Daztur
Hero
When I posted (http://www.enworld.org/forum/new-ho...ombat-war-key-difference-d-d-play-styles.html) I was pretty skeptical about reconciling Combat as War and Combat as Sports in the same campaign, but some ideas in the comment thread lead me to re-evaluate that so let’s see what we can do to make both sides happy. What does each side want and how do we give it to them?
The Sliding Scale of Meaning vs. Balance
Unless a PC death is on the line every single fight, you cannot have both meaningful fights and balanced encounters. To use 4ed terminology, if each combat is followed by an extended rest, then the ONLY way a combat can have any influence on the characters is if a character dies, everything else gets reset. On the other hand, if there is no rest at all between combat encounters then setting up balanced encounters becomes impossible, as how hard the fifth combat is varies massively according to what happened in the first four combats (and varies even more if the party avoids some of the previous combats entirely). It is impossible for the DM to create a balanced fifth combat if the first four combats have effects on the characters that can’t be reset before the fifth combat, it is impossible for the first four combats to mean anything (aside from Character(s) Killed? Y/N) if their effects are reset before the fifth combat.
Of course there is a sliding scale between these two extremes but there is a direct trade-off, you can’t have both at once. So how much of each one do we want?
I believe that having meaningful encounters are more important than balanced encounters (this doesn’t mean that I think that balance isn’t very important, just that it shouldn’t be gauged at the encounter level). This is because there’s a lot of combat in D&D and if there’s a significant chance of characters getting killed in every single fight, then you’re going to have more character death than most people want (even in the hoariest of Old School D&D, lethality declines sharply after you gain a level or two). But if there’s NOT a meaningful chance of PCs getting killed each and every combat then each combat has to have lasting effects that don’t reset or every fight in which PC death is not seriously on the line is basically Harlem Globetrotters Combat as Sports: it looks like a basketball game, but everyone knows what the result will be and nothing is at stake, which is boring. So basically we have only a few choices: lots of dead PCs, Harlem Globetrotters CaS, combat that is much rarer than is standard in D&D or combat that has lasting effects. Combat having lasting effects seems the best option here.
To reiterate, combat having lasting effects subverts encounter balance and the more lasting and important these effects, are the more encounter balance gets undermined. This is because there’s a range of outcomes each combat can have and unless there’s something that returns the results to baseline (like an extended rest) each succeeding combat increases the variance until the difficulty level of the final combat becomes a crap shoot, no matter how you set it up.
The Glory of Attrition
So how to make combat meaningful, even though most combats will be stacked in the favor of the PCs (or at least in favor of the PCs being able to escape)? The only real answer is attrition. In most D&D combats (except in exceedingly lethal campaigns) the PCs will be able to defeat most combats without too much trouble (and/or avoid or escape from most difficult combats), therefore the only way to make most fights mean anything is through attrition. Even if you can kill the goblins without serious risk of PC death, they can still hurt you in a way that’ll make later encounters harder, making you have to care about the encounter with the goblins. This means that you have to use smart tactics against the goblins (or maybe avoid fighting them) even if you know that you’ll win in the end.
To make attrition-based play fun, you have to strip away most of the cushions that players have against attrition. Fistfuls of CLW wands have to go, so do scroll libraries, rope trick, easy teleportation, any form of healing that doesn’t use a healing surge (or other hard to replace source of healing) and the like. Combat also has to be much faster than in 3ed and 4ed to give the DM time to wear down the PCs (bringing back morale rules would be a good start). But the granddaddy of attrition killers is the 15 Minute Adventuring Day. This has to be killed dead for attrition-based play to work. There have been various suggestions offered that work to varying degrees, but I think that the simplest and easiest way to do this is to make getting what 4ed calls an extended rest take a week or so in a relaxing location (so in LotR terms, the hobbits don’t get a single extended rest all the way from the Shire to Rivendell). We can allow characters to recover some resources in the field, but not many and only at significant cost.
Then, in between these rare extended rests (which generally would be impossible get in the field) the characters can enjoy all of the glories of getting worn down by Oregon Trail D&D. Healing surges are a great way of modeling this (as long as they’re not potent enough to ward off attrition entirely). Run out of food? Lose healing surges due to starvation. Get dungeon lung? Healing surges. Exposure? Healing surges. In combat healing would cost healing surges, but downtime healing would give more HPs back per healing surge. This makes spending healing surges in combat a lot like using speed, it gives you energy in the short term but at a long term cost (which is a lot like how going berserk is described in the Icelandic Sagas). Town adventurers would use time as a constraint to prevent 15 Minute Adventuring Day issues (the Dresden stories are a good model here).
By playing up attrition you get rid of combat grind. Even if it’s clear that the PCs have won, being smart is combat is still important since getting hurt now can have a bigger effect on the next combat and the next and the next.
The Glory of Tactical Play
One of the biggest problems of TSR-D&D for me is that one-on-one Fighter duels are often stupefying boring. This isn’t a deal-breaker for me, but in order to get CaS players on board you have to fix that and similar issues. This means that mundane characters need at least the option of fun tactical options to play with and mundane combat can’t be as abstract as in Old School D&D without turning off a sizable portion of the player base.
The hard line to tread here is to be able to make combat tactical without making it take too long. I think that the way of doing this is to make sure that all combat requires tactical smarts, to make all combat situational and to give the players different tactical angles to work with.
Sid Meier’s definition of a game is a series of interesting choices. In all editions there’s stretches of D&D combat in which no real choices are being made (I remember one player who could’ve gone to the bathroom during most fights and left a sign saying “I take a five foot step away from the enemy and do rapid shot at the closest one” and his contribution to combat would’ve been unchanged). Focusing on attrition helps with this problem since even if the players can just grind down a combat in a boring way, they have a big incentive to be smarter than that, but some cues from 4ed design are needed here as well. What would also help a great deal is to give players tactical tools to escape from monsters that are too strong for them (to make fights of all difficulty levels fun) and to have fights in which the PCs have goals aside from “kill them all!” such as: escape! capture one and get away! grab the idol! don’t let them capture the wizard, they want his brain! kill them before they sound the alarm! etc. etc. which help keep things interesting on the tactical side.
In the previous thread one person noted that when players find a method that works they’ll do it over and over again, which takes all the fun out of it. I think that the solution for this is to make combat as situational as possible. That means making sure that the foes and the circumstances have a massive impact on which tactics work best. That means that any modifiers that aren’t big enough to change what kind of tactics the PCs use should be increased until they do or be thrown out entirely (die modifier bloat die!). I think that all editions have useful ideas that can be stolen for this from 4ed’s greater emphasis on terrain, to powers that suck in some circumstances and rock in others (I love Affect Normal Fires) to things like 1ed’s monsters that were completely resistant to non-magical items (“oh crap, I can’t scratch the damn gargoyle with my sword, we’ve got to change our game-plan entirely! We’ve got to find some way of immobilizing the damn thing and then braining it to death with our +1 shield”) what’s important is to not have the special features of monsters elicit a response of “Meh. We’ll just do what we always do; it’ll just take a bit longer.” On the other hand you don’t want to make the thief gimped against all undead, so it’d be better to keep tie this situational stuff to individual critters rather than broad classes (more unique quirky monsters are always good).
Having competing priorities in combat can also spice up the tactics a great deal. Sure you want the enemies to run out of HPs before you do, but there should also other things to keep track of. A lot of the tactical depth in TSR-D&D came from declaring your actions each round before you knew who was going first and I’d really like to see that and easily-interruptible casters brought back in a streamlined form. Morale rules also help here if implemented well since they give something for the players to think about in addition to hit points. Marks (in their most basic form) also are great and a lot of elements of the leader and controller roles need to be preserved in 5ed.
The Glory of Strategic Play
Too keep CaW players happy, strategic play has to be just as important as tactical play. That means that we have to walk the balance between making sure that what the players do outside of combat has a meaningful impact on what happens during combat, but not so much of an impact to make tactics pointless.
This means that devs have to look at mechanical widgets with strategic uses in mind by giving the reader a clear idea of what sort of things various powers can do outside of combat and making sure that these uses are clear. If what abilities (especially spells) can accomplish outside of combat isn’t laid out clearly then a lot of strategic play gets reduced to “does the DM decide that your plan works or not,” so these abilities also need to be carefully bounded or they can slice away whole chunks of gameplay (scry n’ die, etc. etc.). Good examples of the kinds of spells I’m talking about here are (skimming through the first level spells of my 1ed PHB):
Affect Normal Fires, Comprehend Languages, Dancing Lights, Enlarge, Jump, Push, Tenser’s Floating Disc, Unseen Servant and Ventriloquism. All of those spells cry out for creative strategic uses but they have clear descriptions of their effects that don’t rely very much on DM fiat (although Enlarge is clunky rules-wise as much as I love it). I’d like to see this kind of spell be the core of the 5ed Wizard class and have the Wizard be relatively weak at straight-up combat in order to compensate for its strategic utility (with more blasty casters being possible, just not the Wizard default).
Of course other classes need to shine in strategic play as well. Killing the 15 Minute Adventuring Day helps a great deal with this, as does removing spells that steal too much of other class’ thunder and making sure that high level character of other classes have plenty awesome out of combat abilities (see Beowulf et. al.).
In a similar vein, the devs have to write up the monsters with strategic play in mind. Write-ups of ecology and the like (like all of the random percentages of this and that) are useful for strategic play since they give the DM something to go on rather than throwing them back to fiat. Similarly, these needs to be plenty of monsters in which dealing with them through strategic means and never engaging them in a straight-up fight is smart (rust monsters being the classic example here).
Another benefit of strategic play is that is lets DMs uses much more of the Monster Manual at once. At low levels a hill giant is something that players should figure out smart ways of avoiding, after they gain a few levels they can take one down if they come up with a level strategic plan, then they become standard opponents and then at high levels a couple hill giants are still bad news if the party blunders into them while worn down.
One downside to putting an emphasis on strategic play is that it’s hard to pre-plan epic show-downs since it’s hard to hit on the right difficulty when things that the players do before the combat start can cause the difficulty of any one encounter to swing wildly one way or the other. One way of doing this is to let epic fights emerge out of the strategic game, in my experience nothing makes the players smile more after a hard fought battle than the DM saying “what?!? You took THAT on one-on-one and WON? I don’t believe it!” Letting the players find their own climactic final show-downs can be rewarding. For a more CaS-friendly option maybe a rule to let the PCs “dig deep” and get back resources lost to attrition for just one fight, but at severe mid-term cost so that it would only be used for climactic final fights of a session (or heroic last stands).
Balance
OK, now how the hell do you balance THAT? For adventure balance there’s two main methods. First, instead of looking at balance on an encounter level (since attrition-based play and strategic elements kill encounter-based balance) balance it at the adventure level. So if the PCs are going from Rivendell (extended rest available) to Beorn’s farm (extended rest available) through the Misty Mountains (no place to take an extended rest there) then balance is determined by the difficulty of the Misty Mountains taken as a whole, not according to each encounter or even a list of encounters. Maybe set up a 4ed-style adventurer builder to calibrate the difficulty instead of building things encounter by encounter.
The other way is to let players find their own risk/reward balance. The players go out from the town (extended rest available) into the howling wilderness and/or the deep dark dungeon (where, with perhaps a few exceptions, there is no place comfy enough to take an extended rest), and the longer they stay out the more attrition grinds them down and the more dangerous things get. But, if they go back to town to rest after every stiff fight they’ll never get deep enough into the dangerous area to get good loot and they’ll spend a lot of time tromping back and forth through already looted areas running into wandering monsters with no/very little gold. The players can choose to play it safe (low risk/low gold) or gamble (high risk/high gold), with a few helpful doors that lock behind the PCs, enchanted forests in which the trails move around, chutes, teleporters and whatnot to keep things interesting.
What about PC to PC balance? For 5ed, I really don’t think that encounter by encounter balance is the best way to go. First off, we’re not using balanced encounters anyway (see above) and trying to balance things encounter by encounter leads to the MacGyver Wizard Problem. Let me explain. Let’s say that you have a 4ed Wizard and a 4ed Fighter who are both equally useful in combat. Now the question is; does the DM let the Wizard MacGyver his attack powers to let them do useful things out of combat (or off-label things in combat)? If the answer is yes, then the Wizard becomes more powerful than the Fighter since it’s a lot easier to think of creative uses (especially creative uses out of combat) for Fireball than for, say, Sweeping Blow. But if, on the other hand, the DM or the rules are strict in building an impenetrable wall between combat magic (only good for killing things) and out of combat magic (no good at all for killing things) then, at least for me, that robs D&D magic of some of its most fun elements, since all of my favorite D&D spells have always had combat and out of combat applications.
So how do you balance character classes in 5ed then? I think that the best rule of thumb is to make it so that there is no 4-member party that is clearly superior to the traditional one fighter, one thief, one wizard and one cleric over the course of a standard adventure (a “standard adventure” is somewhat hazy, but I think that’s unavoidable). By this standard, 3ed fails badly. A party of two clerics, a druid and a wizard vastly outclass the traditional party at virtually any adventure that you can imagine. Other editions of D&D do better and I don’t think that this is an impossible standard.
There are a number of ways to meet this standard and probably the biggest question is what to do about the Fighter (which mostly boil down to whether a high level fighter looks more like Lancelot or like Hercules, I like the Lancelot option but there’s room for disagreement here) but this is my own personal stab at it:
Fighter: best at tactical combat. Best HPs, best Saves/Defenses/ACs, damn good damage dealing. You need the fighter when the PCs are ambushed and haven’t had time to come up with rat bastard plots and need to fight their way out and they’re the battering ram that makes all of the strategic plans have the oomph to work. Over the course of an adventure, the Fighter should have racked up more HPs in damage than any other class (and HP damage should actually matter). The fighter would have a secondary focus on skills (best bar the thief), leadership and social interaction so they have plenty of things to do outside of combat.
Wizard: best at strategic combat. While the Fighter is the best at playing the game, the Wizard is the best at tilting the playing field. They can do some spike damage, but it’s not their forte and their skills are fairly weak (too much time spent learning magic to learn other stuff) but their magic is good at changing the terms of engagement of fights.
Thief: best at just plain avoiding encounters altogether. The best fight is often the one that never happens (thieves know their Sun Tzu) and the thieves are the best at doing this through sneaking around things and talking their way past them. Their strong skills and stabbity nature give them a strong role in strategic and tactical combat, but they don’t have the raw power of the fighter or game changing magic of the Wizard.
Cleric: best at putting the other threes’ fat out of the fire when they screw up at their roles and a limited buffer against attribution. The Cleric would be useful in combat (second best at a straight up fight), spells that can do some of the strategic shenanigans of the Wizard (although not so well) and OK skills.
Thoughts? Do you think this would satisfy most people or is it too tilted towards my own preferences?
The Sliding Scale of Meaning vs. Balance
Unless a PC death is on the line every single fight, you cannot have both meaningful fights and balanced encounters. To use 4ed terminology, if each combat is followed by an extended rest, then the ONLY way a combat can have any influence on the characters is if a character dies, everything else gets reset. On the other hand, if there is no rest at all between combat encounters then setting up balanced encounters becomes impossible, as how hard the fifth combat is varies massively according to what happened in the first four combats (and varies even more if the party avoids some of the previous combats entirely). It is impossible for the DM to create a balanced fifth combat if the first four combats have effects on the characters that can’t be reset before the fifth combat, it is impossible for the first four combats to mean anything (aside from Character(s) Killed? Y/N) if their effects are reset before the fifth combat.
Of course there is a sliding scale between these two extremes but there is a direct trade-off, you can’t have both at once. So how much of each one do we want?
I believe that having meaningful encounters are more important than balanced encounters (this doesn’t mean that I think that balance isn’t very important, just that it shouldn’t be gauged at the encounter level). This is because there’s a lot of combat in D&D and if there’s a significant chance of characters getting killed in every single fight, then you’re going to have more character death than most people want (even in the hoariest of Old School D&D, lethality declines sharply after you gain a level or two). But if there’s NOT a meaningful chance of PCs getting killed each and every combat then each combat has to have lasting effects that don’t reset or every fight in which PC death is not seriously on the line is basically Harlem Globetrotters Combat as Sports: it looks like a basketball game, but everyone knows what the result will be and nothing is at stake, which is boring. So basically we have only a few choices: lots of dead PCs, Harlem Globetrotters CaS, combat that is much rarer than is standard in D&D or combat that has lasting effects. Combat having lasting effects seems the best option here.
To reiterate, combat having lasting effects subverts encounter balance and the more lasting and important these effects, are the more encounter balance gets undermined. This is because there’s a range of outcomes each combat can have and unless there’s something that returns the results to baseline (like an extended rest) each succeeding combat increases the variance until the difficulty level of the final combat becomes a crap shoot, no matter how you set it up.
The Glory of Attrition
So how to make combat meaningful, even though most combats will be stacked in the favor of the PCs (or at least in favor of the PCs being able to escape)? The only real answer is attrition. In most D&D combats (except in exceedingly lethal campaigns) the PCs will be able to defeat most combats without too much trouble (and/or avoid or escape from most difficult combats), therefore the only way to make most fights mean anything is through attrition. Even if you can kill the goblins without serious risk of PC death, they can still hurt you in a way that’ll make later encounters harder, making you have to care about the encounter with the goblins. This means that you have to use smart tactics against the goblins (or maybe avoid fighting them) even if you know that you’ll win in the end.
To make attrition-based play fun, you have to strip away most of the cushions that players have against attrition. Fistfuls of CLW wands have to go, so do scroll libraries, rope trick, easy teleportation, any form of healing that doesn’t use a healing surge (or other hard to replace source of healing) and the like. Combat also has to be much faster than in 3ed and 4ed to give the DM time to wear down the PCs (bringing back morale rules would be a good start). But the granddaddy of attrition killers is the 15 Minute Adventuring Day. This has to be killed dead for attrition-based play to work. There have been various suggestions offered that work to varying degrees, but I think that the simplest and easiest way to do this is to make getting what 4ed calls an extended rest take a week or so in a relaxing location (so in LotR terms, the hobbits don’t get a single extended rest all the way from the Shire to Rivendell). We can allow characters to recover some resources in the field, but not many and only at significant cost.
Then, in between these rare extended rests (which generally would be impossible get in the field) the characters can enjoy all of the glories of getting worn down by Oregon Trail D&D. Healing surges are a great way of modeling this (as long as they’re not potent enough to ward off attrition entirely). Run out of food? Lose healing surges due to starvation. Get dungeon lung? Healing surges. Exposure? Healing surges. In combat healing would cost healing surges, but downtime healing would give more HPs back per healing surge. This makes spending healing surges in combat a lot like using speed, it gives you energy in the short term but at a long term cost (which is a lot like how going berserk is described in the Icelandic Sagas). Town adventurers would use time as a constraint to prevent 15 Minute Adventuring Day issues (the Dresden stories are a good model here).
By playing up attrition you get rid of combat grind. Even if it’s clear that the PCs have won, being smart is combat is still important since getting hurt now can have a bigger effect on the next combat and the next and the next.
The Glory of Tactical Play
One of the biggest problems of TSR-D&D for me is that one-on-one Fighter duels are often stupefying boring. This isn’t a deal-breaker for me, but in order to get CaS players on board you have to fix that and similar issues. This means that mundane characters need at least the option of fun tactical options to play with and mundane combat can’t be as abstract as in Old School D&D without turning off a sizable portion of the player base.
The hard line to tread here is to be able to make combat tactical without making it take too long. I think that the way of doing this is to make sure that all combat requires tactical smarts, to make all combat situational and to give the players different tactical angles to work with.
Sid Meier’s definition of a game is a series of interesting choices. In all editions there’s stretches of D&D combat in which no real choices are being made (I remember one player who could’ve gone to the bathroom during most fights and left a sign saying “I take a five foot step away from the enemy and do rapid shot at the closest one” and his contribution to combat would’ve been unchanged). Focusing on attrition helps with this problem since even if the players can just grind down a combat in a boring way, they have a big incentive to be smarter than that, but some cues from 4ed design are needed here as well. What would also help a great deal is to give players tactical tools to escape from monsters that are too strong for them (to make fights of all difficulty levels fun) and to have fights in which the PCs have goals aside from “kill them all!” such as: escape! capture one and get away! grab the idol! don’t let them capture the wizard, they want his brain! kill them before they sound the alarm! etc. etc. which help keep things interesting on the tactical side.
In the previous thread one person noted that when players find a method that works they’ll do it over and over again, which takes all the fun out of it. I think that the solution for this is to make combat as situational as possible. That means making sure that the foes and the circumstances have a massive impact on which tactics work best. That means that any modifiers that aren’t big enough to change what kind of tactics the PCs use should be increased until they do or be thrown out entirely (die modifier bloat die!). I think that all editions have useful ideas that can be stolen for this from 4ed’s greater emphasis on terrain, to powers that suck in some circumstances and rock in others (I love Affect Normal Fires) to things like 1ed’s monsters that were completely resistant to non-magical items (“oh crap, I can’t scratch the damn gargoyle with my sword, we’ve got to change our game-plan entirely! We’ve got to find some way of immobilizing the damn thing and then braining it to death with our +1 shield”) what’s important is to not have the special features of monsters elicit a response of “Meh. We’ll just do what we always do; it’ll just take a bit longer.” On the other hand you don’t want to make the thief gimped against all undead, so it’d be better to keep tie this situational stuff to individual critters rather than broad classes (more unique quirky monsters are always good).
Having competing priorities in combat can also spice up the tactics a great deal. Sure you want the enemies to run out of HPs before you do, but there should also other things to keep track of. A lot of the tactical depth in TSR-D&D came from declaring your actions each round before you knew who was going first and I’d really like to see that and easily-interruptible casters brought back in a streamlined form. Morale rules also help here if implemented well since they give something for the players to think about in addition to hit points. Marks (in their most basic form) also are great and a lot of elements of the leader and controller roles need to be preserved in 5ed.
The Glory of Strategic Play
Too keep CaW players happy, strategic play has to be just as important as tactical play. That means that we have to walk the balance between making sure that what the players do outside of combat has a meaningful impact on what happens during combat, but not so much of an impact to make tactics pointless.
This means that devs have to look at mechanical widgets with strategic uses in mind by giving the reader a clear idea of what sort of things various powers can do outside of combat and making sure that these uses are clear. If what abilities (especially spells) can accomplish outside of combat isn’t laid out clearly then a lot of strategic play gets reduced to “does the DM decide that your plan works or not,” so these abilities also need to be carefully bounded or they can slice away whole chunks of gameplay (scry n’ die, etc. etc.). Good examples of the kinds of spells I’m talking about here are (skimming through the first level spells of my 1ed PHB):
Affect Normal Fires, Comprehend Languages, Dancing Lights, Enlarge, Jump, Push, Tenser’s Floating Disc, Unseen Servant and Ventriloquism. All of those spells cry out for creative strategic uses but they have clear descriptions of their effects that don’t rely very much on DM fiat (although Enlarge is clunky rules-wise as much as I love it). I’d like to see this kind of spell be the core of the 5ed Wizard class and have the Wizard be relatively weak at straight-up combat in order to compensate for its strategic utility (with more blasty casters being possible, just not the Wizard default).
Of course other classes need to shine in strategic play as well. Killing the 15 Minute Adventuring Day helps a great deal with this, as does removing spells that steal too much of other class’ thunder and making sure that high level character of other classes have plenty awesome out of combat abilities (see Beowulf et. al.).
In a similar vein, the devs have to write up the monsters with strategic play in mind. Write-ups of ecology and the like (like all of the random percentages of this and that) are useful for strategic play since they give the DM something to go on rather than throwing them back to fiat. Similarly, these needs to be plenty of monsters in which dealing with them through strategic means and never engaging them in a straight-up fight is smart (rust monsters being the classic example here).
Another benefit of strategic play is that is lets DMs uses much more of the Monster Manual at once. At low levels a hill giant is something that players should figure out smart ways of avoiding, after they gain a few levels they can take one down if they come up with a level strategic plan, then they become standard opponents and then at high levels a couple hill giants are still bad news if the party blunders into them while worn down.
One downside to putting an emphasis on strategic play is that it’s hard to pre-plan epic show-downs since it’s hard to hit on the right difficulty when things that the players do before the combat start can cause the difficulty of any one encounter to swing wildly one way or the other. One way of doing this is to let epic fights emerge out of the strategic game, in my experience nothing makes the players smile more after a hard fought battle than the DM saying “what?!? You took THAT on one-on-one and WON? I don’t believe it!” Letting the players find their own climactic final show-downs can be rewarding. For a more CaS-friendly option maybe a rule to let the PCs “dig deep” and get back resources lost to attrition for just one fight, but at severe mid-term cost so that it would only be used for climactic final fights of a session (or heroic last stands).
Balance
OK, now how the hell do you balance THAT? For adventure balance there’s two main methods. First, instead of looking at balance on an encounter level (since attrition-based play and strategic elements kill encounter-based balance) balance it at the adventure level. So if the PCs are going from Rivendell (extended rest available) to Beorn’s farm (extended rest available) through the Misty Mountains (no place to take an extended rest there) then balance is determined by the difficulty of the Misty Mountains taken as a whole, not according to each encounter or even a list of encounters. Maybe set up a 4ed-style adventurer builder to calibrate the difficulty instead of building things encounter by encounter.
The other way is to let players find their own risk/reward balance. The players go out from the town (extended rest available) into the howling wilderness and/or the deep dark dungeon (where, with perhaps a few exceptions, there is no place comfy enough to take an extended rest), and the longer they stay out the more attrition grinds them down and the more dangerous things get. But, if they go back to town to rest after every stiff fight they’ll never get deep enough into the dangerous area to get good loot and they’ll spend a lot of time tromping back and forth through already looted areas running into wandering monsters with no/very little gold. The players can choose to play it safe (low risk/low gold) or gamble (high risk/high gold), with a few helpful doors that lock behind the PCs, enchanted forests in which the trails move around, chutes, teleporters and whatnot to keep things interesting.
What about PC to PC balance? For 5ed, I really don’t think that encounter by encounter balance is the best way to go. First off, we’re not using balanced encounters anyway (see above) and trying to balance things encounter by encounter leads to the MacGyver Wizard Problem. Let me explain. Let’s say that you have a 4ed Wizard and a 4ed Fighter who are both equally useful in combat. Now the question is; does the DM let the Wizard MacGyver his attack powers to let them do useful things out of combat (or off-label things in combat)? If the answer is yes, then the Wizard becomes more powerful than the Fighter since it’s a lot easier to think of creative uses (especially creative uses out of combat) for Fireball than for, say, Sweeping Blow. But if, on the other hand, the DM or the rules are strict in building an impenetrable wall between combat magic (only good for killing things) and out of combat magic (no good at all for killing things) then, at least for me, that robs D&D magic of some of its most fun elements, since all of my favorite D&D spells have always had combat and out of combat applications.
So how do you balance character classes in 5ed then? I think that the best rule of thumb is to make it so that there is no 4-member party that is clearly superior to the traditional one fighter, one thief, one wizard and one cleric over the course of a standard adventure (a “standard adventure” is somewhat hazy, but I think that’s unavoidable). By this standard, 3ed fails badly. A party of two clerics, a druid and a wizard vastly outclass the traditional party at virtually any adventure that you can imagine. Other editions of D&D do better and I don’t think that this is an impossible standard.
There are a number of ways to meet this standard and probably the biggest question is what to do about the Fighter (which mostly boil down to whether a high level fighter looks more like Lancelot or like Hercules, I like the Lancelot option but there’s room for disagreement here) but this is my own personal stab at it:
Fighter: best at tactical combat. Best HPs, best Saves/Defenses/ACs, damn good damage dealing. You need the fighter when the PCs are ambushed and haven’t had time to come up with rat bastard plots and need to fight their way out and they’re the battering ram that makes all of the strategic plans have the oomph to work. Over the course of an adventure, the Fighter should have racked up more HPs in damage than any other class (and HP damage should actually matter). The fighter would have a secondary focus on skills (best bar the thief), leadership and social interaction so they have plenty of things to do outside of combat.
Wizard: best at strategic combat. While the Fighter is the best at playing the game, the Wizard is the best at tilting the playing field. They can do some spike damage, but it’s not their forte and their skills are fairly weak (too much time spent learning magic to learn other stuff) but their magic is good at changing the terms of engagement of fights.
Thief: best at just plain avoiding encounters altogether. The best fight is often the one that never happens (thieves know their Sun Tzu) and the thieves are the best at doing this through sneaking around things and talking their way past them. Their strong skills and stabbity nature give them a strong role in strategic and tactical combat, but they don’t have the raw power of the fighter or game changing magic of the Wizard.
Cleric: best at putting the other threes’ fat out of the fire when they screw up at their roles and a limited buffer against attribution. The Cleric would be useful in combat (second best at a straight up fight), spells that can do some of the strategic shenanigans of the Wizard (although not so well) and OK skills.
Thoughts? Do you think this would satisfy most people or is it too tilted towards my own preferences?
Why not? Uncertainty. It may be unlikely that you're about to die. But you can't know that. You never know what will happen next. 
