fmitchell

BTW, the practical limit for any creature's STR, DEX, etc. is 480% (or old-school 96). At that point it has a straight 94.31% chance of success against all opponents except ones with an equal opposing score (50%) or higher (5.69%). EDIT: I should say 475% / 238 / 95. At that point, all rolls are either criticals, extreme successes, or fumbles.

It's a long one, since I wanted to make it as user-friendly as possible. I called it coc7rt.rb. #!/usr/bin/env ruby ## ## Routines and a command-line program to calculate probabilities of success ## for opposed tests in _Call of Cthulhu_ 7th edition. ## require 'rational' FUMBLE = -1 FAILURE = 0 SUCCESS = 1 HARD_SUCCESS = 2 EXTREME_SUCCESS = 3 CRITICAL_SUCCESS = 5 # Calculate opposed success chances def coc7_opposed_success(a, r, bf=false) bf ? coc7_opposed_success_bf(a, r) : coc7_opposed_success_p(a, r) end # Calculate opposed success chances w/ brute force (counting all possible outcomes) def coc7_opposed_success_bf(a, r) total = Rational(0,1) incr = Rational(1,10000) 1.upto(100) do |ar| as = success_level(a, ar) 1.upto(100) do |rr| rs = success_level(r, rr) if as > rs then total += incr elsif as == rs then if a > r then total += incr elsif a == r then total += incr/2 end end end end return total; end def success_level(target, actual) if actual > 95 then FUMBLE elsif actual == 1 then CRITICAL_SUCCESS elsif actual > target then FAILURE else if actual <= target/5 then EXTREME_SUCCESS elsif actual <= target/2 then HARD_SUCCESS else SUCCESS end end end # Calculate opposed success chances w/ probability calculations def coc7_opposed_success_p(a, r) result = p_critical(a) * (1 - p_critical(r)) + p_extreme_only(a) * (1 - p_extreme(r)) + p_hard_only(a) * (1 - p_hard(r)) + p_ordinary_only(a) * p_failure(r) + p_failure_no_fumble(a) * p_fumble(r) # In cases of a tie, success goes to the higher percentile if (a > r) then result += p_tied(a, r) elsif (a == r) then # coin toss, essentially result += p_tied(a, r) / 2 end return result end # Converts integers to probabilities (0..1) def pct(p) if (p >= 100) then Rational(1) elsif (p < 0) then Rational(0) else Rational(p,100) end end # Probability of a CoC7 fumble def p_fumble(p) pct(5) end # Probability of a CoC7 critical success def p_critical(p) pct(1) end # Probability of a CoC7 ordinary success or better def p_success(p) max = pct(100) - p_fumble(p) result = pct(p) ### Success never more than 100 - fumble ### return (result > max) ? max : result; end # Probability of a CoC7 extreme success or better def p_extreme(p) p_success(p/5) end # Probability of a CoC7 extreme success only def p_extreme_only(p) p_extreme(p) - p_critical(p) end # Probability of a CoC7 hard success or better def p_hard(p) p_success(p/2) end # Probability of a CoC7 extreme success only def p_hard_only(p) p_hard(p) - p_extreme(p) end # Probability of a CoC7 ordinary success only def p_ordinary_only(p) p_success(p) - p_hard(p) end # Probability of a CoC7 ordinary failure or worse def p_failure(p) pct(100) - p_success(p) end def p_failure_no_fumble(p) p_failure(p) - p_fumble(p) end # Probability of tied success levels def p_tied(a, r) p_critical(a) * p_critical(r) + p_extreme_only(a) * p_extreme_only(r) + p_hard_only(a) * p_hard_only(r) + p_ordinary_only(a) * p_ordinary_only(r) + p_failure_no_fumble(a) * p_failure_no_fumble(r) + p_fumble(a) * p_fumble(r) end ### Print out probability table def print_coc7_rt(min=10, max=100, step=10, bf=false) mind = (min/step).to_i maxd = (max/step).to_i puts("") printf("CoC7 Opposed Rolls Probability, from %d to %d\n", mind * 10, maxd * 10) puts('-'*50) puts("") puts("(brute force method)\n\n") if bf printf('| %4s |', 'A\\R') mind.upto(maxd) do |rd| r = rd * step printf('| %5d ', r) end puts("|") printf('|------:|') mind.upto(maxd) do |rd| printf('|------:') end puts("|") mind.upto(maxd) do |ad| a = ad * step printf('| %4d |', a) mind.upto(maxd) do |rd| r = rd * step p = coc7_opposed_success(a, r, bf) printf('| %5.2f ', p.to_f * 100.0) end puts("|") end puts("") end ###### COMMAND LINE ########## def run! require 'optparse' min = 40 max = 80 step = 10 bf = false lookup = false opts = OptionParser.new do |opts| opts.banner = "Usage: coc7rt.rb [options] [-l active resist]" opts.on("-b", "--[no-]brute-force", "Use brute force method") do |v| bf = v end opts.on("-l", "--[no-]lookup", "Lookup a specific entry") do |v| lookup = v end opts.on("-n", "--min [INTEGER]", "Set minimum percentile in table") do |v| min = v.to_i end opts.on("-s", "--step [INTEGER]", "Set step between rows/columns") do |v| step = v.to_i step = 5 if step < 5 end opts.on("-x", "--max [INTEGER]", "Set maximum percentile in table") do |v| max = v.to_i end end.parse! if (lookup) then if (ARGV.size == 2) then a = ARGV[0].to_i r = ARGV[1].to_i p = coc7_opposed_success(a, r, bf) printf("\n%d vs %d: %5.2f%% \n", a, r, p.to_f * 100) else opts.usage exit(1) end else print_coc7_rt(min, max, step, bf) end end run! if __FILE__ == $0

OK, here's the revised table, extended to 200. I'm pretty confident about this one, since I get the same numbers with two separate algorithms. CoC7 Opposed Rolls Probability, from 10 to 200 -------------------------------------------------- | A\R || 10 | 20 | 30 | 40 | 50 | 60 | 70 | 80 | 90 | 100 | 110 | 120 | 130 | 140 | 150 | 160 | 170 | 180 | 190 | 200 | |------:||------:|------:|------:|------:|------:|------:|------:|------:|------:|------:|------:|------:|------:|------:|------:|------:|------:|------:|------:|------:| | 10 || 50.00 | 12.90 | 12.23 | 11.56 | 10.89 | 10.22 | 9.55 | 8.88 | 8.21 | 7.79 | 7.62 | 7.45 | 7.28 | 7.11 | 6.94 | 6.77 | 6.60 | 6.43 | 6.26 | 6.24 | | 20 || 87.10 | 50.00 | 19.66 | 18.30 | 16.94 | 15.58 | 14.22 | 12.86 | 11.50 | 10.64 | 10.28 | 9.92 | 9.56 | 9.20 | 8.84 | 8.48 | 8.12 | 7.76 | 7.40 | 7.34 | | 30 || 87.77 | 80.34 | 50.00 | 25.04 | 22.99 | 20.94 | 18.89 | 16.84 | 14.79 | 13.49 | 12.94 | 12.39 | 11.84 | 11.29 | 10.74 | 10.19 | 9.64 | 9.09 | 8.54 | 8.44 | | 40 || 88.44 | 81.70 | 74.96 | 50.00 | 29.04 | 26.30 | 23.56 | 20.82 | 18.08 | 16.34 | 15.60 | 14.86 | 14.12 | 13.38 | 12.64 | 11.90 | 11.16 | 10.42 | 9.68 | 9.54 | | 50 || 89.11 | 83.06 | 77.01 | 70.96 | 50.00 | 31.66 | 28.23 | 24.80 | 21.37 | 19.19 | 18.26 | 17.33 | 16.40 | 15.47 | 14.54 | 13.61 | 12.68 | 11.75 | 10.82 | 10.64 | | 60 || 89.78 | 84.42 | 79.06 | 73.70 | 68.34 | 50.00 | 32.90 | 28.78 | 24.66 | 22.04 | 20.92 | 19.80 | 18.68 | 17.56 | 16.44 | 15.32 | 14.20 | 13.08 | 11.96 | 11.74 | | 70 || 90.45 | 85.78 | 81.11 | 76.44 | 71.77 | 67.10 | 50.00 | 32.76 | 27.95 | 24.89 | 23.58 | 22.27 | 20.96 | 19.65 | 18.34 | 17.03 | 15.72 | 14.41 | 13.10 | 12.84 | | 80 || 91.12 | 87.14 | 83.16 | 79.18 | 75.20 | 71.22 | 67.24 | 50.00 | 31.24 | 27.74 | 26.24 | 24.74 | 23.24 | 21.74 | 20.24 | 18.74 | 17.24 | 15.74 | 14.24 | 13.94 | | 90 || 91.79 | 88.50 | 85.21 | 81.92 | 78.63 | 75.34 | 72.05 | 68.76 | 50.00 | 30.59 | 28.90 | 27.21 | 25.52 | 23.83 | 22.14 | 20.45 | 18.76 | 17.07 | 15.38 | 15.04 | | 100 || 92.21 | 89.36 | 86.51 | 83.66 | 80.81 | 77.96 | 75.11 | 72.26 | 69.41 | 50.00 | 31.56 | 29.68 | 27.80 | 25.92 | 24.04 | 22.16 | 20.28 | 18.40 | 16.52 | 16.14 | | 110 || 92.38 | 89.72 | 87.06 | 84.40 | 81.74 | 79.08 | 76.42 | 73.76 | 71.10 | 68.44 | 50.00 | 32.15 | 30.08 | 28.01 | 25.94 | 23.87 | 21.80 | 19.73 | 17.66 | 17.24 | | 120 || 92.55 | 90.08 | 87.61 | 85.14 | 82.67 | 80.20 | 77.73 | 75.26 | 72.79 | 70.32 | 67.85 | 50.00 | 32.36 | 30.10 | 27.84 | 25.58 | 23.32 | 21.06 | 18.80 | 18.34 | | 130 || 92.72 | 90.44 | 88.16 | 85.88 | 83.60 | 81.32 | 79.04 | 76.76 | 74.48 | 72.20 | 69.92 | 67.64 | 50.00 | 32.19 | 29.74 | 27.29 | 24.84 | 22.39 | 19.94 | 19.44 | | 140 || 92.89 | 90.80 | 88.71 | 86.62 | 84.53 | 82.44 | 80.35 | 78.26 | 76.17 | 74.08 | 71.99 | 69.90 | 67.81 | 50.00 | 31.64 | 29.00 | 26.36 | 23.72 | 21.08 | 20.54 | | 150 || 93.06 | 91.16 | 89.26 | 87.36 | 85.46 | 83.56 | 81.66 | 79.76 | 77.86 | 75.96 | 74.06 | 72.16 | 70.26 | 68.36 | 50.00 | 30.71 | 27.88 | 25.05 | 22.22 | 21.64 | | 160 || 93.23 | 91.52 | 89.81 | 88.10 | 86.39 | 84.68 | 82.97 | 81.26 | 79.55 | 77.84 | 76.13 | 74.42 | 72.71 | 71.00 | 69.29 | 50.00 | 29.40 | 26.38 | 23.36 | 22.74 | | 170 || 93.40 | 91.88 | 90.36 | 88.84 | 87.32 | 85.80 | 84.28 | 82.76 | 81.24 | 79.72 | 78.20 | 76.68 | 75.16 | 73.64 | 72.12 | 70.60 | 50.00 | 27.71 | 24.50 | 23.84 | | 180 || 93.57 | 92.24 | 90.91 | 89.58 | 88.25 | 86.92 | 85.59 | 84.26 | 82.93 | 81.60 | 80.27 | 78.94 | 77.61 | 76.28 | 74.95 | 73.62 | 72.29 | 50.00 | 25.64 | 24.94 | | 190 || 93.74 | 92.60 | 91.46 | 90.32 | 89.18 | 88.04 | 86.90 | 85.76 | 84.62 | 83.48 | 82.34 | 81.20 | 80.06 | 78.92 | 77.78 | 76.64 | 75.50 | 74.36 | 50.00 | 26.04 | | 200 || 93.76 | 92.66 | 91.56 | 90.46 | 89.36 | 88.26 | 87.16 | 86.06 | 84.96 | 83.86 | 82.76 | 81.66 | 80.56 | 79.46 | 78.36 | 77.26 | 76.16 | 75.06 | 73.96 | 50.00 | I'll provide the Ruby script in a following post, so other programmers can check my assumptions. I'd check it into the download section, but it's probably not interesting to many people.

This evening I wrote a Ruby program to calculate probabilities in CoC7. I can post it for peer review, after I do a little cleanup and retesting, but for now, this looks right: | A\R || 10 | 20 | 30 | 40 | 50 | 60 | 70 | 80 | 90 | 100 | 110 | 120 | |------:||------:|------:|------:|------:|------:|------:|------:|------:|------:|------:|------:|------:| | 10 || 50.00 | 12.74 | 11.99 | 11.24 | 10.49 | 9.74 | 8.99 | 8.24 | 7.49 | 7.09 | 7.04 | 6.99 | | 20 || 87.26 | 50.00 | 19.18 | 17.66 | 16.14 | 14.62 | 13.10 | 11.58 | 10.06 | 9.24 | 9.12 | 9.00 | | 30 || 88.01 | 80.82 | 50.00 | 24.08 | 21.79 | 19.50 | 17.21 | 14.92 | 12.63 | 11.39 | 11.20 | 11.01 | | 40 || 88.76 | 82.34 | 75.92 | 50.00 | 27.44 | 24.38 | 21.32 | 18.26 | 15.20 | 13.54 | 13.28 | 13.02 | | 50 || 89.51 | 83.86 | 78.21 | 72.56 | 50.00 | 29.26 | 25.43 | 21.60 | 17.77 | 15.69 | 15.36 | 15.03 | | 60 || 90.26 | 85.38 | 80.50 | 75.62 | 70.74 | 50.00 | 29.54 | 24.94 | 20.34 | 17.84 | 17.44 | 17.04 | | 70 || 91.01 | 86.90 | 82.79 | 78.68 | 74.57 | 70.46 | 50.00 | 28.28 | 22.91 | 19.99 | 19.52 | 19.05 | | 80 || 91.76 | 88.42 | 85.08 | 81.74 | 78.40 | 75.06 | 71.72 | 50.00 | 25.48 | 22.14 | 21.60 | 21.06 | | 90 || 92.51 | 89.94 | 87.37 | 84.80 | 82.23 | 79.66 | 77.09 | 74.52 | 50.00 | 24.29 | 23.68 | 23.07 | | 100 || 92.91 | 90.76 | 88.61 | 86.46 | 84.31 | 82.16 | 80.01 | 77.86 | 75.71 | 50.00 | 25.76 | 25.08 | | 110 || 92.96 | 90.88 | 88.80 | 86.72 | 84.64 | 82.56 | 80.48 | 78.40 | 76.32 | 74.24 | 50.00 | 27.09 | | 120 || 93.01 | 91.00 | 88.99 | 86.98 | 84.97 | 82.96 | 80.95 | 78.94 | 76.93 | 74.92 | 72.91 | 50.00 | Equal percentiles are 50/50, as in the old Resistance Table. Low percentiles still have a slight chance, probably based on the straight 5% fumble. Note, though, how chances for the less skillful party fall off faster, even in the 190 vs 200 case; I suspect this is due to the "higher skill wins" rule. EDIT: Apparently my probability math is off, because I'm getting weird numbers beyond 200. Among other things I'm not limiting hard and critical successes by the "fumble" limit. I've removed columns beyond 120, and even then the numbers might not be exactly right.

I'll reserve judgement until I see what the opposition looks like. Using the resistance table, STR 15 vs STR 10 has the same chance as STR 25 vs. STR 20. Using opposed rolls in the current rules, STR 125% and STR 100% are a little more evenly matched, and STR 275% vs STR 250% is far more dependent on luck. I suspect creatures will have more tricks in their tentacles than just high stats ... enough perhaps to require pushing rolls and Luck points. (Pushing rolls has an explicit if GM-dependent cost, and I suspect spending luck will be more painful than blowing a few luck rolls.)

I doubt Lovecraft was thinking about hit points when writing that scene, or even force and momentum. It was "big thing crashing into other big thing". A scale mechanic might cover that situation ... or a GM saying "yeah, that might work". In most stories, though, it's humans with human-scale weapons (if that) trying to get out alive. In retrospect, I suspect my supposed "simplifications" really aren't much better than CoC7 Characteristic = CoC6 Characteristic x 5%. Going by the rules available, stat of 500% is essentially a > 95% chance of an "extreme" success. (I hope a skill > 100% reduces fumble chances.) In a way, that might combat stat inflation, which is one of my pet peeves.

That was my big question. A few simplifications occur to me: 1. Adapting the "mastery" mechanic from HeroQuest, every superhuman creature has "automatic" levels of success over mere mortals. 2. A "scale" stat for extremely large creatures would add multiples of 100 or an equivalent to STR, CON, and SIZ. 3. Especially large or superhuman creatures don't have regular stats, just SIZ/scale, HP, AP, and attacks (damage, percentiles). Alternatively, Cthulhu or Abhoth isn't a creature but a series of environmental hazards and dodge-or-die mechanics (hence Luck). All of these are very Lovecraftian, but not very BRP. BTW, I may have been skimming too fast, but I didn't see how skills over 100 are supposed to work. The creature in the sample scenario -- do I have to worry about spoiling "The Haunting"? -- has an attribute over 100, and apart from higher chances of "hard" and "extreme" success it doesn't provide any real advantage.

Backers of the 7th ed Kickstarter received a "Call of Cthulhu 7ed Quickstart" this weekend. Character generation has some major departures from standard BRP, but I kind of like them. Other parts, naturally, are extremely familiar. Is there an implicit NDA for the Kickstarter backers? I was expecting the forums to blow up, and I'm bursting to talk about the rule changes. OTOH, I don't want to step on any toes.

More of a triangle really: a straight line from (2, 2.78%) to (7, 16.67%) and another back down to (12, 2.78%). The standard deviation is 2.42. Here's the probability table in CSV format: 2,2.77777777778 3,5.55555555556 4,8.33333333333 5,11.1111111111 6,13.8888888889 7,16.6666666667 8,13.8888888889 9,11.1111111111 10,8.33333333333 11,5.55555555556 12,2.77777777778

This is why I gravitate towards systems like FATE, HeroQuest, and PDQ. Players (with GM approval) craft their own abilities and disabilities, and mechanically they're all the same. Advantages only provide benefits when the GM agrees that they apply; disadvantages (where available) are only worth the Fate Points or Style Dice they earn. "Balance" comes through GM discretion, difficulty levels, and in FATE the Fate Point economy. Nobody has license to hog the spotlight because they found the killer power combination, and nobody's stuck with a bad choice because the "math" won't work out otherwise. To be honest, I've never been a fan of the superhero genre, so I've never really used the Super Powers system. I'm more comfortable with species-based powers, and there it's easy to come up with balanced packages of abilities and double-edged swords. For example, your alien has thermal vision, but what happens when there's a bonfire or a torch shoved in his face? You're a hulking, four-armed insect man; what happens when you walk through the center of a human town?

Really? In one experiment I created a Call of Cthulhu character in 45 minutes, recording every step as I went; I probably could have cut that down to 25-30 minutes. (For comparison, a GURPS character -- GURPS Lite only -- took me 45 minutes, but I doubt I could have cut that down too much.) I haven't played HERO in a while, but from memory I generally spent at least 45 minutes working on a character, including all sorts of min-maxing, book-flipping, and power modifier calculations. There are people who know HERO backwards and forwards, and can work out a character almost as fast as they can write it down. I'm not one of those people. Granted, you can list all the powers (and modifiers) on a single sheet. However, I'm not one of those people who could remember the details of every power and modifier, so sooner or later I'd have to dive back into the book. On the other hand, the skills of BRP and its relatives are, if not quite orthogonal, distinct enough that I know which ones to use for which situation (Spot/Search/Listen notwithstanding). RuneQuest combat (MRQ/RQ6) occasionally sends me back into the book, but for the most part I only need reference sheets for maneuvers / special effects.

Speaking of GURPS, one thing I like about BRP is that there's NO advantages/disadvantages, qualities/drawbacks, merits/flaws, etc. Let me explain. Once upon a time, I thought GURPS was the bee's knees. I'm not sure what changed: HERO burnout, d20 feat fatigue, the rise of FATE, or the same pattern repeated in every single system since HERO a/o GURPS. In any case I appreciated the virtue of summarizing everything a player needed to know on a single sheet of paper. Imagine a game with no reference books at one's elbow, no meticulously copied index cards or Official Power Cards describing a character's every ability in detail; imagine, instead, a consistent procedure that resolves nearly every situation. (Combat and Magic, as usual, are the partial exceptions.) That isn't D&D 3.x or 4e, or Unisystem, or Ars Magica, or World of Darkness. That is FATE and PDQ, and that was BRP when I played it again after a long absence. The "few rules, many exceptions" philosophy which D&D explicitly endorses, and the others implicitly endorse, leads to games with too many moving parts. At best players and GMs must always keep in mind when Advantages and Disadvantages (or however they're named) kick in. At its worst you have fiddly default rules removed by one of said exceptions, and players who solve problems by using a kewl power instead of thinking through the problem. (And woe to a GM who says the power doesn't work in this situation; they paid good points for that power!) Disadvantages/complications/flaws/etc. can lead to min-maxing and "free points" for disadvantages that aren't. Obviously some systems dodge these bullets: D6 in general and MiniSix in particular uses "Perks" sparingly, and Mutants and Masterminds Complications aren't extra build point but Hero Point attractors. Still, I'd rather avoid the whole mess. Even the Super Powers of BRP make me a little uneasy. I'd rather players focused on what they want to do rather than what their character sheets say they can do.

Arguably it's vital if the players are encountering BRP for the first time. (This is why I think Magic World is brilliant.) For one (unfortunately very short-lived) campaign I put together a player's pack which summarized the BRP rules I was using, listed relevant skills, and included some of my house rules. Luckily BRP character generation is easier to summarize than GURPS.

You could also steal a page from FATE (and C7's DWAiTaS) and fuel some or all superpowers with Story Points. On one level it doesn't matter whether a character dodges a Dalek death ray due to cybernetically augmented reflexes, precognitive abilities, or a fortuitously placed steel cabinet. Sure, it may violate common sense that a 21st-century human survives death rays as well as a 29th-century cyborg -- "where are my steel cabinets" the cyborg cries -- but it's as genre-appropriate for the plucky human to survive by luck as it is the superhuman to encounter a glitch or weakness at exactly the wrong time. ("And suddenly, the BBC practical effects supervisor suffers a heart attack!")

After seeing so many systems' attempts at "base characteristics", I'm increasingly convinced that they're more trouble than they're worth. My heartbreaker would either follow FATE / Castle Falkenstein / GUMSHOE / et al. to place "Might" and "Stamina" on the same level as "Investigation" and "Driving", or take the free-form attribute approach of HeroQuest, PDQ, the Over the Edge system (WaRP), and others whereby players and GM define their own abilities. The BRP family still uses STR, DEX, INT, and the rest, and that's fine. The Mongoose lineage has largely eliminated characteristic tests -- Might replaces STR tests, Resilience/Stamina replaces CON tests, etc. -- but we still need STR, CON, SIZ and the rest to calculate hit points, damage bonuses, magic points, skill defaults, and so forth. Someday I might design a version of BRP that depends only on skills ... probably the same day I redesign GURPS to replace ST/DX/IQ/HT with Talents and advantages, and right after I write a new version of D&D.

Regarding Intelligence stats: Note, however, there are four physical stats in BRP, five if you count APP, but only two mental ones, INT and POW. STR, CON, DEX, and SIZ represent lifting ability, resistance to infection/fatigue/injury/etc., physical coordination (hand-eye AND agility), and height/weight/bulk. All of these are related somewhat -- exercise can increase stamina, strength, and coordination, depending on the type of exercise -- but they're separate stats in the system. Contrast that to INT, POW, and (sometimes) EDU. POW, arguably, is a "mystical" or "spiritual" stat, except that it also measures mundane willpower under the theory that one imposes one's will on the universe. EDU distinguishes general knowledge from "mental flexibility" or however one interprets INT, but in ancient and medieval societies only a select few have any degree of general education (and some of that is flat out wrong). In such societies it's even more important to distinguish the book-learning of a scholar, the observational skills of a "ranger", the perceptiveness and improvisation of a thief or con-man, etc. My main stumbling block is that "intelligence" is a vague and overly broad name. The most recent Doctor Who RPG gives us Ingenuity and Awareness (plus Presence and Resolve), appropriate enough for a series where the title character out-thinks his enemies. WEG's Star Wars RPG used Knowledge, Mechanical, Technical, and Perception; tossing out the two science-related attributes we still distinguish knowing things from making mental connections. (Perception was the basis for the skills "bargain", "con", and "persuasion", among others.) In action-oriented genres, especially in sword-and-sandal or Arthurian legends, "thinky stuff" takes a back seat. While there are a few heroes like Odysseus who depend heavily on their wits, most characters will have a few specialized fields of expertise represented more effectively by skills. A not-so-clever player trying to play somebody smarter needs prompting from the GM and fellow players to be "the smart one", which can be unsatisfying for everyone else. ("I think of a solution!" *roll* *roll* "OK, you see ...") A clever player playing a thick-as-two-planks character is either frustrating or insulting to the mentally challenged community. ("Here's an idea! Why don't we --" "Wait!" *roll* *roll* "Nope, sorry, you didn't think of that.") As you said in a part of the post I cut, "Intelligence" works fine mechanically in BRP. But it's neither necessary nor, depending on genre and setting, sufficient. (BTW, I'm not the only one; at least one blogger things that analogous D&D stats are over-broad. Said post renames "Intelligence" to "Acuity" to reflect attention to detail and adaptability without the baggage of "smart"/"dumb".)

Except that "intelligence" isn't a single number in real life, as social scientists belatedly discovered. Book-learning is just as easily represented as a set of skills ... as are animal-handling, social influence, artistic ability, and any other arena in which some people excel and others fumble around. The real difference between Mr. Bean and Dr. Hawking is common sense. Dr. Hawking knows you can't paint a room by attaching dynamite to a paint can; Mr. Bean doesn't but lives in a world where that somehow works. If you want to be rewarded for doing stupid things ... well, there's always FATE (and PDQ# and Mutants & Masterminds 3).

Magic World has more detailed skills, random AP, and the Major Wounds table, but character generation is a little simpler and there's only one magic system that covers a lot of territory, so I think it balances out. Magic World is a thicker book largely due to larger print and a little more hand-holding. Really, either is a valid choice; it depends on the mood you're trying to create -- old-school Stormbringer or old-school RuneQuest -- and what bits you need.

Ward Location (free combat action) and Choose Location (combat special effect) go away. Otherwise the combat rules don't change, as far as I can tell.

I'm a big fan of FATE's movement via zones. Essentially the map consists of zones defined by line-of-sight, existing barriers, and ease of movement. Characters usually move at most only one zone a turn; they can strike anyone in their zone with a fist or hand weapon, poke people in adjacent zones with a polearm, and shoot up to five zones away with increasing range penalties. This sort of map illustrates who's standing where and roughly how far each target is without having to count out squares/hexes/inches every round. Diaspora (my source for the explanation) extends the concept of zones to strategic combat, space combat, and social combat (i.e. where every party in a negotiation or verbal conflict is in relation to their starting positions and goals).

D&D has been the 800 lb owlbear for a while now. There's no official version of D&D (4e is on its way out, and 5e hasn't hit the street), the OSR is gaining attention, and Goodman rose to prominence thanks to its original Dungeon Crawl Classic modules. On the other hand fans of RuneQuest are (I suspect) mostly older, and Mongoose's version didn't exactly win new players. So a 10:1 ratio sounds about right.

For the most part I never got into mecha anime, but ... In the interview you mention adding '"storytelling" rules' to BRP. Can you give an indication of what those are? For example, I presume a Mecha has a STR, SIZ, and DEX (and CON?). There's probably an option for hit locations, if not a specialized "major wound table" for mechanical devices. Does the "storytelling" aspect come into play when describing the effects of damage? When using the mecha's special abilities? The bond between a pilot and his/her mecha? Thanks.

Some parts of the Pendragon system are fairly brilliant: Removing INT and (except for 4th edition magicians) POW, on the premise that skills and a player's own intelligence replace mental stats. Replacing "weapon damage + Damage Modifier" with basic sword Damage and adjustments for type of weapon (e.g. axes and maces do 1 die less damage, daggers are half damage). As noted above, reducing the skill granularity from d100 to d20 to make experience and criticals/fumbles simpler. Passions and (to a lesser extent) personality traits, both as goads to action and occasional sources of benefits. Granted, some of these elements work best in a game of knights who always follow their hearts/guts/other organs below the neck. However, I've seen some of these mechanics in subsequent games: GURPS (base damage), d20 (skill granularity and unifying mechanic), RQ6 (passions), HeroQuest (d20 skills and passions), ... the list goes on. Except in some really old and obscure games I've yet to see mental characteristics removed, but if I ever design my own heartbreaker that will be a central feature.

Savage Worlds never really grabbed me, so I guess I'd be better off with FATE (or the aforementioned PDQ, with a little work).

If you're looking for material to convert (or just inspire) take a look at Goblinoid Games' Starships & Spacemen 2nd ed. This new version of a 1978 game ports the messier original rules to "old-school" d20, so it's easier to translate into BRP. S&S2 also allocates gear using rank and department, and contains subsystems for exploring strange new worlds and shooting Kling-, I mean Zangids out of the sky. If nothing else, check out the Random Bumpy Forehead table.