First off, sorry for the two week absence. The first one was intentional because I felt like crap, the second was unintentional because my internet was out from Wednesday through Saturday because internet providers in America have the same business model as drug cartels. I think it’s back up now, at least until the next almost certainly avoidable but slightly unprofitable. multi-day outage hits, so on with the blog. 

Since I’m going to need to put together some kind of coherent rules document in the next few weeks, I needed to see how the changes I’m planning to implement affect the die rolls. Those of you who have been here a while know what that means: It’s time for Troll Dice and Spreadsheets! One thing I’ve regretted about the recent (planned) changes is that the d20 went away and I hadn’t been able to think of a good way to get it back without unnecessarily complicating the basic dice-rolling or character creation rules. Fortunately when I started putting together the first spreadsheet, I realized that we actually need the d20. With d12 as the free die, the rolls are just too low. You need a Hero Die of at least d8 to even think about rolling a 20, and at least a skill point or two to get to the 5% chance you have on a plain old d20. Since splitting the effect into 10s and 1s digit happens after subtracting the DN or opposing roll, raises would be way too rare if the 10s digit tops out at 1 for most street level characters. Making 20 the default free die  brings the roll ranges up more into the range I want them to be in, where more ability or a really good roll gives you 1 raise but multiples require being both better and luckier than the other guy in most cases. 

On spreadsheet one, the roll is the die combination you’re rolling and the next few columns are the stats on that die combo. Minimum, Maximum, and Average roll are self-explanatory, Dmin and DMax are the average minus the mean deviation and the average plus the mean deviation, which gives a good idea of the common range for that die combination. For example, if you’re rolling d20+d10, about 60% or rolls will be in the 11-21 range, with the remaining 40% split about evenly between low rolls (10 or less) and high rolls (22+). 

The next several columns factor in modifiers, so if you want to see what chance you have to roll a 10 if you’re rolling d8+d10+3, you’d go over to the “Chance of Rolling 10 by Modifier” column group and cross-reference d8+d10 with the “3’ column, which tells you that you’ll beat a DN 10 81.3% of the time with that die and modifier combo. If you want to find out your chances of a modifier higher than +4, you can just shift one column group to the left for every 5 points. Rolling a 10 with a +5 modifier is the same as rolling a 5 with no modifier, rolling 20 with a +12 is the same as rolling 10 with a +2, etc. The first (DN 5) column group only goes to +3 because dAnything +4 gives you 100% chance of rolling a 5. The results are rounded normally to 1 decimal place except for the extreme ends: 99.95% and up are rounded down to 99.9% since there’s still a very tiny chance of failure and anything under 0.05% is marked with an asterisk to show that the chance of a successful roll is non-zero but tiny. 

The table goes up to a Hero Die of 3d12, which is basically a character with Hero Factor 16 with a +8 Core skill. I stopped the DNs at 50 because it’s a nice round number and because the chance of rolling 55 with a modifier or +4 or lower is less than 1% for even the cosmic being rolling d20+3d12.  

Since the default free die is d20, you’re probably wondering why I included all those d12 and lower results. That’s an alternative I’m considering for the “ganging up” rules in combat. In a combat round, your action is the usual d20. Defensive actions start at d12 and the free die drops by 1 each time you defend. So if you’ve been attacked twice this round, your free die for the third attack is d8. Making defense rolls start at a lower dice means you don’t have to half defense rolls, which was necessary in the previous versions to prevent situations where attacking a more powerful character was basically the same as giving him a free attack. I’m also toying with the idea of making wound levels cap your free die instead of giving a penalty. So if you have 3 wound levels, your maximum free die is d8 (the first would drops your free die from d20 to d12, the next gives you a d10 cap, and the third caps your free die at d8). 

To make all those rolls of percentages a little more illustrative, I created spreadsheet 2, which shows the chances of rolling different DNs for sample characters from different power levels and build types. The Power Levels (which determine base Hero Die and stats) are more or less the same as the ones I’ve been using for Hero Factor ranges all along, so I’ll just explain the Character builds:

• Unskilled: The character is jjust rolling Hero Die with no modifiers.
• Skilled: The characters has 1 point in the appropriate Core Skill (so his Hero Die gets 1 bump) and a +1 Edge modifier. 
• Professional: The character’s Core Skill is about halfway between “Skilled” and “Master” (about 25% of Hero Factor for most characters) and the character’s modifier from Edges for the action is 50% Hero Die. 
• Master: The character’s Core Skill is maxed out at half Hero Factor and the Edge modifier is equal to Hero Factor. 
• Munchkin: Core Skill still maxed out, but the character has min/maxed his way into a modifier that’s 150% Hero Factor. 

Mooks and Regular Joes (Hero Factors 1 and 2) aren’t divided into build types because the differences between the different types is so small that it’s not going to make much difference, and because Hero Factor 3 is still the recommended minimum for PCs. The top set of results (before the blank row) is with a d12 Free Die, so it’s basically the standard defense roll for a healthy character. The second batch of results are for a d20 Free Die, so that’s where most PC rolls that aren’t for defense will fall.The results are color-coded: Red: 1% or less; Orange: 1.1-24.9%; Dark Yellow: 25-49.9%; Yellow: 50-74.9%; Light Green: 75-98.9%; Dark Green: 99.9%+. 

The results look more like the range I’m looking for: DN 5 is so easy that it doesn’t need to be rolled unless there are extenuating circumstances; DN 10 is more likely to succeed than fail for most characters; DN 15 is hard to hit (less than 50%) unless you’re playing a high-power game or have put a lot of character points into whatever you’re rolling for; 20 is more likely to fail than succeed unless you’re a super-hero or munchkin; 25 is outside the range of normal human possibility (unless you’re an action hero munchkin), and 30 or higher is difficult even for Superman, so 40 is probably the “Fly Around The World So Fast You Turn Back Time” DN. 

Next Week: Significantly less math! 

Let’s talk about resources, which at least for this post include things like money and equipment, connections and allies, locations (like crime caves or secret hideouts), social capital (rank, status, fame, etc.), and anything else that’s beneficial to the character but not a personal trait of the character. In QAGS, we called (more or less) these things “trappings” and assumed that your character had whatever trappings were appropriate for your job: A knight has a horse, armor and shield, several weapons, a liege, and a lady to champion; A private detective has a trench coat and fedora, a gun, some informants, and hip flask; A super-hero has a goofy costume, a rogues’ gallery, and a secret identity. If there was any question about a specific trapping--Does the knight have a squire? Does the P.I. have a car? Does the super-hero have a crime cave?--you could either make a Job roll or spend Yum Yums to decide. This works great right up until you get into a situation where resources are important to the character or genre: a game of court intrigue were social capital and connections define where the character fits into the game; a con artist game where money is both the goal and a vital resource to set up the next job; or a cyberpunk game where the line between “character trait” and “equipment” is fuzzy at best.

A very slightly less abstract option is to treat resources as a separate stat (either an attribute-style stat that players roll against or a bonus), so you might have a ratings for Money, Gear, Contacts, Reputation, or whatever. These kind of stats require that a lot of ground rules be set: Is the scale absolute or relative to the character concept? Is there a modifier for resources that are/aren’t appropriate to the specific character? What about when the character is away from his usual area of operation? How to you calculate reputation for Robin Hood type who some people hate and others love? Even once the rules are worked out, the only real advantage this kind of system has is a slightly clearer breakdown on the type of resources the character has the most of and a less direct tie between resources and Job stat (which may or may not be desirable). It’s still all abstract enough that it doesn’t help much when resources really matter. 

The next level of abstraction is the Advantage/Disadvantage kind of set-up, where some resources are hand-waved, but some resources (mostly those with obvious game uses) have to be bought with character points. Since the resource traits tend to be more specific than abstract stats, this provides a bit more detail, but there’s always some gray space between “your character would have this” and “you need to spend points for this” that can cause a problem similar to the Thief skills problem I mentioned last week: if you’re listing stuff, most players are going to err toward listing more stuff than they really need too. It also introduces the question of whether and under what circumstances players have to go into character point debt to pay for resources they’ve earned during play, which can lead to tedious character point math. 

The least abstract option is the “track your iron rations and ammo” option where all resources are tracked, but this only works if the PCs are murderhobos and the players don’t mind the tedium of tracking their food supply. As soon as the players find a place to live or buy a van, the amount of stuff you need to list for a complete inventory becomes excessive. It also inevitably results in a situation where the player forgot to list something that the characters would definitely have. To avoid stretching suspension of disbelief, the GM is right back to hand-waving, die rolls, or other abstract resource tracking. 

Even ignoring things like cyberpunk where resources are central to the genre, specifically defining resources to some extent is really useful, especially when you start introducing things like contacts and player-linked locations. Knowing that the characters celebrate their successful missions at an Elvis-themed Mexican restaurant, have a team dog named Bob Barker, or live next door to a hippie commune ties the characters into the world around them and gives them something to fight for. Even equipment can add flavor to the game if the players have fun with it. For example, probably half of the M-Force games I’ve played or run has had a weird or super-cool team vehicle that had its own personality and in some cases came close to that Serenity “10th Character” territory. Having the characters drive around in an ice cream truck, an old van that’s been modified into a war wagon, or a short bus named “George W. Bus” can definitely add something to the game, even if it’s just a running gag or a little quirk. 

My gut instinct (and probably yours, if you’re a regular reader) is that all this kind of thing can be handled with story and you really just need a simple rule (like the “roll Job” mechanic in QAGS) for when there’s a question or disagreement. For pure flavor, that works. The problem is in that murky area between “flavor” and “resource.” The ideal situation is for players to introduce all kinds of flavor and then let it naturally find its role in the story if appropriate, but that doesn’t happen. Some players won’t bother with the flavor because it’s not “useful.” Others are very good at introducing resources disguised as flavor in such a way that the GM will be hard-pressed to justify not allowing the PCs to exploit it later in the game. If others catch on and start emulating the “stealth resources” strategy, you end up with a setting that feels like a video game where every notable minor character or piece of background scenery is something that either gives the PCs and advantage or triggers a side quest once you figure out how to unlock it. Making resources too abstract or hand-waved also makes it hard to play characters whose “Gimmick” is resource-based (Billionaire Playboy, a character with contacts and allies everywhere, the gadgeteer, and even some characters who don’t have one of Batman’s shticks). Either the gimmick seems watered down because everyone has some version of it by default or the character seems unbalanced because you can buy “Filthy Stinking Rich 3” for the same number of points as “Knives 3.” 

Taking all that together, I think I’m starting to see my “want list” for dealing with resources:

• Resources should not require extensive inventory management. The character has some source income, owns things, and knows people appropriate to his role, skills, and other traits. A cop has a badge and a gun, access to crime lab nerds, and a superior who angrily yells “This time you’ve gone too far!”; A gamer has funny dice and expensive rulebooks and knows people who quote Monty Python and the Holy Grail at the drop of a hat; A politician has expensive suits, corporate donors, and a closet full of skeletons. 
• The potential that something can become a resource should not interfere with player-introduced GMCs or world elements. Players should never have to spend points to establish that something exists in the world and their character has a connection to it, though the GM may require a roll or outright veto anything that’s out of character, unbelievable, or transparently self-serving. (“No, Bob, your character didn’t date Charlize Theron in high school, and she’s not coming to his birthday party.”) 
• There should be some kind of abstract mechanic to answer the question “Is this available?,” but there should also be a clear game advantage to defining specific resources.
• When players want to obtain or take advantage of a resource, they can work for it through role-playing and die rolls (haggling over hard-to-find gear, persuading a GMC to help you, etc.), but there should also be some mechanic for just bringing the resource without the “working for it” scene (which happened off-screen, wasn’t necessary, or whatever). This could help making a “Resource Guy” kind of character more clearly useful, since he can just make a phone call and save a lot of time and trouble. 

I’m thinking that the best bet may be a QAGS-type set-up (like the existing Resource Roll in the most recent draft of the Cinemechanix rules) for establishing the existence or availability of questionable resources, but with one or more (probably at least “Stuff” and “People”) Resource stats that you use in the roll. For actually bringing resources into play (without working for it), I’m thinking maybe the Resource stat also works as a pool that the players have to spend to “activate” a resource without doing the legwork. To encourage players to define specific resource, the stat could also allow the player to establish a certain number of specific resources that are always available and can be called upon without spending resource points or by spending resource points at a discounted rate. 

I’ve tried variations on this kind of thing before and the main problem I’ve run into is that they always get way crunchier than I intended. I think the best way to try to develop resource rules is to make it clear that they’re optional (if hand-waving or a simple roll works, use that) and to make the pieces as modular as possible so the players can use the parts that matter for their game and player group without having to keep track of the stuff that isn’t important to them. In fact, that’s probably a good goal for every section of the rules, especially since I have a tendency to design rules far more crunchy than I actually want to use during play. 

Last week I ended with the statement: “My current list of attributes-that-aren’t-quite-attributes has 6 stats: Combat, Thrilling Heroics, Tradecraft, Persuasion, Brains, and Wits. I’ll get into what they mean and how I came up with the list next time.” About 10 minutes later, I’d already changed the list. That’s part of the reason I’ve decided to hold off on trying to get a stable draft of the rules together for a few weeks. I’m getting to the point where I’m starting to get a grasp on what I want and I’m just kind of brainstorming to find the best way to do it. Besides, I’ve got until DieCon in June before I need a playable draft.  

Anyway, the plan from last week was to replace the Attributes with a set of core adventuring super-skills. These would work basically like Attributes (each point in a Core Skill boosts your Hero Die one step), but they’re more directly applicable to the sorts of things players roll for during a typical gaming session and distinct enough to keep overlap to a minimum. In most cases, there shouldn’t be a question about which Core Skill applies to a particular roll. I also want them to be general enough that they don’t say anything about the character beyond “The character is [average/good/really good] at [thing covered by Core Skill].” A high Thrilling Heroics score could be the result of natural athletic ability and years of training that allow you to do nearly impossible stunts, fearlessness that allows you to defy death without breaking a sweat, or some kind of River Tam ability to intuitively math your way around certain doom. Those details are matters of character design that may or may not be backed up with other traits. 

Here’s the (slightly revised) list:

Combat: Killing people and avoiding getting killed. 

Thrilling Heroics: Anything that’s exciting and might get you killed (other than combat). Most physical stunts fall into this category, but it can also include stupid car tricks (or horse tricks, or dog fights or whatever) and anything else that combines high pressure, low margin for error, and loads of danger. If the director calls in the stunt people and it’s not a fight scene, it goes here. 

Brains: Intelligence, intuition, resourcefulness, and perception. This sounds more like a traditional attribute, but that’s mainly because any variation on “Thinking” either sounds dumb or is too narrow for my tastes. “Brains” tells you the character’s smart without telling you how the character’s smart. The brains of the operation can be a well-educated super-genius or an intuitive dropout with a good memory. 

Persuasion: Again, this one sounds like a more typical attribute because it’s nice and broad. You can persuade someone to do something by being charming and charismatic, by offering a well-thought-out rational argument, by appealing to virtues or vices, or by threatening to break some kneecaps. The word “Persuasion” doesn’t judge. 

Know-How: Know-How is sort of applied Brains and covers actually doing things, whether it’s picking a lock, bandaging a wound, or rewiring a robot. A character with a high Brains can explain how an alternator works, but probably gets skinned knuckles if he thinks too hard about a wrench. A character with high Know-How may not know exactly how an alternator keeps your battery charged, but he can get your car running again. 

Know-How is the combination last week’s “Tradecraft” and “Wits,” which makes a nice transition into how I landed on these five stats as the Core Traits. Combat and Thrilling Heroics were set in stone from the beginning. I also knew I needed core skills for social skills, knowledge, and non-stunt actions, but getting there took some trial and error. As recently as last week, I still had “Wits” broken out as separate still. The idea was that it would represent perception and common sense/street smarts, with Brains representing more scholarly intelligence and knowledge. I cut it out because it broke the “minimal overlap” rule as well as the thing about the stat not making decisions about characterization. Also, once Tradecraft (which I’ll get to toward the end) became Know-How, it covered most of Wits with less gray area. 

The social core skill was the easiest, since it turned out to basically be a matter of coming up with a good name. Once you cut out the Willpower/Ego/Mental Fortitude aspect of it (which probably should be some other thing along the lines of Stamina, if it’s needed at all), nearly all social rolls are attempts to convince somebody of something, whether it’s to give up the combination to the safe or to believe that your character is a cool guy. As noted above, Persuasion works across the board. 

Other possible skills were mostly trying to reconcile different kinds of intelligence and skill. At one point (before I realized the need for the “don’t let the skill define the character” rule) those were combined into thematic groups that could cover knowing stuff or doing stuff. The skills were things like “Technology,” “Streetwise,” “Nature,” and “Subterfuge.” There was also going to be a “Lore” skill that basically covered “things PCs know.” So in a spy game, “Lore” would cover planting bugs, setting up dead drops, and paying attention to ensure you fully understand the mission before the message self-destructs; Hobomancer Lore would cover finding work, basic railroad knowledge, and information about songlines, so-and-sos, and what-nots; Dungeoneering Lore would cover identifying slimes and oozes, listening at doors, skulking around, and disarming traps. 

Actually, the last two (and maybe last three) items in the Dungeoneering list were iffy at first because there was also the whole “Subterfuge” (ie., Thief Skills) option on the table. Then I found a thread on some site (I’ve been reading a lot of threads and posts about game design and theory lately) that was titled something like “The Thief is the worst thing to happen to D&D.” The basic premise was that in really Old School (pre-Thief) D&D, it was assumed that all characters could skulk around, find traps, climb things, etc. When the Thief came along with those skills enumerated, most players took it to mean that only Thieves could do those things, leading to the eventual ruin of gaming by letting players use herbs or something. There were even some decent arguments that all the thief skills were intended to represent extraordinary talent in things everyone could do (anyone can hide behind a barrel, only the thief can hide in a shady corner; move silently didn’t mean “stealth” it meant “move in absolute silence,” etc.). Anyway, that reminded me of something I keep saying over and over but keep forgetting when I fall into a rules tinkering fugue: You don’t need game traits for things everyone can do. So Lore kind of morphed into the “Adventuring” Skill. 

The main problem with Lore is that most people read it as more about knowing things than doing things, which is an overlap with Brain that I didn’t like. I thought about splitting it into “knowing stuff about [whatever the PCs do]” and “doing stuff [that PCs do]” which I decided to call Tradecraft. This is an expansion of the actual definition (“stuff spies do,” more or less), but it seemed to fit. Since that caused Lore to overlap with two things, I ultimately decided to roll Lore up into Tradecraft, which is where we were last week.

Tradecraft covered a lot of “doing stuff” skills, but not quite all of them. Specifically, they wouldn’t cover something like skills from your day job in a game like M-Force. Initially I rationalized that most of the important stuff from “background” are either covered by one of the other Core Skills (combat training, etc) or shouldn’t require a roll (you can automatically fill out your TPS report). On the off-chance you did need a roll and no Core Skills applied, you could always just default to Hero Die (which, as noted last week, is still often a better die than highly-trained professional non-PCs would roll). But that’s not really true, because there are a lot of those “doing things” abilities that don’t fit another Core Skill but you’d still want the player to roll for: B.A. Baracus-style tune-ups, for instance. I could just roll job skills into Tradecraft, but that means the definition of “Tradecraft” varies from character to character depending on background. That’s when I decided to turn it into the more general “Know-How” and make it a generic “doing things in the real world” kind of skill, which means there are very few rolls that aren’t going to fall within a Core Skill. That made Wits kind overlap between Know-How and Brain, so Wits got cut. 

So that’s how we got here, and like all the darlings I’ve killed recently, I really like it right now. One big advantage that Core Skills have over Attributes that I haven’t already mentioned is that they give the player more control over the level of detail, especially if Core Skills and Edges or Trademarks or whatever the non-core skills end up being named are bought from the same pool. If you want a broad character, you can put most or all of your points into Core Skills. If you want more nuance, you can put more points into narrower skills that describe the character in better detail. For some kinds of games (pulp comes to mind), you might even be okay sticking with just Core Skills.  

Next week I’ll either revise this again or talk about some of the other new ideas I’m tossing around. Right now I’m looking at March for Brainstorming, April for planning and working out the big picture, and May for getting a workable playtest draft together. Then in June I’ll take it to DieCon and see how it breaks. 

Before you start reading the blog, you need to go over to Kickstarter and back Josh's Draugr & Draculas zine. 

If there’s one constant in my goal for game design, it’s that I don’t like “grocery list” style skill lists. It was a design goal in QAGS. It was a design goal in Cinemechanix. So imagine my surprise when I realized that Cinemechanix has turned into a “grocery list” style skill system. With no modifier for Role (except maybe a small concept bonus), players are going to trend toward trying to make sure they’ve got skills for all the stuff they want their character to be able to do. I know this because I’ve been doing it myself in thinking through what character creation might look like in the current system. The number of skills a character “needs” is way higher than I’m comfortable with. 

You might be thinking that I’m completely missing the point that between the Hero Die and Attribute scores, most skills are redundant. If an average person (Hero Factor 2) puts all his character points into one skill, he’s going to top out at something like d12+d6+2. A Hero Factor 4 character rolls a minimum of d12+d4 even with a zero attribute and no skills, so even a low-level PC is nearly as competent at literally anything he tries as a bit player with years of training and experience. The problem is that Hero Factor (especially) and attributes (to a lesser extent) are abstract. Most players aren’t going to look at Hero Factor 6 and think “my character is good at [skill of choice]” and even the relevant attribute only makes a weak connection to all the specific rolls that the attribute applies to. If that Hero Factor 4 character has a Strength of 3, the typical analysis will be “my character is big and strong, which helps him in combat,” not “my character is a combat monster even without any training.” It’s more about psychology than system, but it makes you think you need skills for the things your character is good at doing. 

There are also other problems with the “raw talent” kind of attributes (Strength, Agility, etc.). The first is also due abstraction, or at least vagueness. While Strength or Agility are less abstract than “Hero Factor,” there’s still not always a clear connection between a given attribute and a thing the character is doing in the game. You roll Strength to lift a heavy rock, but does it necessarily make sense to use Strength for all melee attacks? D&D’s “finesse weapons” rules suggests “no.” And that’s without even acknowledging that intelligence or instincts or other traits defined by stats might also play a role in combat ability. Depending on which side of the slope you decide to slide down, you either end up with sloppy rationalization, system-bloating derived stats, or letting the players justify using their best stat instead of the expected one for dozens of rolls because they make a good point. 

Another problem is that since attributes are useful, players are more likely to build the character concept around the stats than the stats around the character. If we see a movie where two action heroes played by guys named Chris fight, we’re going to assume they have about the same stats. You need to throw a The Rock (or at least a Hemsworth) into the mix before we mentally tag a character as a “Chris +1” because he’s clearly bigger and stronger than the other Chrises. But if every +1 to Chris raises your combat die, players are going to end up making unnecessary distinctions around average (if Pine’s a Chris 0, Pratt is at least Chris +1), which requires bigger bonuses when there’s a real difference (if Hemsworth’s a Chris +2, The Rock is at least Chris +5). If you can be a good at fighting with a Chris score of 0, and your Chris bonus only applies to roles that are specifically Chrisish, players are only going to pump up those Chris numbers if the character is strongly defined by his Chrisness.

I think most of these problems can be solved by (mostly) swapping the “raw ability” kind of attributes for the more “super-skills” kind of scheme I briefly considered. Instead of Attributes, there’s a small group of core skills (I haven’t come up with a good name for them yet) that boost your Hero Die and cover the majority of rolls players will make in a typical game. So, for example, a character who wants to be good at fighting would buy up his “Combat” core skill. Whether he’s good at fighting because he’s strong, quick, or just a badass is more about characterization and allocation of skill points. If being strong is a defining character trait, you can give him a Strength skill. If not, you spend your skill points somewhere else and he’s got typical Chris-level Strength. 

Giving the character a skill also works for characters who are good at a particular weapon or fighting style, which is also a distinction that has more to do with gamer brain than how we consume fiction. Most characters who are good at fighting are good at fighting. Perceived “specialization” is more a matter of storytelling. Most gamers would give Daryl Dixon a higher skill in “Crossbow” than “Guns” or “Melee” because the crossbow is a key character prop that he uses a lot. If you actually went through and figured out what percentage of the time Daryl hits or kills a walker with different weapons, you’d probably get similar hit-to-miss ratios for the crossbow, guns, and whatever other weapon categories you counted. The reason we think he needs a Crossbow skill is because his 90% accuracy rate with a crossbow comes from hitting 900 out of 1,000 shots, while his 88% accuracy rate with a rifle represents 440 hits out of 500 shots. The real reason to give Daryl a Crossbow skill is because the crossbow is a signature prop for the character. 

I think reworking the attributes as core skills will solve these problems while still providing the benefits of attributes. Here’s the checklist:

• If the core stats cover most of the common rolls needed in a game, the grocery list gets a lot shorter without reintroducing a Role bonus and all the ambiguity about default skills that comes with it. The core skills cover the abilities the character needs, the standard skills are for fine tuning and characterization. 
• The direct correlation between the stat (“Combat”) and specific rolls (“shoot that guy”) brings at least the attribute aspect of the character’s “hidden” competence out in the open. The player doesn’t feel like he needs a guns skill and a swords skill and a boxing skill because the Combat stat has them covered. 
• The direct link between the stat and rolls also cuts down on ambiguity about which stat is more appropriate to a particular roll. 
• Since core abilities aren’t defined by attributes, you don’t need to be The Rock to be good at melee or Jackie Chan to doge. You can just be a regular Chris who’s good at combat. You do need to use skills (or drawbacks) to define “attributes” that vary from the average, but since skill costs are now weighted by how broad or narrow the skill is, they fit in a lot better. 
• While some players will probably still put more points into skills that seem more useful (like combat), I think this scheme is more likely to lead to building around the character concept than building around optimizing the stats. If nothing else, covering the core adventuring rolls with a separate kind of trait frees up the standard skill slots for more interesting traits. 
• Best of all (from a “the guy who has to write the rules” standpoint”), we’ve still got standard roll types that can be referred to by name in the rules. It’s just “make a Combat roll” instead of “make a Strength roll” now. 

My current list of attributes-that-aren’t-quite-attributes has 6 stats: Combat, Thrilling Heroics, Tradecraft, Persuasion, Brains, and Wits. I’ll get into what they mean and how I came up with the list next time.