Saturday, 28 September 2013

Monitors: Imagined and modelled combats

Another long, dry post full of maths that only makes sense when you can consult an extensive set of related tables you don't have access to yet. Hooray!

So, I've had bit of a think about weapon types and examined some of the maths involved in hit and damage rolls. What next? Right, I said I'd look at narrative expectations of results.

I think it's reasonably important to do this at this stage, because disconnect between what you imagine and what happens in play is liable to cause all kinds of... I kind of want to say "cognitive dissonance" but I don't know why and I'm not entirely clear what that is. It's going to throw you off, is what I mean. If you're expecting to wade through storms of bullets and instead go down like a ninepin, that'll wreck your expectations. If you expect your character to cower behind barrels and fire off inaccurate warning shots while trying to calculate an intellectual solution, but they take down heavy infantry at long range with a handgun, that will also change the complexion of the game and tend to redefine your character. If you expect a hard-nosed tactical game where you constantly look for advantage, swap weaponry and cling to cover, then a looser game where your tactician character can't really use those features to advantage will disappoint. So I need to work out what I actually expect from this game and the various characters you might run into.

Going back to basics, the starting points are:

  • Simply in order to qualify for active service, Monitors are all baseline competent with weaponry, and can hit a person-sized target at the average distance for a weapon.
  • Monitors have access to better equipment than ordinary criminals, security guards and local militias, but not all Monitors will take such equipment.
  • Monitors are frequently outnumbered and must be able to defeat a moderate number of opponents, given their skill and equipment.
  • While a range of equipment types are available, Monitors are expected to stick with their preferred gear most of the time, rather than constantly switching for tactical advantage.
  • Almost any weapon is more likely to wound than an unarmed attack, which will be our baseline
  • Almost any armour is more protective than

I will need to revisit soft attacks at some point, as I think under the current rules it's not especially viable to rely mostly on soft weapons: there's simply no way to eliminate a target. Perhaps cumulative attacks can result in a wound? Or perhaps the results of the attack are hampering enough that a target can be readily brought down with the equivalent of the "clean blow to the jaw" so beloved of Dick Barton et al?

Also, some sample characters:

  • Professor Rayner, the noted physicist dandy, who uses weapons only reluctantly and wears nothing but silk
  • Xerxes Hardly, special investigator, handy with a pistol but preferring guile to force
  • Siobhain Greenclaw, hardy adventuress, equally comfortable with rifle, jetbike or antique vase, and geared up for tough situations
  • Captain Ukala, former special forces, a crack shot in military armour
  • Toa, finest shot in the spinward sector, professional rogue robot hunter with the social skills of a sprig of broccoli

Rayner is likely to avoid combat entirely. If a firefight ensues, she will largely stick to covering fire while seeking another option. She's unlikely to engage in long-range shooting under normal circumstances, and wouldn't expect to hit anything. In extremis, and at short range, she would pull a gun in self-defence, and expect to have a reasonable chance to stop one (two at the outside) attacker with fairly basic gear. I wouldn't expect her to take down any well-trained or armoured attacker with straightforward shooting, but to rely on intelligence, or make skillful use of the environment to buy enough time to escape or finally get in a lucky shot. Under fire, she can weather an injury as well as any other Monitor, but I'd expect most accurate shots to cause an injury.

Xerxes uses violence as a backup when stealth fails. He's comfortable enough with a weapon to confidently take on a crook or security guard and expect to win, especially since he can usually get the drop on them somehow. Facing more than a couple of opponents, though, he'd look for other options. He's unlikely to carry anything heavy enough to dent serious armour, and would make special arrangements if he expected that kind of trouble. His tough clothing should soften impacts a bit, but he still wants to avoid getting shot at in the first place. In a brawl, he expects to knock out the average street thug fairly handily, but will go down quickly to a gang.

Greenclaw is willing to engage in a firefight with whatever bandits, aliens or extradimensional horrors care to kick off. That kind of character expects to fight off larger numbers of opponents, to pick off weak enemies fairly readily and tougher ones with a few shots, be they mercenaries or wild beasts. She isn't expecting to tear through military targets, but does expect to survive an encounter with a fairly dangerous enemy, holding it off and getting out of there until a better opportunity presents itself. We'd also want her to weather a few hits with the survival gear she has.

Ukala will probably be drawing a bead as soon as an enemy presents itself. She fully expects to defeat several soldiers single-handed, to go toe-to-toe with robosaurs or Kargbeasts, and have a decent bash at vehicles if she has the kit to hand. In a straight-up gunfight, most shots should hit home, and only well-armoured targets should regularly withstand them. A particularly tough opponent ought to weather a few shots in order to be satisfying, but should still be defeatable.

Toa will be deeply disappointed to miss a shot against any but the trickiest target. He expects to pick off fleeing robots at long range, take out drivers as they race past, and hit that vulnerable thermal outlet valve more often than not. While he probably won't carry the heaviest weapons, he should be able to use accuracy to take on tough targets, and should rarely have cause to worry about any kind of firefight. Facing hostile creatures, there's really no more sensible tactic than finding a good spot to shoot from and beginning as soon as they come into view.


Okay, where does that leave us?

Just in passing, I'm going to say that I don't especially anticipate Monitors going around with a heavy weapon as their usual kit. While I'm perfectly happy for them to do that, they're special agents rather than actual soldiers, and even when confronting bandits and the like, unwieldy heavy weapons aren't usually the best choice. But I'll try to keep those as a viable but not overpowered option too.

Also, for clarity: this is not an exercise in picking the result I want and then allocating modifiers to fit. I am in all cases starting with the descriptions given above (each character has 4 Skill points more than the last) and rules described elsewhere, applying them as the situation would warrant and then looking to see what the numbers do. So while I can't rule out subjectivism, I am at least trying to limit it.

I will be using the model with a +5 Wounding bonus for rolling half the target number to hit, as suggested at the end of the last post.


If we treat Rayner as our suggested Skill 4 character, then assuming a short-range +5 bonus for pistols she'd hit 45% of the time, so can draw a weapon and still have a decent shot. If the pistol is exactly as good as the target's armour, she should successfully hit and wound a target one time in four. Anyone starting a fight is probably as good if not a bit better at fighting, but then ordinary mooks have only one Wound. Thus, Rayner should be able to take on one or two low-quality mooks and have a decent chance at coming out on top. Let's see.

With the +5 close bonus, it will take Rayner an average of four shots (two rounds) to cause a wound.* It will tend to take her eight attacks to drop both mooks. Depending how long the first one takes, and whether they go first or last, two mooks with exactly the same gear can get off anywhere from zero to sixteen attacks. Most likely, each will take about four shots, and so they can get off about seven shots before the first goes down, and then another four. This will tend to inflict 2.75 wounds. Just about perfect, I'd say. Rayner has a very slight advantage against two fairly feeble attackers, or one more competent attacker, but would be taken down very quickly by three.

*this took me several sprawling tables to calculate, which may appear elsewhere on the blog in future.


Allowing Xerxes a skill of 8, a +5 close bonus and a +1 strength weapon, he has a 44% chance of wounding, which translates into just under three attacks for an average wound on a thug. He's most likely to use force when he expects to win, which means he'll likely be getting another bonus somewhere and dropping the thug in one round, but let's not assume that. Against two, he can take them down in five shots, during which they might get off four to eight shots before one falls and another couple before the second. More likely it will be six before the first falls and another two before the second drops, for a total of eight attacks, inflicting 1.6 wounds through his tough coat (2 armour). Xerxes comes out battered but conscious and the odds against him going down are decent.

In a brawl (no weapon mods), he has a 25% chance to wound while theirs is a mere 11% because of his outfit. Being generous, they might have some equivalent protective gear and drop his chances to 21%. Here it's likely to take him five attacks to cause a single wound, but he can still easily take out both of them (ten attacks) while taking only 1.8ish wounds in the same period. About the same. It's going to take more than just a couple of two-bit thugs to drop Xerxes when he's on form, but three of them (or two with weapons) can handle it.

I'm happy with that.


Greenclaw (skill 12 in ranged and melée) expects to fend off, say, four or five competent bandits (Skill 8, armour 3, str 6 rifle). With her rifle and moderate armour, she has a +4 advantage in attacking them, and a 0 modifier in defence. The bandits have a 25% chance of inflicting a wound, while her odds are 50%. It should take her five rounds on average to finish off the bandits. During this time, if they stand their ground, she will suffer a total of 50 shots. With the best will in the world, there's no chance she can handle that. Hiding in cover would improve her odds by making it harder to hit her, but they can do the same, of course.

In practice, we're not expecting Greenclaw to stand in the open and shoot; she should be using some of her actions to move around for advantage, getting out of the line of fire for some bandits and forcing them to move around too, so she can handle them piecemeal. Assuming that this will allow only half the remaining bandits to attack each round, and that Greenclaw and the bandits each spend one action moving per round... she'll have to weather a more reasonable 25 shots, which is just slightly more than she can expect to survive. With suitable application of cunning, cover and tactics, and allowing for things like cowardice on the part of the bandits, I think it's reasonable.

Greenclaw expects to drive off a mob of cultists. Not being properly trained, the cultists have a puny skill of 2 and their rusting weapons are -5 against Greenclaw's armour. They have only a 4% chance of inflicting a Wound. Meanwhile, Greenclaw's skill 12 and +6 weapon grant her a 56% chance. She can mow down the cultists in 18 shots, but really they will probably scatter once half go down (nine shots, five rounds). However, that's plenty of time for them to get off 148 shots, if we allow them all to fire each round, twice as many as they need. To succeed here, Greenclaw will have to rely on not all the cultists having ranged weapons, on blast weapons like grenades, or on some being distracted each round by jostling for position, by clambering over obstacles or by crazed chanting. Not awful, but not amazing. Using auto-wound minions as discussed earlier would make virtually no difference here, as the issue is hitting the cultists. But a mob of cultists should not be modelled individually; such puny and numerous opponents should be a mob entity that's modelled individually. So this is a false issue right now.

Greenclaw expects to defeat a Kelithan Rockchewer (three Wounds, armour 6, str 4) and to survive an encounter with a Kargbeast (four Wounds, armour 8, str 9) but not defeat one in a straight-up fight. Her ranged weapon has about a +2 on the Rockchewer's armour, while the laser knife she'll probably use in melée will only be a str 3-4. If she can open fire before it reaches her, her odds are about even - she'll cause about 1.3 wounds in three shots, while it'll take five melée attacks to cause the remaining wounds. The Rockchewer is probably better in combat than she is (skill 16) but even so, will take four rounds of combat to defeat Greenclaw - too long! However, if the Rockchewer gets the drop on her, she will be in trouble. On the other hand, if it's a juvenile no more skilled than her then it's even odds.

What about the Kargbeast? Well, this isn't really a maths question but one of actions. Greenclaw needs to use her actions for evasion and escape - or for delaying attacks, like slowing darts - and rely on the Kargbeast spending some on attacks. As long as the Kargbeast isn't likely to take her down in a single strike, she should be okay.


Ukala expects to take half-a-dozen mercenaries single-handed. Mercs are skill 12, with rifles like Ukala's (but probably a little less good - str 6) and armour 6. Ukala has skill 16, a str 8 rifle and armour 8 - as I said above, most of her shots will find a target. With a 58% hit chance, it'll take Ukala eleven shots to drop them all if she relies on straight-up shooting, though as a competent soldier she shouldn't be. As previous examples will demonstrate, this is far too long (they get 66 shots, and need only 10), but Ukala should be using suppressing tactics and throwing grenades and so on. Nevertheless, I don't think this is good enough purely off the numbers, though it's hard to tell how it would work out in play with many more factors involved. The classic situation tends to have the hero picking their moments, ready and waiting when a searching merc rounds the corner, or spinning out from behind a pillar. In short, Ukala shouldn't reasonably expect to the mercs as a group, but by taking on a couple at a time. Having done some checking, the most she can realistically take on even terms is only two, simply because they get more shots than she does and are moderately competent. One to keep an eye one.

Ukala expects to go toe-to-toe with a robosaur or Kargbeast and have a decent chance. The Kargbeast has four Wounds, armour 8, str 9 and probably melée of about 15. The Kargbeast has a substantial advantage once it gets into combat, which means Ukala needs to get some shots off before combat starts. If she does, all well and good; she should be able to get a wound in, possibly two. If she uses a soft attack (which would be sensible) she can reduce the beast's offensive ability in the long run, and improve her odds. On the other hand, if she leaps into melée the Kargbeast is likely to emerge battered but victorious.

Ukala expects well-armoured targets to weather her shots and prove tough to defeat. At present, that's entirely possible if armour values are basically unlimited. If they're capped around 10-12, she can expect to do around one wound in every two or three rounds of fire.


Toa has a skill of 20, and a dedicated anti-robot rifle (shock weapon) designed for mid to long-range fire. He would have military armour, but not heavy stuff, since he expects to be the one doing the shooting. I'll grant that armour of 7, and assume that a mob of security bots have weak weapons with about a +2, so Toa has advantage by 5 points.

With no attack modifier for range and a weapon that's likely to be at least +5 against the robots compared to their armour, Toa can expect to damage a robot a full 88% of the time, dropping five robots in three rounds of focused fire. He can take out a ten-bot squad in six rounds, which is probably not enough to stop a distant target from closing into melée (if it wants to). If it's a firefight, during that time Toa would be exposed to... wow, let me just do some maths...

Okay, Toa is going to be taking 136 attacks before he drops the robots - assuming they choose to fire rather than move. I think we need to assume that a robot with a weapon has at least some idea what it's doing, and more so if in a pack, so let's give them a "militia" score of 10. They're looking at a 19% wound chance. Even allowing for all Toa's advantages, ten robots are likely to drop him in the first round with things as they stand (16 attacks, to be specific). If he's in a bit of cover, they're looking at two rounds of fire. Toa is a great robot hunter, but can't simply stand around blasting at overwhelming numbers of militia-grade robots with his current gear. He's going to need to try long-range fire (where their weapons will be penalised or ineffective), some proper hunting tactics (divide and conquer) or just some really serious armour if he wants to weather that kind of situation.

What about a single high-grade military bot? Let's allow the robot Armour 10, and Toa's rifle Strength 5 against it (a total penalty of -5). Due to stellar marksmanship, he will score a wound a full 38% of the time, allowing him to take the heavily-armoured military bot down in four rounds of sustained fire - providing he stays standing that long.


Okay, I would say this isn't awful, but also has some obvious weak spots. One prominent one is that I haven't really established a high-end power level for dangerous beasts, so it's hard to tell if that stuff works (the Kargbeast is only slightly better than a Monitor, which is hardly "terrifying alien monstrosity" level). Strength in numbers very rapidly allows competent enemies to overpower a Monitor despite the difference in wounds, because three one-wound enemies with two shots each can hugely outshoot a three-wound Monitor with two shots. This is true at very small numbers of competent enemies, well before you get into hordes.


Here's the other question: what would it take to survive a mob of ten enemies for several rounds? That's a classic of fiction, after all. Incidentally, I am well aware that this is not the way to handle large groups of enemies, I'm just curious.

A mob of poorly-trained Stormtroopers may have Skill of only 2-3 and low-quality weapons. It will take them 50 shots to take down the Professor, 60-odd for a lightly-armoured Monitor, or 75+ to drop a target with military armour of grade 6 or better. However, 60 shots is only three rounds of fire for a squad of ten Stormtroopers! You'd need to reduce their wound chance to a mere 3% before it'll take 100 shots (5 rounds for them) to drop you on average. Meanwhile, Rayner will take 77 shots to destroy the robots (39 rounds) and Xerxes will take 38 shots (19 rounds) - neither stands a snowball's chance in a kiln. Heavy armour does a limited amount to help here - what you really need is penalties to hit, which calls for evasive action and/or cover. I don't think this is that outrageous to be honest - characters surviving mobs do typically use speed, mobility or cover to survive - but if I want tank-like characters soaking up damage then I'm going to have problems in this model.

So mobs are something I need to look at again, and like just about every other (combat-including) RPG out there, I will probably end up with some swarm rules for handling large numbers of ineffectual targets. I have also mentioned the idea of minion NPCs, who don't even have wounds but go down to a single hit from any kind of weapon, which would perhaps be a better fit for the Stormtroopers.

There might also just be some standard rules for mobs of weak enemies, allowing them to look nasty while presenting less of a threat than expected. For example, it makes sense that in normal circumstances only part of the mob can get a clear line of sight, while others are distracted, cowering under suppressing fire, clambering over obstacles and so on. Similarly, it may be easier to hit a mob if you aren't aiming at a specific target.

I could add rules for suppressing fire. They can be pretty broad-brush and affect only certain enemies (one-wounders, who come in mobs). Alternatively, taking cover might be a part of the rules for such NPCs, allowing GMs to present a mob of them while having standard rules that prevent them being too much of a threat.

Of course, this doesn't deal at all with the actions issue, because the idea of actions is you sometimes do things other than standing still and shooting. You use cover, manoeuvre for advantage, run from overwhelming force and try to pick off isolated targets.

The other thing is that, impressive as Monitors should be, I don't especially want individual characters to be fighting off large numbers of enemies on a regular basis. That sort of thing should be restricted either to genuine swarms, to very particular situations with very particular gear, or to streaks of outrageous luck and cunning. Apart from any other considerations, if a single Monitor that's anything other than a minmaxed combat machine with outrageous gear can wade through ten enemies worth differentiating in a straight-up fight, the GM is going to be faced with managing huge numbers of NPC combatants and combats will grind to a screeching halt. There is no problem with a Monitor fighting, say, ten pirates, as long as they appear in small groups or are otherwise unable to bring their numbers to bear. Similarly, there's no problem with a Monitor defeating a horde of crazed hoover-bots that pose virtually no individual threat.

On top of that, a lot of the time (though not all) Monitors should be working in groups, and three or four can handle a significantly larger number of enemies far more effectively than one, because it's a smaller multiple. Broadly speaking, if one combat-ready Monitor can handle two or three enemies then a group should be able to deal with eight to twelve at once, which is a nicely impressive number.

Next steps

I want to try out an alternate model (suggested, as so often, by Dan) which would exchange Wound rolls for an unmodified Armour save. This would emphasise the idea that Monitor weapons do hurt if they hit you, and only some decent armour will protect you. Powerful weapons would not modify the roll, but simply penetrate armour of a particular value or worse. This can greatly speed up play because there's no roll for high-pen weapons, while allowing armour to be really very good against weapons that are even slightly weaker than it.

I also want to take a look at mob rules.

One day, I will actually get round to publishing some actual armoury again!

Back in the saddle

So after a four-year hiatus, my original Librarians and Leviathans 4E game is restarting. Astonishingly, despite people leaving town, it has three out of four original players.

This isn't really a resurrection of the campaign as such. Scheduling issues abound, and with ongoing health and personal issues I'm not at all confident that I'd be up to a campaign. Instead, I'm planning to do occasional one-shots loosely linked by narration and handwaving. This should reduce the chance of being unable to schedule a game for months and then ending up DMing for eight hours straight, which proved pretty devastating previously.

One-shots are also much more conducive to group changes: people can drop in and out for individual sessions, either because of scheduling or not fancying a particular module. Perhaps we'll even have someone else DMing now and then, who knows? I certainly suspect it's less intimidating planning one single game session than taking over for an unknown period. Similarly, if people fancy trying out a different character sometime, that should be easier to handle without complicated explanations.

Planning the game

It's been long enough since I did this that I'd forgotten just how much I enjoy prep, and for some reason D&D (and perhaps 4E in particular) seems to scratch that itch better than Call of Cthulhu. It's probably partly down to how straightforward the planning is; you're looking for plausible background and fun creatures in a fantasy setting, and a reasonable game balance. That contrasts pretty strongly with the complexities of Cthulhu, where the real-life setting and semirealistic characters means there can be lots of issues you need to consider. Moreover, it's a little easier to predict how the party are likely to react to situations (though not always) and whather that's what you wanted. On top of that, D&D adventures are reasonably universal, whereas Cthulhu scenarios can easily be either in completely the wrong continent, a very different genre from what I want, have very different assumptions about playstyle from my group, and so on.

This session I'm going for the most straightforward adventure I can find. It's by Aeryn Rudel ("Blackdirge"), called Stick in the Mud, and appeared in Dungeon #171. As often, the adventure needs a bit of reskinning to fit into my campaign world, which doesn't currently feature the Chaos Scar (though I might add it in future). I also wasn't particularly sold on the adventure hooks and backstory, and my players do tend to question their motivations.

Since I'm trying to run this as a one-shot, I've decided to keep things quick by having the actual play start on the threshold of the valley, skipping lightly over the introductions rather than roleplaying it as I usually would, though I'll happily flashback to discussion with the questgivers if people seem keen. The party won't be exploring generally, but actively seeking out a stolen artifact on behalf of a powerful Guild, who trust them to handle it discreetly because of an unspecified "previous relationship". They have plenty of reason to go through with this because doing favours for the Guild will be socially and professionally useful, let alone any reward they're likely to accrue. The need to find the artifact should explain why they'd keep exploring despite finding hostile creatures, which is often a tricky question in exploratory adventures.

I'll give some more explanation of the changes I made once I've actually run the scenario, as more will probably appear.


So one of the things I always enjoy is coming up with stuff, and this adventure is no exception.

I decided fairly early on that I wasn't going to stick with the original monsters. There were a couple of reasons. The main one was that I'd altered the background, and so the precise powers and nature of the artifact seemed to call for a change in the nature of the monsters. Rather than being trigged by the monsters, the artifact itself is the reason for their presence. As such, I wanted the monsters to emphasise the story I'm going to give the players. The original monsters, while fine in themselves, didn't really fit my background or campaign, but the Chaos Scar setting. I felt like they just felt somewhat arbitrary, and while strangeness is fine in fantasy, I'm trying to avoid things getting completely hotchpotch.

The second reason was that I tend to find encounters quickly get repetitive when you're facing very similar enemies, even when (as here) there are slight variations in the precise make-up of the enemy forces, and some creativity in terms of the tactics and environment. The first stages of Keep on the Shadowfell, with its interminable kobold battles, are burned firmly into my memory. Having different monsters for most of the battles adds more variety, while simultaneously allowing me to create a progression that should give a sense of what is happening in the... location.

To avoid unnecessary complication, I just reskinned some of the adventure's original monsters, slightly altering descriptions and keywords for the most part. I still had a lot of fun with that, and to some extent I think it's the best way to handle this kind of thing - the published adventure's supposed to be balanced and fun, and I don't have any specific problems with the monsters, so why change them too much? The idea is just that the fights will feel slightly different, and description is a lot of that.

I also had to not only pick out loot for the party, but (due to background) devise a number of lesser but still interesting items to discover for flavour. This is one of those interesting little niches: items that are curious enough to be fun to find, and potentially even slightly useful, but without reaching the level of genuine "magic items". This was particularly important as I'm not entirely sure of what gear the PCs had when I last saw them, and whether they were 2nd or 3rd level! I'm keen to avoid overloading them with magic items if they were already reasonably equipped, and so my plan is to keep the adventure itself slightly conservative lootwise, since the Guild can always be more or less generous in their rewards when I work out what they've already got.

It's really fun sitting down and trying to come up with very minor, interesting magical trinkets that aren't likely to have much use in game but are nevertheless fun. If the players do end up getting some use out of them, all the better.


Anyway, all this called for a new version of the adventure, to avoid having an illegible mass of crossings-out to deal with. The text was a bit of a pain to transfer, what with copy-pasting from PDFs being unhelpful, but the main issues were maps, items, and monster stat blocks. Thankfully, my PDF reader (Foxit, amongst others) includes a tool for selecting and copying bits of a PDF as images, which avoids too much faff.

However, this didn't solve the issue of my homebrewed monsters. In theory, this is exactly what the Wizards Adventure Tools are for; in practice, they require a subscription to D&D insider, I believe they no longer offer a downloadable offline version, and their Monster Builder wasn't very good last time I did use it.

While I did have a D&DI subscription originally, mine lapsed four years ago, and I'm not keen to fork out for another one. It's only really worthwhile if you're playing more than I ever managed, and I've no idea how many sessions we'll get in this time around.

Moreover, while $US70 a year isn't a huge amount, it's still a fair chunk of money. I've already spent quite a lot of money on D&D materials, mostly rulebooks but also a couple of handy things like wipe-clean maps and a fair number of PDF supplements. I don't begrudge it, but I'm not particularly keen to keep spending money on things that are largely for the benefit of players, which was certainly the case last time I saw DDI - the character builder and the Compendium's updates on powers were the really useful bits. If we end up doing a lot more playing, and I'm feeling strong, I suppose I could float the idea of a group subscription. I imagine this is a fairly common question for DMs, and I'm sure I've seen it mentioned before, though I haven't been able to actually find anything now I thought of it.

After some frustrating messing around with tables and trying a couple of online tools that turn out not to actually do what they give the impression of doing, I finally discovered a brilliant one in the shape of Asmor's Monster Maker. This not only generates statblocks, puts information in the right place and does some of the calculation for you, but it can auto-generate generic monsters based on your choice of level and role, and import files from elsewhere. So if you do happen to have a DDI subscription, you can actually get monsters from there and then adjust them.

One thing I would say is that the statblock as produced is technically accurate, but hugely oversized and fairly ugly. I did a quick hackjob on the .css file it uses for its HTML in order to get it looking more like a normal one, then print-pasted it into my Writer doc. I'll copy the adjusted .css below.

dark green: 48542e
light green: c7c7b0
background: e7e5d3
at-will: 5b8f62 (91, 143, 98)
encounter: 811c32 (129, 28, 50)
daily: 414342 (65, 67, 66)
power gradient beginning: dad9c7
power gradient end: ffffff

/* whole table */
table.power, {
padding: 0 0 0 0;
margin: 0 0 0 0;
font-family: tahoma;
width: 300px;
} {
background-color: #e7e5d3;

div.sectionPlain {

/* Headings */ thead {
background-color: #48542e;
color: #ffffff;
thead.atwill {
background-color: #5b8f62;
color: #ffffff;
thead.encounter {
background-color: #811c32;
color: #ffffff;
thead.daily {
background-color: #414342;
color: #ffffff;

/* Applies to everything */, span.role, span.type, span.exp, span.initiative, span.senses, span.aura, span.hp, span.regeneration, span.defenses, span.damagetypemodifiers, span.savingthrows, span.speed, span.actionpoints, span.powerheadspan, span.powerbodyspan, span.alignment, span.languages, span.skills, span.ability {
text-indent: -15px;
margin-left: 20px;
margin-right: 5px;

/* also applies to everything */ td, table.power td {
vertical-align: top;
font-size: 8pt;
} tr.powerbody span {
text-indent: 0px;

.title {
font-family: fairydustb, arial;
font-weight: 900;
font-size: 16pt;
/*display: block;
width: 100%;
border-bottom-style: solid;
border-bottom-width: 5px;
border-bottom-color: #48542e;*/
color: #48542e;

.sectionTitle {
font-family: fairydustb, arial;
font-weight: 600;
font-size: 12pt;
/*display: block;
width: 100%;
border-bottom-style: solid;
border-bottom-width: 5px;
border-bottom-color: #48542e;*/
color: #48542e;

/* name line */ {
font-weight: bold;
color: #ffffff;

/* role line */
span.role {
font-weight: bold;
color: #ffffff;
float: right;

/* type line */
span.type {
color: #ffffff;

/* exp line */
span.exp {
float: right;
color: #ffffff;

/* embedded tables */
table table {
padding: 0 0 0 0;
margin: 0 0 0 0;
width: 100%;

/* first block of stats */
tr.mainstats {

/* heading of powers */
tr.powerhead {
background-color: #c7c7b0;

/* body of powers */
tr.powerbody {

/* seconda block of stats */
tr.secondstats {
background-color: #c7c7b0;

/* equipment */ {

span.class {
float: right;
color: #ffffff;

tr.color, tr.indentcolor {
background-color: #dad9c7;
tr.backpng, tr.indentbackpng {
background-image: url(powerback.png);
background-repeat: repeat-y;
tr.backgif, tr.indentbackgif {
background-image: url(powerback.gif);
background-repeat: repeat-y;
tr.indent span, tr.indentcolor span, tr.indentbackpng span, tr.indentbackgif span {
margin-left: 35px;

span.flavortext {

span.action, span.powertype {
font-weight: bold;

span.dndicon {
font-family: 'D&D 4e icons';

p {
font-family: tahoma;
font-size: 8pt;
line-height: 120%;
width: 300px;
h2.sectionTitle {
font-size: 10pt;
font-weight: bold;

Thursday, 26 September 2013

Musing on Maths

I have been running quite a lot of maths recently to try and work out whether different combat models would work in Monitors, let alone what numbers to use. And this has given me a huge amount of sympathy for people faced with doing this sort of thing professionally.

The thing is, it's relatively simple (by which I mean to say, very complicated) to do calculations for what would happen (on average, of course) if hero X and enemy Y stand still in an enemy room with no features whatsoever and shoot or stab at each other. You can, if you want, tweak the numbers to produce different results, or different distributions - maybe you want pretty reliable combat, perhaps you prefer it dramatically swingy. You might introduce a complicating layer of different attacks, or perhaps multiple combatants, but while these make things even more difficult it's still possible to see how you model it with maths, even if the execution is a pain.

The problem really comes about when you try to introduce other stuff. That is, anything that allows combat to be remotely interesting.

In combat, we want heroes to use cunning and stealth and tactics. They take cover, launch surprise attacks, use suppressing fire. People don't fire endlessly at static targets, but duck and crouch and move around for perceived advantage. Large groups try to encircle smaller groups, while the smaller groups try to use terrain and technology to create bottlenecks, allowing them to face only a few enemies at a time.

How much of an advantage can the hero gain by doing this? Exactly how much do we want the hero to do it - or any particular heroic archetype? Should they be taking cover frequently for a small bonus, occasionally for a large one? How successfully should they be able to bottleneck enemies, and how does that relate to different enemy types? How effective is that suppressing fire?

Very quickly, you end up in the situation where you'd have to picture the whole fight in your head in detail in order to work out exactly what kinds of factors you want to come in and how effective you want them to be. You end up with an unmanageable number of things to take into account, and potentially with a massively-spiralling set of rules to handle it all. Or else you resort to alpha-testing each possible set of rules in detail, each time having to decide not only what effects you care about, but the numbers that should be attached to them. Real game designers have my sincere sympathy. I have no idea how they handle this stuff.

Which is to say nothing of Grapple rules.

Tuesday, 24 September 2013

Looking for Ogre

So both Dan and Arthur are talking about Quantum Ogres, which inevitable prompted me to waffle about it myself. This is a very OSR-centric debate, which is not particularly my area, but that never really stopped me opining before.

Quantum Ogres seem to originate in Beedo’s 2011 post here. Essentially, given a choice of three Woods, the Quantum Ogre always turns out to have been in the first one you explored. This idea prompted a number of posts, many of them critical.

Dan’s article is prompted by this one by Cat at Hack & Slash, which is less focused as it's mostly a response article. Cat asserts that:

  • Quantum Ogreing is bad for players because "the players never have the experience of choosing correctly and skipping right to the end (which is fun for them)"
  • Fun comes from exploring objective worlds and one-in-a-million rolls rather than from lovingly-crafted encounters, so Quantum Ogreing does not increase fun
  • Players will grow to resent Quantum Ogre encounters and will become aware of them
  • Quantum Ogreing as a time-saving measure is a fallacy, because you can effortlessly do all the prep you need by cribbing and using random content

Kevin at KORPG Games discusses whether or not Quantum Ogres can be considered a false choice. Essential quote: "Remove the foreknowledge of the possible outcomes and you’ve removed all knowledge of the palette shift… and with removal of that knowledge, the palette shift itself is actually removed because the knowledge is a requirement of the shift by definition. What robs the game of fun is when the palette shift is known and is in direct opposition to the meaningful choice expectation of the player."

Dan mostly talks about the nature of choices in RPGs and how this interacts with game content and with player fun. He also has some interesting points about the validity of random encounters versus human-genned encounters.

Arthur discusses the nature of interactivity and the different assumptions people have about it, and the way this relates to their games.

I will instead present a rambling set of loosely-connected ideas with no firm conclusions. As usual.

What is a Quantum Ogre anyway?

People have a couple of different takes on this. Beedo, the proposer, says: "Regardless of which woods the players choose first, he'd like the party to have the opportunity to encounter the ogre. The MacGuffin will be somewhere else."

Commenting articles seem to have read this as something like: "Regardless of which woods the players choose first, ... the party ... encounter the ogre. The MacGuffin will be somewhere else." This is subtly different. Moreover, quite a number of people read this in an even more heavy-handed way, bringing in assumptions that the party will a) confront the ogre; b) in the manner prescribed by the DM; c) regardless of their wishes.

Cat does this very specifically in a comment on Beedo's next article.

The problem is not even when the DM plans to put the ogre encounter in front of the players! There is nothing wrong with this. In a game where the players can go anywhere, improv is necessary and not necessarily abhorrent.

The problem comes when the players attempt to gather information, and then the DM becomes evasive, blocking their attempts because he is vested in his encounter. This leads to the known bad case of the McGuffin always in the last wood.

Leaving aside the interesting question of whether "the MacGuffin always in the last wood" is necessarily bad, I think it's worth noting that Beedo does not seem to hold any of these assumptions. I should also note that Beedo later decided "I ultimately came around to the position that any kind of illusionism is a bad practice" in a series of edits, though the reasoning wasn't clear to me and seems to be part of some OSR school of philosophy I'm not familiar with.

I am happy to agree that actively thwarting player choices is a bad thing (though there may be some circs where I'd accept it), and the related idea that forcing players to handle a situation in a prescribed manner is a bad thing. Both of those tend to reduce players to dice-rolling machines making purely cosmetic decisions about their reactions to your plot, and controlled by a mixture of DM fiat and random chance. I can do that more efficiently by playing a CRPG.

In contrast, to me Beedo's original Quantum Ogre seems to be saying that the party will always a) become aware of the available possibilities; and b) have the least final option presented first, such that it can make IC sense to tackle it rather than bypass it. Going back to Dan's point about IC and OOC goals, this seems a pretty good way to satisfy both the IC goal of finding the artefact, and the OOC goal of fighting the ogre, without having to compromise IC motivation. In contrast, an artefact-hunting PC given the option of Ogre Wood or Artefact Wood has little motivation to visit the ogre, nor does a PC who finds the artefact first through a blind choice or through a random table result.

In general, the Quantum Ogre discussion seems to have rapidly wandered off into what I might instead call the Massive Roman Achilles Ogre discussion: all roads lead to the Ogre, the Ogre's gravitational field is too large to escape from, and the Ogre can only be overcome in one way. This is, quite frankly, a completely different issue.

The Quantumness of Ogres

Dan makes some points about how random encounter tables are not necessarily qualitatively different from DM-decision outcomes. The dice thing is kind of hard to get your head around, but... okay, here’s some examples vaguely related to it.

  1. A DM plans three encounters: Nothing, Ogre and Artefact. These are allocated to Woods A, B and C respectively. (planned known geographical)
  2. A DM plans three encounters: Nothing, Ogre and Artefact. These are allocated to First Wood Explored, Second Wood Explored and Third Wood Explored respectively. (planned known chronological)
  3. A DM rolls on a random encounter table and gets Nothing, Ogre and Artefact. These are allocated to Woods A, B and C respectively. (random known geographical)
  4. A DM rolls on a random encounter table and gets Nothing, Ogre and Artefact. These are allocated to First Wood Explored, Second Wood Explored and Third Wood Explored respectively. (random known chronological)
  5. A DM plans three encounters: Nothing, Ogre and Artefact. These are put on cards, shuffled and allocated to Woods A-C, so their order is unknown to anyone. They are revealed during Wood exploration. (planned unknown geographical)
  6. A DM plans three encounters: Nothing, Ogre and Artefact. These are put on cards, shuffled and drawn in order during Wood exploration, so their order is unknown to anyone. (planned unknown chronological)
  7. A DM uses a blind random selection method to obtain three level-appropriate encounters from an arbitrarily large set. Unbeknownst to the DM, they are Nothing, Ogre and Artefact. The cards are allocated to a particular Wood A-C and revealed during Wood exploration. (random unknown geographical)
  8. A DM uses a blind random selection method to obtain three level-appropriate encounters. Unbeknownst to the DM, they are Nothing, Ogre and Artefact. The cards are drawn in order during Wood exploration. (random unknown chronological)
  9. There are three Woods A-C. When players express an interest in exploring a Wood, the DM invents an encounter out of whole cloth. (planned spontaneous chronological)
  10. There are three Woods A-C. When players express an interest in exploring a Wood, a random encounter is rolled from an arbitrarily large set. (random spontaneous chronological)

Which of these undermine player agency? Which are okay? Which enable meaningful decision-making?

I'd argue that mostly, we don't have enough information to judge. How far in advance will the DM check "unknown" results so as to provide clues and details? How easy does the DM make it to shortcut or circumvent results that the players don't want? What do the different 'encounters' even mean? There is an enormous difference between flipping over a card as you approach the wood, and then providing all the relevant details, clues and even NPCs necessary for the players to make meaningful decisions about whether and how they wish to interact with it, and flipping over a card to determine whether they find an artefact in a clearing or are attacked at night by an ogre.

I'd also make two broad points. Firstly, the more random a result is, the less likely it is that players (or the PCs) can predict it based on existing knowledge. This may impact their ability to make informed choices. Secondly, the more unknown a result is to the DM, the more difficult it is to prepare the ground, and thus the more difficult it is for players to anticipate it and make an informed decision. This affects pre-existing encounters where the DM is unsure of their location; random encounter results that a DM knows may come up or not; and spontaneous encounters.

In contrast, a DM who knows precisely what is coming up (either chronologically or geographically) can enable a firm decision from the players. Moreover, in a group with a strong DM-player relationship, players can often judge from metagame clues and OOC chat the direction things are going, and make their decisions based on those factors. As such, I am actually inclined to say that randomness undermines player agency, while prepared aspects enable it.

There is another point to make here, which is that a DM assigning content on the fly (whether premade or spontaneous) has the option to twist that content to suit the state of the party and the group's mood at that specific point. While this can always be done with small details, enabling a frustrated group to skip filler content or a badly injured party to not wander into the trap-ridden ettercap settlement just yet seems to me a valuable option and one that, as a player, I appreciate. I don't want to play a game where even if frustrated and bored, I have to spend an hour of game-time combing a forest to discover that it contains nothing of interest; and if I've just fought my way through three random encounters, I might well prefer not to blunder into yet another fight with an Ogre just because a DM had decided there'd be one here.

Cat's article

Cat's article includes a couple of arguments I wanted to discuss, while hopefully not descending too far into just picking at someone else's post.

Ogres should be foreseen and not heard

If you always pre-ordain 'your precious encounter' then the players never have the experience of choosing correctly and skipping right to the end (which is fun for them).

This contains a fairly major assumption: that "skipping right to the end" is a) fun, and b) probably more fun than encountering the ogre.

Let me be unreasonable for a minute, while also calling on Dan’s example of the treasure-hunting PC. A really canny bunch of characters given enough information can work out where treasure might be, and go straight there evading all ogres. Many adventurers would like nothing more than to stroll from hoard to hoard, stuff their pockets with diamonds, and then retire to a life of indolence.

From my perspective as a player, though, that would be incredibly boring – it’s the equivalent of a mystery game where you immediately identify and arrest the culprit and then drive off into the sunset. I have yet to sign up to a campaign on the premise of “plucky heroes collect artefacts from the places they would logically be and retire wealthy after a minimum of hardship”. In fact, I would almost never choose to avoid the ogre unless I had reason to think that encountering the ogre would be less interesting that other things I could do with the time. Honestly, “skipping to the end” doesn’t seem especially fun to me, apart from the brief satisfaction of having correctly interpreted some clues. Quite often, I don’t want any particular outcome to a situation, but simply want it to be fun. To get into some nasty grammar: if it is going to have been more fun for me to have run into an Ogre before I found the treasure, I want the DM to have interposed an Ogre. Usually! And that, of course is the other thing. We don’t necessarily know what we want. We don’t necessarily want the same things consistently.

The other point is the flipside of end-skipping: if you present three woods, where one contains an ogre and one contains the treasure, then there is at least a one in three chance that players will pick the empty wood first, and if you are handling these as though they are equally important, a considerable amount of time could be spend on it. That is okay if the group are interested in playing some people exploring a forest, but if they are keen to get on with quest-relevant activities, then it could be very frustrating. On the other hand, if you dispense with the forest in a few sentences, then was it really worth including it at all? Why does this forest even “exist”, from an adventuring perspective? That’s not to say I couldn’t enjoy myself just exploring a forest, but these things vary between players, genres, and even individual play sessions. I know at least one player who generally just wants to move from dramatic event to dramatic event, without faffing about.

I suppose another thing is, if the DM (always a friend of mine) has spent time and effort devising an Ogre for me, I am generally pretty keen to encounter that Ogre. My DMs can usually be depended on to devise pretty damn good Ogres, and I’m happy both to get the chance to appreciate them, and to give my DMs the satisfaction of seeing everyone get a kick out of their Ogre.

Just occasionally, spotting a way to evade all kinds of challenges and achieve an objective can be immensely satisfying. But I want those occasions to be a matter of player and PC cunning, and the feeling that I have in fact achieved something. I don't want to have bypassed the bit of the adventure where I would do stuff because I picked the right card. And I will only want it occasionally, because mostly I'll be playing the game in the hope of encountering ogres.

Quantum Ogre and the Jerk DM

The flaw of the Quantum Ogre is that, if you have a party who plays smart, he won't be quantum long before you enter the woods, and then you've wasted time by not assigning him to a location already or you become the jerk DM where ESP doesn't work, the ground doesn't hold tracks, and if you try and teleport - suddenly anti-magic fields everywhere.

This is a reductio ad auteuram (to coin a phrase). A DM who will prepare encounters and then assign them on an ad-hoc basis so that PCs interact with each one to some degree, is by no means the same as a DM who will block tactics adopted by the players. There are very different things going on in each case. For a start, the Quantum Ogre situation does not necessarily involve good information on the choice being made. Frequently players are left making choices with only quite broad information (go to the town or the forest?), and while they can draw some conclusions about what they're likely to encounter and the opportunities they'll have, there are many things they can't reasonably predict without performing detailed research. Secondly, arranging for an ogre to be in the forest or in the town does not in itself involve contradicting common assumptions about the setting, the physical laws of the gameworld, or the game mechanics; this is a sharp contrast to banning teleportation or blocking the use of skills in order to make players handle an encounter in a particular way.

I would also say that, in small doses, preventing the use of specific abilities in a specific situation can offer opportunities for creativity and make things much more interesting, providing it's done with some kind of in-game logic and isn't aimed at forcing a particular approach, but at preventing one.

It also seems strange to me for Cat, who shrugs off the time needed for sandbox preparation (of which more later), to quibble about wasting time by not assigning de-quantising Ogres. If you have forests A-C prepared, and locations 1-3 on a map, just how much time is being wasted by associating one with the other during the game rather than beforehand?

Finally, how quantum is an Ogre? Some “ogres” can’t reasonably be predicted by PCs, even by asking locals or carrying out research. A ruin or notable monster might be learned of, but there are plenty of other Ogres that are either commonplace or impossible to discover without going there. Other places aren’t frequented by anyone, so nobody’s there to warn you about the ghost, artefact or quicksand.

A Choice of Ogres

Looking again at Dan's points about choice, you could also bring up the question of what precisely the choice is the players are trying to make.

Are they trying to choose “do we go and find the MacGuffin, or encounter an Ogre?”, or are they trying to choose one of three locations to nose around in? If it is the latter, then arranging for a Quantum Ogre is not in any way infringing on that choice. Or perhaps they might be choosing “do we go for some undemanding content and mess about in the woods, or push on to another stage of the plot?”, which is different again, since getting to MacGuffin Wood doesn’t necessarily enable them to advance the plot, nor does meeting the Ogre preclude it.

If the players choose “undemanding content”, I personally don’t see any failing of a pregenned Ogre over a random encounter; a random encounter might prove interesting and exciting, by throwing out something the DM wouldn’t have thought of, but could equally well seem nonsensical. Odd results on random tables can seem out of place narratively, either being so petty that it seems pointless to waste time on them during your epic quest, or so substantial that they shouldn’t be an unexpected diversion.

I might also say, if you’re going to argue that PCs should have the chance to learn about ogres in the wood and avoid them, they also deserve a chance to learn about many kinds of random encounter, which in effect means pre-rolling the random encounters and planting some clues. Your prep time just went up.

I think this is my main take-home message from Dan's article: “It is only wrong to force the players to encounter an ogre if they specifically want to avoid encountering an ogre.”

I don’t care one jot about any supposed loss of agency in depriving me of the ability to discover that a wood, chosen blindly out of three, contains nothing whatsoever of interest. This is not a meaningful choice (and I think the use of that term on various posts in that chain is highly misleading). For one thing, much of my real life consists of doing things that are not particularly interesting; I don’t need to do them in my gaming time too. I don’t generally feel the need to roleplay the time spend on long stretches of dull road, waiting for the kettle to boil or asleep. For another, if empty wood is called for, why not just use the 99% of Ogre Wood that does not contain the Ogre? We can do all the same things there, it’s cool, seriously.

Essentially, this choice is only relevant if I have some information that Wood B contains an Ogre, and decide that I would like to explore the other two woods in the hope of finding the artefact without encountering said Ogre. Of course, I might believe that the Ogre is likely to have the Artefact, in which case I might go there first. If there’s plenty of information indicating the location of the Artefact, then perhaps there’s no point having the three woods anyway, and you might as well combine them. At that point, the question stops being a case of Wood-choosing, and becomes a tactical one: there is an Ogre in the Wood where the Artefact is, what is our best strategy for dealing with the situation?

Another point is that people want different things at different times. Sometimes I want to see interesting content of whatever type. Sometimes I'm quite invested in finding the artefact to push the plot forward and don't care about the side content. Sometimes I'm not that interested in the plot and would rather fight an Ogre. Sometimes I'd rather explore the setting my DM has created. And sometimes I really don't care what I do IC, so long as I can spend time with my mates OOC. Arthur talks a lot about types of interaction, which is also quite relevant here.

Agency and Cooking with Ogres

Broadly speaking, in playing a game, the agency I'm looking for is ability to control my interactions with the content of the gameworld. I may choose not to explore an element as fully as the DM expected, or to dive right into it. I may not interact with it in the expected way. If I show no interest whatsoever in interacting with some content, and especially if I (as a player) show signs of avoiding it, I expect my DMs to respect that wish within reason and within the expectations of the group as a whole. I expect freedom in the way my character behaves, the ability to shape events and to affect the plot to at least some degree; though all of these are subject to the genre, the game system and the kind of campaign I signed up for. However, I am not bothered about having 100% total freedom to explore any part of the world as I choose, because this strikes me as being both a bit pointless, and kind of arsey. I am particularly not bothered about having an objective world from which I select elements to explore, because while I very much enjoy exploring the creations of my DMs, I don't think this experience would be enhanced by knowing that if I'd chosen the wood over to the right I would have discovered something different and missed out on the current wood.

As a player, I don't want to feel that I am obliging my DM to devise a vast swathe of original and interesting content so that I can completely miss 95% of it, and that I may well miss the most interesting elements (to both my PC and myself) because I chose to take the left door rather than the right. Cat (I don't mean to pick on Cat, they just happen to have written the article that mentions this) suggests that this isn't a problem: content can be quickly generated through web resources, recycling and random content. However, I'm not sure this would really fly outside the OSR. While I’m really not that experienced as a DM, I've never been in games based around walking from hex to hex, or strict geography. Groups I've been in tend to want to go somewhere specific, and are more interested in what happens there than in the intervening geography and any semi-random content that might have been allocated to it. This means it’s more important to devise narrative locations than geographical ones. The Dark Tower and the Big Lake are locations. N13E04 is a datum.

Being creative does not, in fact, make you less creative. The more you create, the more your output increases! Let's ignore that there's enough free material on the web to stock 1 millarn over 9000 hexes and dungeons with no more effort than hitting print... (Cat again)

Random unconnected stuff does not fit well in my games. I want things to make sense, both as a DM and a player, as do other members of my group. They want to know what people were doing somewhere, and why. They like to investigate the history of things, or ask about their religious significance. They generally aren't very interested in being attacked by six generic bandits, or even investigating a cave several miles away because “it’s there”. As such, it takes not just time but also creativity to come up with things that fit the location and the situation (let alone the setting material as developed so far, the tone of the campaign, and so on). It's difficult to transpose material that's quite specific, particularly into a new campaign that may not share the same kinds of assumptions. And there is a vast amount of free most things on the web, but as I can say professionally, the main problem with information is excluding the bits you don't want. If your campaign can readily accomodate absolutely anything you grab off the internet, I don't think I want to play in it. The vast majority of stuff will need adapting at the very least, perhaps rebuilding from scratch, and that means time. Time is precious.

Random encounters are not everyone's cup of tea - players I know aren't generally keen to just fight something because they saw it on the road somewhere, and predesigned tables may well not fit the tone of your campaign. There are some very large tables, with truly interesting and creative possibilities, but it’s a bit of a devil trying to make these conform to your own campaign. I’d also say that the less your group is interested in Killing Things and Taking Their Stuff as prime game content, the harder it is to come up with good random content that will enthuse people and make sense to them.

Quantum Ogres are an attractive prospect for exactly this reason: a DM can produce something interesting and relevant, and ensure that the players will see it. There is nothing about Quantum Ogres – even as defined by Cat, who dislikes them – that obliges the players to interact with them in any specific way, or indeed at all. All you are doing is removing the possibility of never knowing that the Ogre exists.

Sunday, 22 September 2013

I Defend the Castle (Against Magic)

So, as promised, this is a defensive counterpart to my post about the issue of magic and sieges.


So what I'm suggesting is that fortifications would adopt a range of measures against the threat of magic (and some associated threats), including physical changes, magical countermeasures, and practices designed to limit the danger posed by wizards. This would be a natural development, in the same way that fortifications adapted in the face of gunpowder.

I'm not suggesting that castles should be entirely immune to magical attack, any more than they are to mundane attack. Everything is in balance, and gold weighs a lot. However, I think in general a fortress should have similar resistance to magic as it does to force. It doesn't especially matter if a high-level wizard can breach the castle fairly readily, because a high-level wizard is probably equivalent to a bevy of siege engines. Nor do I think each castle should be resistant across the board; as with other features, there should be strengths and weaknesses, and a canny wizard will test, discern and exploit those weaknesses. That being said, I think it's also reasonable to say that getting through those defences ought to take time, attract attention and require some genuine effort. If you think about it, if a decent wizard can breach a castle's defences and leave it vulnerable in the space of a few minutes, they render castles and siege engines entirely redundant.

Changes in approach

Some existing features of castles would be much less useful in a magical setting. Traditionally keeps would have a door only on the first floor, with wooden staircases leading up that can be demolished to keep attackers out: this would be much less useful in a world where wizards can conjure up magical stairs, create ramps from earth or repair the shattered stairs with a word and a gesture.

On a bigger scale, people tend to think in terms of bottlenecks, relying on really basic things like the difference between walls and doors. But doors become massively less crucial as strategic points if people have the ability to pass easily through walls; they can avoid bottlenecks and kill-zones, and it's far more difficult for small numbers of troops to hold off larger numbers by controlling the terrain. While it's always possible to knock holes in walls, this tends to be pretty obvious, and also carries a risk of collapsing the wall entirely, which might kill you or leave your routes even more blocked than they were before. In a magical world, people would need to think of new doctrines that didn't focus so strongly on control of routes.

Tactical changes


Disguises, shapechangers, and above all invisible creatures are major problems for a fantasy castle.

A very straightforward countermeasure is to have a significant number of guard dogs, or equivalent creatures, relying on hearing or scent to detect and locate unwanted visitors. I can also envisage buckets of flour or paint kept at regular intervals around the site, or perhaps even remove invisibility or glitterdust items.

A wealthy castle might have permanent true seeing items, such as scopes or helms. They might even be built into a viewport in the main games. There might be a number of them, perhaps one at each of the watchtowers, or they might be portable items passed from watch to watch.

A magically-sophisticated civilisation might have more complex effect; for example, a kind of security gate might detect invisible or disguised creatures and cast counterspells on them, or simply alert nearby guards.

Speaking of which, I think it's likely that any castle worth its salt would use the cheap and cheerful alarm spell, and even mage's faithful hound. Wards would likely be placed around the perimeter, as well as in restricted areas, just as they are nowadays. Magical wards such as glyphs of warding offer the chance to identify individuals, demand passwords and so on. Wards might be purely for detection, but in the harsher sorts of societies (and considering these are military installations) they're quite likely to imprison, hurt or even kill trespassers. For some more thoughts on lethality of traps, you might want to consider this post.


Having guards taken out quietly by a long-range sleep spell is a bit of an issue - at least they might scream if an arrow gets them - and countermeasures would be sensible. One that occurred to me was actually to instigate a dead man's handle on guard posts. If the person on shift lets go of the handle, or (in another design) steps off their platform, it will immediately sound an alarm. Given magic, you could have a magical 'handle', or simply a more subtle kind of alarm. It would be entirely possible to have other guards insta-summoned to the spot.

Another option is simply to ward guards against things like sleep, perhaps with a side-effect of shielding them from physical tiredness. Wealthy rulers might provide helms, amulets, rings or simply potions that will keep their guards alert throughout the night. Some will be in a position to provide at least some sleepless guards, be they fey, undead, golems or elemental watchers - many of whom are also immune to the other cheap and cheerful enchantments that pose such a threat.


There are several ways to magically spy on a castle, but magic also offers some options to foil each one.

Much as fortresses may use camouflage paint and thermal paint to protect against aerial photography, they can invoke techniques against flying spies. One fairly simple one would be a roof. In a world with easily-available magical light, you don't especially need to have natural light. It'll be harder to use the inner castle to raise animals or crops, but perhaps you're willing to accept that (or call on further magical resources to compensate). For very swish places, you might have a roof imbued with magic, so that it's transparent from below but opaque from above. Another option is a big old illusion that blocks the castle from above - you could adapt hallucinatory terrain or just devise your own spell.

Thinking about it, a really huge sheet of gauze would actually tend to work. The light from above will come through the fabric, but (as with net curtains) you won't be able to see through to the inside unless it's dark and there's loads of lights on. Magically-reinforced gauze, anyone? On the downside, it would cut down on the light getting through.

For scrying, countermagic is needed. Any ruler worth their salt is likely to have invested in developing versions detect scrying that will protect a whole area, as well as permanent versions of screen. A cheaper (lower-level) variant might dispense with deception and simply blur an area or cover it with obvious illusory patterns. A nastier version might involve a full-blown counterattack: a counterscry on the scrier, a psychic backlash, domination tracing back the magical thread to its source, or summoning a couple of fire elementals on top of the wizard in question.

Even detection is a good start - this would warn defenders that something was up, and give them a chance to prepare for attack and start trying to track down the wizard.

Worried about charms and other enchantments? Key personnel could be equipped with anti-enchantment gear, so that even when wandering in town at night there's no chance of them being compromised. For a subtle version, don't block the enchantment, but rather have the ward immediately alert castle personnel so they can prepare suitable countermeasures while leaving the attackers thinking they've succeeded. For a surprise version, have the amulets reflect enchantments back onto their casters!


One very obvious area for magical intervention is in making the castle plain old tougher. I've discussed a number of ways that magic allows easy damage or destruction to castles, but by the same token magic can prevent that sort of thing.

For starters, it's very likely that any magic-heavy universe would be using magic for construction. This would allow for structures well beyond the capabilities of the mediaeval period: bigger, taller, more complicated. If you can teleport stone, transform the local geography or levitate segments into place, you can do a lot more, very likely in a shorter time and on a smaller budget. Magic might allow the use of something like reinforced steel frames without centuries of research into metallurgy, or develop concrete earlier. Vast moats can be excavated, and filled with water, spikes shaped from the living rock, horrific poisonous plants or fire direct from the Elemental Plane. Almost any castle can be placed on a 100m-high plateau, even in the middle of your Netherlands-equivalent. They can be surrounded by marshes, and reachable only by bridges that magically detect your religious affiliation or ancestry.

Next, you can soup up the castle in various ways. Spells that grant resistance to attacks are widespread, and would certainly be adapted for use against siege engines. Imagine a castle wall that reflects boulders and cannonballs like rubber, hurling them back on attackers. Wooden palisades could easily be rendered fireproof, or imbued with druidic magic that causes them to sprout vicious thorns when someone attempts to climb them - perhaps they regrow when damaged, too.

Attacks coming from above are a problem for any castle, let alone one faced with flying enemies. As a castle-builder, I'd immediately have my wizards develop spells that produce some kind of force bubble, somewhere between a tiny hut and a resilient sphere. A permanent force wall would do the trick, too.

Similarly, mining is a traditional anti-castle tactic that magic can deal with far better than force. Have your transmuters turn all the ground beneath your castle into granite, so mining will take several decades.

Plus, poisoning and infecting castles is one of the favoured ways to destroy a garrison. But there's no need to worry about poisoned water if you have portals to the elemental planes, or magical water. Similarly, a few suitable artefacts or clerics can render ordinary disease a non-issue.

Defensive magic

At some point, you need to get into actually defending the castle.

For starters, you want to ward the castle against the most obvious attack spells. These will by no means provide absolute protection, any more than a wall is absolute protection against physical attack, but the wards will make it much slower and tougher to demolish your castle. The wards will be long-lasting in most cases, and wizards aligned with the castle-owner will come to do regular maintenance, just as architects do. Quite likely, several powerful wizards will create the initial ward, but less skilled ones can maintain it by following the plans. Perhaps have the walls reflect spells if possible, so that fireballs and so on will rebound to fry the wizards in question. If you're very good, perhaps you can actually absorb and use the magic for your own purposes.

Conjure up earth elementals and similar beasts, and bargain or force them into patrolling your territory and devouring any would-be miners. Bury objects with protective glyphs all around the castle's outer walls, enchanted to activate when miners approach and cast the nastiest spells against them - although subtler ones, like messing with their sense of direction, could be handy too.

Next, consider ordinary attackers. Some dock-off enchantments around the walls could be more effective than any physical defences, disheartening or bewildering soldiers that come nearby. I'm picturing this as just massive glyphs carved into the walls - your symbols of fear or whatever, but in a siege-specific form that can be activated by the defenders.

Then, too, your attackers aren't the only ones that can use summoning spells. A thinly-garrisoned fortress could receive reinforcements immediately with a few well-placed triggerable summoning spells, ranging from troops stationed elsewhere, to simple swarms of rats, all the way up to major demons. Similarly, a teleportation gate could allow you to travel rapidly between a network of fortresses. This could be a weakness, of course, and would need careful handling. I'd suggest having it in a location that can itself be locked down, where allied troops must be allowed out, and enemy intruders can be penned in. Of course, something like an alignment lock on the gate would potentially do. Or have a password that controls whether you end up in a deep pit covered by a grating, or in the main courtyard.

Flight has been a major concern, and so you may want to take special action against flying enemies. A nice idea you could borrow is the doubling crows from Jack of Fables (the only volume I read in that series), which are basically crows that split in two repeatedly and devour anything they see in the sky when unleashed. Alternatively, just have standard nasties like air elementals, giant eagles, small dragons, griffons and so on patrolling. Air elementals are a good option simply because they're tireless, largely invisible and don't require food. A flying golem could also be good.

Worried about teleportation? Add in some spells to mess with that. Perhaps any teleportation spell that doesn't incorporate a special control word with drop people into a pocket dimension. While "inside a volcano" is tempting, remember that your own wizards will have to bespell this stuff, and will also be the ones most likely to cast teleportation spells around the place - they aren't likely to like a setup where a slip on their part will be fatal.

General stuff

In a setting like this, you'd be likely to have magic items crafted specifically for use in sieges, as well as siege-specific spells. As well as their tactical value, important members of the garrison might carry items that let them override certain spells, pass guards and so on, which provides a new point of vulnerability.

It also occurs to me that many magical creatures might act like Swiss mercenaries in the old wars, taking on contracts to defend or attack castles for a fee. The ability to hire enough earth elementals, or even ghosts, might be a crucial factor in deciding wars. Clerics and wizards of certain spheres might offer a vast advantage in particular tactics: dwarves' tendency towards Earth magic might well give them impregnable fortresses and make them terrifying besiegers.


If you're planning to introduce these kinds of setting elements, you'll eventually want some way to handle the mechanics. I'm inclined to say you should add more stages to things like, as in our first example, using transmute rock to mud to demolish walls. Perhaps there is a top layer of general protective spells, which must first be dispelled, and then a specific anti-spell to overcome. Each stage in the process might call for an extended magical "duel", with the caster matching their strength against the original warder to try and break down the protective spells. Alarms might be built into the spells, and so care is needed (perhaps an Intelligence, Spellcraft or even Stealth check) to avoid and defuse them - you might want to allow some means to shut down or intercept these spells if the check is failed, just to avoid one-roll failures. Basically, make it a more complex process that calls for more interesting decisions and - crucially - takes some time. It is, in many ways, the speed of spellcasting that made it an issue in the first place.

There's probably a lot more you could say here, but I'm tired now and this seems long enough. Hope someone enjoys it.

Monday, 16 September 2013

Deathwatch: The Price of Hubris, part 02

Contains spoilers for, funnily enough, The Price of Hubris from The Emperor Protects.

As always, be aware that the podcast is not really family-friendly, and features a fair amount of background noise, if that sort of thing bothers you.

The Episode

I don't have much to say about this session. It does highlight some of the complications when some players have in-depth knowledge of a campaign setting (in particular, details that greatly affect how their character is expected to act and think) and others have very little, but I think we managed pretty well in the circs.

Link to Episode 02.

Monitors: combat maths


So far I've got one injury model I'm mostly looking at, with one backup. I need to look at the numbers for weaponry, to try and get them about as effective as I want. Fair warning, this is a fairly long and relatively dry post, so if you're looking for reptile jokes and shoehorned song references, I'd wait for another one.

Let's start with some assumptions.

The system needs to cover targets wearing anything from 'nothing' to heavy battle armour, with common targets including criminals (with ordinary clothing), pirates (lightly armoured), wild beasts (unarmoured) and military threats (moderate armour). In addition, many creatures will have some degree of innate armour from scaly skin, barklike flesh, thick blubber, being a robot and similar protection.

Weaponry needs to cover a range from bare-knuckle brawling, through primitive clubs and slings, focus heavily on futuristic handguns and blasters, and allow breathing space for lizard-portable heavy weaponry and vehicle-mounted hardpoints. While the setting features starship-mounted artillery and the like, I don't need a system to rigorously model what happens when you fire those at a person: I think we can guess.

Attack Roll, Wound Roll model

Attack Rolls

  • An attack roll involves a roll under the attacker's skill. The defender may roll to evade if in melee or if they spent an action for evasion on their previous turn. If they fail, they are affected by the attack.
  • A wounding model handles most physical attacks, while blinding damage is modelled with a blind die and slowing with a slow die. Restraint by ropes, webs and so on requires a Strength check to escape (with occasional allowance for Houdini stunts).
  • A wound requires a roll of 11+ on the d20, with weapon strength as a bonus and armour (or other defence, depending on the weapon) as a penalty. This number should be tweaked so that Monitors have around a 75% chance of wounding an average target - coupled with the need for skill rolls to hit, they should end up with about one wound per two actions, which is a single round.
  • An unsuccessful roll leaves the target pinned: they suffer a penalty until they spend an action to recover.
  • Monitor-grade characters have 3 wounds, civilians 1, and hulking alien monstrosities 5 or more.

We're looking at 1d20 rolls of 11+, with armour and weapon as opposing modifiers. If we want to allow some scope for any weapon to affect any target, then the greatest possible difference between armour (including hide) and weapon strength must be 9. At a pinch, we might allow heavy armour to provide complete protection from unarmed attacks and possibly from primitive weapons like clubs - not sure about this. Given the system is non-lethal anyway, I think allowing a mob of animals or civilians to overwhelm a character is probably okay. Moreover, because we have Pinning rules, we could allow very minor creatures to have a bigger differential than that, overcoming the classic Wizard/Cat phenomenon while allowing for particularly puny NPCs to, in fact, be temporarily disabled by a housecat.

Let's assume that armour 0 represents soft squishy creatures with no armour, 1 indicates thick civilian clothing or scales, 3 is basic protective gear, 5 is riot armour and 7-8 is military grade. Armour values above that are restricted to heavy infantry, vehicles and serious monstrosities and will be occasional occurrences rather than regular opponents.

See below for number-crunching. I couldn't easily get the table to look how I wanted; while the colours provide a broad indication, it's important to note that we're actually not aiming for the attractive green zone of 100% kill rates, but actually more of a fetching yellow.

Chance of causing a wound by modifiers

Weapon bonus
Armour penalty   0 1 2 3 4 5 6 7 8 9 10
0 50% 55% 60% 65% 70% 75% 80% 85% 90% 95% 100%
-1 45% 50% 55% 60% 65% 70% 75% 80% 85% 90% 95%
-2 40% 45% 50% 55% 60% 65% 70% 75% 80% 85% 90%
-3 35% 40% 45% 50% 55% 60% 65% 70% 75% 80% 85%
-4 30% 35% 40% 45% 50% 55% 60% 65% 70% 75% 80%
-5 25% 30% 35% 40% 45% 50% 55% 60% 65% 70% 75%
-6 20% 25% 30% 35% 40% 45% 50% 55% 60% 65% 70%
-7 15% 20% 25% 30% 35% 40% 45% 50% 55% 60% 65%
-8 10% 15% 20% 25% 30% 35% 40% 45% 50% 55% 60%
-9 5% 10% 15% 20% 25% 30% 35% 40% 45% 50% 55%
-10 0% 5% 10% 15% 20% 25% 30% 35% 40% 45% 50%


Looking at this table, I think we're looking at a weapon strength of 7-8 for the go-to Monitor weapons if we want a ~70% wound rate on average targets. This will allow... okay, someone with Skill 4 (basic training only) will hit 20% of the time, and so cause a wound 15% of the time overall, which with two actions a turn means they're likely to wound someone every third turn; if they can get the drop on them or otherwise gain an advantage, though, that'll increase significantly (because accuracy is the main problem, with a +4 bonus to hit, every second shot is likely to cause a wound). A more military character who actually takes some weapons training (Skill 10) has only a 39% chance of failing to wound with two shots.

Of course, there will be other factors to consider. Some weapons should be weak against certain enemies and strong against others. Large weapons are unwieldy, but may be more powerful or simply affect more targets. Light weapons can easily be concealed, drawn and even used in a brawl. Some weapons are useless in the wrong circumstances, or have side-effects.

A problem here (in one sense at least) is that the numbers involved mean weapons will be heavily concentrated towards the top of the table, simply because of probability. With a minimum skill of 4 for Monitors, rolling under skill on 1d20 to hit and needing an 11+ on a d20 Wound roll, they have only a 10% chance of causing a Wound. This means that as soon as armour comes into the equation, weapons need to be at least equally good to offer a significant chance of success. A +4 overall modifier will increase the chance to 15%. Of course, one way to look at this is that the minimum skills are exactly that - representing a Monitor whose skills lie elsewhere, and who uses weapons as a fall-back. If they've upped their skill to 8, they're already succeeding 20% of the time when weapon strength equals armour, which tends to mean about a 35% Wound rate per round if they're just attacking. Still not huge, admittedly.

Sanity check

Pause a minute. Is this actually a problem? What success rate will be satisfying?

D&D is my go-to for combat success rates.

  • AD&D. A 1st-level fighter has THAC0 20 with a +1 specialisation bonus, and most likely no Strength bonus. Against a level-appropriate goblin (AC 6) this calls for a roll of 13, giving a 40% chance of goblin-brutalising. A wizard has a 35% chance. Either will likely kill the goblin in one hit.
  • D&D 3.5. A 1st-level fighter probably has Str 16 and a +1 BAB, for 1d20+4 overall. Against a level-appropriate goblin (AC 15) this gives exactly a 50% chance of hitting, and almost certainly killing it. A wizard is more likely to have Str 8 and a +0 BAB, for 1d20-1 overall, which gives a 25% chance of success.
  • D&D 4E. A 1st-level fighter probably has Str 18, 1/2 level (+0) and a +2 proficiency bonus , for 1d20+6 overall. Against a level-appropriate goblin (AC 16-17) this gives a 50%-55% chance of a hit. A minion will be killed instantly, but other types may easily have 30 hit points, and will survive three or four hits (at 1d8+4) on average. This means the fighter will take around eight attacks to kill the goblin! That's a very dramatic difference from earlier editions, and results in extended fights that I for one found tedious. I think three to four is more my speed for your thug-level creatures. Of course, fighters aren't the big damage-dealers... a striker will add extra damage, around 1d10+4 to 2d8+4, dropping that to two or three hits. A wizard functions very much like a fighter here so there's no real difference.

At this point, I'm inclined to think that it's not a huge problem if unoptimised characters have a low chance of success. I'd perhaps prefer a 20-25% level, simply because Monitors are supposed to be competent and it's annoying to constantly fail. On the downside, that's actually more or less impossible in the current model because they have only a 20% chance of hitting the target, and so would need a more or less 100% wound chance. The other problem here is that differentials have different effects at different skill levels: each +1 to weapon is roughly +1% for a skill 4 character, but a full +5% for a skill 20 character. Of course, I could simply up the skill level of Monitors. Or I could decide that if a professor of physics who passed basic weapons training ,but mostly investigates space-time anomalies and negotiates with local officials, can only drop a hostile moving target one time out of five, that isn't actually a problem. That decision is likely to come in a future post.

Obligatory table of wounding chances:

1d20 wound roll cumulative wound chances by target number and skill

Target number (as modified)
Skill 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20
1 5% 5% 5% 4% 4% 4% 4% 3% 3% 3% 3% 2% 2% 2% 2% 1% 1% 1% 1% 0%
2 10% 10% 9% 9% 8% 8% 7% 7% 6% 6% 5% 5% 4% 4% 3% 3% 2% 2% 1% 0%
3 15% 14% 14% 13% 12% 11% 11% 10% 9% 8% 8% 7% 6% 5% 5% 4% 3% 2% 2% 1%
4 20% 19% 18% 17% 16% 15% 14% 13% 12% 11% 10% 9% 8% 7% 6% 5% 4% 3% 2% 1%
5 25% 24% 23% 21% 20% 19% 18% 16% 15% 14% 13% 11% 10% 9% 8% 6% 5% 4% 3% 1%
6 30% 29% 27% 26% 24% 23% 21% 20% 18% 17% 15% 14% 12% 11% 9% 8% 6% 5% 3% 2%
7 35% 33% 32% 30% 28% 26% 25% 23% 21% 19% 18% 16% 14% 12% 11% 9% 7% 5% 4% 2%
8 40% 38% 36% 34% 32% 30% 28% 26% 24% 22% 20% 18% 16% 14% 12% 10% 8% 6% 4% 2%
9 45% 43% 41% 38% 36% 34% 32% 29% 27% 25% 23% 20% 18% 16% 14% 11% 9% 7% 5% 2%
10 50% 48% 45% 43% 40% 38% 35% 33% 30% 28% 25% 23% 20% 18% 15% 13% 10% 8% 5% 3%
11 55% 52% 50% 47% 44% 41% 39% 36% 33% 30% 28% 25% 22% 19% 17% 14% 11% 8% 6% 3%
12 60% 57% 54% 51% 48% 45% 42% 39% 36% 33% 30% 27% 24% 21% 18% 15% 12% 9% 6% 3%
13 65% 62% 59% 55% 52% 49% 46% 42% 39% 36% 33% 29% 26% 23% 20% 16% 13% 10% 7% 3%
14 70% 67% 63% 60% 56% 53% 49% 46% 42% 39% 35% 32% 28% 25% 21% 18% 14% 11% 7% 4%
15 75% 71% 68% 64% 60% 56% 53% 49% 45% 41% 38% 34% 30% 26% 23% 19% 15% 11% 8% 4%
16 80% 76% 72% 68% 64% 60% 56% 52% 48% 44% 40% 36% 32% 28% 24% 20% 16% 12% 8% 4%
17 85% 81% 77% 72% 68% 64% 60% 55% 51% 47% 43% 38% 34% 30% 26% 21% 17% 13% 9% 4%
18 90% 86% 81% 77% 72% 68% 63% 59% 54% 50% 45% 41% 36% 32% 27% 23% 18% 14% 9% 4%
19 95% 90% 86% 81% 76% 71% 67% 62% 57% 52% 48% 43% 38% 33% 29% 24% 19% 14% 10% 5%
20 100% 95% 90% 85% 80% 75% 70% 65% 60% 55% 50% 45% 40% 35% 30% 25% 20% 15% 10% 5%

In case anyone needs to do anything similar...

The chance of wounding is calculated as (chance of rolling X or better on 1d20)*(chance of rolling Skill or less on 1d20).

For a one-die roll, the chance of rolling X or better is calculated as (1 - (X-1)*(1/sides)) and the chance of rolling Skill or less is (Skill*(1/sides))

For multi-die rolls, I produced a table of possible 2d6 rolls, and used a CountIf function to tot up occurrences of each roll. A secondary row then simply added up each number's probability and the previous cell's cumulative probability (zero in the first cell), to produce a cumulative probability.

Maybe I'll do another post just for this stuff...


Curved model

One possibility would be to switch damage rolls to a 2d10 roll or similar. What would that do?

With a 2d6 roll needing a 7 or better, a skill 4 character would need a +5 on weapons to get a 20% Wound rate. That being said, a mere +1 or +2 would allow around a 15% rate, which is better than the previous option. To hit 50%, you'd need to have about skill 12 and a +2 weapon.

Changing the die size doesn't improve matters - you need a bigger modifier with d10s. However, one thing it does do is create a bell curve. Is this a good thing? Well... if I'm not using variable damage then a bell curve is irrelevant, since we're modelling only change of overall success rather than degree of success, and a flat graph with cutoffs does that perfectly well. If I do want to use variable damage, then a bell curve would be worth investigating, but at the moment, no.

2d6 wound roll cumulative probabilities

Target number (as modified)
Skill 1d20 2 3 4 5 6 7 8 9 10 11 12
1 5% 5% 5% 4% 4% 3% 2% 1% 1% 0% 0%
2 10% 10% 9% 8% 7% 6% 4% 3% 2% 1% 0%
3 15% 15% 14% 13% 11% 9% 6% 4% 3% 1% 0%
4 20% 19% 18% 17% 14% 12% 8% 6% 3% 2% 1%
5 25% 24% 23% 21% 18% 15% 10% 7% 4% 2% 1%
6 30% 29% 28% 25% 22% 18% 13% 8% 5% 3% 1%
7 35% 34% 32% 29% 25% 20% 15% 10% 6% 3% 1%
8 40% 39% 37% 33% 29% 23% 17% 11% 7% 3% 1%
9 45% 44% 41% 38% 33% 26% 19% 13% 8% 4% 1%
10 50% 49% 46% 42% 36% 29% 21% 14% 8% 4% 1%
11 55% 53% 50% 46% 40% 32% 23% 15% 9% 5% 2%
12 60% 58% 55% 50% 43% 35% 25% 17% 10% 5% 2%
13 65% 63% 60% 54% 47% 38% 27% 18% 11% 5% 2%
14 70% 68% 64% 58% 51% 41% 29% 19% 12% 6% 2%
15 75% 73% 69% 63% 54% 44% 31% 21% 13% 6% 2%
16 80% 78% 73% 67% 58% 47% 33% 22% 13% 7% 2%
17 85% 83% 78% 71% 61% 50% 35% 24% 14% 7% 2%
18 90% 88% 83% 75% 65% 53% 38% 25% 15% 8% 3%
19 95% 92% 87% 79% 69% 55% 40% 26% 16% 8% 3%
20 100% 97% 92% 83% 72% 58% 42% 28% 17% 8% 3%

2d10 wound roll cumulative probabilities

Target number (as modified)
Skill 1d20 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20
1 5% 5% 5% 5% 5% 4% 4% 4% 3% 3% 2% 2% 1% 1% 1% 1% 0% 0% 0%
2 10% 10% 10% 9% 9% 9% 8% 7% 6% 6% 5% 4% 3% 2% 2% 1% 1% 0% 0%
3 15% 15% 15% 14% 14% 13% 12% 11% 10% 8% 7% 5% 4% 3% 2% 2% 1% 0% 0%
4 20% 20% 19% 19% 18% 17% 16% 14% 13% 11% 9% 7% 6% 4% 3% 2% 1% 1% 0%
5 25% 25% 24% 24% 23% 21% 20% 18% 16% 14% 11% 9% 7% 5% 4% 3% 2% 1% 0%
6 30% 30% 29% 28% 27% 26% 24% 22% 19% 17% 14% 11% 8% 6% 5% 3% 2% 1% 0%
7 35% 35% 34% 33% 32% 30% 28% 25% 22% 19% 16% 13% 10% 7% 5% 4% 2% 1% 0%
8 40% 40% 39% 38% 36% 34% 32% 29% 26% 22% 18% 14% 11% 8% 6% 4% 2% 1% 0%
9 45% 45% 44% 42% 41% 38% 36% 32% 29% 25% 20% 16% 13% 9% 7% 5% 3% 1% 0%
10 50% 50% 49% 47% 45% 43% 40% 36% 32% 28% 23% 18% 14% 11% 8% 5% 3% 2% 1%
11 55% 54% 53% 52% 50% 47% 43% 40% 35% 30% 25% 20% 15% 12% 8% 6% 3% 2% 1%
12 60% 59% 58% 56% 54% 51% 47% 43% 38% 33% 27% 22% 17% 13% 9% 6% 4% 2% 1%
13 65% 64% 63% 61% 59% 55% 51% 47% 42% 36% 29% 23% 18% 14% 10% 7% 4% 2% 1%
14 70% 69% 68% 66% 63% 60% 55% 50% 45% 39% 32% 25% 20% 15% 11% 7% 4% 2% 1%
15 75% 74% 73% 71% 68% 64% 59% 54% 48% 41% 34% 27% 21% 16% 11% 8% 5% 2% 1%
16 80% 79% 78% 75% 72% 68% 63% 58% 51% 44% 36% 29% 22% 17% 12% 8% 5% 2% 1%
17 85% 84% 82% 80% 77% 72% 67% 61% 54% 47% 38% 31% 24% 18% 13% 9% 5% 3% 1%
18 90% 89% 87% 85% 81% 77% 71% 65% 58% 50% 41% 32% 25% 19% 14% 9% 5% 3% 1%
19 95% 94% 92% 89% 86% 81% 75% 68% 61% 52% 43% 34% 27% 20% 14% 10% 6% 3% 1%
20 100% 99% 97% 94% 90% 85% 79% 72% 64% 55% 45% 36% 28% 21% 15% 10% 6% 3% 1%

Bigger modifiers

Another possibility is to decide that armour and weapons can have substantially bigger modifiers. Maybe +5 is a weak weapon, +10 a normal one and +20 a heavy weapon.

With a normal +10 weapon and a +5 lightly armoured target, you'd have a... 15% chance again. A solid skill 12 shooter would have a 45% chance per shot, and a 16 skill sharpshooter a 60% chance. That's actually not too bad. Against an unarmoured target you'd automatically succeed. A heavy weapon would injure anyone without at least 11 points of armour. This might work okay.

Sharpshooting model

We could also allow some interplay between the hit roll and the wound roll. At the most basic level, if you roll half your skill or less, you get a +5 bonus on the wound roll. I'm keen to avoid odd little aspects of fixed-roll Critical Hit rules, such as the situation where all hits on a tough monster are critical, so let's see how this one might work out.

Under this model, our skill 4 professor needs a 4 to hit (a 9 against a straightforward target, or a 14 against an easy target). Normally she needs an 11 to wound, but half the time she hits she'll score a critical and need only a 6+. This looks a bit fiddly to model, but broadly speaking we should be able (I think) to average the values for a 4/11 and a 4/6 (slashes are successive rolls), which gives us a 13% in d20/d20 or a 15% in d20/2d10. For higher values, a skill 8 gets 25, a skill 12 34% and a skill 16 50% in d20/d20. 29%, 40% and 58% in 2d10.

I'm actually quite pleased with this one. I think we can work with this.

Percentage chance of wounding with +5 bonus for half-skill hit

Skill d20/d20 d20/2d10
1 3% 4%
2 6% 7%
3 9% 11%
4 13% 15%
5 16% 18%
6 19% 22%
7 22% 25%
8 25% 29%
9 28% 33%
10 31% 36%
11 34% 40%
12 38% 44%
13 41% 47%
14 44% 51%
15 47% 54%
16 50% 58%
17 53% 62%
18 56% 65%
19 59% 69%
20 63% 73%

Single Roll model

Single Success

The Single Success Roll model might be something like: 1d20 + AttackerSkill + weapon modifier - DefenderSkill - armour = damage. So 1d20 + 12 Ballistics + 2 (heavy blaster) - 8 Dodge - 4 armour = 1d20 + 2 damage (however that translates). My main objection here is it's fiddly maths! In practical terms, the player would roll 1d20 + skill + weapon = damage, and the GM would apply damage - defence - armour = result. So it's two separate lots of maths, but it's still something to bear in mind. I'm a bit tempted by this model, though. It's straightforward.

Let's see how this version might play out, adapting it for the current wound model so the overspill from the attack roll becomes the Wound roll.

1d20 + AttackerSkill + weapon modifier - DefenderSkill - armour = Wound roll

Assuming our professor's weapon and armour cancel out, the average will be 10.5 + 4 + 0 - 4 - 0, giving a score of 10. For an equally puny opponent (DS4), that's about a 50% chance of a successful Wound. More likely, the enemy will have a bit of an edge here, and the odds will drop somewhat. A skilled character with skill 10 has around an 80% chance to take down an ordinary criminal (DS4), and around a 50% chance to drop a mercenary (DS10). A real sharpshooter with 15-20 skill can drop a mercenary (DS10) with military armour (+5) one shot out of two and drop henchmen like flies.

Another version would scrap DefenderSkill (for simplicity), and would allow a wider range of armour values to compensate. I think basically this would work either way around, so we can decide later whether or not defensive attributes are a good idea.

Here there are a lot of factors at play in a single calculation, and so it's a bit tricky to decide how they interact. The relative frequency of various armour types will make a significant difference to success over time. Monitor weaponry is likely to make less difference, as PCs tend to broadly stick with the gear they have except where there's a level-based upgrade model.

There's not a huge amount of maths I can usefully do here. Broadly speaking, though, I'm slightly less happy about it for some reason. Not sure why. Any ideas?

That really is an awful lot of stuff (and it takes freakin' ages to do those tables if you're fussy like me and insist on peeling off all the revolting useless bits of code that importing tables adds, and making the cell/row/table structure halfway respectable) so I'll leave it there for now. I still need to think about what sort of narrative results I'd want from various situations, in order to try and juggle the numbers appropriately.