposed or even associated. Their goals and actions are independent, and their actions change the setting even if the PC's are not there. Each town, dungeon, forest etc is not just a setpiece of creatures and NPC's that suddenly comes to life when the party shows up. These things are in motion just as the party is.

[Uncle Pine]

Words of advice: do not rely on your players fleeing from a fight. Players never flee from fights. You could be lucky and have a single savvy player that would like to run away, but everyone else is going to make him shut up and charge to their death. You don't want that to happen.

That said, if Team Opposition needs a win for any reason - something I don't usually plan in advance and build my story on, but let's do it anyway - you necessarily don't have to set up an encounter in which the party loses. If you manage to direct the party's attention to Location A (i.e. sending a couple dozens of orcs to set some villages on fire) or you know that they'll leave for Location B in two days and that they'll spend 3 weeks there (i.e. scrying on them until you get the required informations), they won't be around in Location C to stop Team Opposition's plan.

Also, see the quote in my sig.

[Rhyltran]

Words of advice: do not rely on your players fleeing from a fight. Players never flee from fights. You could be lucky and have a single savvy player that would like to run away, but everyone else is going to make him shut up and charge to their death. You don't want that to happen.

That said, if Team Opposition needs a win for any reason - something I don't usually plan in advance and build my story on, but let's do it anyway - you necessarily don't have to set up an encounter in which the party loses. If you manage to direct the party's attention to Location A (i.e. sending a couple dozens of orcs to set some villages on fire) or you know that they'll leave for Location B in two days and that they'll spend 3 weeks there (i.e. scrying on them until you get the required informations), they won't be around in Location C to stop Team Opposition's plan.

Also, see the quote in my sig.

I completely agree with this and it's what I was trying to say in my final paragraph. You shouldn't try to force the players to fail because it doesn't work. Let the big bad play as he would. You don't have to pull punches or make it super lethal for the party in question just make the BBEG a tactical or even strategic minded opponent. If they kill him great! Well done! But he may have set in motion some events that will persist after his death or as I mentioned in one of my options strikes a blow at the players while they're dealing with him or one of his key set pieces. As long as the villain isn't mindless he probably has his own plans/setup and if the player is unaware of everything he's doing it's very possible to have him strike at the party from an angle they don't foresee.

As the person I quoted said and as Red Fel mentioned in the sig never ever "expect" the party to make a certain move. If you expect players to do something they will always do the opposite. As long as the players aren't actually bored there isn't really a problem. As mentioned, I was player in a campaign where we won every step of the way (some were very hard fought) but we felt satisfied through the whole thing. It was a blast and we still talk about that campaign to this day.

[Trasilor]

Option 1: There's several ways. Have the pcs believe something but it turns into a trap. The BBEG tricks them into positioning themselves where they are surrounded and completely outgunned.

The counter problem? Your PC's may think they can win and may try to fight their way out when it's just not possible resulting in their deaths.

Option 2: You have them encounter the BBEG too early and he might be too powerful. Maybe his champions. The group ends up defeated but the enemies might have motivations/orders not to kill them yet.

This scenario can also go badly because no matter how powerful the enemy is as long as it has a stat block the players have a chance to come up with a way to kill it.

Option 3: The players find out where the enemies are hiding but find when they get there the place is mostly vacant. Upon fighting their way to the BBEG they realize only one of his lieutenants is left behind and that the BBEG had taken his real threats to go deal with the kingdom/whatever you're protecting. Resulting in the death of some favorable npcs or the ransacking of some important town.

This doesn't require a party's defeat but does let them feel it was a loss. It doesn't make it entirely pointless either because the death of the lieutenant could be a bittersweet victory. The problem with this method is the players may feel cheated if they did a proper investigation and felt stuff was purposely left out or they may not even take the bait.

Every method comes with it's own flaws and problems but that's true of anything but if handled carefully it can be done. We kind of frustrated our DM in our last campaign. Even fights that we weren't meant to win we won. This wasn't the DM pulling punches (he's wiped parties before based on their choices. He believes the death of the heroes can be a good story just as their victory) but we really crushed his threats. Despite this there were several very memorable encounters. Sometimes the heroes winning the whole way doesn't end up boring and it can be fun. Sometimes unless you say "Rocks fall everyone dies/loses" the party WILL come up with ways to win.

I have employed variations on options one and two with some success. Currently the PCs have something the BBEG wants, so he is reluctant to kill them without getting the item first.

I may employ something like option 3...


I don't generally set my players up to fail. Protagonists that never lose can be boring as a story, but it's different as a game. A lot of the useful things protagonist failure does for you, like broadcasting how dangerous the opponent is, or contributing to a character arc, are accomplished better in other ways when there are multiple authors and no audience. For instance, if they don't know already, my players will learn how dangerous that goon is the first time it power attacks somebody for a whole bunch of HP. I don't need to have them lose in order to create a moment of dramatic tension.

Now, that doesn't mean that enemies can't win sometimes. If you give them resources and play them smart, they'll beat the PCs occasionally even with all the party's advantages. When that happens, try to avoid a tpk but do create some kind of failure consequence that leads to something interesting, like an escape or a tense negotiation, or a new quest.

Second, you can guarantee the bad guys win sometimes through the medium of other NPCs. Just because the players win all or most of their fights doesn't mean their allies do. Take a bit of time doing worldbuilding at the table and show them a set of NPCs with a particularly fearsome reputation--a knightly order or a college of wizards or the evil empire, whattever. This org could be a background for one character, or maybe a semi-regular quest giver or even an antagonist. When you want to introduce the new baddy, have the players come across a wiped out team from that group, or maybe visit the headquarters looking for advice or assistance and find nothing but a blasted shell.

One way to play with "failure" is to have the PCs fight their way out of an overwhelming attack on a plot important site. This only works with players who are willing to retreat, but if you have that sort of group, it can be a fun way to make a bad guy seem powerful and dangerous while also making the PCs hate his guts.

You are getting the crux of my problem - I am trying to avoid TPK - while maintains some verisimilitude.

Your suggestion of using other NPC organizations is definitely an interesting idea I might be able to employ.

I guess part of my problem is my players like to be railroaded. They are casual players and we play about once a month (families and jobs and whatnot) so they just want to show up and fun with friends (which we do). That means, protagonists have to be over the top. Subtle hints and delicate foreshadowing tends to be lost on the players (even with a campaign log back-up).

As it stands, my party has met one lieutenant - which they think is the BBEG.

[Rhyltran]

I have employed variations on options one and two with some success. Currently the PCs have something the BBEG wants, so he is reluctant to kill them without getting the item first.

I may employ something like option 3...



You are getting the crux of my problem - I am trying to avoid TPK - while maintains some verisimilitude.

Your suggestion of using other NPC organizations is definitely an interesting idea I might be able to employ.

I guess part of my problem is my players like to be railroaded. They are casual players and we play about once a month (families and jobs and whatnot) so they just want to show up and fun with friends (which we do). That means, protagonists have to be over the top. Subtle hints and delicate foreshadowing tends to be lost on the players (even with a campaign log back-up).

As it stands, my party has met one lieutenant - which they think is the BBEG.

Is the item on them or is it hidden? If it's hidden my option three idea might be perfect. They face the lieutenant only to realize that's all he was. In the meantime, the BBEG could either have a very powerful wizard in his services or be one himself. He then uses divination to find the location of the item that he wants. While the party is wasting their time on what they believe to be the BBEG the enemy gets the item that they have.

If it's important enough that the BBEG wants it and the players are keeping him from it that would be a devastating blow for the players and reveal the BBEG to be competent and resourceful. Of course, this assumes the players don't have the item in question on them and if it's as important as I'm guessing it to be. Just make sure if you go this route the BBEG should discover the item in question through legitimate means. That way the party finds it clever.

It would also teach them a lesson that thinking they can just hide something important to the enemy in order to protect themselves by using it as a bargaining chip in case of capture or threat of death won't work with this guy. He will find it through other means. This suddenly makes him a lot more dangerous. It will also mean they'll need to jump through more hoops to hide something similar or keep it on them.

Now if it's on them.. all is not lost. Have the enemy strike at the players from "behind" while they enter what they think will be the lair as described above. The enemy has captured some important npc and threatens them with this choice "Return the item or ___ dies." If the players refuse? Have the npc killed. If they comply? Let the NPC go. That way in the future they won't just attack under the same thing thinking "They'll just kill said npc anyway."

[Taelas]

You can set them up, but you should never set them up so their only choice is to fail utterly. There's nothing interesting in arbitrary failure.

The bad guy could threaten their loved ones or a group of innocents to distract them from stopping him. That's a classic.

[Rhyltran]

You can set them up, but you should never set them up so their only choice is to fail utterly. There's nothing interesting in arbitrary failure.

The bad guy could threaten their loved ones or a group of innocents to distract them from stopping him. That's a classic.

I second this. Especially since the group are casuals looking for some fun. Classic bad guy tropes work especially well for players like this.

[Trasilor]

You can set them up, but you should never set them up so their only choice is to fail utterly. There's nothing interesting in arbitrary failure.

The bad guy could threaten their loved ones or a group of innocents to distract them from stopping him. That's a classic.

Yes, absolutely true. And I would never do this. I can however make it such that:

a) PCs don't realize they a failed until later.
b) Its a Pyrrhic victory for the PCs
c) PCs win, but find they are working on the wrong side.

My hope is that the PCs will have a visceral hatred for my BBEG without resorting to murdering PC families or killing babies (not that those are off the table specifically :smallamused:).

[Trasilor]

I second this. Especially since the group are casuals looking for some fun. Classic bad guy tropes work especially well for players like this.

You know the more I think about it, the more i should just go to this route - find every classic bag guy tropes and incorporate them with abandonment.

might be time for a new thread :smallbiggrin:

[Flickerdart]

Words of advice: do not rely on your players fleeing from a fight. Players never flee from fights. You could be lucky and have a single savvy player that would like to run away, but everyone else is going to make him shut up and charge to their death. You don't want that to happen.
PCs never flee because the game is rubbish at enabling it. Monsters are typically faster than PCs, and turns are so chunky that by the time you realize the fight has gone sour, it's too late to do anything about it. Maybe the wizard can teleport out, but can he reach the rest of the party to take them along? Probably not.

Similarly, defeat of the non-TPK variety is not really supported, so the DM has to do workarounds with NPCs and treasure and stuff.

[ComaVision]

One thing that I didn't see mentioned in this thread: Time pressures. Make the time requirements challenging, and the requirement can be implicit or explicit (probably explicit if your party likes railroading). Make the group consider every night they spend regaining spells, and every side quest they do. Make them decide between going out of their way to save a town under siege or continuing on to their main goal. The difference that failure makes should be obvious, whether the BBEG and crew take over another town, get a mcguffan, or manage to open a portal for fiendish reinforcements.

[Trasilor]

PCs never flee because the game is rubbish at enabling it. Monsters are typically faster than PCs, and turns are so chunky that by the time you realize the fight has gone sour, it's too late to do anything about it. Maybe the wizard can teleport out, but can he reach the rest of the party to take them along? Probably not.

Similarly, defeat of the non-TPK variety is not really supported, so the DM has to do workarounds with NPCs and treasure and stuff.

This has been my experience as well. If PCs wish to flee, it must be done at the beginning of combat, not the end.

[rockdeworld]

To answer the OP, yes, all the time. That's part of a heavy-roll-playing campaign, where the gm's job is to give reasonable challenges. Unless you mean like "absurdly-high-cr", and then no. That's for cutscenes and suicidal/skilled players only.
To answer this:


So how do other DMs set up the PCs to fail? Or, set it up in such a way in which the bad guy wins?
-give the group 4 equal-cr encounters per day. Make resting between encounters hard to come by.
-pick more from monsters with SLA's, or otherwise absurdly good (how is your group on magic items? If low, an Allip. If high, Rust Monster, Gelatenous Cube, or some swarms that aren't in the MM are good), or the CR: Awesome monsters (dragons, be holders, etc)
-give the enemies better tactics/location/motivation than the MM expects them to have

[Uncle Pine]

This has been my experience as well. If PCs wish to flee, it must be done at the beginning of combat, not the end.

Or before the combat even starts. One of the few times I've been a player, even if we knew there was a boss ahead and I - the group's Barbarian - was stable at -9 and other two characters were injured rather badly they decided to leave me somewhere safe and go ahead anyway, even if the DM hinted heavily that it was a bad idea. Needless to say that the boss wiped the floor with them.

[SowZ]

I sometimes set up encounters based on prior events and I expect the villains to succeed in that encounter based on circumstances, but I will play by the rules and let them lose if the players can stop them. In such cases, I usually leave more blatant outs for the players to survive even if it means letting the town die/losing out on a quest/loot. To my surprise, the encounters I expect to be a win for team evil end up with the PCs overcoming by the skin of their teeth, and the deadliest encounters are more typical ones. This can be due to overconfidence, bad luck, or a PC enacting some scheme that ends up inadvertently, (or sometimes advertently,) getting the party in trouble. As far as unlucky dice making an otherwise trivial encounter difficult goes, it makes sense that some of the hardest encounters are supposed to be average difficulty as opposed to boss fights. After all, one streak of good rolls on behalf of the bad guys can totally change an encounter and there are a lot more normal encounters than super challenging ones. Meaning average encounters have a lot more chances to get those lucky streaks. It's just a numbers game.

Of course, when the bosses get their lucky streak, which will happen eventually, it can spell the death of the PCs if they aren't careful. I don't like TPKs, which is why I'll usually try and leave some sort of out if I really think the players are going to be in over their heads. I don't believe in fudging dice, however.

[ShaneMRoth]

I never set PCs up to fail in an encounter.

I rarely set my NPC antagonists up to fail, either.

I was running the Return to the Temple of Elemental Evil a while back.

I established a timeline of what would happen in the setting if the players negotiated the adventure with no sense of urgency.

The Big Bads had a plan that hinged on a ceremony taking place during Needfest (which is the closest thing in Greyhawk to Christmas... kinda...) which would cause the dormant volcano on which the the dungeon was sitting to re-activate and erupt. I got a sense of how many civilian lives would be lost, how much property would be destroyed, and what the general impact of this event would have on the political balance of the continent.

All of the impacts were bad.

But, furthermore, as far as the players were concerned, it would destroy all unclaimed treasure and experience points left in the dungeon.

I didn't tell the PCs about this plan or this timeline. It was up to them to stumble upon it.

I left the occasional hint. Like earthquakes would happen with increasing frequency in the area.

Not enough to have a direct impact on the characters, but enough to cause any savvy player to ask, "Why is the DM making a point of telling me about earthquakes in the area?"

They sometimes discovered written instructions from the Big Bads to the general population making it clear that something big was going down during Needfest. There were no details on what that thing was, but a savvy player would have guessed that it wasn't going to be Secret Santa.

At the end of that in-game year, a magically induced weather system (a stationary blizzard) formed over the caldera of the volcano and the immediate surrounding area.

The NPC patrons who tasked the PCs to look into this adventure in the first place would make an occasional appearance during the PC's downtime and ask for a report on how things are going. (Is there any actionable intelligence on the activity of the villains? Any big plans? Anything we might need to know about? Nothing? Really? Nothing at all?) As time passed in-game, the NPC patrons would point out the earthquakes and magic blizzard are evidence that things are going in the wrong direction and maybe the PCs should pick up the pace.

It fell to the players to respond to all of the information I provided to decide to take enough of a break from the Murderhobo Shuffle to figure out what all of these monsters were really doing and why.

And what did the players do?

They treated the dungeon like it was a set piece in a computer game.

They acted like they had forever and a day.

They took their damn sweet time murderhoboing their way through every square inch of the dungeon.

They 'exploited' the Take 20 rule to painstakingly search every area they explored.

I actually made a rough estimate of how many 5' squares were to be searched in that dungeon. It would have taken thousands of murderhobo-hours to perform a Take 20 search on every inch of the place.

Even better, the players would take long breaks from the dungeon crawl to let the wizard craft magic items. This alone ate up weeks, going on months, of in-game time.

The volcano erupted, the dungeon was destroyed.

The players weren't in the dungeon when it went up.

The PCs never lost a battle, but they lost the war.

The NPC patrons gave them a very sad Nice Job Breaking It, Hero (http://tvtropes.org/pmwiki/pmwiki.php/Main/NiceJobBreakingItHero) speech followed by a pat on the back and a "Forget it, Jake... it's Chinatown."

The Forces of Evil were one step closer to freeing Tharizdun (Think Zuel from Ghostbusters... kinda...) and tens of thousands of people are dead, but hey... that's some pretty sweet armor you scored there, dude.

Then I prepared their next adventure.

[Brookshw]

Oh absolutely depending on how you mean fail.

I've put them up (indirectly) against hire level opponents they weren't capable of beating in a straight fight and they had to be clever in their approach to overcome. (and to dovetail on earlier discussions about pc's going the unexpected, in the instance that comes to mind I think they overcame the situation by helping the supposed opponent in accomplishing their goals).

I've had them be hired for a delivery job and handed a fake mcguffin to deliver (while they were actually being used as bait in a mole hunt).

I've had them spend a whole campaign building up alliances or a mass pitched battle and given them binary choices for certain parts of the battle where they had to choose which allies to sacrifice where all the allies they had and choices they made affected the campaigns epilogue.

My answer depends on how what fail means in the context.

[ZamielVanWeber]

I believe the goal would be to give them the opportunity to fail. Have it have consequences. The PCS should not be allowed to ride rough shod over the world without consequences, but make sure they have the opportunity to affect real change. It is alright to have variance in difficulty in challenges, even single (non-boss) unwinnable encounters (just be sure to justify the players escaping).

Shane has a great example of how to do it.

[QuickLyRaiNbow]

Put together situations that might be a trap and put the players under some time pressure. Then keep strict track of time, and tell your players you're doing so. Make them make a snap decision. If they delay too long then the ritual completes/the portal opens/the McGuffin McGuffins. If they rush in, surprise, it's a trap. Use this extremely sparingly.

Put situations together that are pretty lethal but make sure to leave the PCs an out so they can retreat. Retreating is a learned behaviour. You can't make your PCs do it without killing them off once or twice. After players have lost characters they'll start being more careful.

Give the PCs mutually exclusive choices. Do they save the little girl if it means BadGuy_01 is free to complete a terrible ritual? You can set these up so that, for instance, if the PCs choose to go stop the ritual, then the BBEG doesn't have access to demonic allies in the final battle, but the PCs don't have access to the kingdom's resources because the little girl was the king's niece. Combine this with things possibly being traps and making the players make snap decisions. Again, use sparingly.

[frogglesmash]

You could always try running it where the BBEG's apparent goals are just a cover to keep the PCs distracted from his true intentions Of course this kind of thing isn't really fair to the PCs unless you give opportunities to potentially learn about what's actually going on.

[PlatinumVixen]

I wouldn't say I set up the PCs to fail, but there are definitely situations that come pretty close. I usually try to leave things open for the PCs to come up with a solution that I didn't expect that resolves things in their favor though, which usually turns out way more rewarding if it works.

[jiriku]

I never do, but when I run games, the NPCs are capable, the encounters are hard, and consequences are severe. They fail often, enough so that when the victories come, they have the sweet savor of being well-earned. Note that because I'm not engineering situations to be automatic failures or successes, it's as much a surprise to me when they lose as it is to them. It does require a certain level of improvisation to keep the world moving in response to unexpected events.

[rockdeworld]

<snip>
I love this story. I want to hear the player's reactions too.

[Telok]

All the resources and knowledge needed to succeed at anything plot related are present in my setting/campaign from the very first play session. Failure comes from the players.

Insult the king. Drop the artifact in the portal to hell. Fireball the avalanche zone. Ignore the sage and his advice. Don't ask about the powerful reclusive witch. Avoid libraries and bards. Attack the demigod dragon. Fail to cover your tracks after raiding the enemy army camp. Leave the magic hammer on the dead body because it isn't blinged up.

Failure comes from the players.

[Mendicant]

This has been my experience as well. If PCs wish to flee, it must be done at the beginning of combat, not the end.

One way I've found to get around this problem, with varying success, is attacking in waves, and making it very, very clear that they're going to be attacked in waves. If there's a short 2 or 3 round breather that gives them plenty of room to decide to cut and run. You still have to have players who are actually willing to do that though. Some players just wont.