Blog Archives

Dark Souls: The D&D Video Game that Isn’t Actually D&D

If somebody asked me “what video game today best represents old school D&D,” Legend of Grimrock comes to mind, but ultimately I think Dark Souls would be a more interesting and equally qualified answer. Given the raft of actual D&D games, branded with D&D on the box and everything, why on Earth would I think this? Hopefully I’m convincing.

Megadungeon Format

Dark Souls very much takes place in a megadungeon, and a well-made one at that. If you strip away the visual properties and various idiosyncrasies of the game and focus only on the fundamental design tropes of megadungeons you’ll find this to be the case.

Branching, Looping, Linear Pathways

Megadungeons consist of corridors with walls, meaning you are generally on a linear pathway. Sometimes these are broken into larger areas (rooms) filled with content, but the rooms only have a low number of exits (usually no more than 4 and often as few as 1 or 2). Dark Souls, perhaps simply because games are limited in this way but also probably due to design intention, is similarly made up of linear pathways punctuated by expansive “rooms” or, more video-game-appropriately, “arenas”, in the parlance of the Half Life 2 developers at Valve.

At the same time, though, these linear pathways tend to branch and loop. For more analysis of the types of exploratory pathways in games you can see this post on ENWorld. Further writing on the merits of designing good layouts of linear pathways is this piece by The Alexandrian. These are together probably two of the best examples of analyses on how to design combinations of arenas and linear pathways as far as it concerns roleplaying games, so they make good reading and say basically everything I ever could regarding background.

In summary, Dark Souls doesn’t offer an open world like Skyrim – you don’t have 360 degrees of motion between most locations – but it does offer multiple paths, shortcuts, secret passages, and the like. In Dark Souls, like in most good megadungeons, arenas are connected to each other in multiple discrete linear ways, and those connecting paths themselves are often interconnected.

Re-Stocking Enemies and Re-Treading Terrain

Megadungeons do it every time you leave (or in some cases even when you roll a random encounter, to some extent), while Dark Souls does it every time you rest at a bonfire (or die). While Dark Souls’ enemies don’t change the way a megadungeon’s might, they do still come back, and that means that just like a D&D megadungeon the megadungeon of Dark Souls is designed to not be permanently “clearable”. In Dark Souls, just like in megadungeons like the labyrinth under Castle Greyhawk, you also unlock shortcuts between areas (mentioned above) meaning what begins as re-treading dangerous ground over and over eventually becomes a much more streamlined adventure, much like higher level PCs skip straight to the lower levels of a megadungeon via some shortcut or direct access they have found, thereby reducing the amount of time the players spend re-treading familiar terrain and dealing with re-stocking enemies. Additionally, in both cases, as you level up you will find re-treading the same ground to be much easier than it originally was, but also providing much less reward. These two factors combine to encourage you to push forward outside of any story motivation – mechanically you are drawn to areas that are of appropriate difficulty, since these areas will provide reward commensurate with your out-of-game skill and in-game power level.

Subtle Narrative

There are a few ways that narrative happens in a megadungeon, and most of them aren’t what we think of when we think about exposition in modern RPGs. In a D&D megadungeon you can only reliably get information in a few ways. You can ask NPCs (most creatures wouldn’t be willing to help, though) if you manage to find any (or rarely you can ask NPCs outside of the megadungeon, but their chances of possessing significant knowledge is much less), you can look around for messages, writings, maps, and that sort of thing left behind by some precursor to yourself, you can learn by personal experience (you walk in the room and there are orcs? You know there are orcs in the room now, congratulations), or you can learn by some method of divination of non-language artifacts (either literally by using magic or more like an archaeologist by piecing together what you find laying around into a coherent story).

Dark Souls does all of these things in a very similar framework. You learn about the lore of the Dark Souls world in a few ways –

  • A short introductory cut-scene.
  • Rare, stilted, dialog with non-hostile NPCs.
  • Messages left by other players that fit a purposefully limited format (“[thing] ahead”, or “beware of trap”, for instance).
  • Non-contextual item descriptions of loot you find (you’ll get the story of the item in a few sentences, but no explanation of how it fits into the larger picture).
  • Seeing something for yourself (“oh, shit, there’s a hydra in the lake!”)

All of these are distinctly limited and present to you sufficiently less than the whole picture (and, often, aren’t even entirely accurate, as in the case of player messages). They’re generally lacking in overall context, and have a direct 1 for 1 with how you learn things in a traditional D&D megadungeon. Messages on walls, messages attached to items, observation of the environment… all of these are regular elements of what modern D&D designers term the “exploratory pillar” of D&D – finding out new stuff without it all being dumped on you at once by McLecturePants the Plot-Giving Elf.

People have eventually compiled histories of Dark Souls’ lore, just like players can eventually piece together the history of a megadungeon, but it took significant effort and there are still gaps that need the interpretation of these game-world historians, just as in a megadungeon environment.

A Land Where Stuff Is Weird

As one would be informed by the entry under W in The Dungeon Alphabet, megadungeons are all about being weird. They often (though not always) represent a mythic underworld (see page 22), where things have a sort of internal sense that defies complete scientific classification no matter how hard you try. It’s a brutal landscape for no apparent reason. In OD&D monsters can pass through doors without a care in the world but those same doors are stuck for the players a full third of the time, and even if the players get them open they have a habit of closing themselves and becoming stuck again unless spiked. In OD&D monsters can see in pitch black, unless they’re in the PCs’ service, and then they miraculously lose this method of vision. It’s designed to make the game challenging, sure, but the realistic repercussions of these game design decisions is that the world is seriously weird and seriously hostile in the narrative as well.

Dark Souls is no exception here, either. It’s designed to be difficult compared to most modern games. This game design decision (difficulty) needs justification in the world’s fiction, however, and so by playing as an undead monstrosity yourself you can’t truly die permanently no matter how many times you get sent back to the last checkpoint. The brilliance of this associated mechanic that explains “restart at checkpoint” in game-world terms aside, when’s the last time you played as an undead monster in a game? Probably never if you’re most people. When’s the last time you fought a huge wolf with a sword in its mouth or a giant magic butterfly or a flaming spider-centaur with a pyromancer witch as the torso? This kind of unique strangeness, generated partially by strictly game-centric design decisions being reflected in the narrative, is paramount to enhancing the feeling of exploration made possible by the previous section.

Level & Sublevel Theming

A common thing to find in most megadungeons are highly thematic “sub-levels” (smaller areas bound within a close proximity), or slightly less thematic (but still differently themed) whole levels. Put at its simplest, Dark Souls does the same, and is even kind enough to label them for you (Valley of Drakes, Blight Town, Sen’s Fortress, Anor Londo, and so on and so forth). At its most basic, this is just about creating differently themed “zones” so that things don’t become monotonous.

Under a Big Tent

The concept of the “big tent” campaign comes from the very early days of D&D with Gygax at the helm. Since he knew many people interested in playing games of this sort, they invariably all hounded him to run them – and the games grew truly to unwieldy proportions. Perhaps some 20-odd players inhabited his home campaign, even sometimes sitting at the same table! But, more generally, they drifted in and out as the mood suited them, and the more prolific players whose characters’ names live on in infamy were just the ones who played more often and were thus more successful.

This drifting in and out of players, and the discussions the players would’ve had about what they found in the dungeons beneath Castle Greyhawk, mimics the multiplayer aspect of Dark Souls (or perhaps I should say Dark Souls mimics it, chronologically speaking) almost perfectly. The fact that Dark Souls encourages you to come together to solve the problems of a shared environment (via cooperative summoning or messages, or in the meta aspect by simply talking to other people who play the game on regular internet message boards or otherwise) is extraordinarily similar in concept to how a giant rotating cast D&D game worked back then. Some days some people went into the dungeon, other days different people went in, and sometimes they compared notes, and mixed and matched companionship of their characters as the situation warranted.

This sort of individual shared experience (you were never guaranteed to play with the same people consistently, but you all played in the same world) is a quirky similarity, but one nonetheless.

A Land Where Danger is Everywhere

The final point here is that megadungeons are not amusement park rides with fake scares and cheap thrills – they’re designed in large part to be a game and therefore a test of skill. Not only this, but their early-D&D rulesets were pretty unforgiving of mistakes or even of bad luck. Dark Souls is no exception here, either: it is to most games what gutter bowling is to bumper bowling – in some cases literally, as bottomless pits with no safety rail will regularly attest; it’s a fair, but punishing, alternative. It embraces the war-game-like feel of early megadungeon play, where the trick is for the players to outsmart the environment and progress towards a distant point of actual distinct victory (name level and making a fortress in the wilderness, enabled by their earlier successes in the megadungeon).


As I’ve demonstrated here, Dark Souls is simply the video game that gets megadungeons, and by getting megadungeons it is probably the best representation of “old school” D&D that I can think of in video game format. It has some other elements that it shares with D&D, like leveling up and weapons and armor with plusses on them, but those aren’t what makes it like D&D; they’re just superficial similarities. It has things like PvP combat, too, and that makes it no less D&D all the same. The trappings – the PvP additions of Dark Souls or the weapons with plusses – aren’t what makes the games what they are. I think that reasoning belongs to the framework of play itself: the environment.

If you care about seeing and experiencing a representation of the design aesthetics all those OSR(Old School Renaissance)-types are constantly going on about, therefore, Dark Souls is probably just about the best and clearest alternate version of that experience (complete with moving pictures and sound effects) available in a box, period.

Until Next Time,
The Hydra DM


Power Options, Status Effects, and Mutual Assured Destruction – Second World Edition, Part 1

While the majority of D&D players, statistically speaking, live in nations that are known as the First World (a term invented during the cold war to mean “the major industrialized non-Communist nations”), there’s another set of critters with power options and status effects running around. They are, of course, the players’ major opposition, and to keep our cold war analogy running, they would clearly be analagous to the Second World (a term you don’t hear much these days, meaning, obviously, the major industrialized Communist nations). They’re more numerous, scary, and they’re going to slippery slope all of southeast Asia into the Communist regime! Well, alright, maybe the monsters are a bit more innocent than that.

If anyone is confused about this extended metaphor, I’d suggest taking the time to read The Id DM’s latest article: Power Options, Status Effects & Mutual Assured Destruction, not only because it’s an amazing piece of statistical game analysis, but also because if you don’t read it first you’ll probably find this article somewhat confusing. Not least of all because that introduction was a play off of his use of Mutual Assured Destruction. Go ahead, I’ll wait.

Done already? Man, you read pretty fast! Careful you don’t light your screen on fire from your laser scanner eyes!

Alright, alright. Jokes aside, let’s get into it.

We’ll begin with our core assumptions, or hypotheses if we want to get scientific about it. They are:

Hypothesis 1

  • Combat includes too many moving parts, and this is the monsters’ fault.
  • Premise 1: As level increases monsters get more Power Options.
  • Premise 2: As level increase monsters get more powerful Power Options.
  • Premise 3: The above increase in Power Options is at least equal to that players experience.

I expect this hypothesis will be proven false on account of premise 3, but true on premises 1 and 2. We shall see if the expected result matches the statistical one, and if so (or not) by how much.

Hypothesis 2

  • Very few Power Options are intrinsically tied to the different monster classes.
  • Premise 1: Assuming “intrinsically tied” is equivalent to appearing 50% more often than the average both per power and per monster, and that the Power Option appears at least 10% of the time overall both per power and per monster.
  • Premise 2: Assuming that the monster classes are Controller, Soldier, Brute, Lurker, Artillery, and Skirmisher irrespective of their Minion, Standard, Elite, or Solo status.

I expect this hypothesis will prove true for monsters as we see a similar result from the analysis of player characters. The reasoning behind premise 1 is pretty simple (albeit potentially arbitrary) – a signature power should reasonably appear 50% more often in a class of monster that has it as a signature than it would in a class of monster that does not. Why 50%? Well, if only half the defender classes had ways to punish some sort of mark you’d be a bit suspect about the mark being core to a Defender, no? I hold this principle true for monsters as well. Similarly, a feature that is a trend should be appearing more than 10% of the time overall. If less than 1 in 10 controllers blinds you, blinding may be a thing controllers do more often, but it is probably not a controller’s “thing” that controllers just do. Why the 10%? It seemed like a reasonable cut-off as I was gathering the data.

General Questions

While this is obviously not a formally given hypothesis, I would like to know a couple of things personally. Most importantly of those: how the common conception of “anti-grind” monster usage (that is, not to use soldiers or controllers and to instead focus on artillery and brutes) fits into the sorts of Power Options monsters have. 





My method here is pretty much as identical to The Id DM’s as possible primarily because I want my results to be compatible with his for purpose of cross-analysis between both the players and the monsters (whom all too often in 4th edition are treated under different sets of rules and assumptions). For this week’s analysis I decided to start at the very beginning, a very good place to start, and use monsters only from the original Monster Manual 1. Next week I will use Monster Vault and see how the two compare given approximately 2.5 years of design innovation (which should be especially apt considering they have many of the same monsters).

The status data is collected on the following premises (basically the same as the way The Id DM did it for reasons of compatibility) –

If a power inflicts two different conditions (example: prone and dazed) I record both. If a power inflicts one condition or another condition (example: prone or dazed) I record both. If a power inflicts two different kinds of ongoing damage I record both. I have attached at the end of the article a link to the notes for what the “other effect” row recorded. I have not counted powers that read along the lines of “this creature makes two Claw attacks”, unless it adds a new qualifier, such as “this creature makes two Claw attacks, and if both hit the target is grabbed”, in which case I recorded the new effect, in that example grabbed. Similar to the original table, this means that not only is it possible that the number of Power Options (things a power does that isn’t just damage) exceeds the number of powers, but also it is likely that this is the case.

During the collecting of the data I tried my best to use the same categories that were used to form the charts for the players’ options, however over the course of collection I found that there were some Power Options that did not fit in the categories selected. Not only that, but I found that some effects were not used at all. Furthermore, I found that there were some things that fell outside the criteria I had selected for recordable powers that needed their own category, although I refrained from adding it partway through. I will discuss these more at the end of the article and make the necessary changes when collecting the data from the Monster Vault.

Finally, it should be noted that for powers, much as The Id DM only selected At-Will, Encounter, and Daily powers (with no Utility, Item, Paragon, or Epic powers), I have decided to only select Standard action powers. I made this decision figuring that a monster is as likely to have a “schtick” in its standard powers as a player is likely to find their “schtick” in their At-Will, Encounter, and Daily powers. While I did find some notable examples of powers (notably minor actions) that had interesting and powerful effects on them, I figured that certain Utility powers would mirror this fairly well and, again, I wanted the charts to be as compatible as possible.

Results (and Discussion)

I will present the majority of the data I acquired (although not all of it, as that would be far too many charts) in this section. If you want access to the excel spreadsheet and rough notes I took while making the excel spreadsheet then you can find links to those at the bottom.

Also, you’ll have to forgive my tables for not being nearly as pretty as the players’ companion tables were.

For your convenience, this first table is highlighted with a dark purple in categories where no monsters had any of that particular Power Option, and in blue where a certain role has none of that Power Option across all tiers.

 (click the image for a larger version)

Whew, that’s a lot of data huh? Let’s try to make something intellegible out of this.

So, to do my best to mirror the analysis of the players’ options, I’m going to start with the same thing my colleague did – I’m going to see how many powers (as a percent of all powers) have no effect besides damage. These numbers will be rounded to the nearest whole percent in terms of significant digits, because it’s entirely possible I made a couple minor math boo-boos on the way here.

  • Controller: 25%
  • Soldier: 32%
  • Artillery: 53%
  • Lurker: 27%
  • Brute: 46%
  • Skirmisher: 41%

As compared to the 2% of the Druid, the Controller monster type has 25%. Similarly, instead of the 6% of the Fighter the Soldier has 32%. The Rogue has 7%, at the peak of the PC roles, while the monster equivalent is Artillery with 53%. Finally, the Leader has 3%, and none of the remaining monster types come even remotely close to that low. It’s a fact: monsters have a lot of powers that just deal damage compared to PCs.

Next up, we’ll follow the course of the model article once again and take a look at Power Options and Effects, not only per power, but per monster
(another useful statistic in my opinion). This chart (cleverly using the exact same page layout that the last article used) shows the power options per power
in what is frankly too many significant digits. Still, I’m no statistician, and if Excel’s formula functions want to give me way too many significant digits I’m not about to say no. (Yes I know you can fix that in the options menu somewhere). The point here is, as expected, the power options that each power has goes up per tier, as should’ve been expected (with a few outliers, notably the Skirmisher’s options decreasing in paragon tier and the Artillery monsters’ options decreasing epic tier). The more interesting notion, however, is the comparison between the weighted averages of monster powers and those of the players. While the monster powers have only .75 options per power in heroic tier, the player powers have 1.375. While the monster powers have approximately 1 power option per power in paragon tier the players have approximately 1.6. In epic tier this trend continues, with the monsters peaking at around 1.17 power options per power and the players peaking at 1.635. The interesting thing here is that the ratio of monster options to player options decreases as we rise in tier. In other words, while the players become more complicated, the monsters become more complicated at a faster rate than the players! Perhaps this is a symptom of not including item, utility, and paragon path and epic destiny powers in the player study, but this holds interesting implications for one of our hypotheses. Still, let’s take a look at one more table before we call this one case closed.

The first thing you may notice is that I have revised my chart’s color scheme. I think this one looks a bit snazzier than the last one. At any rate, my professionalism of presentation aside, this chart shows interesting results as well – they are notably similar, in fact, and that’s a good thing. What this tells is that while the powers are becoming more complicated (notably they go from .75 power options to 1.17 per power, as per the last chart, which is a ~56% increase) the monsters are getting more powers, too. If they were not getting more powers, the increase in complexity here, too, would be ~56% (going from 1.58 to 2.46). However, this is distinctly not the case, and in fact we have an approximately 120% increase! While at an average options per power of .75, and average options per monster of 1.58, the monsters of heroic tier have an average of only two powers (roughly), in epic tier with an average options per power of 1.17 and an average options per monster of 3.48 the monsters have, in essence, an average of approximately 3 powers (that have power options). Remember, these are only standard action powers. We can probably assume that move, minor, immediate, etc. powers increase in complexity similarly.

How can we apply this to our hypothesis?

  • Combat includes too many moving parts, and this is the monsters’ fault.
  • Premise 1: As level increases monsters get more Power Options.
  • Premise 2: As level increase monsters get more powerful Power Options.
  • Premise 3: The above increase in Power Options is at least equal to that players experience.

Regarding premise 1, it is definitely true. Monsters get more power options as they level up, this much is clear. However, not only do they get more power options, but they also get more powers! Monsters get more complex by an increase of approximately 120% all told, so this is definitely clear support for the premise that as monsters increase in level they become more complex – a revised premise from the original to take advantage of new information we have uncovered since the prediction.

As for premise 3, this is false so far (in terms of the volume of increase, if not the quality), but not as false as I had predicted. While players do have more options, and that their increase is larger, it becomes less large with each tier. The monsters basically play catch-up, which is not something I expected to see. Still, the increase in power options, in terms of sheer volume, favors the players, so premise 3 is on shakey footing at best so far.

What about premise 2? Well, let’s look at the quality of conditions, starting with the “Big Three” – the hard action deniers. Hard because when subjected to these conditions you cannot somehow ignore them and continue to use your action (for instance, Dazed reduces you to one action no matter what, while with Prone you can ignore the temptation to stand up and retain your move action for something else), and action deniers because they, well, deny actions by literally not allowing the player to use some or all of the actions their turn normally entitles them to. In other words, we are going to look at charts with Dazed, Stunned, and Dominated.

Firstly, I’d like to note that this chart is in percent, not a straight X over Y decimal. This is because the numbers start to get too tiny to really mean anything without having to think about it, and I’m all about aproachability (I say in the article with a squillion pieces of data that had to use excel formulae to achieve anything resembling success). The first thing to note is how there is an apparent outlier in the amount of Daze as tiers go on. First it increases, substantially in fact, and then it sort of decreases back between the first two numbers. This is in clear contrast to the way things work for our four player classes, which have an increasing quantity of daze as the tiers go on. The other thing to note is how few powers actually daze, especially as compared to the player classes. Referencing the charts from our player companion article, we can find that the player powers of those four classes have powers that daze 5.2% of the time in heroic tier, monsters have powers that daze only 4.1% of the time. This difference exaggerates as the tiers increase, with players having 10.9% chance of a daze in one of their powers selected at random, while the monsters only have 7.9% in paragon tier, and in epic tier it gets even wider, with a 12.3% chance of a daze in any given power for the players and only a measly 6% chance for daze to be in any given monster power.  Since I don’t think we’d get a lot out of a “Dazes per Monster” chart in this instance, let’s move on to the next of the Big Three, Stun.

The stunned % per power chart basically tells the story you’d expect. Not only do stuns increase with each tier, but they increase faster with each tier, too. In other words, the rate of stunning is accelerating. One possible reason for this is that the number of dragons (notably still with their stunning dragon fear power in Monster Manual 1) increases in density as tiers increase, but that alone probably doesn’t explain it. At any rate, the story of increase here is different from that of the story of increasing daze when you compare the results to the players. While for daze the players started with more, then accelerated away to create an enormous daze gap by epic tier, in this case it begins with a gap of 0% players to 2.5% in favor of the monsters in heroic tier, slightly closing (player acceleration) 3.4% to 5.7%, and then 6.6% to 10.6%, with the monsters pulling away into a clear lead by a wide margin. Why so many stunning powers in epic tier? Besides the dragon density theory I don’t know, but there they are. The final of the big three, we’ll next look at Dominated.


Now then, Dominated is interesting because it seems to exhibit characteristics shared by the dazed chart. The amount of the dominated condition per power basically doubles between heroic and paragon, then in epic dips down to a figure between the first two – just like in the dazed chart. Meanwhile, the fluctuations in the amount of domination that PCs get per power goes from .65% to .75% to 1.2% – higher than the monsters overall, and specifically in both heroic and epic tier. That being said, dominated is an extremely rare condition, not only for the monsters, but also for the players.

Now I’ll share some charts detailing similar progressions for other powerful conditions (blinded, removed, restrained, unconscious, and weakened), as well as how they compare to the players’ options of the same.

Here’s the chart for blinded, and it, too, exhibits that sort of strange dazed progression, where it peaks in paragon tier. The comparative percentages per power per tier between the players and monsters are 1.7% to 1.1%, 2.3% to 1.3%, and 5.8% to 1% respectively. In other words, the players’ blinded options per power increases tier (and accelerates), while the monsters’ just sort of hangs out around 1.1%-ish.




Meanwhile, removed as a condition exhibits a growth similar to that of the stunned condition, increasing in prevalence per tier. When compared to the progression that the PCs get, it is, PCs first, .22% to .23%, 0% to .44%, and 1.2% to 1.0% per tier respectively. There seems to be a bit of an outlier in paragon tier, but otherwise the PCs and monsters get a similar amount of restrained and also a similar amount of increase (again with the exception of paragon tier).





Next up there’s restrained, which displays another one of those quirky paragon tier maximums, with epic tier falling between paragon and heroic. Compared to the PCs, again with the PCs first and respectively by tier, are .87% to 1.4%, 1.1% to 2.9%, and 2.9% to 1.5%. The monsters begin ahead, then stay ahead, until finally the PCs take first place in epic tier.




Penultimately there’s the unconscious per power chart, with the monsters displaying, again, their epic tier not progressing reasonably onward, but instead peaking at paragon tier. However, this time at least, it also displays another curious behavior – namely that for the first time epic tier is not only less than paragon tier, but also less than heroic tier. Hmm, curious. We’ll discuss that momentarily, however, and for now I’ll just note that the player vs. monster comparison is .22% to .23%, .38% to .88%, and 0% to 0%.



 The final chart of serious effect per power in percent (always avoid alliteration) is, as mentioned, weakened. This one seems to increase as one would expect, unlike several notable exceptions, and additionally the PC vs monster per tier comparisons are .87% to 1.6%, 1.9% to 4.1%, and 5.3% to 5.0%. While the monsters had the PCs on the ropes for a while, it seems that team PC pulls ahead in the end once again.




Now, what does all of this mean? Well, it could mean a lot of things, but the first thing I noticed is that neither side really has a decisive advantage in terms of the big status effects – ones that cause large penalties or the like. The monsters’ effects where they have a slight advantage seem to be the slightly more dangerous ones (removed, stunned, and weakened), but the PCs aren’t exactly far behind, and to boot have an advantage in dazed and blinded. In other words –

  • Premise 3: The above increase in Power Options is at least equal to that players experience.

Premise 3 is looking fairly supported in terms of rate of increase regarding powerful effects. The monsters and players both get increases to their effects at roughly the same rate, with perhaps a very slight edge going to the players in terms of rate of increase, but the edge going to the monsters in terms of the power of the condition inflicted. Of course, just a little while ago I was saying how premise 3 was unsupported since the players get more options at a more rapid rate overall, so what gives? Obviously the players must be getting options that are not the effects I just listed in much higher quantities than the monsters are, and, on the whole, this is true. You can see it most clearly in the case that monsters get many powers that deal only damage, while players get very few of those powers – the powers that, were they monsters, would deal only damage must, therefore, be acquiring lesser power options (like bonuses or movement) that are not major status conditions, while the monsters have many powers that deal only damage. In other words, while the sheer power of the options, and the amount of the powerful options, aren’t in either party’s favor, the less powerful options are distinctly giving the edge to the party, and therefore overall premise 3 seems to be unsupported in regards to quantity.

As for –

  • Premise 2: As level increase monsters get more powerful Power Options.

Well, frankly that one looks kind of supported and kind of busted. Of the eight status effects listed, the monsters had a peak in their relative quantity in paragon for five of them (dazed, dominated, blinded, restrained, unconscious), while they had a peak for three (stunned, removed, and weakened) in epic tier. The trend is definitely there as the monsters roll into paragon tier, but premise 2 seems to be weak as the monsters reach epic. In fact, the difference in percentage points between heroic tier in these eight status effects is, all together, 14.59. On the other hand, the difference between paragon tier and epic tier is, net, only 1.28. The premise is still supported so far, but only barely. Meanwhile, the players have a difference in percentage points between heroic and paragon tier for these eight effects of 11, and paragon and epic tier of 14.57. In other words, while the players get more acceleratively powerful (that is, their power increases, and increases at a more rapid rate than heroic) in paragon tier, and continue this acceleration through epic tier, where they gain power at a faster rate than paragon tier, the monsters… well, they don’t. The monsters start to slow down again as they reach epic tier. This brings us back to –

  • Premise 3: The above increase in Power Options is at least equal to that players experience.

which is now definitely busted. The players gain power at a much faster rate as the game goes on than the monsters, with rough parity in paragon tier, then the players pulling away like the first place horse at the kentucky derby once they reach epic tier, not only overall, but also for the eight most powerful (debateable I know) status effects.

Therefore, hypothesis 1 is, primarily, false. While the monsters do have a lot of complexity to them, and this complexity is comparable up until paragon tier, at the end of the day the players pull ahead far and away. Epic tier combat is most definitely more complex on the player end by a wide margin in terms of status effects and other power options within the limited At-Will, Encounter, Daily / Standard Action scope of this study.


Now to begin on hypothesis 2 – that certain kinds of power options are intrinsically tied to certain kinds of monsters. This is useful to see if the roles are really different, and if so, how they are different in terms of the power options they possess. The player study showed that this was less the case than one could imagine, so it’ll be interesting to see how monsters work out. To this end, I have created two charts. They are both color coded for convenience, with associative power options labelled in red (>50% increase in prevalence compared to average), dissociative power options labelled in blue (>50% decrease in prevalence compared to average), and null options (those that do not appear at all) labelled in purple. All other boxes (results that fall within +-50% of the average) are gray. The first chart is the percent of monsters that have a certain power option, and the second chart is the percent of powers that have a certain power option, both charts broken down by role.

Basically, we’ve got a lot of results here and we need a way to cull from the herd. A good place to start, or so I think, are where the associative results line up. In other words, where both a lot of the monsters have powers with that result, and a lot of the powers the monsters possess have that result. For instance, the Controller role has several that line up: Blinded, Dazed, Dominated, Immobilized, Removed, Slowed, Stunned, Combat Advantage, Healing, and Penalty. A rough theme here are heavy action denying effects and numerical penalties (dazed, dominated, removed, and stunned denying actions hard, immobilized, combat advantage, slowed, and other penalties denying softly). The controllers exert control, who would’ve thought? They are also defined by being above average in the most effects and power options overall (both individually and together). However, what truly defines a trend? Is it a power option appearing 5% of the time? 10% of the time? 20% of the time? I decided to arbitrarily make a cut off at 10%, because in my mind if every other encounter made entirely of that role doesn’t necessitate showing a power option, then the power option is probably too rare to truly define the class of monster with. Do you think that another percentage should be used? Feel free to comment below and justify your response! However, using my arbitrary cutoff for now, we can see that a controller is “truly” defined by only a few power options: Slowed and Penalty on a per-power basis (with Dazed and Immobilized coming close), and Bonus, Penalty, Healing, Movement, Weakened, Stunned, Slowed, Immobilized and Dazed on a per-monster basis. Therefore, a rough “most defined by” to “least defined by” for controllers might look something like this (using only associative results):

  • Controller – Defined by: Slowed and Penalty, Dazed and Immobilized, Movement, Stunned and Healing, Weakened and Bonus

The other roles are defined a lot more tightly by their power options than are the Controllers, with the possible exception of Lurkers who come close. Therefore, let’s just work from top to bottom. Lurkers have as associative on the first chart: Blinded, Grabbed, Removed, Unconscious, Weakened, Penalty, Bonus, and Other Effect. Other Effect? Well, that’s not good. Let’s find out what that effect tends to be. According to the notes I took while creating the charts of data, Other Effect for Lurkers fell into approximately 3 major categories: caster becomes Insubstantial, caster becomes Invisible, and far behind those two, caster creates Line of Sight blocking Zone. Since Other Effect covers a wide range of effects in this case we’ll ignore it, but make a note that Lurkers are prominently able to become Insubstantial and Invisible. From the second chart the associative results are Blinded, Grabbed, Removed, Unconscious, and Weakened, as well as Penalty, Bonus, and Other Effect. The options that meet the 10% cutoff are Penalty and Other Effect on chart 2, as well as Blinded, Grabbed, Weakened, Penalty, Bonus, and Other Effect on chart 1. As a result of this, a Lurker defined by power options might look something like this (as above):

  • Lurker – Defined by: The ability to become Insubstantial or Intangible, Penalty, Bonus, Weakened and Grabbed and Blinded, and (barely) Removed and Unconscious

To continue on to the next most defined by its power options in terms of quantity is the Artillery role but with a distinctive dip in the amount. Artillery is defined on chart 1 by being associated with Dazed, Dominated, Petrified, and Unconscious as well as No Effect, and on chart 2 with Dazed and Dominated. In terms of the 10% cutoff, on chart 1 the most prominent associative options are Dazed and No Effect, and on chart 2 there are no options that meet the cutoff, although Dazed is close. When defining artillery, normally I would put Dazed first, as it meets the cutoff on one chart and is also associative on both charts, but the No Effect power option on chart 1 is so far and away above 10% occurance (in fact, rather than occuring once in every 10 monsters as 10% indicates, it occurs approximately 1.5 times per monster) that it only seems prudent to use it as Artillery’s defining aspect. Therefore, Artillery is most clearly defined like so –

  • Artillery – Defined by: No Effect, Dazed, Dominated, Petrified, and Unconscious

Next is Soldier, again with many fewer defining options than Controller or Lurker. In the Soldier’s case, it is associative on chart 1 with Grabbed, Marked, and Petrified, and on chart 2 also with Grabbed, Marked, and Petrified. In terms of the 10% cutoff it has Marked on both charts and Grabbed on the first. Therefore –

  • Soldier – Defined by: Marked and Grabbed

 I’ll skip boring you to tears by listing out the remainder and simply present them in their abridged, bolded, bullet-pointed format (together with the ones that are already done). They were done in the same manner.

  • Controller – Defined by: Slowed and Penalty, Dazed and Immobilized, Movement, Stunned and Healing, Weakened and Bonus
  • Lurker – Defined by: The ability to become Insubstantial or Intangible, Penalty, Bonus, Weakened and Grabbed and Blinded, and (barely) Removed and Unconscious
  • Artillery – Defined by: No Effect, Dazed, Dominated, Petrified, and Unconscious
  • Soldier – Defined by: Marked and Grabbed
  • Skirmisher – Defined by: Movement, Combat Advantage
  • Brute – Defined by: Prone

So, in essence, what are we looking at here? Well, contrary to the players, it seems that the monster classes are usually all pretty clearly defined. The controller slows and imposes penalties, as well as restricting movemement and action in other ways, while the lurker can become insubstantial or intangible and inflict wicked penalties. Artillery is primarily about the damage, soldier about marking, skirmishers about movement, and Brutes about prone. However, associative power options only tell one half of the story. What about the dissociative power options? Well, here are those, listed again for your convenience (and determined in much the same way, this time ordered from most anti-definitive to least) –

  • Controller – Anti-Defined by: Marked
  • Lurker – Anti-Defined by: Marked and Petrified and Combat Advantage, Immobilized,  Movement
  • Artillery – Anti-Defined by: Grabbed and Marked and Removed and Combat Advantage,  Bonus, Healing, Prone
  • Soldier – Anti-Defined by: Dominated and Unconscious and Combat Advantage, Blinded, Penalty
  • Skirmisher – Anti-Defined by: Blinded and Dominated and Immobilized and Marked and Removed, Grabbed, Dazed, Penalty, Stunned
  • Brute – Anti-Defined by: Blinded and Dominated and Marked and Petrified and Removed and Unconscious and Combat Advantage, Immobilized, Bonus, and Slowed

So, regarding what makes a monster class unique, what can we learn from all of that? Basically, that Controllers are defined by their ability to detain enemies from doing as they please (movement penalties and action penalties, as well as just regular old penalties) as well as able to do a lot of different things overall – the only condition they cannot inflict at least some of the time, besides those that no monster in this study can inflict, is Marked. The Lurker, too, is defined by its penalty infliction, but in a different way – it inflicts primarily numerical penalties (and bonuses to itself), as well as having the ability to become invisible and/or insubstantial very frequently. The Artillery monster is defined by its ability to … have very few abilities. It primarily just deals damage, but it has a touch of control in there as well (most notably Dazed). Meanwhile, it has quite a few things it can’t do without any real solid theme it seems. The Soldier, on the other hand, has very few things it can do that define it: basically just Marked and Grabbed. In contrast, it also doesn’t have many things it can’t do, with the majority of its options being gray rather than red or blue. Skirmishers are clearly defined by movement, as one would expect, as well as combat advantage (although much less clearly defined in terms of inflicting it as a condition), also as one would expect. Meanwhile they have very little ability to impede the movement or action of others. Finally, Brutes knock things prone a lot, but otherwise are fairly limited in what they can accomplish. Their rate of No Effect powers, while not as prominent as Artillery, is basically tied for second with the Skirmisher. How does this all affect Hypothesis 2? –

  • Very few Power Options are intrinsically tied to the different monster classes.
  • Premise 1: Assuming “intrinsically tied” is equivalent to appearing 50% more often than the average both per power and per monster, and that the Power Option appears at least 10% of the time overall both per power and per monster.
  • Premise 2: Assuming that the monster classes are Controller, Soldier, Brute, Lurker, Artillery, and Skirmisher irrespective of their Minion, Standard, Elite, or Solo status.

Well, using premise 1 we created those charts and analyzed what they were all about, and also using premise 2 we ignored the monsters’ relative worth to their same-levelled comrades. In terms of the hypothesis itself, I would say that, contrary to my expectations, monsters are fairly well defined by, and tied to, their power options on a role basis. Controllers clearly use action and movement denying effects much more often than any other role, Lurkers are the only role that really has any consistency in its ability to become insubstantial or intangible from a power, Artillery has tons of powers that are just damage and sometimes daze, Soldiers are defined by their mark and grabbing (and also by their relative ability to do anything), and Brutes and Skirmishers both have a schtick to (prone and movement/taking advantage of CA respectively), as well as being very clearly planted in the second place spot of No Effect. The answers aren’t as definitive as I might’ve liked to see, but on the whole the trend of these monsters seems to point towards certain power options being tied heavily to certain classes. In fact, at no point do any of the red (>50% more prevalent) power options go to more than two of the monster classes at a time. I’d say that’s a pretty big success on the part of the design team for the monsters in terms of making them distinct from one another based on role.

That leaves only one point of discussion, namely how the concept of anti-grind monsters fits into this general scheme alongside the relative strengths and weaknesses of various power options to that end, and how that relates to the existing data.

The most common advice for using monsters that “don’t cause combat to be a grind” basically runs along the following lines: Whatever you do, don’t use soldiers or controllers, and instead use artillery, skirmishers, and brutes. Lurkers are kind of a wild card that are situationally useful, primarily because they will often be the first to run away thereby not lengthening combats needlessly.

So, do I see any statistical trend that might support this qualitative analysis? In fact, I do. Artillery, skirmishers, and brutes are all very high on the No Effect category. To pull some numbers from earlier in the article, the artillery role has a full 53% of its powers not inflicting any kind of status effect or power option of any kind. Similarly, brutes take second place at 46%, and skirmishers a close third at 41%. In contrast, soldiers only have 32%, and controllers only 25%. Ouch! Lurkers, the “wild card” are sitting at only 27%! But that’s just in terms of powers, let’s take a look at it in terms of monsters, which may paint an even more damning picture. Across all three tiers an artillery monster will average 1.64 powers each that do not inflict any power options, with skirmisher taking second this time at .96 (only 4 monsters in a hundred do not have a power with No Effect), and brutes at .88 (12 monsters in 100 don’t have a No Effect power). In sharp contrast stands the controller at .78 (22 monsters in 100 don’t have a No Effect power), the soldier at .69 (31 monsters in 100 don’t have a No Effect power), and the “wild card” lurker at .67! If you want to select a controller, soldier, or lurker monster they are sabotaging your efforts to avoid effects in comparison!

Furthermore, when analyzing the stunned effect in the same manner, it can be clearly seen that so many controller monsters have one or more powers that stun it pulls the average up such that the average amount of stuns is higher than the amount of stuns on any other class of monster. Removed? Controller and Lurker. Unconscious? Lurker. Marked? Soldier. Immobilized? Controller and Soldier. Each of these conditions that denies or heavily suppresses actions has one of the “no-no” classes linked directly to it such that it pulls the average up so far it is greater than the individual statistics of any other class. The link between the “grindy” classes and power options is clear and irrevocable, not only in the direction of certain classes of monsters having fewer power options in general, and more powers that have no effect, but also in the direction of certain other classes having many different power options feature prominently, many of those being the most powerful at denying actions. Similarly, the Lurker only has its wild card status precisely because of the fact that it is commonly the monster that runs away. If it stayed it would prolong combat immensely with its insubstantial conditions, ability to hide due to invisible, and a plethora of action denying effects that commonly appear.

So, why then do these status effects increase so heavily? Well, the monsters themselves are getting more complicated, it’s true, but, perhaps more importantly, something else is changing besides the powers’ complexity. Specifically, the kinds of monsters available is changing. Observe –


As this chart clearly illustrates, one potential reason that monsters become more complicated at higher tiers is simply because the kinds of monsters with the No Effect powers go away, and in their place we are given monsters with tons of power options. While of course epic tier has fewer monsters overall, the majority of those monsters are soldiers, brutes, and controllers. Notice how there are brutes there, though. This gap is particularly large in paragon tier, where brutes are at an all-time lowest percent of the monster make-up, and in fact is probably the most likely explanation for why some power options seem to peak in paragon tier and decrease in epic tier – the kinds of monsters that have them, controllers and soldiers, are most prevalent in paragon tier and therefore so are their conditions!


Hypothesis 1

  • Combat includes too many moving parts, and this is the monsters’ fault.
  • Premise 1: As level increases monsters get more Power Options.
  • Premise 2: As level increase monsters get more powerful Power Options.
  • Premise 3: The above increase in Power Options is at least equal to that players experience.

Neither proved nor disproved, although with fairly clear results to each premise.

Premise 1: Monsters become more complex by acquiring more power options as they level up on the whole. This is true. Additionally, they acquire more powers in general (although the rate of power options per power still goes up). The rate of power option increase (overall) is approximately linear as evidenced by this graph –

Premise 2: Monsters do get more powerful Power Options on the whole as they level up, but not by a lot in epic tier. Monsters mostly only get non-standard action non-power option bonuses as they enter epic tier, with their standard action power option related statistics remaining fairly close to their paragon tier assets.

Premise 3: This premise is false, as I predicted, thereby invalidating any total conclusion that could be made about the first hypothesis. Looking at a graph of player power versus monster power (by role), we can see some pretty clear trends here –

and also clearly see that monsters are all over the place while players are not. While this bodes well for monsters being definitive and iconic (and players having a crapload of power options at their disposal), it bodes poorly for any average comparison. Still, using the previous chart showing the linear increase of average monster power options, we can take a look at the player chart and see that –

Power Options Per Tier (1, 2, 3) (Player)

it is quite different looking, and definitely not linear. While at first it might appear exponential, were you to analyze only the rise from Heroic and Paragon tiers, taking all three tiers together the graph appears very much to be logarithmic. So, while it’s fun to yell about how “monster progression is linear and player progression is exponential!”, it turns out that, in fact, monster progression is linear and player progression is logarithmic. What does that mean for premise 3? Basically that player power options per power increase faster than they do for monsters, most noticeably in heroic and paragon tiers, while over the course of epic tier player and monster progression is roughly similar in terms of quantity, although by then the damage has already been done, and both gaining a between 2 and 2.7% increase just serves to further exacerbate things to the breaking point.

Hypothesis 2

  • Very few Power Options are intrinsically tied to the different monster classes.
  • Premise 1: Assuming “intrinsically tied” is equivalent to appearing 50% more often than the average both per power and per monster, and that the Power Option appears at least 10% of the time overall both per power and per monster.
  • Premise 2: Assuming that the monster classes are Controller, Soldier, Brute, Lurker, Artillery, and Skirmisher irrespective of their Minion, Standard, Elite, or Solo status.

This hypothesis, surprisingly, was proven false. While the four player classes in The Id DM’s study did not experience a significant divergence in the amount of power options they had available to them the monster classes most certainly did and, indeed, several of them are practically defined by their relative lack (or relative frequency) of power options, and still further many of the classes of monster are defined by the amplitude of certain key power options. (While the same can be said for the players in terms of amplitudes of specific power options, it is not quite so exaggerated).

General Questions

What, then, does this mean for us? Well, a lot of things. Firstly, that basically any player class can do complicated things but that monsters are much more often defined by a lack of power options. Similarly, the anecdotal evidence from experienced dungeon masters seems to tell us the same story that the statistics do – if you don’t want to slow down combat then stay away from the monsters with tons of power options and especially the power options that deny or suppress actions.

And now, to ask and answer the same discussion questions that The Id DM posed (they’re still applicable to this article, after all, and while good DMs borrow, great DMs steal!) –

  • Should status effects be tied to specific classes and roles to increase identity?

In the case of monsters, I think that definitely yes. My experience with monsters tells me that having each monster have a clearly defined role on the battlefield not only helps me plan ahead, but also gives the players clearer information. If that monster attacks at a range and is doing a lot of stunning, it’s probably a controller, and therefore it’s probably a priority target. If it’s attacking up close but isn’t marking? It’s probably a brute. For the player classes, however, I’m not quite so sure. While one would suspect some iconic powers to make clear appearances in certain classes, like the marked condition for defenders, a lot of the power options available aren’t particularly tied to the ideal of any one class or role. This is a very delicate in large part, I think, because it’s important to give players the option to play the game the way they want to. If they want to be the defender, do they want to mark? Or do they want to spam prone and immobilized? Both are effective ways to defend your fellow party members, but they both work very differently. I think, maybe, that following the Essentials idea moving forward is probably the best way to handle it, and by that I mean that an essentials class is, in essence, a focused build. The mage isn’t very focused, aside from its school encouraging you, that’s true, but a controller’s thing is having access to tons of different options, right? Meanwhile the Slayer’s schtick is No Effect, No Effect, No Effect, and the Cavalier has two very clear choices to make: either you’re Virtue of Sacrifice and get Healing and Bonus (to defense), or you’re Virtue of Valor, which gets primarily just Bonus to offense. In short, I’m in favor of tightening up the looseness of the power spread, and making controllers feel like they’ve got the run of the gamut while their friends probably don’t.

  • How special do you view Status Effects as a DM and as a player?

Well, this one really depends for me, at least, on the status effect (or other power option). It also depends on the way I’m playing the game, which these days is most often online over Map Tool. One thing I’ve noticed, for instance, is that while every other power option was definitively belonging to a monster role in my analysis, ongoing damage and restrained weren’t. That’s a problem in my book. If it’s too muddied to be iconic it needs sprucing up, especially if it’s a prevalent as ongoing damage is. What is the point of ongoing damage, anyway? It’s backloaded damage, so that takes a bit of planning, and I like that in terms of how it can increase depth and reliability, but it relies on a saving throw to go away, so it’s still totally random when you get to the point in the game where every other leader power is suddenly granting saving throws. Will the ongoing damage go away? Won’t it? How can we know for sure? There went reliability and predictability, and as a result there went depth. It’s still adding complexity to the gameplay, but it’s a bad kind of complexity, the kind that can’t be predicted or accounted for with any certainty. Similarly, while Stunned is a ridiculously overbearing effect (it removes your entire turn!) it never felt that menacing to me, just annoying. My players have never said “oh shit, he can stun!” What they have said is “it rolled HOW MUCH DAMAGE?” I think a lot of this stems from the fact that unlike in some other games, most notably Star Wars: Saga Edition or even Spirit of the Century, you cannot control somebody to death in 4th edition. It’s all about the damage. If you stun somebody six times in a row, they are still at full hit points, which means the instant they roll their 55% chance of success saving throw, they’re right back into the fight like nothing ever happened. In other words, being stunned has unintuitive consequences. In SWSE you can stun somebody two or three times, and suddenly their condition track hits bottom and they’re unconscious. That’s a fight winner. Similarly, in Spirit of the Century, if you hit somebody with knockout gas a couple times and their composure bar fills up past a major consequence? They’re taken out just the same as if you shot them with a gun. So why does control not have a cumulative effect in 4th edition? I’ve heard people say “oh, well, SWSE’s control track is a death spiral, and that’s not very heroic,” but what about Spirit of the Century? It’s a game about pulp action where solving problems with your fists is just as viable as with your wit and that has cumulative control effects out the wazoo. I think that’s a major problem with 4th edition right now, actually, the more I think about it. If being Stunned twice suddenly turns you unconscious, Stunned is a lot more scary and a lot less annoying. “Oh no I cannot act whatever will I do” is a lot less enthralling than “if he stuns me one more time I’m done for!”

  • Should the majority of powers have additional benefits or simply cause damage?

This one, for me at least, is the trickiest to answer. Simply causing damage seems to work just fine for the monsters and the slayers of the world, so why not? On the other hand, lots of players complain that things like the slayer feel too boring or simple (and since they have to/get to play the same character for weeks on end while as DMs we constantly change our oppositional cast to new things, I can see why this complaint comes up). I think that, probably, my ideal balance here would be somewhere between the essentials material and the core material (pre-essentials). Basically I think that maybe they went a little overboard with the new essentials material to make it more distinct than it necessarily needed to be, and the real sweet spot would be toned back just a little bit.

  • What did you find interesting or surprising in the results above?

At slightly over 8,000 words, I’m going to have to take the cop-out answer on this one. Chalk one up for “all of it” from The Id DM’s article, and in terms of my own? I’m most surprised by how, while the players continue to increase in power options at a rapid rate among the most powerful of the options (the so-called top eight I listed earlier) the monsters really slow down to an absolutely glacial pace. They get more power options in epic tier, and more powerful power options in epic tier, but the former is less than what the players get and the latter is less than what the players get by a LOT.

Also, whoever designed the Bodaks somehow managed to sneak the most crazy overpowered creatures to ever exist into a folio of fiends most would call underwhelming in the danger department at best.

How about you, dear reader? What do you think of those questions? Do you have any other thoughts to share? Do you want to contribute to the ongoing call to arms to acquire statistics started by The Id DM and maybe even help with figuring out how the game is designed, why it’s designed that way, and how it could be designed better? Please leave a comment below!

As promised, here is the original excel spreadsheet with the data I have collected, and here is the “glossary” for the excel spreadsheet with my notes I took as I was taking down the data. Both are very helter-skelter, I’m afraid.

Tune in sometime next week when I analyze the changes that slightly over two years of design innovation and revision have brought to our beloved monsters when I do a follow-up article on the D&D Essentials Monster Vault.


The Hydra DM

Exploring Complexity (Player Characters)

Hello again!

Much like another recent article I wrote, I wanted to take some time to expand on a topic that a fellow blogger (or several, although in this case just one) came up with. Specifically, a piece that The Id DM wrote titled “Player Characters Are Gods“. In that post, which I highly recommend reading for yourself in addition to my short summary, he basically puts forward that the number of options, and the power of those options, make player characters in Paragon tier and beyond get completely out of hand, potentially slowing combat to a halt with analysis paralysis and option overload (link courtesy of an old article written on the topic of option overload by AngryDM) in addition to their numerical superiority (in the sense of more plusses) to monsters. I will be focusing on expanding the former claim rather than the latter.

In this article, I’d like to take the example set by The Id DM and take it to its logical conclusion. While he presents J’hari Wrex (his level 12 dragonborn rogue scoundrel) in a fashion that did not include quite a few of his options, most notably those not related to combat I’ve decided to go whole hog and steal liberate a core fighter (also known as the weapon master) build from the 4th edition character optimization forum for sake of having a “good build” (i.e. a pretty reasonably strong level 30 character that you might expect a good player to make if he had a lot of time on his hands and happened to enjoy character optimization in addition to following the “recommended wealth” guidelines in the DMG). I also built this same character as a level 1 version of himself to see how the stat blocks compare, and made sure to include all of his powers and various options. For sake of thoroughness, since I included traits that were granted by feats, I also included traits such as the base proficiencies, etc. Are you ready to see the difference in complexity between a level 1 and level 30 character? Here we go:

Level 1

Level 30

Before I go on, I’d like to apologize for the inability to include one of the paragon path powers in the level 30 stat block, which if you take the time to read it thoroughly you may notice. The monster builder tool actually couldn’t manage to wrap itself around such a complicated power (with a tertiary attack) without crashing, so an approximation will have to do. I could have included a few other things for clarity (like a specific opportunity attack power), but most monsters rely on traits to tell the DM about changes to their opportunity attacks so I followed the same sort of paradigm here as best I could.

The Conclusion

Not only do the number of options increase (even with many powers being replaced by better ones instead of supplemented by them at higher levels), but their complexity does as well as many of them begin to become triggered actions, which is a common complaint among some of the more prominent bloggers like Mike Shea and Robert Schwalb (that is, the complaint is about too many out-of-turn actions slowing combat way down in epic tier).

The Numbers –

Level 1 Fighter

Traits – 7

Standard Action Options – 6

Triggered Action Options – 1

Other Options – 1

Level 30 Fighter

Traits – 31

Standard Action Options – 12

Move Action Options – 1

Minor Action Options – 7

Free Action Options – 6

Triggered Action Options – 6

Other Options – 2

The grand totals?

Level 1 Fighter – 15

Level 30 Fighter – 65
A level 30 fighter, although this might be a bit presumptuous, looks to be around four times as complicated as a level 1 fighter, going with the definition of complexity as being “number of parts”. With a character four times as complicated, it’s reasonable, then, that perhaps it can take up to four times as long to figure out which one(s) to use… and that hypothesis seems to bear out, according to the experiences of several epic tier running DMs.

Plan of Action

How can you reduce the ridiculously overbearing complexity? Well, there’s really only a two notable options, both of which came out closer to now than they did to 4th Edition’s launch, which bodes well for the designers recognizing some of the big problems.

1) Essentials material.

Essentials characters do not get this complicated, that’s basically just how it is. There are a few exceptions, like the Mage, that can… but most don’t. This, of course, has the drawback that players will feel somewhat slighted as you are stripping them of potential options. Of course, whether we know it or not, sometimes fewer options is better. In fact, sometimes having no choice at all is the best thing for us, whether we believe it or not!

2) Fixed Enhancements/Restricted Magic Items.

With some vigilance, you may note that of the 65 options that exist to the 30th level fighter, 6 traits, 2 standards, 2 minors, 5 free actions, and 3 triggered actions of those (for a total of 18 options, approx. 28% of the options, and further 5 of the 8 added out-of-turn options that were gained since level 1) are directly related to his complement of magic items. That’s quite a lot! Cutting down on magic item options may make you somewhat more popular with your players than cutting down on class options.

Put together, the two of these can seriously reduce complexity at epic tier, which can result in seriously enhanced play because there is much less of an issue with the players being so overwhelmed with options they don’t know which end is up.


A level 30 fighter is, approximately, four times as complex as a level 1 fighter and that can cause a lot of issues as the players become overwhelmed in a deluge of options. A lot of these options come from the At-Will, Encounter, Daily, Utility (AEDU) design structure of the first three Player’s Handbooks, and a lot of them come from the absolute downpour of magic items that is normally accumulated (per the DMG’s suggestion, at least) by level 30. By removing these two major contributors to the ballooning of complexity (or at least one of them) you can drastically improve the play experience in epic tier. In theory, anyway.

The Hydra DM

Building a Better Skeleton

Well, I was going to do a post about my latest set of sessions… but, frankly, they aren’t that interesting. The intrepid heroes are sucked into the mind of a crazed aspect of death, who in turn is stuck inside a prison composed of a Tiefling’s body, who is in turn stuck inside a prison composed of – well, you get the point. Pretty usual stuff.

Anyway, I thought I would make this post about monster design, one of my favorite topics by far. Why is it my favorite? Because, designing monsters brings me back to that time in my youth when the defense scores really meant something individual and unique, and I try to hold on to what’s left of that feeling while I design new monsters, villains, creatures, and other such things with stats for my players to interact with. It’s a lot more work than simply slapping some numbers down from a chart and away you go, but it definitely makes you really feel like the creation has a rationale behind it if you have to justify every point in each of its abilities.

So, without further adieu, I present to you three of my new and improved 4th edition skeletons, including all of the reasoning that went into each of the decisions.

Okay, well, a little adieu may stand between you and the stat blocks. I have to give a qualifying statement: these were designed with a bit of a hardcore streak in mind. They are on the tougher side of balanced, at least so far as I can tell without having used them yet, since they are designed for a campaign that is is made with a sandbox feel and “hard but fair” attitude as its core concepts.

So, the first place I decided to start was with a skeleton archer of some sort. When I think skeleton, for some reason that’s the first thing that springs to mind. But, before I could just start hammering out what a skeleton archer’s abilities are, I first had to decide what it meant to be a skeleton at all. For this I went through some works of fiction, not least of all the 3.5 edition monster manuals, pathfinder monster entries, and 4th edition monster manuals. I figure the best place to start figuring out what it means to be one of these creatures is in their ability scores and defenses, so that’s where I began.

First of all, 3.5 skeletons were rather frail (though resistant to several common types of damage, such as piercing, not to mention their enormous undead immunity, which is a rather large battery of effects against which they are flat-out immune), but they were very fast little buggers indeed. Their speed was quite high and their reflex and AC were, too. Like most undead their will defense was rather spotty and their fortitude defense was certainly nothing special. I then proceeded to pathfinder, where to my surprise they had decided something different: their skeletons would have a fairly GOOD will defense, in addition to their reflex defense, while leaving the fortitude defense high and dry as usual. This was very interesting and I can see where they were coming from – it’s pretty hard to convince a creature whose sole purpose is to maul you out of existence to, in fact, not maul you out of existence. On the other hand, 4th edition seems convinced that, like late 3rd edition, it was a good idea to have the will defense be low on all of the undead critters. They all seemed to agree that their hitpoints weren’t too sturdy, once you hit them they went down, and that their reflex defense and AC were quite good, as well as their other battery of defenses. It seems that skeletons, therefore, should specialize in avoiding damage, not taking it. As for the question of will defense, I decided after some deliberation that if they were easy enough to control that a necromancer could keep a stable of the things it seems likely that they would be easy enough to harm mentally that their will defense should be pretty low, thereby going along the 3.5 and 4th edition routes, rather than that outlined by Pathfinder. So, I had figured out that their HP, Will, and Fort defenses should be low, and their AC and Reflex should be high, which coupled with their battery of immunities and resistances would probably keep them rather safe from a lot of damage, but not so safe from other types. I didn’t want to reinstate their crazy immunity to piercing damage, though, mostly because of things like rangers tying rocks to the heads of their arrows in a vain attempt at dealing damage without resorting to fisticuffs. I figure a ranger should always be good at shooting things with arrows, since that’s kind of their thing, and to take that away from them would be a bit too on the mean side. Sorry damage reduction/resistance/immunity traits, but you’re going to have stay with previous editions on this one. As a final point, skeletons, unlike other undead, were usually keen on using weapons and armor (albeit rotten), so of course making an unarmed skeleton seemed kind of pointless since what makes them unique is being users of tools, unlike most zombies or ghouls or wraiths and the like.

Once I had their defenses figuring the ability scores out was pretty easy, really. A high dexterity score gives you a good cause for high movement speed, and a fairly low charisma and wisdom score makes sense for such a low will defense. I figured they were pretty much average-ly strong and tough (much more brittle than zombies, that much is for certain, though for a time I debated even that since skeletons are technically magical constructs more than they are animated corpses, but I figured if you broke the bones you broke the magic so they were more brittle), so they got pretty middle of the road strength and, since this is 4th edition, constitution scores (undead with Con? Blasphemy I know :p)

Once that was done the next part of the process was coming up with more flavor for the creations (although the Dungeon Master’s Guide emphasizes assigning the stats before going in depth on the fluff, I figured if I knew what I wanted to make right down to the letter that figuring what stats to give them would come much more easily). On a bit of a joke I suggested to one of my colleagues that “man, the skeleton archers just aren’t as good ever since they lost both of their eyes, no depth perception anymore…” which is when I had my first idea: skeleton archers would reasonably use a much easier to use weapon than a normal bow. That is, they would use a crossbow. Point and click is a lot simpler to aim, and since there’s no bow oscillation there’s not as much drift introduced into the arrows. This would let the shoddier skeletons who aren’t imbued with the best of sight-giving magic to hit things with more ease, and as a side effect to deal significantly more damage (taking my inspiration from the historical crossbow), albeit at a slower rate of fire. Still, skeletons are plentiful, which led to the inevitable comparison of peasants with crossbows in the middle ages killing knights to skeletons in D&D with crossbows killing… well… knights! With this perfectly sensible plan for our surprisingly intelligent and genre-savvy necromancer (apparently), I had the basis of my skeleton archers – a “more advanced” model, that is, a standard enemy with a crossbow, and a “less advanced” model, that is, a minion enemy with a shortbow (if you had to give a longbow to every skeleton that’s like a 40% budget increase on wood! Simply too much for our thrifty necromancer to handle, unfortunately).

Selecting a template at that point was rather easy: an artillery role would suit this monster obviously (as that is, in essence, the “ranged dps” monster role in 4th edition), and for the minion I would simply minionize it by giving it an average quantity of damage and 1 hitpoint. Off to make some abilities, then (after assigning stats, of course, which you can see at the end of the article in the stat blocks).

The first ability I made was, reasonably enough, the crossbow-wielding skeleton’s crossbow bolt attack. Since a real crossbow, where I got my inspiration, fires at around a 60% rate of fire compared to a longbow of the same period (which fires 12 arrows per minute), I figure that I would reduce the amount of times it could attack. But how should I do that? Should I give it a reload action, much like a PC’s ranged weapon would have? Probably not, since that means it could fire every round, which means I couldn’t justify a damage increase statistically speaking. On the other hand, utilizing a new kind of ability not available in previous editions of the game (though it’s old-hat by now, especially to 4th edition DMs) known as a “recharge” power set to recharge on a roll of 2+ I came pretty close to simulating the rate of fire I wanted. It couldn’t shoot EVERY round, but it would shoot MOST rounds on a per-minute basis. Since it was a recharge power, this also gave me the statistical lee-way to increase its damage more than what an artillery monster, with its fairly high accuracy, would normally enjoy. I figured that was important because, as I said before, the crossbow was modelled after a historical crossbow, and those things packed some punch! Additionally, a crossbow is a much more stocky piece of equipment than a normal bow, which meant when it came time to make a melee attack such that the monster could defend itself at close range there really was no alternative: it smashes the PC with the butt of its weapon! Now that this monster could throw down it was time to give it a couple fluffy features or traits that would identify it as a skeleton beyond the mere (pardon the pun) skeleton of its defenses and ability scores. For this I opted to re-use an existing zombie trait from the Monster Vault, namely that the skeleton would be destroyed instantly from a critical hit. I figure if their fluff is that they’re fragile, why not make them quite fragile indeed? At LEAST as fragile as the zombies. So, of course, I did. But, that wasn’t quite flimsy enough for me, so I figured I’d add a bit of a surprise while I was creating the minion. After all, if the minion sucks a critical it’s not like the extra damage matters, right? So, in its place, I gave it another trait: one that made it go off like a grenade when it was destroyed by radiant damage, as though the thing is so violently smashed that it fragments and flies everywhere, potentially doing a lot of damage! This takes the classic “death burst” type effect but flavors it very specifically for the skeleton, I think, and I liked it so much in fact that I applied to the other skeletons (the non-minions) as well. That’ll encourage some tactical thinking, too, since overkilling the skeletons with radiant damage may not necessarily be the best course of action when a slightly more controlled destruction could be possible. Finally, for the archer skeletons I finished them up by giving their minion buddies a pretty simple minion-like set of attacks for both range and close in. Artillery minions are, after all, some of the best monsters for reasons I’ll explain in another post (this one is pretty long as it is already).

The third, and final, kind of skeleton I decided to create for the time being is a classic skeleton soldier type using an old rusty/rotten sword and shield combo along with a little bit of armor. I again assigned its stats as being very defense-favoring, but while I was originally tempted to make it a soldier type (due to emphasis of defense scores over hitpoints) I decided that because one of their biggest defining features is mobility that I would instead opt to make it a skirmisher with abnormally high AC defense (and reflex, too, with lower fort and will to compensate). With the template of skeleton grenade and critical weakness already in place, and going very light on the hitpoints, I gave it a high speed just like the archers, as well as an additional movement power (the ability to shift two squares as a minor action) to emphasize how darn fast it is – it can just slip right through where ordinarily you’d be able to get a hit in edge-wise. I then decided to give it a bit of an ability to fluff out its shield a bit (a one-time free counter-attack from a good block made with the shield), and then I set on making its primary attack, which I envisioned as a sort of wild, ravenous flailing with its somewhat dull but still dangerous blade (while real arming swords had sophisticated martial arts built around them this is a skeleton we’re talking about, and its sword is dull and rusty to boot). As a result I made its attack a very wide swing (2d8+2 rather than the more consistant and recommended 2d6+6), but to compensate I decided that its combat advantage feature (which is something it would constantly and relentlessly seek to have, as it is highly mobile) would grant it unconditional damage, assuming it had combat advantage. This is to represent there being simply too many attacks from a bad angle, and you can’t fend them all off, rather than something more roguelike, which would be one really good attack.

Anyway, after around 2200 words we’ve finally reached our conclusion, which means it’s time for the helpful summary.

What we learned –

1) When creating (or reinventing) a monster, villain, or anything with stats make sure to look for the inspiration and fluff (and any mechanics it had) before you begin. You’ll be surprised what help a little inspiration is. In my design process I used older editions and the current edition of the D&D roleplaying game to see how the skeleton as a creature had evolved over the years, and how those things fit my vision of what a skeleton should be as its own distinct entity, different from the other undead like zombies. Specifically I learned that it (usually) had a low will defense, a low fort defense, relatively low hitpoints, high AC and reflex defenses, and a high movement speed in addition to its usual battery of undead immunities, and I decided to take this general template and exaggerate it slightly to make how skeletons behaved more pronouncedly different from things like zombies in a 4th edition context, as well as using historical context (the crossbow) to give them their own, fairly realistic I hope, place in the game’s world.

2) Make sure to weigh each point you give to the creatures you make. The chart may tell you that their AC should be level + 15 and their other defenses should be level + 12, but make sure that their defenses really reflect their strengths and weaknesses. Be certain to assign dice values and static modifiers to appropriately represent the kind of result you’d like to have (including using a utility like Troll Dice Roller and Probability Calculator to analyze the statistical spread of each potential result of rolling the dice you have chosen if you’d like!), and don’t be afraid of doing unorthodox things like including automatic damage or not giving your monster a true at-will attack to use (in moderation, of course) in order to emphasize the fluff you discovered in the initial research. If you don’t have a lot of time then don’t spend forever, but do try to take at least a minute to address most aspects of the numbers you’re going to be using and why they are what they are.

3) Finally, make certain that your creation has its own distinct feel to it when you’re done. If your skeleton behaves just like a zombie then you may as well use a zombie. While it is an old DM’s trick to take one stat block and call it something else (an orc can be pretty much any humanoid thing with a weapon, after all, if you just rename it), when you’re doing real design work without a very short time budget forcing you to make concessions it’s very important that you create iconic results. Everybody remembers Stormtroopers because they are THE definitive evil faceless minion of science fiction, and that’s because they had a very distinctive look and behavior to them. Be sure that your creations, unless you specifically want them to blend into the backdrop, are just as recognizable. A delayed monster is only bad until it’s done, then it’s good; a bad monster, once released, is bad forever – to paraphrase a certain well-known game developer. If you have time constraints remember that finished is always better than perfect, but when you’ve got that Big Bad boss fight coming up in three or four weeks and have plenty of time then make sure to SPEND that time. A combat is going to be an entire hour (give or take) of time your players spend with this creation, assuming you don’t re-use them (which you should, usually!), so make sure that hour is unique and enjoyable because of the monsters involved, not just the environment.

Now that I’ve bored you to tears, and actually without further adieu this time, here are the stat blocks –

Now in Cherry!

The general idea behind the skeletons was that they were supernaturally quick (and very fragile). As a result I gave this particularly hardy warrior specimen things like a high base speed, minor action to shift, made its damage reliant on positioning (combat advantage), and even gave it the possibility for a free attack (which can be a charge, if necessary, I might add).

The story is pretty much the same for this creature. It moves around, gets in a good spot, and then unleashes its powerful and precise crossbow on unsuspecting adventurers. It won’t try to run up and smack them if it fails to load its crossbow (although if you insist on playing a very optimized combat from the DM’s side you may wish to), and that’s purposefully done so that its damage is in leaps and spikes, with rare but occasional dips. Spikier damage makes for more exciting combat for the players, too, since it’s unpredictable round to round exactly what’s going to happen to them.


Finally, here’s a minion skeleton archer. These are designed to show how, while skeleton archers with normal bows may be plentiful, they are inferior to those with crossbows since they lack the expertise required to aim the bow as precisely, and additionally their subpar equipment deals much less damage than a crossbow of equivalent cost (once again, to our corporate necromancer who is clearly on a budget). Originally the shortbow could crit and deal 8 damage on a natural 20 but adventure tool was being a bit irksome when it came to putting that into the stat block, so feel free to make that modification on your own if you like.

As a final note, while these skeletons were originally designed for use as level 5 creatures I’ve used them as level 3 creatures to great effect (make sure to adjust defenses, hitpoints, attack bonuses, and damage appropriately!), so feel free to adapt them to your campaign at a level you feel is appropriate.

That just about wraps up my discussion on the latest three monsters I created. Hopefully you enjoyed the commentary, or at the very least the free monster stat blocks.

– The Hydra DM

Waiting for Theseus or: How the players react to a Labyrinth

So, in the latest adventure the poor heroes have been thrown into I decided on using a rather traditional mythological concept. You see, their time had come to defeat one of the Big Bads (well, technically an aspect of him) and I decided to plumb through mythology for something fitting. As it turns out I decided on using the classic myth, which you’ve no doubt guessed, by the title: Theseus and the Minotaur. In this case it happened to fit perfectly because the Big Bad was a predatory hunting type of semi-intelligent entity, similar to how the minotaur is classically depicted, but after constructing the maze I learned a few things.

1) How much tension is too much?

I decided to have them explore the maze on initiative count, and this was both a good idea and a bad idea, and going into it I knew it would either work well or flop horribly. It mostly did the latter, but hear me out. Using combat initiative run exploration is a technique that can prove useful in a very specific environment. To be precise it is useful in an environment where at any moment something may attack the party and you don’t want to deal with the awkward interruption of saying “no, wait, stop, move back, there was a trap…” and having the thrill already be gone by the time you’re halfway through it. The original intention of the labyrinthine maze the heroes were subjected to was to be riddled with traps (mostly of the movement-restricting variety) such that a Combat Out could be tricking the “minotaur” (in this case a horrific far-realm mutated abomination about ten feet tall, covered in tentacles, and possessing rather wicked claws and many eyes) into getting itself restrained by the trap (not literally, those of you jumping to the “restrained” condition, but in the sense that it is rendered helpless, which I do mean literally) and then declaring victory when their version of Ariadne popped in and vaporized the critter for them.

Anyway, the original decision was made figuring that there would be plenty of interruptions as they navigated the labyrinth. This turned out, due to time constraints, not to be the case. There was tension, alright, and it was there for about thirty minutes… but once they realized nothing was going to attack them and that the enforced-initiative was a farce they were less than impressed.

This method also resulted in a dramatic slowdown of exploration since they figured something was going to jump out at them. This was not good, as the boss was at (or near) the center of the maze, and it turns out that, even after some extreme encouraging, they took the better part of three hours to find him and I had to wrap the session as the monster revealed itself rather than when the monster was defeated. While the timing lined up nicely, part 2 of the 2 part boss-killing-series was about a thousand times more exciting and about a thousand times less a waste of time than the slow trawl through a maze that couldn’t hurt them in any way.

In retrospect what I probably should’ve done was the age-old staple of playing at tabletop. Simply put, the same way a text-based roguelike works. You ask the party where they’re going, then you move them there, and when there’s a decision to be made or they reach where they’re going you present it to them. Unfortunately, this method of movement presents its own problem, which I’ll expound upon under header #3 momentarily. But, for now –

2) How much MAZE is too much?

Well, the maze was beautifully designed. The visual representation the players got to see was well constructed and fully decked out with things like vision-blocking layer (to show realistic character vision for immersive purposes) and very nice looking layers of visual assets. The walls were wrought gold in intricate patterns and the floor… well, alright, I’ll stop waxing poetic about what it looked like, that’s not too important. What is important is how much maze is too much maze. In other words, once combat with the boss began, how far would the heroes REALLY be willing to trek from the spot he attacked them?

As the answer turns out: not very far. Despite the boss’s habit of repositioning party members through walls, the others refused to budge to go help them. In other words, if I kept playing the boss brutally, the party was going to die because the design was not conducive to them doing what I wanted them to do – have a running fight all across the maze with the boss who can phase through (and later break down) the walls.

The maze, in retrospect, was probably twice as big as it needed to be, and the combat only took place across a patch of terrain around 15 squares on a side. While the combat turned out to be interesting anyway, the rest of the maze went to waste. In fact, they didn’t even explore a full 25% of it! The things the players never see is just wasted effort on the DM’s part, unfortunately.

3) Some things don’t work as nicely with a map…

It’s true, and this is one of them. A lot of people who play this game at a real table have a fond memory of their party having a “party cartographer” who maps the dungeon as they walk through it to keep from getting lost and wandering around aimlessly, much like what might occur in real life. Similarly, in real life, you can get easily lost and turned around in a labyrinth or maze and this behavior is accurately reflected at a table with only your imagination to guide you (or, again, a cartographer-drawn map). Exploring in a roguelike text based game is very similar to a tabletop experience in that it doesn’t give any visuals and it’s easy to become lost. The problem with this, of course, is in the format of a virtual tabletop. Since nobody is near each other you can’t easily keep a map as a cartographer, so of course nobody does (that I know of, though perhaps I can convince them… more on that in future posts!) and as a DM it falls on you to provide the visuals. If you provide a top-down map of a labyrinth, though, each dead end is not a chance to get lost, it is instead simply an annoyance as the party turns around and easily backtracks on the map you provided for them.

As a result, I would not run a maze again with a map provided by myself to the players simply for visual purposes, and instead I would (try) to convince them to make a party cartographer, perhaps sharing the drawing as they go on something like Google Docs.

So, let’s break it down. What did we learn?

A) Try to pick an appropriate pace for the exploration phase of the game. This is often depicted in example form whenever the players want to travel between cities and nothing ambushes them along the way – they, in essence, simply skip over the blank part since nothing eventful happened. If nothing eventful could even possibly happen then don’t make the players take their time going through it. If I had included a bunch of traps or puzzles along the way then the pace of exploration would’ve been much more appropriate while on the other hand since I hadn’t I needed to change the pace of exploration from what I had intended.

B) Don’t make things bigger than they need to be. Wasted space leads not only to frustration on your part (since your work was never seen or used by the players), but also on the players’ part as they have to trawl through a ton of filler hallways to reach the end rather than having a more exciting time. There is an argument to be made that filler hallways provide a natural low point to contrast the natural high point of a fight or big negotiation, and they do, but so do things like puzzles, so there’s no excuse for overusing empty hallways. One now and again is a good idea, but making an entire 3 hour session out of (practically) empty hallways is not a good plan for success, especially if –

C) You need to remember the medium you’re working with. A maze totally devoid of enemies CAN be a fun mental problem to solve… assuming you aren’t presented with a map that fills itself in as you go, in which case there is absolutely no way you can make an error and become lost or anything that might impede your progress and act as an adversary.

Finally, as a fun anecdote to end this analysis, I’d like to point out the inclusion of the fifty kilogram key-ring. The keys didn’t go to anything and they were three feet long and made of wrought iron… but that didn’t stop the players from assuming they would be useful and hauling them along behind them for a couple of hours until their Ariadne informed them that they were entirely useless for comedic effect. Don’t be afraid to throw in a red herring now and again, if only to give yourself a laugh and make a go at studying player psychology along the lines of “well, the DM spent time describing it so it must be important…”

– The Hydra DM