ESCARGO: Exponential Skill Costs: A Radical GURPS Option (GURPS 4e/3e)

ESCARGO is an experimental look at an unusual option for GURPS: skill and attribute costs that just go up, up, up. It looks odd, but offers a surprising number of nifty benefits. Put on your game engine-hacking cap and read on…


v1.0: Creation date forgotten.

v1.1 update (2003/01/24): With help from reader feedback, have clarified conditions that do and don’t work for setting skill cross-defaults. Added PDF skill cost table. Made other small improvements. Some reader feedback comments are noted in dark red text. Special thanks to J. Schipper, P. McCurry, D. Weber, and D. Cole.

v1.2: Moved to main Games Diner site.

v1.3 (2008/11/26): Cleaned up.

v1.4 (2009/08/16): Minor revision for 4e compatibility. 


There’s a handy skill cost table (PDF file) for your reference. See link at bottom of page.

Doing strange things to skills 

ESCARGO? Like the gourmet garden pest, but spelled wrong? Yep. Picture a snail’s shell, expanding its diameter exponentially as it spirals outward, and you have a . . . er . . . okay, you have a barely relevant metaphor for the topic at hand, and a really lame epithet to boot. But forget about the mollusk; everyone knows it’s the garlic and butter that are the tasty parts anyway.

Huh? Well, as long as Your Author is already out of his mind, he’ll go fully bipolar and play two roles in this article: the loopy progenitor of yet another unasked-for rules widget (dull main text), and the skeptical reader wisely poised for escape (snappy questions in headers).

Come on, this’ll be as exciting as a real escargot race! Here goes!

(Warning! This is a working paper, and it’s real geek stuff to boot. It’s not a finished system. Join in if interested!)

Redefining skills – why?

You’re messing with GURPS skill costs? What the heck for?

Well, the existing system isn’t so simple. We have Easy, Medium, Hard, and Very Hard modifications to skill level (essentially +1, +0, -1, and -2 modifiers) – and in 3e, three different cost progressions (for Mental, Mental/VH, and Physical skills) as well. We have the addition of many required specialties and (using different rules in 3e and 4e) optional specialties, plus a separate system for the specialties called techniques (maneuvers in 3e), which are further broken down into Average and Hard. And on top of that we have some instances of multiple skills clumped into single units, including those optional Wildcard skills.

Below is an idea that replaces all the above with only two components: a single skill cost progression, and a level adjustment to differentiate degree of difficulty. Yet thanks to the secret ingredient in the cost progression – exponential increase – that simple little difficulty indicator models skills, subskills, specialties, techniques, and even “superskill” groupings of individual skills.

How wonderfully vague! Do go on while we file our nails.

Okay. Here are the main features:

1. The cost per level of skill increases with each level, as it does with the first few levels in GURPS, but continues to increase in that exponential manner. Forever. The cost does not plateau at 4 points per level (or 2, 4, or 8 points in 3e, depending on skill type).

2. All skills use the same cost progression. In 3e terms, the cost distinction among Physical, Mental, and Mental/VH skills is lost (in terms of maximum cost per additional level).

3. Skill difficulty is differentiated by simple adjustment of final level: the same 2 points that buy Skill A at base level buy the easy Skill B at base +1, the hard Skill C at base -1, the harder Skill D at base -2, etc. This mirrors the GURPS distinction among Easy, Average, Hard, and Very Hard, but with any level of difficulty possible (not only GURPS‘ four levels).

4. Defaults are mostly unchanged: a skill can default to an attribute or another skill, at the appropriate penalty.

Lovely. Now quit being abstract and show me the goods.

You bet. Here’s a sample system – not the One Way to set things up, but one possibility. It’s used throughout the rest of this article.

1. For skill cost progression, I’ll borrow the progression from the Size and Speed/Range Table: +6 levels = x10 cost, properly extended (as in 3e) to both above and below 0. (I’ll clean up a few oddities as well on the low end, for a smooth progression.) Here, the point cost for each level is roughly 1.5 times the cost of the previous level.

2. I’ll define a skill of generic “average” difficulty as costing 2 points for base level, i.e., DX, IQ, or HT level.

3. Here’s our cost progression for such a generic skill:

Level  Cost
+10 100
+9 70
+8 50
+7 30
+6 20
+5 15
+4 10
+3 7
+2 5
+1 3
0 2
-1 1.5
-2 1
-3 0.7
-4 0.5
-5 0.3
-6 0.2
-7 0.15
-8 0.1

And so on, extending in either direction (though you likely won’t need to!).

4. The names Easy, Average, Hard, and Very Hard aren’t sufficient as labels for skill difficulty, as any level of difficulty is possible. Instead, let’s use the name “Ease” plus the level adjustment (eschewing ugly names like “Difficulty Factor”).

“Average” difficulty is “Ease 0”. An Ease 0 skill costs 2 points for IQ level, 1 point for IQ-2, 7 points for IQ+3, etc.

An “Ease -4” skill would be a very difficult one. Purchase as above, then adjust level down by 4: spending 2 points only gets you DX-4, not DX.

An “Ease +2” skill would be an easy one: purchase using the costs on the table, then adjust skill upward by two levels. Skill at DX+2 costs 2 points, DX costs 1 point, etc.

Comment: D. Weber suggests: Why not drop the logarithmic “Ease” business, and simply use a cost multiplier? That is, an Ease 0 skill becomes “x1 difficulty”; Ease -1 = x1.5 difficulty, Ease -6 = x10, Ease +2 = x1/2, etc. When buying skills, the cost is the cost shown on the table, times this difficulty multiplier. When combining skills, multipliers add directly: a x2 difficulty skill and a x5 difficulty skill combine to a x7 difficulty skill. Breaking down skills involves equally simple subtraction. (This is similar to the use of Cost Factor (CF) in some GURPS lines.)

It looks sound, and could be an improvement! I’m sticking with the current ESCARGO write-up until further reader input. 

What about existing GURPS difficulty levels?

Easy enough. Call Easy skills Ease +1, Average skills Ease 0, Hard skills Ease -1, and Very Hard skills Ease -2.

With any level of skill difficulty now possible, make whatever changes you like. If you’ve always wanted to make some skill easier than Easy (Ease +2 or better) or harder than Very Hard (Ease -3 or worse), go for it.

It’s also possible to re-engineer existing skills to arrive at uniquely-determined Ease levels. More on that much later, in the discussion of combat skills.

For 3e games, remember that ESCARGO makes no distinction between costs for Mental and Physical skills. If you want to do so, though, it’s simple enough: add a point or two of Ease for Mental skills, or subtract a point or two for Physical skills. Given ESCARGO’s cost for high skill levels, I’d lean toward leaving Physical skills as is, and handing Mental skills an extra couple points of Ease. I’ll leave any such adjustment up to you, though.

Comment: J. Schipper notes that the above Ease adjustment makes Mental skills in 3e cost half as much even at low levels – e.g., only 1/2 cp for an M/E skill at IQ level, not a GURPS-like 1 cp. Yes, that’d be an unavoidable side effect of the adjustment.

For 3e, ESCARGO can handle a system where the relative difficulty of Mental and Physical skills is fixed at all levels – i.e., they’re equally difficult at all levels, or one is more difficult than the other by a fixed degree at all levels (twice as hard, etc.).

ESCARGO can’t handle a system where this relative difficulty changes in mid-course – i.e., the GURPS 3e-like way that Mental and Physical skills start out with no difference in difficulty, but then Physical goes on to become four times as hard. ESCARGO is no good for that. (Which is fine with me; I don’t want that odd Mental vs Physical pricing distinction! Neither does 4e, so we’re all cool with that, right?)

Great. I kind of see the what, but I don’t see the why. What does it all mean?

The important question! Here are the implications of the system:

1. Any level of skill difficulty – any degree of breadth or narrowness – can be expressed by a single modifier, Ease.

2. Any skill can be compared to another and labeled “x times easier/harder”. For example, any given level in an Ease +2 skill will cost half as much as that same level in an Ease 0 skill. The Ease 0 skill clearly represents twice as much “stuff” to learn as the Ease +2 skill. It’s twice as hard. The same holds true for any two skills separated by two levels of Ease.

Likewise, a four-level difference in Ease represents one skill five times as hard as the other, requiring five times the points or time to learn at a given level. (Check the cost progression to confirm this for yourself.)

This difference – where a given level of one skill is x times as costly as another – occurs only on a small scale in GURPS. There the cost of Brawling is half that of Boxing, yes, but only for the first few levels; after that, there is an absolute cost difference of 4 points (8 points in 3e), not an unchanging relative difference.

3. You can divide skills into “subskills” with a simple Ease adjustment. For example, take any skill, and divide it equally into three component fields of learning. Whether you call this a single skill of Ease X or three simpler skills of Ease X+3, the cost is the same.

4. Likewise, you can combine skills into “superskills” with an Ease adjustment. For example, take seven skills of Ease X, and combine them into one “superskill” of Ease X-5. The cost is the same.

5. The big point: Despite the above usages, there are no real “subskills” or “superskills” under ESCARGO. There’s no real distinction among skills, superskills (Wildcard skills), subskills, techniques/maneuvers, or specialties. There are only skills, plain and simple, with floating levels of difficulty that roughly represent amount of knowledge/practice required.

Give me the basics on defaults.

They’d work as they do now. You could set a standard default to -5, plus Ease: an Ease -1 skill would default to Attribute -6. Change this “default default” for individual skills as appropriate.

Some points to note:

1. A default between two skills represents overlap between the skills. It indicates that two skills are, to some partial degree, the same skill. Hours invested in one skill equate directly to a fixed percentage of study of a defaulting skill.

If Broadsword and Shortsword cross-default at -2, they are partially – half, under ESCARGO – the same skill. Study of one is always equivalent to half as much study of the other. In terms of point costs, that relationship doesn’t hold for GURPS.

In this way, ESCARGO not only removes the lines between skills, specialties, techniques, and so on, it even blurs the lines between separate skills themselves.

2. The default is determined by that degree of overlap. If about one-fifth the content of two skills of equal Ease is content common to both skills, they should cross-default at -4.

3. As a default indicates overlap of content, defaults under ESCARGO should be cross-defaults, not just one-way defaults. No exceptions.

Sounds good, but I see trouble for skills of differing Ease. If one skill is Ease -2 and the other is Ease +3, and they cross-default at -6, you’d be nuts to not buy the cheaper skill only, and get the hard one almost for free.

True. A simple and logical adjustment fixes this.

The default between two skills is the base default based on degree of overlap, as above. That gives you the initial, “base” cross-default. But now, do this:

Increase the cross-default by the difference in Ease levels (i.e., make the default harder) when defaulting from the easier to the more difficult.

Decrease the cross-default by the difference in Ease levels (i.e., make the default easier) when defaulting from the more difficult to the easier.

A close look at what exactly is going on will make clear why you’d do this.

Example: Skill A is Ease -1. Skill B is Ease +1. You decide that about one-tenth of each skill is not unique, and could be applied to the other skill.

If the skills were of equal Ease, we would assume one-tenth overlapping material going either way, and a cross-default at -6. But from the differences in Ease, Skill B only covers half as much “stuff” as Skill A to begin with. One-tenth of the content of Skill A should have an appropriately bigger “bang” when injected into the narrow Skill B, making up one-fifth of Skill B’s skimpy content. Likewise, a tenth of Skill B’s content dropped into Skill A will only represent a twentieth of that wider skill’s broad knowledge.

You get that result by adjusting the cross-defaults for two levels of difference in Ease: Skill B defaults to Skill A at -4, and Skill A defaults to Skill B at -8. Check the table, and juggle some sample point costs and hours of study in your head; you’ll see that it all makes sense.

What would it mean if a third of Skill A applied to Skill B, but Skill B were only a third as broad?

That would mean Skill B is entirely a subset of Skill A, the exact equivalent of a third of Skill A.

In game terms, Skill B defaults to Skill A at -3, but is also three Ease levels higher. From the above default rules, Skill B will have a net default to Skill A at no minus.

Going the other direction, Skill A would default to Skill B at -6.

And if a third of Skill A applied to Skill B, but Skill B were only a tenth as broad?

Let’s see. That means you’re devoting a full third of your study on the very narrow content of Skill B, and two-thirds of your study to other content. We’d expect very rapid progress in Skill B, and the math will show this: adjusting for the difference in Ease, Skill A defaults to Skill B at -9, while Skill B defaults to Skill A at +3.

Huh? Can a skill default to another at a plus? Sure. This jumped-up level in Skill B simply reflects what happens when the number of points in question are directed toward the narrow Skill B. Try the math and see.

(The initial release of ESCARGO suggested that something was funny here, but I take that back. Upon re-examination, this situation is perfectly valid, positive defaults and all.)

So pretty much anything goes.

Not quite. Here’s a snail trap for the unwary:

Say you decide that Occultism is ten times broader (six Ease levels lower) than Ghost Stories, a simple Lore or hobby skill for campfire fun. And you decide that Occultism defaults to Ghost Stories at, oh, -4; that sounds about right. Everything a-OK?

No. If the final default to Occultism from Ghost Stories is -4, then the initial “base” cross-default must have been +2, before adjustment for Ease differences. And that’s just not kosher: you can’t have a base cross-default that’s positive, ESCARGO or no ESCARGO. That would indicate two skills whose content overlaps by over 100%, and that’s meaningless!

That shows up a key side of ESCARGO: If you want it to work its mathemagic, you have to feed it proper numbers. With straight GURPS, you can make up skill difficulties and cross-defaults willy-nilly, creating no more pitfalls than already exist. But ESCARGO requires a three-step method to set proper defaults:

a) Decide the degree of overlap between two skills;
b) Set Ease of each skill;
c) Adjust a) by b) to get final cross-defaults.

So overlap, Ease differences, and defaults all have to work together.

Yes – and it’s because they work together that things are all nifty. Here’s a summary of what degree of overlap and differences in Ease mean:

If the base (non-adjusted) cross-default between Skill A and Skill B is larger than the difference in Ease, then the skills overlap partially. Each contains content common to both, as well as content not contained in the other. Cross-defaults will both be negative.

If the base cross-default between Skill A and Skill B is smaller than the difference in Ease, again the skills overlap partially. The broad skill’s default from the narrow skill will be negative; the narrow skill’s default from the broad skill will be positive.

If the base cross-default between Skill A and Skill B is the same size as the difference in Ease, then the narrow skill is entirely contained in the broad skill. The narrow skill defaults to the broader at no penalty; the broad skill defaults to the narrower at twice the difference in Ease.

Actually, it shouldn’t be hard to check whether you’ve got things set up right or not. Look at the cross-defaults you’ve set for two skills, and ask two questions:

  1. Is the difference between the two cross-defaults equal to twice the difference in Ease?
  2. Is the mid-point between the two cross-defaults (which is the “base” cross-default) a negative number?

If “yes” to both, then all is well. (Hmm, that’s easier than I expected!)

Okay. What else should I know about defaults?

1. In GURPS, if a high-level skill and a defaulting skill have both reached the max cost per additional level, you have zero incentive to put points into the latter. You can continue to study the higher skill only, and will continue to progress in both skills equally for no extra effort, pulling the default skill along for free. This effect is especially pronounced with Mental skills in 3e, where even defaults quickly reach the low maximum cost (2 points per level).

But under ESCARGO, there is always some incentive to study a defaulting skill instead of skill it defaults from. If you’re planning to desert the Cardinal’s Broadsword Guard and join the King’s Shortsword Legion, you’ll progress in shortsword ability twice as fast by choosing to switch your concentration to Shortsword, rather than by continuing your Broadsword study and defaulting from that.

2. Skills defaulting to other defaults? No problem! Unlike GURPS, the ever-increasing cost of the “lead” skill means you can’t take advantage of a point crock, pulling along a huge train of multiple defaults for a mere 4 points per level (2, 4, or 8 in 3e). Under ESCARGO, x% of study of Skill A counts as study of Skill B, and y% of study of Skill B counts as study of Skill C… so it only makes sense that an appropriately tiny percentage of study of Skill A should apply to Skill C. Go ahead and allow the multiple default; both the effects and the point costs will make sense.

3. You can be more liberal with defaults in ESCARGO. Does 4e‘s Broadsword and Two-Handed Sword cross-default of -4 (no default in 3e!) seem too harsh? Reduce it to -2; remember, the cost of the “lead” skill will keep rising and rising. Hmm, how about allowing a Broadsword and Axe/Mace cross-default, at -4 or so? Sounds crazy, but hey, they’re just balanced and unbalanced versions of sticks. Done.

Heck, you could let any melee weapon skills, no matter how disparate, cross-default at a maximum hefty penalty of -8 or so, recognizing that combat expertise with one weapon will afford some benefit with any other. (See the Appendix for more on this.)

The point is it’ll work in ESCARGO – if you can pay the hefty points to get your main skill up to a high level, you deserve to have all those defaulting skills tagging along for “free”! Have no fear of point crocks. Embrace the default. 

4. Chains of spell defaults raise whole new possibilities for gaming magic. Too much to even start playing with here. But I think it could really overhaul how spells are bought in GURPS.

Give me examples, dagnabbit!

Enough theory; let’s see some stuff in action. Show me a purchase of a skill from default, to make sure I have that right.

It’s no different from buying up a skill in GURPS. Pay for the difference in cost between the current level and the desired level.

But just as when you buy a skill in the first place, make the purchase straight from the table, and then add Ease to skill levels when you’re done. Remember, Ease is just a final adjustment to a skill’s level.

Example: Skill A is Ease -1. You have it at IQ+3 level. Skill B is Ease +1 and defaults to Skill A at -4, so you have it by default at IQ-1.

Drop Ease from both skill levels, setting Skill A to IQ+4 and Skill B to IQ-2.

You’ll see you can boost Skill B by two levels for only 1 point, or three levels for 2 points. You choose the latter, bringing it up to IQ+1.

Now add Ease back to both skills. You have Skill A at IQ+3 and Skill B at IQ+2.

How would I buy an optional skill specialization?

Easy enough. The main point: You’re not stuck with “specialization rules”, limited to GURPS‘ specific mechanics. A specialization in ESCARGO is just another skill, and can be bought at any level.

For a Chemistry specialty that touches upon about a third (GM call) of the broader Ease -1 skill, use an Ease +2 skill (i.e., three Ease levels of difference). Buy that up as little or as much as you like.

Let’s check out an example in detail. Say the GM calls History an Ease 0 skill, and decides that Ancient History is as challenging as about a fifth the study of History overall. Only a fifth of the study of History goes into Ancient History, so the latter defaults to History at -4. On the other hand, at only one-fifth History’s breadth, Ancient History is an Ease +4 skill. (Net effect: it defaults to History.)

You buy History at IQ+2 (5 points) and so have Ancient History at IQ+2 (default), but want to boost the latter by two levels. Ignoring its Ease, you’re buying Ancient History up from IQ-2. You’ll see that you can buy two levels for only 1 point, gaining the skill at IQ. Add its Ease back in, and you’ve got Ancient History at IQ+4.

That’s that. To add an extra twist, it should be obvious that if History were an Ease -1 skill instead of Ease 0, everything would work as above, but with all final skills a level lower. You’d end up with History at IQ+1, Ancient History at IQ+3.

Incidentally, can History ever default to Ancient History? Sure. They cross-default at -4, and Ancient History is 4 levels easier. Per earlier notes on defaults, History should default to Ancient History at -8.

Detailed examples require a lot of words on paper, but in short, buying up a specialty is no different from the usual GURPS procedure for buying up a skill from the appropriate default.

How about a required skill specialty?

In GURPS these are just separate skills with a cross-default. Treat as such.

Give me an example of buying up a technique.

Once again, buying up a technique (maneuver in 3e) is the same process as buying up a default. Example: For a single technique of an Ease +1 combat skill, representing about a tenth of the skill’s techniques, use an extra six levels of Ease, or an Ease +7 skill.

But note this difference from GURPS:

In GURPS, if you have Shortsword defaulting to Broadsword-2, you can buy up your Shortsword so it trails Broadsword by only one level – but when you next raise Broadsword a level, Shortsword will stay right where it is, not tagging along. In other words, you’ve improved Shortsword’s level to Broadsword -1, but its default is unchanged at Broadsword -2. That’s how skills work.

Conversely, if you buy up the GURPS technique Kicking so it trails Karate by only a level, Kicking will tag along with later increases in Karate, always staying one level behind. You’ve improved Kicking’s default itself to Karate -1. That’s how techniques work.

Since ESCARGO makes no such distinction between skills and techniques, the latter work like any other skill. You can pay a bit to improve your Kicking from its Karate default, but later improvements in Karate won’t automatically bring Kicking up along with it (until the gap is large enough that the default matters again). That’s a difference from GURPS.

That kind of stinks. If I go through the trouble of buying my Kicking technique up to my level of Karate, I don’t want to have to re-boost it with fresh points every time I improve Karate. I want my improved Kicking to automatically stay level with my Karate, like in GURPS.

That can be done too, Grasshopper: you can choose to improve not just a defaulting skill’s level, but its default as well. Read on.

Stupid skill tricks

You earlier mentioned combining skills. Examples, please!

You bet. Try these on for size:

For a combat skill that combines the two Ease 0 skills Axe/Mace and Two-Handed Axe/Mace, create an Ease -2 Axe/Mace! skill.

For all-round swordfighting mastery that incorporates about five Ease 0 sword skills, create an Ease -4 Sword! skill.

Comment: D. Cole notes a benefit of all this skill-combining: In skill-heavy settings, ESCARGO skill groupings allow easier design of high-powered PCs, with much less character sheet clutter. In a game of Black Ops, for example, quick-and-easy Guns!, Unarmed!, Melee! and similar skills may be much more in-genre than the painfully exhaustive skill lists GURPS currently requires.

Okay, it’s possible to combine multiple skills of equal Ease into one skill of lesser Ease. Now try a harder task: can you combine skills of differing Ease into a single skill?

Sure. Anyone who’s messed with a “logarithmic” scale like ESCARGO’s knows that you can’t just add the numbers here — that is, an Ease +4 and Ease +1 skill don’t combine to an Ease +5 skill. Rather, you need to jump from the logarithmic to the linear (i.e., from skill levels to point costs), add those linear values, and jump back to the log. Easier done than said, fortunately:

Let’s look at the cost to buy skills at a given level – base level, for simplicity. What’s the cost of an Ease +4 skill at base level? 0.5 points. What’s the cost of the Ease +1 skill? 1.5 points. These combine to a neat 2 points, the cost of an Ease 0 skill.

There’s your answer: an Ease +4 and Ease +1 skill combine to one Ease 0 skill. 2 points will buy you both skills at DX level. 15 points would buy you both skills at DX+5. And so on.

Let’s get weird. Take ten skills: Skill A (Ease -2), Skills B and C (Ease +1), and Skills D through J (Ease +2). You want to combine these into one skill, but keeping the same relative levels: Skill A remains three levels below B-C, which remains a level below D-J. No go?

Well, there may or may not be a reason for doing this, instead of purchasing each separately, but to answer your question: it’s a go, and it’s simple.

Like the previous example, you need to ask what’s the cost to buy each skill at the desired level. But here, your desired levels are what you’d get if you put an equal number of points into each skill. So do that. To get a comparison with base level, put 2 points into each: total 20 points, the cost of an Ease -6 skill at base level.

So, for the cost of an Ease -6 skill, you can have Skill A at the purchased level -2, Skills B-C at the purchased level +1, and Skills D-J at the purchased level +2. For example, spending 10 points would get you Skill A at IQ-4, Skills B-C at IQ-1, and Skills D-J at IQ.

Now, for the sake of contrast: What if you did want all those skills at the same level? Combine the total costs to purchase each at the same level: 5 points to buy Skill A at base level, 1.5 points each for Skills B-C, and 1 each for skills D-J. Total: 15 points, or the cost of an Ease -5 skill at base level. For example, spending 10 points would get you each of the skills at IQ-1.

All you’re doing in “combining” skills, really, is totaling up the costs to buy the skills individually at the desired levels.

Sure. And that’s not different from computing the combined cost of a bunch of skills in GURPS.

What’s different is the shorthand that ESCARGO allows. If the cost to buy some big mess of skills at base level is the same as the cost to buy a single Ease -5 skill at base level, then that will remain true for any desired level. You can shout, “The combined cost of these skills is that of an Ease -5 skill!”, and buy that group to whatever level you like by purchasing only one skill. Because they really do combine to that skill.

In GURPS, you’d always have to look at and combine the costs for each separate skill; in that sense, they don’t really combine.

Mm-hmm. Let’s go back to defaults. How about the combination of Skill A and Skill B, which defaults to Skill A at -3?

Well, there’s nothing to combine; Skill B is already a part of Skill A. The cost of A alone includes B, three levels lower.

Ah, right. Make it tougher: I buy Skill A, which is Ease -1. Skill B defaults to this at -3. I want to buy up Skill B so it’s at the same level as Skill A, and slap these together as one skill. Can or can’t do?

Can do. This addresses the earlier question of how you would buy up a default itself, not just a defaulting skill’s level.

What’s the cost of Skill A at base level? 3 points. What’s the cost to bring Skill B up to the same level? You need an extra 2 points (check the table and see). Add these together: 5 points, which happens to be the cost of an Ease -2 skill at base level.

In other words, pay for an Ease -2 skill at whatever level you desire; you’ll get Skill A at that level, as well as the default Skill B bought up to that level. Pay 50 points, and you get both skills at DX+6. Notice that this is the same total cost you’d pay to buy Skill A alone at DX+6, and then buy Skill B up to A’s level. (Try it and see.) 

Use this method to buy a subskill so it works as GURPS techniques do, always staying some x levels above or behind the parent skill. Using ESCARGO, that’s how you’d handle a Kicking technique bought up to always equal Karate skill level. 

Incidentally, an earlier example suggested you could combine Axe/Mace and Two-Handed Axe/Mace into a single, all-round Axe/Mace! skill. That remains true, but if you also gamed these two skills as cross-defaulting to begin with (and ESCARGO encourages such defaults; see Appendix), then the combination would follow this example. The result would be a slightly lower cost for the Axe/Mace! combo, which makes perfect sense if Axe/Mace and Two-Handed Axe/Mace are partly the same skill anyway (as indicated by the existence of the cross-default).

Speaking of costs: your examples always churn out results that perfectly match numbers in the skill cost progression. Coincidence?

No, it’s to keep this discussion simple. Computed costs won’t always be as neat in practice. Depending on what level you’re buying, for example, the cost of buying Skill A and its default B separately in the above exercise may not exactly match the cost for the two “combined”.

That’s fine. The table’s costs don’t follow a perfect logarithmic progression, and round for neatness and simplicity, but the functionality holds true. If you compute a cost that falls between values in the progression – say, a cost for combined skills that works out to 6 points, not 5 or 7 points – treat as the next higher cost. Rounding costs down would create point crocks, letting you tack all sorts of low-level skills on to another skill, for free.

Let’s play Stump the Geek once more. Say that I, all inspired by ESCARGO, decide that Demolition is really an Ease 0 combination of two Ease +2 skills: Chemistry (Explosives), to make the stuff, and Explosives Handling, to use the stuff.

Okay, you could do that…

Obviously, that first one is a “subskill” of Chemistry. So, should Demolitions default to Chemistry? I see trouble…

Well, it’d make sense that the Chemistry (Explosives) portion of Demolition would go up along with skill in Chemistry, but there’s no reason the Explosives Handling portion would. So it wouldn’t be right to have the combined Demolitions skill default to Chemistry.

Maybe this example should serve as a warning that you don’t necessarily want to combine everything and anything just because you can.

In any case, I think that such situations won’t be an issue. The GM need only rule, “Sure, I’ll allow a Demolitions default from Chemistry if you want to mix up an explosive, but you don’t get that default to attach a timer and place the charges skillfully.” That sort of common-sense ruling is needed under any skill system. (What’s nice about ESCARGO, though, is that the GM has the choice of breaking Demolitions into its chemistry and non-chemistry portions, should the players want to improve these separately.)

Combat skills

Okay. Let’s step back and see what we’ve got. It looks like combining skills works for just about anything. We can start with skills of equal or differing Ease, as many or as few as we want, and the resulting combined skill can smooth these out into a single equal level, or can preserve differences in level. We can even combine skills and their bought-up defaults into a single skill with a unique cost. Right?


And how about breaking down skills into component techniques, specialties, whatever? I know ESCARGO can take an existing GURPS skill and break it neatly in half, into thirds, etc., but how about breaking it down into multiple skills of unequal Ease or level, or into combinations of skills and their bought-up defaults, or some other complex mix?

Sure, it works – it’s the reverse of all the above. I can bore you with examples if you like.

Um, maybe later. For now, I’ve been thinking. While we’ve been slogging through this, an idea’s been tickling at the back of my mind… and I now know what it is. With ESCARGO, couldn’t we take a bunch of skills and techniques, and combine them to create a single skill that represents…

A complete martial arts style!

Dang, how’d you guess?

Scary, isn’t it? To answer your question: You betcha. Take Judo, Boxing, a few bought-up techniques, and a handful of associated skills, and slap them together into one Superjitsu skill. It won’t be cheap, but under ESCARGO, purchasing levels in the whole style will be as easy as buying any single skill.

Show me!

Sorry, not yet. There are a lot of design decisions to make here.

First, before you start combining combat skills, you hit the question: what makes up a combat skill itself? Do we want to treat skills as GURPS‘ fixed, ready-made units, or start by designing each skill from scratch, from its component techniques/capabilities?

The latter temptation is big, but there’s a lot of prep work. You’ve got to decide first what techniques make up each skill (a question tougher than it appears), and what the relative cost of each should be. That includes basic abilities like the unadorned, generic hand or weapon strike and parry. Likewise, built-in abilities like Retain Weapon or Feint are easy to overlook, but if they’re part of the skill, they need to counted in its cost. And a skill may cover only a single grip and usage (swing or thrust), or multiple grips and usages (see Appendix for more on this); again, the latter needs to cost more.

More difficult are the abilities that GURPS doesn’t count as techniques. How do you account for the improved defensive abilities of Karate or Fencing? Make that an extra technique, or increase the difficulty of the regular Parry? The same question goes for Fencing’s unusually fast Parry, Karate’s damage bonus, Judo’s ability to replace DX with skill, and so on. Then there are the skills – Main Gauche, Brawling, Shield, and several others – that include off-hand use bought up to full level; that needs to be accounted for. There are a lot of special considerations like these.

Skills can also contain limitations, such as encumbrance penalties, that would presumably decrease the difficulty of some components. But which components, and by how much?

Finally, are the countable techniques/actions all that make up a skill? Shouldn’t there be some additional “core” component added to the mix, representing the basic stances and forms? If so, what should its cost be relative to other components? (This consideration extends to skills outside of physical combat arts. ESCARGO can break off Egyptian History from the rest of History, sure, but could you really learn that alone? Is it a meaningful skill without the core background and methods of analysis included in the full History skill?)

I’ve played with things aplenty; see COSH for a look at breaking combat skills down into component parts. But my forays require more tweaking to fill holes, and to convert to ESCARGO-ready results. One tough spot is getting finished skills to resemble GURPS in relative difficulty. For example, the large number of techniques and special abilities attached to all unarmed combat skills cause Brawling to come out as a more difficult skill than Broadsword using COSH. Maybe that’s as it should be, but to get a more GURPS-like feel, we have to arbitrarily make some weapon-based actions (like a basic strike) much more difficult than their unarmed counterparts.

Once we’ve made all the hard decisions, we can settle down to the more enjoyable task of building combat skills. And even that’s a task best suited for spreadsheet assistance – definitely a “between play sessions” activity! (But that’s the designing stage only; during play, grabbing and using a pre-designed skill would be no harder than it is now.)

Okay, say you did all that. Then you can combine skills into styles, right?

Sure. It’ll just be a bigger version of building a single skill, combining an even wider range of components. Again, there may be some questions related to “what is a style?”, but it should all work.

A style will also have non-Skill components – that is, Perks and other traits that don’t use Skill mechanics. You would use ESCARGO only for combining the skills into a single Style Skill, while purchasing the non-Skill components separately. That’s nothing unusual. 

Hmm. Well, if you’re not going to flesh things out here, how about at least tossing out some of the possible benefits of playing with combat skills like this? I need some reward for enduring all this blather…

Understood. First, on the individual skills front, you can design unique variants. Your variant will have a unique Ease and cost, depending on what’s contained in it.

Design a cheaper Axe/Mace with no Parry component, if you never parry anyway. Build either a one-handed, two-handed, or both-ways version of Katana (and even arrive at a cost that’s balanced compared to other weapons!). Learn Shield with or without offensive striking and Feinting ability. Create an exotic version of Shortsword with increased Parry and Off-Hand usage. Develop a versatile Karate with all the listed techniques, or forget the fancy spins and just focus on your basic crushing punch.

Very narrow skills can be interesting. Blackjack needs nothing more than a single striking motion, and maybe Targeted Attack – no Retain Weapon, Feint, Parry, or other fancy stuff. The cost will be less than a GURPS Easy skill, and appropriately so. Similarly, an Axe/Mace skill incorporating only a striking swing, and nothing else, would form the very cheap and limited “combat” skill of a peasant woodsman.

Moving on to styles, the benefit will again be the ability to combine as many techniques/abilities as you want, but only what you want. One irksome point of GURPS‘ styles is wasteful double-costing. You pay full cost for Boxing in Jeet Kune Do, just to add its few unique abilities; you waste points on abilities already covered by your Karate, such as the damage bonus. Under ESCARGO, you would pick and pay for just those parts that interested you, and not pay twice for that damage bonus. You’ll essentially be building a unique Jeet Kune Do skill (perhaps removing any difference at all between skill and style).

The process should feel much more like developing a truly unique style, with possibilities much more flexible than choosing which of a handful of cut-and-dried skills you’ll toss together. Again, see COSH for a working, if simple, look at what could be possible for single GURPS-like skills (if not combined superskills).

Sounds promising.

Yes, but more development is needed – and who knows, the whole thing could fall on its face during testing. Let’s move along to other stuff for now.

Attribute stuff (or, “It costs what?!”)

Hold the presses. Something’s been bothering me all along: costs. The cost for high-end skill levels is huge – and still rising?

Yes, additional skill levels at the high end become much more expensive than additional levels in standard GURPS. But remember that here, absolute differences in skill level represent multiples of knowledge or learning: Just as your Skill-10 represents twice as much knowledge as Skill-8, your Skill-22 represents twice as much knowledge as Skill-20. Under ESCARGO, consider each level of skill as meaning a lot.

In any case, the cost of additional skill levels surpasses GURPS costs only around the +4 range (or in 3e, the +4 range for mental skills, the +7 range for physical). In many games, PCs don’t often buy levels that high; the effect won’t hamper too many characters. (And those who do spend on those higher levels should enjoy the previously-mentioned benefit of generous defaults.) 

Also note that ESCARGO affords a little help in purchases at the low end, such as the two-level skill jump that occurs between a cost of 1 point and 2 points.

But, yes, very high levels in a skill get expensive. Rather than reliance on a stellar level in a single skill, ESCARGO’s skill cost progression will encourage specializations, techniques, skill-boosting traits (such as Animal Empathy; see many suggested new 3e traits here), and achieving goals through clever use of complementary skills.

It’ll also encourage players to seek situational bonuses. For example, a spy PC will likely buy a more modest level of Stealth than he would in GURPS, and instead work to boost his skill roll through good equipment, careful planning, and distractions (probably invoked through other skills). These are, in fact, the things we would expect a master sneak artist to do.

Creative use of the above factors, plus generous use of skill defaults (including multiple defaults), will help keep expertise affordable under ESCARGO. But in the end, only heavy playtesting will tell us whether the system is too strict an environment for highly-skilled PCs.

But we have the cost of additional skill levels exceeding the cost of additional levels in attributes! That’s crazy!

The solution: Put attributes on an exponential cost as well! The skill cost progression by itself won’t work as a cost progression for attributes, but a nip and a tuck make it possible:

a) Slap the skill cost progression onto attributes, matching things up so an attribute of 10 costs about 10 or 20 points.
b) Subtract the cost of that 10 attribute from all costs, so that a 10 costs zero points.

That’s a start, but not good enough. Try it and look at the costs, especially the puny points awarded for attributes below 10 – it’s very un-GURPS-like. So carry on with these steps:

c) Let the above cost cover only the benefits of attributes’ effect on skills. Now add a cost for the miscellaneous goodies that attributes provide aside from that. This could be a flat +/- x points per level above/below 10.
d) Round the results up, down, or how you like, to the nearest 5 points. There’s no reason why attribute costs have to stick exactly to some formula; round ’em and make ’em pretty.

That’s the recipe: an exponential cost for effect on skills, plus a flat cost for other effects. Now, watch the chef at work, young gourmet:

a) Let’s attach the skill cost progression so that an attribute of 10 matches a cost of 10. (An attribute of 9 pairs up with 7 points; an 11 pairs with 15 points; a 12 pairs with 20 points; and so on.)
b) Subtract 10 points from all costs, so the cost of an attribute of 10 is 0 points.
c) For the non-skill benefits of attributes, let’s try a flat 10 points per level above 10, -10 points per level below. That covers the miscellaneous goodies of IQ, DX, and HT: generic sense or memory rolls, generic DX tests, your base Move and Dodge scores, survival rolls, Hit Points (3e) or Fatigue (4e), etc.
d) We’ll round down to the nearest 5 points where necessary (i.e., for low attributes).

Here’s the result:

Score   Cost
20 590
19 380
18 270
17 210
16 150
15 110
14 80
13 50
12 30
11 15
10 0
9 -15
8 -25
7 -40
6 -50
5 -60
4 -70
3 -80
2 -90
1 -100
0 -110

Fiddle with the chosen parameters, and costs can vary from the above as little or as wildly as you like. Whatever your preferred numbers, the point is this: the cost of skill-base attributes incorporates the same progression as the cost of skills. Buying x levels of such an attribute will always cost more than buying x levels of a skill.

That looks more GURPS-like, I agree. But like skills, the cost for high levels gets very expensive. Comment?

A few:

1. Much more than standard GURPS, this system presumes that each and every additional level of an attribute or skill has significant meaning. The cost for IQ or DX quickly rises to very high levels, but attributes of 15 or more should represent incredible extremes, with 20 a godlike level (and each and every level above 20 another godlike step).

ESCARGO promotes what some call “normalization”, both in stats and skills: just a few levels of IQ, DX, or skill above the norm represent awesome ability.

2. The rapidly-increasing cost of attributes provides real incentive to put points into skills or ability-enhancing advantages. (And to put effort into clever use of situational modifiers!)

3. Under ESCARGO, we can view attributes in terms of multiplying or dividing the cost of learning. If Jim Quick has a DX two points higher than John Slow, any given physical skill level will cost Jim half as many points as John would spend. He’ll achieve any given level of ability in half the time. GURPS‘ attributes don’t currently support that distinction.

Yes, the cost of IQ 16 may look high in ESCARGO – but it doesn’t feel so extreme when you realize that, compared with IQ 10, it means one-tenth the time and points needed to learn any given level of skill!

You know, some 3e GMs will actually like this increased cost for DX and IQ. But the initial levels are low for 4e.

True. When updating ESCARGO for 4e, I didn’t change the suggested progression; as it is, it’s arguably balanced in that DX and IQ become a bit cheaper on the low end, before becoming more expensive at the high end. The key factor still holds: buying x levels of an attribute will always cost more than buying x levels of a skill.

I heartily endorse tweaking, though, such as changing the base 10 points/level for non-skill benefits of attributes to 15 or 20 points/level, for IQ and DX. Or “slide” the skill cost progression component relative to the starting stat of 10. Experiment, and let me know what you come up with that’s better. 

Should ST use this same progression?

No way! The reason’s simple: the above costs incorporate the cost of attributes as bases for skills. But no skills (even defaults) should ever be based on ST. See 3e-centric reasons here, here, and here. (I think 4e abolishes ST-based skills in practice, if not in theory. Let me know if you find an offender!)

That leaves only non-skill-base costs as the cost of ST. Per 4e, a flat +/-10 points is grand. Of course, there are so many ways that ST differs from the other attributes, besides the skill base difference; you may prefer some other unique cost. The ideal cost for ST is a matter unrelated to ESCARGO; I’ll note some worthwhile reading for the interested at Rules Nugget (GURPS): A Better Cost for ST and Design notes: Implementing “log ST in a game and then move on.

Other stuff

What’s with skill costs at the low end? I don’t like all those little decimal numbers…

Those numbers faithfully stick to the cost progression, and that’s a good thing. Fortunately, you won’t need to bother with them.

First, generous skill defaults will often keep your minimum purchases up high, where the numbers are clean.

Second, even when you would purchase skills way down there, you can stick with neat-numbered purchases. Don’t be stingy with the fractions of a point; put a whole integer point into that skill, Big Spender! Or scrimp with half-points, as GURPS 3e does. There’s no need to even look at other fractions.

Those tiny costs are useful, though, if the GM wants to simulate micro-training, with time in study enabling each and every level of improvement above default.

Any special advice for Eidetic Memory?

The 4e advantage requires no special consideration. The following applies to 3e, where a lot of GMs already twiddle with the advantage’s troublesome particulars.

Under ESCARGO, there’s no need to double or quadruple points put into Mental skills; skill bonuses provide that exact effect. So treat the advantage as a skill bonus, but only for skills involving extensive memorization of facts (GM call; does not include spells or psi skills). A roll vs Eidetic Memory-adjusted IQ also allows feats of memory recall.

On balance, the value of these bonuses seems equivalent to the cost of additional points of IQ. Pay the cost of such.

How about skills that already encompass varying breadth, like Area Knowledge?

ESCARGO seems perfectly suited for the sort of trade-off between breadth and depth that takes place in skills like these. But to get the simulation right, it’s worth taking a moment to think it through. There are two ways we could run a skill like Area Knowledge. Picture a country called Kingsland, divided into 9 or 10 equal-sized regions, including one called the Barony:

Method 1: Area Knowledge (Kingsland) and Area Knowledge (Barony) are skills of equal Ease. This is the current GURPS method (obviously, as the system really has no alternative). Say your purchase of AK (Kingsland) includes detailed knowledge of a hundred or so “country-level” personalities. The same level of AK (Barony), costing the same number of points, would have to mean the same amount of knowledge: a hundred or so “barony-level” personalities, at the same level of detail.

The two skills would cross-default at -6 (based on ten-fold difference in “coverage”).

Method 2: Area Knowledge (Barony) is Ease +6 for limited coverage, relative to Area Knowledge (Kingsland). This is more of an ESCARGO-inspired method. To learn the skills at equal level, you’d pay only a tenth as many points for AK (Barony), but this purchase would have to cover a tenth the knowledge: detailed knowledge of only the ten or so “country-level” people who live in the Barony, not the hundred “country-level” people that the same skill level in AK (Kingsland) would afford.

If you pay an equal number of points, you get AK (Barony) at six levels higher than AK (Kingsland). This would bring you back to detailed knowledge of a hundred “barony-level” people, including ten who are also “country-level”. (Obviously, there’s not a clear distinction here between “breadth” of knowledge as determined by Ease, and “depth” of knowledge as expressed by skill level; both of these point to how much stuff you know.) AK (Barony) defaults to AK (Kingsland), as it’s contained within it. AK (Kingsland) would default to AK (Barony) at -12.

Compare this to the similar discussion on Basic Set 176 (4e). I may be missing something, but I think the above approach is more flexible and is clearer on what knowledge of a skill does and doesn’t include. 

Hobby skills?

These are half price in 3e; that’s two extra levels of Ease.

In either 3e or 4e, ESCARGO is great for modeling “trivial skills”. Tea Ceremony and Tournament Law may be legitimate and fascinating skills, but even GURPS‘ designation of Easy seems too high. Narrow skills like these may be excellent candidates for Ease +2 or higher.

I see. And I also grow tired and hungry. Any final advice on using ESCARGO?

Just because there’s a lot of potential for folding and spindling skills, doesn’t mean you have to go nuts. Keep things simple. The idea here is to use GURPS skills as is, saving ESCARGO’s bag of tricks only for where it adds some interest.

Most importantly: Much of the above looks good on paper, but it’s not properly playtested. Problems are bound to appear. If you see a flaw, rip into it!

Okay. So, to wrap this up: after additional work to finish up and playtest ESCARGO, we’ll have a Great New Improvement for GURPS that you want us all to implement, right?

Heck no! ESCARGO doesn’t address any whopping flaws in the skill system. The potential for cool tricks aside, there’s no pressing need for it that I’m aware of. This is all a mental exercise in game design, a wacky idea which in the end may be of interest to rules-geek dilettantes only. Do with it as you will.

Check. Oh, one final thought. Wouldn’t this system’s name more properly incorporate “Logarithmic”, not “Exponential”?

Quiet. After thinking up even one ridiculous acronym, Your Author shakes with exhaustion for days. There will be no talk of creating another.

Comment: D. Weber suggests “Logarithmic Adjustable Skill Tabulation system, or the LAST system for skills you’ll need”. Hey, I like that! Wish I’d thought of LAST first.


Flailing About: Choosing and using the wrong weapon

Allowing greater defaults among weapons, as suggested earlier, sounds like fun. But how should Shortsword default to Axe/Mace? Here’s a suggestion that makes generous use of double defaults. (You could use this with regular GURPS by ignoring its ban on multiple defaults, but watch out – sans ESCARGO, there’s a lot of potential for abuse.)

Each weapon skill covers one unique configuration of the following properties: length, balance, handedness, flexibility, and method of use (thrust, swung, or thrown). For example, Shortsword is the skill covering use of short, balanced, one-handed, rigid, melee weapons.

To go from one configuration to another, combine the penalties for each and every difference listed; that’s the total cross-default. (Adjust defaults after that for Ease differences.)

An important point: A weapon default is really just your adjusted ability in the parent skill, using an inappropriate weapon. When you pick up a fencer’s foil, your Shortsword skill may allow a default to Smallsword at only a small penalty, for use of a weapon of different (medium) length. But techniques and abilities (like the improved retreating bonus) that are not included in your Shortsword skill will not suddenly appear in your Smallsword default. Your “Smallsword” default is only Shortsword skill with the wrong weapon, nothing more.

See GLAIVE for a weapon-construction system that meshes with these rules.

Weapon skill characteristics


Hex size is half user’s height/length, or about a yard for a human. Two-handed swords, hammers, and axes, along with most spears and staves, are Medium weapons. Only polearms, or unusually long specimens of other weapons, are likely to be Long.


Halve the default penalty if the weapon is used to thrust instead of swing (balance makes less difference here).

GURPS does not make use of Semi-balanced weapons (see GLAIVE for a system that does), but this could be used for unbalanced weapons wielded by ST high enough to remove penalties. For example, a Broadsword user, with an unbalanced axe and enough ST to remove the 3e Ready requirement, would use Axe/Mace at only a -2 penalty from Broadsword.


One-Handed (Close Grip) is a typical sword-like grip at the butt of a weapon; One-Handed (Center Grip) is a spear-like grip. Two-Handed Close Grip means two hands at one end, per a two-handed sword. Two-Handed Wide Grip means two hands spaced apart, per a spear, staff, or polearm. Multi-Handed refers to three or more (decidedly non-human) hands.


Single-jointed refers to a weapon with one, short flex in the middle, such as nunchaku. Flexible refers to weapons with great flexibility, such as a chain, whip, or long-chained flail.


The difference between Thrust and Swung is obvious to GURPS players. Many skills (like Broadsword) include thrust and swing techniques equally; other would reasonably include thrust only (like Spear) or swing only (like Axe/Mace).


Some examples of cross defaults, before any adjustments for Ease differences:

  • The only difference between Broadsword and Shortsword is the jump from Medium to Short length. Cross-default: -2.
  • The jump from Shortsword to Knife is a similar -2.
  • The jump from Broadsword to Knife goes from Medium to Short to Close, for a total cross-default of -4 – and an additional -6 should the swordsman throw that knife.
  • Going from Broadsword to Axe/Mace, assuming a one-hex axe, means a cross-default at -4 for the difference in balance. Make that Flail, and tack on another -4 for a flexible weapon.
  • Two-Handed Sword could make use of a long-hafted battleaxe at -4 for unbalance, gripping it at the end. To switch to an un-swordlike wide grip, though, such as required to go from two-hex to one-hex range, would take an extra -2 for Two-Handed Wide Grip.
  • Spear skill could thrust a polearm at -2 for unbalance (the -4 penalty for swings, halved). Swinging that polearm, Spear skill would take the full -4 for unbalance, plus an additional -2 if we assume that Spear skill doesn’t include training in swings. (Compare this with the all-round -4 penalty in 4e for using a polearm with Spear skill. Here it’s only -2 if you thrust with the polearm, per Spear skill’s proficiency, but -6 if you swing with the polearm, something Spear skill doesn’t cover.)   
  • A staff held like a two-handed sword will exhibit no differences from the sword, skill-wise. If it’s held like a staff, though, that’s a -2 default from Two-Handed Sword, for grip.

Note that under these rules, one-handed use of Short Axe/Mace (for a hatchet) is a different skill from one-handed use of Medium Axe/Mace (for a battleaxe), just as Shortsword differs from Broadsword! And Medium Axe/Mace Two-Handed Close Grip (for two-hex reach) differs from Medium Axe/Mace Two-Handed Wide Grip (for one-hex reach). Similarly, One-Handed Spear becomes a different skill from Two-Handed Spear. And the skill to use a one-handed, two-foot, balanced nunchaku becomes a considerably different skill from that for a two-handed, two-yard, unbalanced war flail.

How to reconcile these? If you want to get technical, you do the ESCARGO thing and either buy those narrower skills as is, or combine them. Combine various Axe/Mace lengths and grips to get what GURPS calls Axe/Mace and Two-Handed Axe/Mace. (Further combine those two for an all-round skill.) Combine One-Handed Spear with Two-Handed Spear to get what GURPS calls Spear. And so on.

Points to remember

Very different weapons will lead to very poor cross-defaults – essentially, no worthwhile default at all. As always, use a simple DX-based default if that’s higher.

Don’t forget to adjust cross-defaults for Ease. Knife and Broadsword cross-default at a base -4 for difference in use. But if Knife is also a level of Ease higher than Broadsword, the net cross-default will be -3 going from Broadsword to Knife, -5 going from Knife to Broadsword.



  • Ryan Pettigrew

    Pricing should relate to understanding of game system & campaign. New players should be encouraged to generalize. Whether new players picking up a new game system, or old players participating in a new campaign, their incomplete understanding of the game, its world, and their role in the party, means that, early on in the campaign, they should be encouraged to generalize; and later on in the game, they should be encouraged to specialize, once they know what the GM is likely (and unlikely) to throw at them. “If I’m a mercenary, does that mean I need the NBC suit skill?” A new player cannot be expected to know, nor should they have to spend points early on, trying to out-guess the GM. It should just be more efficient to spend points on attribute levels for the first couple sessions, and take advantage of the base defaults.

    This means a few important things… the cost of raising or lowering attribute levels should be the same for new characters, even for characters at radically different point totals, and thus base campaign attribute levels, as long as those characters are both at the start of their respective campaigns. This may be why GURPS has a flat attribute level cost. I don’t necessarily agree with a flat attribute cost, but so long as the player is buying attributes near the base level of the start of the campaign, regardless of what that base level is for another campaign, it should cost about as much. I think this means that buying Advantages that increase the attribute levels (or effective attribute levels) to the base campaign attribute level should be at a flat rate. Buying regular attribute levels above the campaign level should cost more, and returning attribute levels should return less. But the point totals of “more” and “less” are equivalent for new characters of different base campaign attribute levels. Skills should be bought at or near their lowest levels, except for experienced players, in those situations for which it makes sense.

    At that point, buying new attribute levels should be cheaper than buying up your skills, but the cost should quickly rise at a rate where it becomes less efficient than buying up your skills. This should encourage players to specialize more. The cost of skill levels will probably still rise per level, but will also have to balance out with the cheapness of buying base campaign attribute levels, so that characters of different point totals stay equivalent… “older” characters will simply be more skillful. New players (or maybe even new characters for an old player) should probably start with an increased base campaign attribute level, based on the current point totals in the party. Even if the player has figured out the game system, and the GM, he may not yet have figured out his new place in the party. Since the skills of the older players were bought at costs balanced with base campaign attribute levels, the matching point totals should still make for equivalent characters. What this means for the costs of skills, I’m not sure. But I think a balance can be reached.

    The GM may need to play a bigger role in order to maintain this balance, and keep character creation working. But I think this line of thought sheds new light on your musings. This cost structure emphasizes learning the game quickly, easy character development, and improves player readiness at the start of the campaign. In return, it asks the game designer to do some math, to write some clear instructions for the GM so that he can maintain character balance, and the GM to follow those instructions (within reason) when assisting players with character creation. I suspect a system like this can be built, and I believe, perhaps with some tweaking, it can mesh with Escargot, or at least its principles.

    • tbone

      I agree about generalization early in the game. The players (especially if new gamers) don’t know yet what they need for the setting, or even exactly what character they want to play. Creating highly detailed characters also eats up a lot of time when players are anxious to get adventuring.

      Perhaps that’s one plus behind D&D’s eternal popularity: new characters are quick to make and easy to write up, without a lot of feats or skills or spells or other unique stuff. Beginning GURPS or HERO characters are great in the flexibility and variety they can have, but boy, those character sheets can be packed full of detail…

      On specific points:

      1) Attribute (or other) costs rising after character creation: I agree, I see no reason to raise costs after character creation. Why punish players for the expected and fun step of character growth?

      2) Cost schemes that encourage generalization at the start, specialization later: As long the schemes don’t get too complex, I agree, it’s a good goal.

      3) ESCARGO in all that: The big lists of advantages that can clutter a GURPS character form isn’t really a game play problem for new players, IMO; it’s simple enough to package a bunch into one lump (“Knight Package” etc.), and hand that to a new player (“Here, this covers most of your knightly stuff”). But the game’s skills don’t lend themselves to that easy treatment. I like the potential for ESCARGO to create a single, super-simple “Knight skill”, whose cost still reflects the number of components that went into it. Sounds like you do as well, or at least a variant system built on ESCARGO.


  • Bill

    What about the existing devaluation of point investments? Imagine an accountant PC with IQ 10, and Accounting-12 (a decent enough skill level). He’s paid 12 points for Accounting-12 (IQ/Hard). For this skill level, he gets a 74.1% chance to succeed under normal adventuring conditions (doing the Don’s books during a gunfight with Gangbusters, for example). He gets some character points and is interested in raising his skill. For 4 points, he can increase his skill to Accounting-13. Another 4 points, another +1, forever. Sound linear? Not quite.

    For 4 points, he can raise his skill to 13 (83.8% chance to succeed). He’s just bought himself (83.8% – 74.1%) or 9.7% more likelihood to succeed on a typical roll. But once he’s there, another 4 character points to skill 14 buys him only 6.9% better odds. How about skill 15? A net increase of 4.7%. And at skill 16, the net gain is only 2.7%. Each time, he’s only gained about two-thirds as much benefit from the same point cost. Beyond this, the utility of high skill to him is nothing, unless he starts rolling at penalties on a regular basis.

    • tbone

      You’re of course right that when skill level gets high, each additional level confers a shrinking number of added percentage points to the chance of success on a skill roll. I always find it interesting, though, to also turn such probabilities on their head and look at them again. In this case:

      Instead of chance of success, look at chance of failure. It’s 50% at Skill-10. Increase that to Skill-11, and the chance of failure drops to 37.5%. That’s 12.5 percentage points, but a 25% decrease in the size of the chance of failure itself (50 points x 3/4 = 37.5. I welcome suggestions from the more learned on how to better express in words this difference between an absolute change in percentage points of a probability, and a percentage change in the degree of probability itself.)

      This welcome percentage decrease in probability of failure gets better with each added skill level! At Skill-14, the chance of failure is 9.3%. Raise skill to 15, and the chance of failure plunges by half to 4.6%. That’s a greater decrease in chance of failure than the improvement from Skill-10 to Skill-11 bought!

      The increase to Skill-16 further slashes the chance of failure to 1.9%, less than half of Skill-15’s chance. Skill-17 cuts the chance to less than a third that of Skill-16, or 0.5%. And finally, Skill-18 divides that remaining chance of failure by an effective infinity, to achieve 0% chance of failure.

      So, that’s another way of looking at the matter of whether increasing skill levels are worth less and less, or not. Yet that doesn’t do away with the original viewpoint. They’re both good observations. In the end, what’s the deal – is each added level of skill actually worth more than the previous level, or less?

      I wrote my thoughts on just that topic in an earlier article, noting how arguments can be made for each side. Here’s the article; tell me what you think.

      Game design musing: Point-cost scale for stats

Leave a Reply

Your email address will not be published. Required fields are marked *

This site uses Akismet to reduce spam. Learn how your comment data is processed.