"Essential Clues"

This one's going to be a bit of a rant. I'm sure you're overjoyed to read that. But I've just read something that really bugged me, and I was trying to sort out just why that was.

The topic in question was the GUMSHOE mechanic whereby characters are just given the appropriate clues if they investigate a scene - if the Medic examines the corpse then he just gets told about the disparity between the gunshot wound and the time of death, or whatever. Fair enough - I actually think that's a pretty good idea.

But what bugged me was the discussion of the opposite approach, where the player is required to roll, and how a "good GM" will operate in the GUMSHOE manner anyway.

The argument goes as follows: if the ongoing story requires the PCs to find an essential clue, and they have to roll for it, the game comes to a shuddering halt if they fail that roll. Therefore, the GM must either arrange things so the player can reroll until he succeeds, or otherwise arrange for that information to fall into their laps. In effect, he's having to bypass the roll anyway. And, since that just wastes time and generates frustration, the "good GM" will simply skip all that and give the PCs the essential clue.

That's fine as far as it goes, except for one thing: "if the ongoing story requires the PCs to find an essential clue..."

What we have here is a form of railroading. Whatever else happens in the game, the PCs must do something. They must visit a certain location or they must talk to a particular NPC, or whatever. You've created a choke-point in the adventure. To proceed, they must go this way. And each essential clue represents another choke-point - the PCs have to hit A, and B, and C to proceed; all that matters is the order (ABC, ACB, BAC, BCA, CAB, CBA). That's not a railroad in the classical sense, but it's not far off it.

The key thing is that the "good GM" in the roll-for version of the game would not have any essential clues in the first place. Instead, he'd have liberally scattered lots of clues about the place, any of which might or might not be found (but where, statistically, something like 70% will be found assuming 'good' play), and where a suitable subset will give rise to the right conclusions.

And, actually, the "good GM" in the roll-for version would also not require that the players come to the 'right' conclusions anyway. If they decide to proceed against the wrong man, well that's their business - it just leads to another set of circumstances, no more or less valid than the other. (Specifically, it most likely results in the real villain getting away to strike again, the PCs creating a new enemy, and the PCs losing some of their hard-earned prestige. None of which is ideal from this adventure, but in terms of the wider campaign it's not actually all that bad a thing.)

Now, you'll note that I've not said you must play this way. And, indeed, if you are going to have essential clues in the game then I agree you really should just give them to the PCs. I'm just not convinced about the need for essential clues.

Untested Death and Dying Rule (also healing!)

This is actually aimed at Nutshell Fantasy, my never-actually-to-be-completed pet d20 system, but it's adaptable to other d20 games, so here it is.

As noted, characters have a number of hit points per level: 7 for 'tough' classes, 5 for 'medium' classes, and 3 for 'weak' classes (in D&D, Fighters, Rogues, Wizards, respectively). In addition, they have a 'recovery die' assigned by their class: d12 for 'tough' classes, d8 for 'medium' classes, and d4 for 'weak' classes. Characters then 1 die at 1st level, plus 1 die at each odd-numbered level thereafter (so a 3rd level Fighter would have 2d12).

(Multiclass characters are a bit tricky - I'm not sure whether they should simply use the biggest recovery die, or if they should maintain a mixed pool. My gut feeling is the former would be better.)

Damage works as normal. In particular, note that in Nutshell a character who takes damage (except from "friendly fire") may choose to become Bloodied - this moves him 1 step down the Condition Track, but also allows him to be healed.

At the end of the combat, a character who is Bloodied and takes a short rest regains hit points automatically - roll one (and only one) Recovery Die* and add this to his total. Then remove the Bloodied condition.

The other use of the Recovery Pool comes when a character is knocked unconscious. Immediately on reaching 0 hit points, and each round thereafter on his turn, he must roll his entire Recovery Pool (so, 2d12 for that 3rd level Fighter). Any dice that come up a '1' are removed from his pool until he takes an Overnight Rest. Otherwise, this roll has no effect... until the character's last Recovery Die is removed, at which point he dies.

(If the character takes further damage while unconscious, he must make an immediate, additional roll of his Recovery Pool. This is done instead of tracking negative hit points, which are just a pain.)

* Other types of healing would likewise allow the character to regain hit points by rolling a number of Recovery Dice - 2 dice for cure light wounds, 3 for cure moderate wounds and so on. This closes up the oddity where a tougher character actually benefits less as a percentage of his total from healing spells than does a weaker one.

So, About Those Weapon Proficiencies...

SWSE, being at heart a d20 system game, inherits the weapon and armour proficiency rules from most d20 systems - you get a bunch of proficiencies from your class, you can buy more using feats if you want, and if you use an item without the proficiency you suffer a penalty. That's all reasonable enough.

Where it falls down is in the groupings of proficiencies, and in who has them.

Here's the thing: in the six Star Wars films (not to mention all the EU material I've seen), there is precisely one instance of a character using a weapon without proficiency, excluding scenes where the character is specifically shown to be training. That scene occurs in "Empire Strikes Back", when Han uses Luke's lightsaber to cut open his dead Tauntaun.

That's it, that's the only one. Padme, who in "Phantom Menace" is a pampered child politician with no formal training, quite happily makes use of blaster pistols in the assault on the palace. Luke is shown to be "good in a fight" - knowing how to use blaster rifles, blaster pistols, and even the weapons on an X-Wing without any difficulties. And Obi-Wan, in "Revenge of the Sith", is able to swallow his contempt, pick up a discarded blaster, and shoot Grievous in the heart several times, all while hanging off the edge of a platform.

So, that's one issue I have with the weapon proficiency rules in SWSE.

The other issue concerns the groupings: weapons are listed as pistols, or rifles, or whatever. Once you have proficiency with the group, you can use all weapons in the group without penalty. Now that's fine, mostly. But when the group contains both a blaster pistol and a heavy blaster pistol, where the latter is strictly better than the former - more damage, access to autofire, with the same range and same magazine capacity. There's a very slight differential in cost (but given the priorities of most players, not enough), and a differential in 'legality' (which never shows up in-game). So, the 'standard' version of the weapon is just redundant.

IMO, a better solution would be revert to the standard three-level proficiency arrangement - character's can be proficient in 'normal' weapons, 'military' weapons, and 'signature' weapons.

The 'normal' weapons designation would include all simple weapons, all the so-called "advanced melee weapons", and also the 'normal' uses of pistols and rifles - that is "point and shoot". However, switching the weapon's mode (set to normal vs set to stun) would require a standard action, and they'd remain non-proficient in autofire (and burst fire). All classes would give this level of proficiency.

The 'military' weapons designation allows the character to switch the weapon's mode as a swift action, and grants proficiency in autofire. Only the soldier class would gain this proficiency for free.

Finally, the 'signature' weapons are heavy weapons, lightsabers, and similar. And they work as you would expect - without a proficiency you take a -5 penalty on attack rolls; with the proficiency you do not. Done.

As for the weapons themselves... these should be separated out into 'civilian' weapons and 'military' weapons. The civilian weapons would have a single mode (normal damage or stun damage or ion damage), would be limited to the lower damage level (3d6 for pistols; 3d8 for rifles), and would not have autofire. Military weapons, by contrast, could support multiple modes, grant more damage, and allow autofire. In general, PCs should only be allowed to buy military-grade weapons if they're actively in military service or via the black market (at a hefty mark-up, say x5 cost).

For the record, I'm pretty much happy with the way armour and armour proficiencies work in SWSE - armour is basically useless, but that's in keeping with the depiction in the films.

Happy Birthday D&D!

Apparently, today marks the 40th anniversary of the day when Gary Gygax invited the world (presumably, the movers and shakers in the wargames business) to come and experience Dungeons & Dragons for the first time. In effect, this is as close to a birthday as something like D&D has.

So, happy birthday D&D!

Surprising Errata

The rules for stunning weapons are a bit tricky in SWSE. It seems that somewhere along the line, someone decided they should work differently, and didn't change everything consistently.

Most of the blasters in the game have two damage ratings listed: one for 'real' damage and one for stun damage. So far, so good. When used on stun setting, first compare the damage rolled with the Damage Threshold (and if it beats it, move -2 steps down), then halve the damage and subtract it from the target's hit points. That's not too difficult.

However, in the errata, they make one significant change to this: instead of listing separate damage for the two types, weapons instead simply have a damage rating and an annotation "Yes" to indicate they have a stun mode. Pretty much everything else remains the same. That's also not too hard - in fact, it's a bit easier.

But where it gets wierd is this: all the weapons published after the Core Rulebook promptly ignore the errata, and go back to having two different damage codes for the two modes.

So I'm confused. Did they try the errata and find they didn't like it? Did they simply forget that they'd changed that? Or did the change cause so much confusion that they quietly abandoned it?

--

For what it's worth, I would instead prefer the following:

Weapons have a single damage rating. Some weapons have a single fixed mode (normal, stun, ion), while some can switch between them, though at any time a weapon can only be in one of the three modes.

When firing in stun mode:

1. Roll an attack vs Fortitude (not Reflex!)
2. On a hit, move the target -1 step down the Condition Track.
3. Roll damage, and divide by 2. (Remember to round fractions down.)
4. Subtract the (halved) damage from the target's hit points. If reduced to 0 hit points, the target is knocked unconscious.
5. Compare the (halved) damage with the Damage Threshold. If it is higher, move a further -1 step down the Condition Track.

(Ion weapons, when used against vehicles, droids, or cyborgs, work exactly like stun weapons. When used against organic creatures ignore step 2 above.)

This of course introduces a slightly oddity to the rules, in that almost all attacks target Reflex where this one attacks Fortitude, but I think it's a small enough discrepancy to ignore in favour of speed. (Plus, SWSE probably puts too much emphasis on Reflex, so applying stuff to the other defences is A Good Thing.) In trade, however, it gives a guaranteed effect for a successful stun attack, where currently there's likely to be no meaningful effect, and once you've halved the damage it actually works (almost) exactly like a normal attack - the only slight difference being that a normal attack can kill where a stun attack merely stuns.

There are a couple of other nits I wish to pick with the weapon rules, but that's another rant...

Result!

Last night we had the first session of the "Imperial Fist" campaign for 2014, and it proved to be far more satisfying than the three sessions to date.

Some highlights:

• The issue with the unreliable player has been resolved, with him dropping out of the campaign. This is obviously not the ideal resolution, but at least it is a resolution, and something I can now plan for accordingly.
• The session was much more action-packed than previously. And I think the idea of going for a much more fringe-based campaign is also better. (Coincidentally, this should also allow for a more sandbox-style play, which is something that has been mentioned recently...)
• The 4e-style stat blocks, and the associated methodology, is a massive boon. For the most part, it works indistinguishably from the 'real thing', only it's much easier to prepare for and it allows for some combat stuff that isn't really supported by the rules as written.

Having said that, there were two things that became apparent:

• Vehicular combat doesn't quite work. One of my realisations when setting up the 4e-style stat-blocks was that an enemy on a vehicle could be considered to be nothing other than a 'threat', albeit one with a higher movement rate than normal (and perhaps a few other combat options). Certainly, though, it didn't require a full rewrite of the stat block. However, this only works if PC vehicles work the same way - if the PCs benefit from much higher defences, much higher hit points, and a high level of DR when they get into a vehicle, then they're working on an entirely different scale from the threats, and the system fails. So that will need to be my next fix.
• The game works okay with three players, but it would almost certainly work better with at least one more. Indeed, a full complement of five would probably be best. That said, it seems highly unlikely that that will happen, so I'm going to plan for three going forward. This may or may not result in the campaign being curtailed somewhat compared with my original vision, but that's not necessarily a problem.

All in all, I was well pleased with the way this session went. My hope for the next session is to tie up the current plot-line and set things up for the revised campaign outline, so that the PCs can move forward in a somewhat less constrained manner. This may well involve having them level up a bit sooner than initially expected, but I doubt they'll mind that too much either!

The End of White Dwarf

So, back in April I posted about having bought White Dwarf #400, in which I commented that it was a travesty that it had reached 400 issues while Dragon ended at #359. It appears that Games Workshop agreed, because they have decided to kill off White Dwarf. Well, sort of. What they've actually done is decided to turn White Dwarf into a weekly mag available only through themselves, while also launching a new "Warhammer: Visions" monthly magazine that will be a direct (albeit larger) replacement for the existing magazine.

Someone should probably have told them about the internet, and how it's busy killing off printed magazines.

They will no doubt deny it, but I'm reasonably sure that the new WD weekly magazine will mostly be adverts for their new releases, adverts for the new releases for the next week, and then some rules material that is nothing more than early access to whatever is going to be in a book one or more months down the line. I don't think the magazine can offer unique rules, because that's a quick way to make the game unmanageable (as WotC and Paizo found out with D&D 3.5e). This means that the weekly magazine is essentially without value - the adverts are better accessed online, while if you're interested in the new rules you're better off waiting for the 'real' book anyway - that way, you get the playtested version, rather than the one they've just slapped together.

Meanwhile, the new monthly magazine, "Warhammer: Visions" is a new magazine, and will suffer accordingly. While those few people who still buy actual magazines knew to look for "White Dwarf", the same isn't true of "Warhammer: Visions". And so, they'll lose a lot of their casual readers. They'll also be fighting for space alongside all the video games and sci-fi magazines, most of which will sell better. Basically, only a fool launches a new print magazine these days. (And, incidentally, that's why the printed Dragon will never, or at least should never, be back.)

The other problem they have to deal with is that both these magazines will be competing with their own online versions. This puts them in line for the exact problem D&D 4e had with the DDI and its printed books: in order to keep costs down, you need as large a print run as possible (many of the costs of a print run are fixed, whether you do 100 copies of 10,000), but you also can't afford to have lots of leftover copies, because that's wasted money.

The likely consequence of offering both a printed and online version is that they'll see a split in the customer base - some people will buy print (and refuse to buy online), while some will buy online. Virtually nobody will buy both. Pretty quickly, then, it will become apparent that the print version is a non-starter - enough people are buying online, and thus enough print sales are lost, that it's just not worth printing (and if they push up the cost to compensate, they lose yet more sales, and enter a detah spiral).

But when they take the inevitable step of cancelling print and going online-only, they'll find that the sales of that alone aren't enough to justify all the effort of making it - to be worthwhile, it needs the people who were buying the print version. Alas, they'll find that there are plenty of luddites who would buy print, but who will not buy an electronic magazine.

(Actually, that's not entirely unreasonable. An online mag has to compete with a whole lot of other online material, great swathes of it free, and some of it of very high quality indeed. A physical product, to a certain extent, can sidestep this. So, while the physical copy might offer enough value-add to be worth paying for, the electronic version has to compete with free offerings that are 'good enough', and probably can't. And that's also why the internet is going to kill quality newspapers, but that's another rant.)

So farewell Grombindal. Looks like you finally found a dragon (or troll, vampire, or whatever) you couldn't Slay.

Aha!

Ages ago, I did a post entitled "Three GMing Tricks" in which I repeated three things I'd found on various blogs that I'd found very helpful. Unfortunately, at the time I couldn't remember where two of these three had come from.

Well, I've found the blog in question, which is by Charles Ryan and is found here.

Specifically, here's the post about "Five Things You Know About..."

And here's the one about Bullet Points.

I'm now going to nick two more of his ideas.

The First Session Rule

This one is something that I pretty much did with the Eberron Code campaign anyway, except that I offered the opportunity between Volumes One and Two, and again between Volumes Two and Three. The idea is simple - you run the first session of the campaign as a 'taster', and after it you see how things played out. If a player finds that his character has some poorly chosen feats, or two players find their characters are too close for comfort (or anything else, really), the players get to make changes to their characters. Up to and including a total replacement.

You then play on as if things had always been that way - sure, Takashima has been replaced by Ivanova, and Delenn has an entirely different look (that may include a sex change - that's never quite been tied down), but you just ignore that going foward. It's been retconned out of existence. (But the XP, and memories of key exploits, remain. They're just considered part of that fuzzy "stuff that happened".)

And, of course, the GM likewise gets to make changes as well, though these should be communicated to the players where they should rightly be known.

GUMSHOE's Clue System

This one didn't originate here, nor is it even the first place I've heard about it, but that post provides a good summary of the system. And it's a good system, and a good complement for the "Three Clue Rule".

The idea is very simple: in a location where there are clues to be found, if a character who is at all trained in a relevant skill does anything even remotely like "looking for clues", they should be given the clues - with no roll. If need be, a roll can be added to find further clues, or more detailed clues, but not for the basic ones. Those should just be given.

(So, if a character with even one rank in Gather Information goes to the pub and starts asking around, they should just be given the relevant local rumour. Don't expect them to interrogate Old Pete specifically, and don't expect the to roll on the Random Clue table.)

This is mostly just about speeding up the game, and cutting out the frustration that is often inherent in investigations - the required data is all available, but the players have to spend an age hunting it down, quite possibly by asking to speak to an NPC the GM has forgotten to mention!

(Actually, I'm inclined to do one better than even this. If the PCs are venturing into the "Temple of Elemental Evil" for the first time, in addition to giving them the "Five Things..." list for the location, the DM should ideally give out one or more clue cards to each PC who has invested in ranks in a Knowledge skill. If the PC has Knowledge (religion), tell them about the factions of the Elemental Evil cult. If the PC has Knowledge (history), tell them about the depravations of the cult in the past. And so on and so forth. Again, they can (and should) get more information later by making rolls, but giving them stuff up-front has several benefits: it directly rewards players who invested in non-combat skills, it suggests possible approaches to the adventure beyond just "kill everything", and it can help build flavour. There's not really a down side, other than a requirement to put together those clue cards.)

Oh, yes: the other two nominees on that last blog post are good ones as well!

Revised Threat Generation Process

Okay, this should be my last post on this topic for some time. I've given the matter some careful thought and revised some of the values. The updated process for generating a Threat is given below. Since it's too long to re-read the whole thing, I'll note that the changes are to points 2, 3, 4, and 7, and mostly involve just small tweaks to the numbers - only the attack mode modifiers (#2) and the hit points (#7) have had a significant change made.

Last night I revised the stat-blocks that I had already generated and then compared my go-to Threat (the vanilla Stormtrooper) with the version from the SWSE core rulebook. The result was close enough for me to be happy with. So hopefully this is good to go...

1. Choose the threat's level, role, type, and dominant attack.
2. Determine the creature's attack mode modifiers. These are (4 + level)/8 for the two non-dominant attacks and (16 + level)/8 for the dominant attack. Round all fractions down.
3. Assign Defences. Ref is 10 + level + Ranged modifier. Fort is 10 + level + Melee modifier. Will is 10 + level + Force modifier. If the threat is a Soldier, add +2 to Ref; if a Brute or Artillery then subtract 2 from Ref. At this point, you may wish to add points to one defence and subtract them from another on a 1-for-1 basis. It is recommended that you add or subtract no more than 2 points to any single defence, and that you do so to increase the disparity between best and worst defences rather than equalise them. The threat's Damage Threshold equals its Fort defence.
4. Determine hit bonuses. There is one of these for each attack mode, equal to 4 + level + the appropriate modifier. For Artillery, increase the Ranged hit bonus by +1.
5. Determine the damage for attacks. The base average damage should be 8 + level. This should be increased by +25% for a single-use effect, increased by +25% if the threat is a Brute, decreased by -25% if the attack affects multiple targets, decreased by -25% if the attack also drives the target down the condition track, and reduced by -50% for a minion. For all creatures except minions, this should then be converted into a dice range + modifier. For example, if the base average damage is 10, this could become 1d8+6. (Note that, when counting dice ranges, any leftover fractions should be assume to round down.)
6. From the hit bonuses and base damage values, assign some specific attacks. Most creatures should have at least one 'standard' attack, and probably at least one special attack, probably a per-encounter attack.
7. Determine hit points. Minions have 1 hit point, but don't take damage on a miss. Otherwise, the base hit points are 20 + (5 * level). For Brutes, increase this by 25%; for Artillery or Lurkers, reduce it by 25%. Elites have double hit points; Solos have quadruple hit points. Finally, round all fractions down.
8. Determine Initiative. This is (level / 2) + Ranged bonus. Increase by +2 for Soldiers or Skirmishers, or by +4 for Lurkers.
9. Determine Perception. This is (level / 2) + Force bonus. If appropriate, assign low-light or darkvision to the threat.
10. Assign a skills bonus. This should equal (level/2) + 2. Note that Initiative and Perception are skills, but are handled separately, as indicated above. Note also that the Use the Force skill does not affect the threat's Force-based attacks, if any.
11. At this point, you may wish to assign a few trained skills to the threat. In most cases, this shouldn't need to be done. However, if the threat is particularly noted for some skill, this should be done. It should also probably be done for solo and elite creatures. For trained skills, add +5 to the bonus for those skills. For elite threats, you may wish to assign a skill focus, increasing a single skill by a further +5 (over and above the 'trained' bonus). For solo threats, you may wish to add two different skill focuses. You may wish to increase Initiative and/or Perception in any of these stages; if so, increase the appropriate modifier.
12. At this point, all the cold, numeric data has been determined for the threat. You should now look to flesh the threat out to make it more interesting. This may involve adding one or more special powers to the creature. Minions should usually have a single special power related to its role. Otherwise, Brutes and Artillery should have 0-1 powers, Soldiers and Skirmishers should have 1-2 powers, and Lurkers and Controllers should have 2-3 powers. Leaders should have one extra power, Elites should have one extra power, and Solos should have two extra powers. (Thus, a Solo Controller Leader could have as many as 6 powers!)
13. Assign equipment, descriptive text, or similar to the creature.

More on those 4e-style Threats

I tried out my process last night, and mechanically it works like a charm - whereas the existing process takes about half an hour per creature, mostly spent choosing feats and skills, and ending up with a less-than-satisfactory result, this process requires five entries to a spreadsheet, four of them choices from fixed lists, and then it was just a matter of statting up some suitable powers. Easy.

However, there is one thing that's problematic: my quick scan by eye seems to suggest that the Threats thus generated are way too powerful - they hit more often than PCs of the same level, do more damage, and are just generally better. I'll obviously need to verify this, but it's not the best sign.

It does appear that all the 'offsets' are fine - there's a nice distinction between roles and types, and the three dominant attacks are all sensible. It's just the baseline that needs some work. Perhaps the attack modifiers need reduced, or the rate at which things increase by level. Or something...

4e-style Threats for SWSE

One of the things that is a real pain in 3e and its derivitives is generating stat-blocks for monsters, NPCs, and other threats the PCs might encounter. Because such opponents are built using much the same structure as the PCs, there's a need to juggle ability scores, levels, feats, talents, skills, etc etc... all for a creature that might not be fought at all, and even if it is it gets used for a handful of rounds before it dies.

One of the things that 4e got very right, by contrast, is in its monster design. Each monster is given a role (Brute, Skirmisher, etc), a type (Minion, normal, Elite, Solo), and a level, and from that you could basically stat out the creature. This was quick, simple, and gave monsters that mostly just worked.

Star Wars Saga Edition sits in a slightly odd position in that it incorporates many of the enhancements from 4e, but the 'threats' (the equivalent of monsters) are mostly built according to the 3e paradigm. This actually makes them very lame - in order to be effective, pretty much all monsters have to be built in much the same way, which costs any interesting diversity that the 3e model might potentially grant.

So, for the last day or so I've been considering the possibility of 4e-ing the SWSE threats. And, given the relatively small number of threats detailed in the books, it does look like it might well be possible.

So...

There are six monster roles in 4e:

• Brute: A straightforward melee combatant, a brute hits hard, soaks up damage, but is otherwise unremarkable.
• Soldier: A more involved combatant (melee or ranged), soldiers don't hit so hard, but they have better defences to compensate. Additionally, soldiers probably work best in squads.
• Artillery: A straightforward ranged combatant, artillery hits hard, but is generally both immobile and easily killed.
• Skirmisher: Mobility is key for the skirmisher. They tend to use terrain to their advantage, remain mobile... but are easily dealt with when pinned down.
• Lurker: The 'gotcha' threat, these guys remain out of sight, spring up, do massive damage, and disappear. Or they die.
• Controller The most complex role, these guys tend to alter the nature of the battle or the battlefield itself.
• Leader: A special-case role, a Leader may be a modified version of any other role. They have the same traits as the main role, but tend to add one or two more powers to their comrades, allowing them to benefit from teamwork.

There are also four monster types:

• Minions: Minions are the cannon-fodder monsters. They have only 1 hit point, but never take damage on a 'miss'. They tend to be very simple monsters, but tend also to have one key power to make them a little more interesting.
• Elite monsters are effectively two monsters at once - they're tougher, hit harder, and act more.
• Solo monsters are effectively four monsters at once - like elite monsters, but moreso.

And, of course, there are 'normal' monsters, but they don't need an entry.

Finally, a threat can make its dominant attack in one of three ways: Melee, Ranged, or through the Force. Of course, these need not be the only way the threat attacks, but the main attack will be one of these three.

So, to create a threat:

1. Choose the threat's level, role, type, and dominant attack.
2. Determine the creature's attack mode modifiers. These are (6 + level)/4 for the two non-dominant attacks and (12 + level)/4 for the dominant attack. Round all fractions down.
3. Assign Defences. Ref is 12 + level + Ranged modifier. Fort is 12 + level + Melee modifier. Will is 12 + level + Force modifier. If the threat is a Soldier, add +2 to Ref; if a Brute or Artillery then subtract 2 from Ref. At this point, you may wish to add points to one defence and subtract them from another on a 1-for-1 basis. It is recommended that you add or subtract no more than 2 points to any single defence, and that you do so to increase the disparity between best and worst defences rather than equalise them. The threat's Damage Threshold equals its Fort defence.
4. Determine hit bonuses. There is one of these for each attack mode, equal to 5 + level + the appropriate modifier. For Artillery, increase the Ranged hit bonus by +1.
5. Determine the damage for attacks. The base average damage should be 8 + level. This should be increased by +25% for a single-use effect, increased by +25% if the threat is a Brute, decreased by -25% if the attack affects multiple targets, decreased by -25% if the attack also drives the target down the condition track, and reduced by -50% for a minion. For all creatures except minions, this should then be converted into a dice range + modifier. For example, if the base average damage is 10, this could become 1d8+6. (Note that, when counting dice ranges, any leftover fractions should be assume to round down.)
6. From the hit bonuses and base damage values, assign some specific attacks. Most creatures should have at least one 'standard' attack, and probably at least one special attack, probably a per-encounter attack.
7. Determine hit points. Minions have 1 hit point, but don't take damage on a miss. Otherwise, the base hit point depend on role and level. For Skirmishers, Controllers, and Soldiers, hit points are 24 + (8 * level). For Artillery and Lurkers, they are 21 + (6 * level). And for Brutes, they're 26 + (10 * level). Elites have double hit points; Solos have quadruple hit points.
8. Determine Initiative. This is (level / 2) + Ranged bonus. Increase by +2 for Soldiers or Skirmishers, or by +4 for Lurkers.
9. Determine Perception. This is (level / 2) + Force bonus. If appropriate, assign low-light or darkvision to the threat.
10. Assign a skills bonus. This should equal (4 + level)/2. Note that Initiative and Perception are skills, but are handled separately, as indicated above. Note also that the Use the Force skill does not affect the threat's Force-based attacks, if any.
11. At this point, you may wish to assign a few trained skills to the threat. In most cases, this shouldn't need to be done. However, if the threat is particularly noted for some skill, this should be done. It should also probably be done for solo and elite creatures. For trained skills, add +5 to the bonus for those skills. For elite threats, you may wish to assign a skill focus, increasing a single skill by a further +5 (over and above the 'trained' bonus). For solo threats, you may wish to add two different skill focuses. You may wish to increase Initiative and/or Perception in any of these stages; if so, increase the appropriate modifier.
12. At this point, all the cold, numeric data has been determined for the threat. You should now look to flesh the threat out to make it more interesting. This may involve adding one or more special powers to the creature. Minions should usually have a single special power related to its role. Otherwise, Brutes and Artillery should have 0-1 powers, Soldiers and Skirmishers should have 1-2 powers, and Lurkers and Controllers should have 2-3 powers. Leaders should have one extra power, Elites should have one extra power, and Solos should have two extra powers. (Thus, a Solo Controller Leader could have as many as 6 powers!)
13. Assign equipment, descriptive text, or similar to the creature.

That's the process. It's probably in need of some serious tweaks, since it's basically lifted from 4e and SWSE is a bit different. It's also quite counter-intuitive that the damage ranges will be so different from those the equipment lists say they 'should' be. However, I'm going to suggest that's because SWSE as it stands is too lethal at low level and too lenient at high; this system may well improve the game.

My next step is to convert all this into an auto-calculating spreadsheet or similar, check it for accuracy, and then start converting some threats. But that's for another day.