Leveling Up Monsters

In Chris Perkins' latest article (the same article that I linked to in my last post on NPCs) he mentions that the stat block for one of his NPCs is a level 9 Succubus leveled up to 25.  Now, the DMG specifically warns against leveling monsters up or down more than a few levels from the base stat block.  Though I'd never thought about this before reading about Chris' over-leveled Succubus, I think this is an example of the DMG being too conservative rather than Chris doing something crazy and unconventional.

Because the math for monster scaling is so simple and transparent in 4e all of the basic numbers (defenses, HP, attack bonuses, and damage) will add up with no problems.  That's part of why customizing monsters is so easy and painless in this edition (cuts down on prep time, too!).  So what's the problem perceived by the DMG authors?  Most likely, it's status effects and riders that are tacked onto powers, or other traits such as auras, conditional damage, and off-action attacks (or number of attacks).  Not surprisingly, monsters get more dangerous at higher levels than their raw base stats can account for, which makes sense because players get more access to feats, items, powers, and paragon path/epic destiny features.  Leveling an Epic monster down for Heroic PCs would make for a disproportionately challenging encounter (possibly impossible in some cases), and leveling a Heroic monster up to Epic might make for a cakewalk. 

For example, lets look at the Efreet Flamestrider, a level 23 Skirmisher.  I picked this guy at "random" since he was the highest level standard monster in the MV.  He has an aura that multiplies ongoing fire damage, a MBA that inflicts ongoing fire damage on a hit, and an at-will ranged attack that both immobilizes and deals ongoing fire damage (save ends both).  And of course there's the at-will 20 square teleport (even if it does require fire, which any DM will build into the encounter while using this creature).  Even leveled down, Heroic characters don't have enough ways to trigger saving throws to easily counter his ongoing damage and immobilization (and teleports will be much more rare as well).  Their healing ability is also more limited, so PCs within the aura can easily find themselves overwhelmed by raw damage.  Finally, Heroic PCs will be more dependent on status effects like immobilized, slowed and restrained because they probably won't yet have wide access to dominates, stuns, and blinds, meaning that the efreet's teleport makes him very tough to hinder.  Likewise, a level 1 Goblin Cutthroat with just straight damage and some minor shifting tricks won't be much of a problem for Epic tier PCs.

After pointing out the problems of over-leveling, I'm still not going to advise against doing it, at least assuming you're reasonably familiar with the system.  The thing is, after pointing out the problems with the efreet simple common sense (and possibly a quick comparison with the Goblin Cutthroat) can inform which further changes must be made.  It's clearly too powerful when leveled down without further adjustments, but what if we made the following changes on the fly:

• The ranged attack becomes a recharge or encounter power so it can't constantly immobilize PCs.
• The teleport becomes an encounter power so it can't constantly escape the common status effects and re-position itself.
• Either get rid of the aura, or get rid of the ongoing damage with the MBA.  Specifically, either have ongoing 5 damage and no aura, or have no inherent ongoing damage with the attack but have the aura inflict ongoing 5 damage if a creature is hit with a fire attack.  These 2 methods look identical, but keep in mind that with the latter option a ranged fire attack from a second monster can gain ongoing damage against PCs in an efreet's aura.  

BAM!  Your Efreet is now ready for Heroic tier play.  Similarly, you can level a Goblin Cutthroat up to epic and give him additional abilities based on a quick glance at several other epic creatures.  You can even just make stuff up pretty easily thanks to 4e's concise list of status effects; call his blade poisoned and have it inflict ongoing poison damage and possibly another status effect (blind and weakened seem thematically appropriate for poison).  Maybe give him an encounter or recharge power that stuns.  Rogues get to do stuff like that at higher levels, so it's not too big a stretch.

So now that I've demonstrated that you can level monsters across the entire spectrum of levels for just a little bit more work than with a more narrow level range, why should you?  It's mostly a matter of personal taste.  Do you want a specific monster for your adventure?  Use it regardless of its base level!  What if an off-level monster has a set of abilities that you really want to use for something else?  Level it down/up and re-flavor it, or tack its abilities onto an appropriate-level monster.  Are your PCs sick of fighting the same orcs over and over again?  Find a different monster with some interesting mechanics, level it as appropriate, and call it an orc!  Or, simply modify the existing orcs by giving them different abilities.  One of the great strengths of 4e is its modularity.  It's not only possible to mix and match elements from different monsters and different levels, but it's fairly easy as well!  So when you plan encounters, don't necessarily limit yourself to the pool of monsters in a given level range; branch out for more ideas and give the players something unexpected (especially important if your players also DM, and thus have access to the same monsters that you do).

Fleshing Out NPCs

Chris Perkins says it better than I could, in his new article on 3D NPCs.  I would love to play in one of his campaigns!  Anyways, I'd highly suggest anyone who DMs read this article.

In particular, I'd like to highlight the parallel between an ongoing D&D campaign and a serialized TV show.  Obviously with the PCs being the main characters a lot of the complex relationships will grow organically as the players interact amongst each other, thus making the comparison a lot less daunting for DMs!  However, DMs still must think about this since the PCs aren't the only people populating the campaign world.  Each "adventure" should feature at least a few fleshed out NPCs (more than just background NPCs, like the random shopkeeper or barmaid), and you should have a vague idea of these characters' personalities, motivations, and backstories.  This obviously doesn't need to be too in-depth, but at least summarize each of these 3 traits in a sentence or 2.  It doesn't have to be elaborate either; most NPCs are much more mundane than the Heroes! 

For example, let's look at one of the main NPCs from my Talamhlar campaign, Gaknar (this campaign ended prematurely when I moved away for a seasonal job, and due to the months that passed before I came home we just never resumed it).  To re-cap, that campaign started with the PCs captured in a castle, regaining consciousness with the place in chaos.  A group of mages were doing various arcane experiments in the castle, including the creation of super-soldiers.  Not surprisingly, the mages lost control of the super-soldiers, were largely massacred, and the magical containment was broken, freeing the PCs.  In the room their was also a goblin named Gaknar, and he promised to show the PCs to where their equipment was and how to get out of the castle if they freed him (he'd been a servant for the mages).  He accompanied the PCs throughout the campaign, becoming a valued member of the party (well, most of the PCs liked him).  There were a few times where he acted somewhat suspiciously and was caught lying, and so was never 100% trusted.  He even disappeared "on business" once, refusing to tell the PCs what the business was.  So what was the deal with Gaknar, and what ideas did I have floating around in my head for him?

• Backstory:  Gaknar was a runt in his tribe, always picked on for it, and some semblance of a conscience (moreso than most goblins, anyways) didn't help matters.  During a raid he left his tribe, but finding it difficult to fend for himself ended up working for the mages.
• Personality:  Sycophantic, and has quirky ways of showing affection.  Sly and sneaky like most goblins, but with a greater sense of loyalty.
• Motivations/Secret:  On his errands away from the mage's castle, Gaknar began to see what life was like in the civilized worlds.  He became a spy for someone in the nearby town of Marblemount, helping to gather info on the mages to help shut them down.

Gaknar's secret never got a chance to be revealed in-game, and in fact I hadn't even fleshed out who he would be working for in Marblemount, what their exact goal was, or even whether it was a single person or an organization.  I didn't need those details until an opportunity arose for me to incorporate them into the game, and they would ultimately depend on what direction the campaign was headed.  I didn't plan on when that might be, but for things like that you just sort of know when the time is right.  I just really liked the idea of a goblin double-agent traveling with the PCs, who was ostensibly good but with a shadow of doubt.  And that concept came built in with a possible future story hook for later, even if I didn't know what that hook would be.

Holy Statistical Anomolies, Batman!

I'd like to qualify this post by saying that I was bored tonight, and really in the mood for D&D. Given that this wasn't a night that the group is available, that means that I decided to do a random playtest. Feeling particularly ambitious, I decided to transcribe everything that happened in this encounter (which I normally don't do). The PCs are all level 6, and the enemy forces are a 9th level encounter for 4 PCs.

In one corner, we have our PCs (some of which you may recognize from the Red Frogs campaign): Lyra Cinderfield (Human Staff Wizard), Berylis Lindelenon (Elf Panther Shaman), Albanon Izariel (Eladrin Assault Swordmage | Charisma Warlock Hybrid, with focus on teleportation), and Unit 27 (Warforged Weaponmaster Fighter). In the other corner, we have the Raven Roost strike force: 3 Raven Roost Outlaw Veterans, 2 Raven Roost Sharpshooters, and 4 Human Thugs (minions). I used the poster map that came with MV2, specifically the one in the woods near the water, with the small cabin (which is where the PCs started out).

This encounter turned out to be the epitome of an encounter-gone-wrong due to cold dice. Albanon missed a total of 10 times out of 16 attacks, for a hit rate of 38%! Doing a rough calculation assuming Eldritch Strike is used, he should have had a 70% hit rate versus the Sharpshooters, 65% vs the Thugs, and 80% vs the Veterans. But the dice just were not on his side, even with attacks granted by Aegis of Assault and Berylis' Claws of the Eagle. Normally I hit the mark pretty well with these playtests, defeating encounters in 4-6 rounds. This one took EIGHT! Healing Spirit was used both times, and Hearth Spirit was also used (both Lyra and 27 took advantage of the 2nd wind as minor despite poor positioning, all made possible by a strategically placed enlarged Winged Horde). Still, nobody ever fell unconscious so the encounter wasn't too challenging, it was just turned into a grind by all of the missing (specifically on the part of the party striker). Surge expenditure (including post-battle recovering back to full health) was as follows: Lyra used 2 surges, Berylis 1, Albanon 2, and 27 3. For much of the encounter Albanon was in danger (bloodied), and Lyra spent several rounds getting between enemies and him (fortunately she has both Shield and Staff of Defense, both of which negated attacks).

Now, I'll fully admit that cold dice weren't the only factor. I'll usually roll randomly before a playtest to determine which PCs have action points available and what, if any, Daily powers can be used. I didn't do that this time, so 1) nobody had action points, and 2) the only dailies I used were Icy Terrain (from Lyra's wand, which I hadn't used in play yet since picking it up with some leftover cash) and Hearth Spirit (because either Lyra or 27 or both would have gone down if I hadn't). Upon further reflection I also didn't focus fire as well as I should have, and tactics were more focused on action denial than anything (Albanon probably could have gotten more consistent CA if I'd tried harder...though to be fair I usually rely on a combination of Stalker's Strike and Claws of the Eagle for that, but they always kept missing).

Unfortunately, this isn't the first time this has happened. At the start of the Red Frogs campaign (before it was even named) the party was TPK'd in a very close fight; one in which my Hunter (a very accurate class) had an absurd missing spree where I rolled 5 or 6 twos and threes. It doesn't seem like anyone else I play with is quite as prone to such severe bouts of bad luck; perhaps I need to stick with a single d20 and develop some superstitious dice rituals (our player who rolls the most crits notably sticks his dice in his mouth a lot...maybe I'll try that?).

So what's a player to do in such circumstances? Well, aside from picking up the standard accuracy boosters (expertise, accurate implements, +3 weapons) one of my strategies is to favor controllers. The more attack rolls you make, the more likely you are to hit something. Definitely enjoying playing Lyra more than that poor Hunter. I'm also fond of implement users that can be flexible about which defense to target. Lyra does exactly this, with an emphasis on Will, which is just perfect. Auto-damage is also nice. Though Lyra hasn't been struck with a major missing streak yet, I think as I level her I may enchant a Wand of Beguiling Strands (my least-used at-will) and trade out that power for Heroic Effort. For the same reason I tend to favor Elf PCs as well. Interestingly, in this playtest encounter I ended up critting with Spirit Hunt in round 1 thanks to Elven Accuracy (talk about a hard-hitting power to crit with!), though lady luck balanced that out when I rolled a 20 for Charm of Misplaced Wrath and a 2 for the attack that it grants. Yeesh.

Multiple-Arc Campaigns

Chris Perkins has a column on the WotC website called The Dungeon Master Experience.  It's by far the best column that the website currently has (if you're a DM, that is) and this week's pertains to something I've been thinking about lately:  the balance between a "sandbox" style and a "railroaded" campaign.  Most campaigns occupy a middle ground between the two, and I think that's best overall.  From a player's perspective (I enjoy that I play almost as often as I DM) too much railroad makes me feel like my character's actions don't matter, and too much sandbox can make the whole venture seem aimless (not to mention the fact that it could easily catch the DM off guard, and I don't care how good of a DM you are, you're not going to be running as optimally if you're trying to improvise a random direction as opposed to something you've planned, or at least considered). 

I have a feeling that most DMs think about this spectrum in relation to the current adventure, because after all the players are going to react to things happening in the present.  It's tougher, however, to compromise throughout the span of an entire campaign arc (at least I think so).  Generally it's either unplanned and the DM just goes with the flow (often leading to a string of semi-related adventures) or the DM has a good idea of where he wants the campaign to go.  Planning for multiple (in this case, three) different arcs gives the DM focus because he's thought about the possible arcs and in general knows where they're headed (he just has no idea when and in what order the players will tackle them in), but it also gives the players a lot of choice in what they want to pursue because the DM has thrown several well-thought options out there.  Obviously this will work best if the DM allows even the arcs that weren't pursued to "progress," perhaps making the situation "worse" and providing the PCs with more and more difficult choices.

Anways, here is a link to the article.  It's a good read.

Recovery: Viral's Diary (Session 1)

Cast of Characters

Lyra Cinderfield (my character):  Human Staff of Defense Wizard (Helo - Dragonling Familiar)
Berylis Lindelenon (my other character):  Elf Panther Shaman (Legadema - Spirit Companion)
Rosil:  Elf Brutal Scoundrel Rogue
Unit 27:  Warforged Weapon Talent Fighter (Hammer and Board)
Zeus:  Dwarf Wrathful Invoker

The Amber Farmhouse

The Red Frogs were hired by a man named Aramis Emis to find the diary of an archaeologist named Viral.  It is reputed that Viral discovered a powerful magical artifact, and the diary is thought to provide clues to its whereabouts.  The last man known to have seen Viral was Amoz Gib, but he lives in a secluded village called Deadwood Falls located deep within the dangerous and labyrinthine Deadwood Forest.  Few find their way through the forest alive, but the party does have one lead:  a local farmer/merchant by the name of Athos Amber regularly visits the village, and so the party travels to his farmhouse in order to enlist his services as a guide. 

As the party approaches the property, they see a golem with a sack of corn on its back standing in the middle of the path.  Upon approaching the construct, it animates and plays back a recording of a man's voice.  Assuming the man is Athos Amber, it sounds like his wife was killed by some unknown assailants, and that he was able to hide his daughter in the "attic" before he too was killed.  Lyra examines the magical workings of the golem while Rosil investigates its mechanics, and they determine that it's non-sentient, and essentially a piece of sophisticated farm equipment.  So they investigate the farm house (which has an attached barn) cautiously.  Lyra sends Helo ahead to peer through the windows, but nothing can be seen.  The party goes into the barn which houses not farm animals, but a workshop containing another (inactive) golem as well as a bunch of random junk.  They enter the house through the door that connects to the barn, and find nothing of interest on the ground floor.  Upstairs Rosil scouts the individual rooms; the first contains a desk and a blue chest, the second is a bedroom that appears to be the site of a struggle, and the third contains the horribly mangled body of the wife.  The party returns to the first room where Lyra detects the presence of magic in the blue chest, which Rosil picks the lock to revealing a book, several loose pieces of paper, and a magical key that's warm to the touch.  The book seems to be an atlas with random notes scrawled into it (including the names of some influential merchants), and fortunately a route through Deadwood Forest is laid out.  Among the loose papers is the will of Athos Amber, who appears to be very rich for a simple farmer.

As the party makes their way back to the barn, the key grows warmer in Lyra's hand.  It reaches its hottest in front of a very tall pile of junk, which turns out to be an illusion hiding a ladder and a trapdoor in the ceiling.  The secret loft contains many crates, and the party suspects that Millie Amber (the daughter) is hidden in one of them.  Lyra uses Ghost Sound to emulate the voice that the sentinel golem recorded, and Millie (about 10 years old) emerges from a crate thinking that her father has returned.  The party asks her what happened, and she reports that about 5 days ago she was woken up in the middle of the night by her father and told to hide in the (well-provisioned) attic until he came to get her.  The party doesn't tell her that it looks like her parents are dead, but offer to escort the girl to Winterhaven where her grandparents live (it's on the way to Deadwood).  As they get back to the ground floor of the barn, the doors slam shut and Lyra orders her to return to the attic.  Lyra fires several enlarged Winged Hordes out of the windows in an attempt to catch their ambushers off guard, but it doesn't sound like she makes contact.  Three regular Wraiths and 2 Mad Wraiths float in through the windows and initiative is rolled. 

Before anyone's turn came up Lyra yelled for Zeus to focus on the 3 Wraiths, and the two of them largely succeeded in keeping them out of the fight while the Mad Wraiths were dealt with.  Berylis led off with Spirit Hunt after the Mad Wraiths caused Unit 27 to attack Lyra.  Zeus caught 2 of the Wraiths with Rebuke Undead, after which they moved forward (unable to do anything else thanks to the daze).  Lyra hit them both with Icy Rays, immobilizing them.  Rosil critted a Mad Wraith, and Unit 27 stuck close to the Rogue in order to protect him.  The controllers double teamed the Wraiths with a combination of area damage and status effects (eventually Berylis finished two off with Twin Panthers), while 27, Rosil, and Berylis took care of the Mad Wraiths. 

Once the battle was over, the party escorted Millie to Winterhaven where they found her grandparents.  Lyra made sure that they were capable of taking care of the child, and offered to check up on them on their way back through (Lyra was herself an orphan).  They then traveled the rest of the distance, with Lyra studying the book (and their route) throughout the journey.  Finally the made it to the border of Deadwood Forest, and it was at this point that we ended the session.

Commentary

The DM for this adventure made a "rookie" mistake, which was forgivable since this was only his second time DMing 4th edition.  The mistake was that he used wraiths from the Monster Manual instead of the new ones from the Monster Vault.  BIG DIFFERENCE!  MM Wraiths are infamously grindy, with the insubstantial property (they take half damage from most attacks) and attacks that inflict the "weakened" condition (target deals half damage).  Thus, damage was often quartered when fighting these guys.  Also, aside from the recharge power on the Mad Wraiths (that force PCs to attack their allies) their damage is pathetic.  Oh, after looking at the stat block those wraiths also have regeneration, which the DM forgot to apply.  Anyways, these monsters easily make for some of the most boring combats in 4e being non-threatening and yet dragging out combat by being difficult to actually hurt.  The DM ended up calling the fight into round 4 when it was blatantly obvious that the party was going to win (even if it would take a couple more rounds) and sustain no meaningful damage. 

I advised the DM to use Monster Vault monsters in the future, and so a very important lesson was learned; 4e had some pretty crappy monster design at first, and an inexperienced eye can't always tell which monsters are the worst offenders.  For example, the Monster Vault wraiths don't have regeneration and they don't weaken, so insubstantial is the only damage-reducer that PCs need to worry about.  The wraiths also deal almost twice as much damage as their MM incarnations, and they gain the ability to become invisible.  If invisible when they attack, they more than double their already improved damage.  In short, they actually function as Lurkers.  Though MM wraiths are labeled as the same, they actually function more like skirmishers (and not very effective ones at that).  I honestly hadn't examined the differences between the 2 versions that closely (I learned early on to simply pretend that wraiths didn't exist precisely because of the aformentioned problems), so I'm actually glad the mistake was made because now I realize that MV wraiths are not only useable, but look pretty fun! 

As for my characters, I'm really happy with them so far.  I've been agonizing over Lyra's power selection a bit, and before this adventure decided to switch out Color Spray for Icy Rays at pretty much the last minute.  Even though Icy Rays isn't enlargeable (and doesn't synergize with Helo's "breath" ability), I think it'll be a better choice.  In playtesting multi-target daze often doesn't reliably deny standard actions (it's tough to get a set-up where everyone is denied charging), and the unfriendliness was a concern with relatively inexperienced 4e players.  Plus now that Icy Rays slows on a miss it's a phenomenal opening move, and I generally like powers with very flexible targeting capacity (Icy Rays can be used against two creatures that are far apart from each other, meaning that you hamper exactly which enemies are most dangerous regardless of positioning relative to each other).  We'll see how it goes. 

I also gave Berylis Spirit Hunt shortly before starting play with him, replacing Spirit of the Slavering Bloodlust.  Without being an Eagle Shaman, the buffs from SotSB are unlikely to reach the damage potential of Spirit Hunt, so personal damage won out in this case.  Besides, I'm waiting to see how well the other players respond to my Spirit Boon before I add more buffs that rely on Legadema's proximity into the mix.  So far so good on that front, though; the Spirit Boon came into play almost every round, even if I did have to remind the players about it.  Between the Spirit Boon and Strengthening Spirit, I don't think it will be too long before everyone realizes that it pays to be next to Legadema (though I certainly need to do my part and make sure that I actively position her next to people).  Anyways, between high personal damage, granted attacks (Claws of the Eagle), and the Spirit Boon it seemed like Berylis dealt comparable damage to Rosil, perhaps being slightly lower (hard to say, as Rosil rolled the only 2 crits of the combat).  This being Heroic tier (where Rogues shine) and Rosil being fairly well-optimized means that Berylis is probably exceeding the damage of some strikers, at least this early in the game.  As he levels his damage will probably be more indirect (from buffs), though I'm sure personal damage will remain competitive as well.  I'm already looking forward to a combat where several enemies are bloodied on the field and Lyra and Zeus simply park themselves next to the spirit and blast away (with the backing of a hard-hitting spirit boon).  You can get a lot of mileage out of the Spirit Boon via area damage.

"Filler" Encounters

In my previous post I talked about why it's important to give your combat encounters narrative significance, however big or small it may be.  This is largely because the complicated nature of 4e combat makes it a significant time investment, and "breaking away" from the story to fight some random battle can easily be perceived as a waste of time.  Just because the "default" 4e encounter doesn't lend itself to minor scuffles or "filler" encounters doesn't mean it can't be done, though.  Such encounters just need to be designed very deliberately.  Indeed, these types of encounters should be included from time to time because it's admittedly impossible to make every encounter relevant.

My focus here will be creating encounters that are resource-draining but tactically (and usually narratively) irrelevant.  Perhaps irrelevant is too strong a word, but in any case these are encounters that don't pose much of a threat and that are over very quickly.  There are several ways to evoke this feeling, the simplest probably being an at-level or lower encounter, designed more or less "normally."  The problem with this is that monsters that are too low level are a complete non-threat (PC attacks will virtually auto-hit them, and they'll be unable to hit the defenses of the PCs), and those that are a threat tend to have enough cumulative hit points to drag the encounter out to a decent length.  One way to keep combat short is to simply keep the at-level monsters, but use fewer of them.  If the PCs outnumber weak monsters 2:1 it shouldn't last more than a couple of rounds, but during that time the PCs should take some damage (which they'll need to spend surges to fix), and might even blow a daily if they misjudge the encounter's difficulty. 

Here are a few more ways to create quick and dirty "filler" battles:

1. Use terrain that gives the PCs a massive advantage.  Describe it so that it's really obvious that the PCs can use it to quickly dispatch monsters.  If you want to maintain the level of danger present in a standard encounter but make it happen much more quickly, have the monsters use this advantage first, but allow the PCs to quickly turn it against them.  This way the PCs take boatloads of damage in round 1 (or even a surprise round), and then end the fight quickly by subjecting the monsters to the same effects.  Some examples include pools of lava or acid that creatures can be pushed into, buckets of lava or acid that can be thrown, large and heavy objects that can be knocked onto creatures, or containers of explosive substances that can be ignited to create a nasty explosion.  
2. Use cowardly monsters.  After a couple of rounds, perhaps even before most of the monsters are bloodied, have them flee.  Maybe they'll lure the PCs into a trap or ambush, or perhaps they'll simply reinforce a later encounter.  Depending on the situation they might just run away, and sometimes the PCs may simply let them.  If the PCs insist on chasing them down, instead of rolling loads of dice to whittle away at their HP simply describe the situation narratively:  "knowing they're outgunned, the bandits use double moves to run full speed, and the party finishes them off with arrows, javelins, and spells."  It might be more interesting, however, to use creatures that can actually flee effectively (flight, fast speeds, invisibility, phasing, burrow speeds).  
3. Finally, use copious amounts of minions!  These guys were practically built for encounters like this.  You'll definitely need to tailor encounters to your party's capabilities, though; a party with strong AoE capabilities might be able to kill a dozen or more minions in 1 round, whereas parties lacking AoE would end up being in for a really long battle, which is the opposite of what you're going for!  Also keep in mind that if there's a disparity within the group, as there almost certainly will be, AoE-capable PCs will be the stars of the show for a while, whereas high-damage single-target strikers might feel quite useless.  Wizards with Enlarge Spell can be particularly problematic, as the areas they're capable of affecting can be quite large.

You can make the argument that easy encounters increase verisimilitude, as they give the impression of a dynamic, living world that exists beyond just the PCs and their current adventure.  The orcs that the PCs fought in Heroic tier may still be there in Paragon tier, and it's appropriate that the PCs have a much easier time in dealing with them.  You might also consider sprinkling impossibly difficult encounters into your campaign using the same argument, but if you elect to do this make sure that your players know about it ahead of time!  It's no fun to fight to the death just because "the DM wouldn't let us fight it if we couldn't beat it." 

Encounter Narrative

Because combat in 4e is very tactical (moreso than previous editions of the game), it also has a tendency to take pretty long in real time (though your mileage may vary).  This isn't always the case, but for tactically minded groups combat can almost be a game in and of itself (heck, it's one of the reasons why playtesting isn't as boring as it sounds in practice).  A hefty length of time and a different strategic mindset separate you from the "roleplaying" aspects of the game during combat, which is great for creating a varied play experience but it could also be a bit jarring for the pacing of the story.  For these reasons, the vast majority of major encounters should have some narrative significance.

In previous editions of D&D encounters with "wandering monsters" and dungeon crawls through miles of subterranean passages populated by all manner of random creatures were highly encouraged.  From what I understand, combat in earlier editions was much more swift in real-time, though I only have experience with 3rd edition (indeed, the longest combat encounter I've experienced was in that edition, though save-or-die spells could make encounters equally swift and arguably inconsequential).  If a fight ended up being uninteresting it didn't really matter, because it would probably be over quickly.  Because of that, an encounter didn't need to have any particular importance because you were back to the story (or the next encounter, in more of a hack 'n slash style campaign) in no time.  Because the default 4e encounter is assumed to be somewhat involved, this edition isn't quite as friendly toward that old style of play.

Ultimately, the take home message is that when designing an adventure, every encounter needs to have some kind of significance.  Some encounters can be justified by being exceptionally interesting in and of themselves; monsters with cool powers, traps, terrain, puzzles, or engaging tactics can all make an encounter interesting enough that it doesn't matter if it doesn't have a strong link to the story.  If you need "filler" encounters to give the players something to do as they make their way through the dungeon--an obstacle to their eventual goal--then they better be interesting.  But every encounter can't be over-the-top and exception, because if that becomes the norm then it's no longer very exceptional!  This is why most encounters should have a direct relevance to the adventure narrative. 

But what does that mean, exactly?  Obvious examples are encounters with the foes that the PCs set out to confront in the first place.  This can be as epic as a final "boss battle" with the BBEG (big bad evil guy), or as simple as tracking down the group of bandits that have been harassing the local merchant caravans.  Other ways to make combat relevant are to use the encounter to introduce NPCs (whether they're combatants or non-combatants, allies or enemies), to provide PCs with some kind of information, item or clue, to introduce a new conflict (perhaps a side quest), to foreshadow a future plot point, or to characterize an NPC (perhaps an NPC betrays the party, or maybe they die defending the party who will thereafter seek revenge).  A more intricate encounter might present 2 or more decision points where the choices that the PCs make affects the story.  Perhaps the choices are meant to reveal a character flaw, tempt, or highlight the chivalry of a PC.  Any time you give the player the means to help develop his/her character more fully is time well spent.  An encounter's relevance could also be as simple as providing a direct obstacle that stands between the party and their goal (i.e. they want to get into the Wizard's keep, but first they need to deal with his golems). 

From a pure mechanical standpoint, encounters exist to challenge the party and put a strain on their resources.  At its most basic, combat is the reason why characters have healing surges, encounter powers, and daily powers and any encounter will allow (or force, depending on how you look at it) the characters to use these resources.  It's up to the DM and the players to use these mechanics to tell a story, and as the one designing encounters the DM shoulders most of this load.  The story is what immerses the players in the world, and keeps them coming back for more.  While a delve style game where players simply move through the map and kill a random assortment of monsters is certainly valid, it's simply not enough for some players (or groups).  Those that care about story will enjoy the game much more if the story informs combat; an hour-long scuffle that turns out to be pointless can be an annoying distraction if it happens too much.