So over at SR3R we've been kicking around the issues related to Karma Pool, most notably that it mucks up metatype-balancing something fierce insofar as humans are handicapped in a short/low-karma game (where KP barely accumulates) but become extremely powerful compared to their KP-starved metahuman pals in long/high-karma games. One of the team mentioned that they'd heard people complaining about large karma pools in general, so I wanted to take a straw poll—who has observed this issue, and can you describe what the problem was?

A second obvious issue is if earned karma varies highly across the team—a character with 0 karma (1 KP) joining a group of 8-KP 'runners is going to be feeling sorta left out, especially if the character isn't extremely well-differentiated from the rest of the team role-wise.

Both of those issues seem addressable, but is there anything else people have seen?

~J

At 400 karma, this had become a pretty huge problem in the long-running game I was part of. Especially since, at 400 Karma, most of the distinct advantages of being metahuman had long since disappeared. There was a secondary issue, as well: it became extremely difficult to challenge the PCs in certain ways, since our Karma Pools were so huge that we could reasonably succeed at pretty much any task just by throwing karma at it. Sorta the die pool equivalent of "hit it with a bigger hammer."

The GM's solution was to decrease Karma pool awards over time; it made bookkeeping somewhat harder, but solved a lot of problems. It worked like this:
0-50 Karma: 1 KP every 10 Karma, as normal.
51-150 Karma: 1 KP every 20 karma.
151-300 Karma: 1 kp every 30 karma.
Et cetera, et cetera: every 5 KP earned, the number of Good Karma required ticked up by another 10. Didn't entirely eliminate our ability to hurl karma at problems, but it sure cut down on the "Fuck it, I have 40 karma pool, I'm gonna skydive naked, without a parachute!" moments.

Back when I still ran 3rd, I ended up with groups that had large enough personal and group karma pools to give me fits when trying to concoct something "challenging". As I grew as a GM I realized that at a certain point in a non linear character progression game like Shadowrun, the characters are just going to be unchallengeable from a pure mechanics stand point. At that point the challenge has to come from choice and character dynamics. Your stories have to grow with the characters if you are going to allow them to grow as well. And I've never let a game rule override common sense or campaign stability, but since you are trying to create a better set of rules I'll say this. For newer, or adversarial GM's, the large Karma pools will give them headaches, especially if you are trying to keep things reasonable when it comes to curtailing the superman syndrome some players get when they are good at calculating odds. I would limit the size of a pool to their race (a la Earthdawn) or something like that. If you do something like that, than the same should apply for enemies and the like as well to keep things balanced.

I always thought it's a good thing to limit Karma Pool to a max. size of, say, 10.

It also makes permanently expending Karma Pool dice a bit less harsh.

Bye
Thanee

This primary issue sounds like it's mostly under the "karma-pool-disparity" heading, am I right there?

Also, how did the distinct metahuman advantages disappear? I think in general they get buried under the massive Human karma pool, but the Racial Max means that karma simply can't close the gap—the degree to which this matters varies by metatype and character, but a human simply can't get that twelfth point of Quickness, or the twelfth through potentially eighteenth point of Body.


Even in the face of the TN system and the triangular cost progression of rerolls? With 41 KP, you can only make 8 rerolls on a single test, which assuming you're throwing base 10 dice isn't enough to get you above ~29.4% odds of a single success on a TN 20 task; this only gets worse if one success won't cut it, an overestimation of your odds at at least two successes on a TN 15 task still only gives you ~49.8% success odds. Note that this involves spending 36 of your 41 KP, meaning that you'll only be attempting this kind of feat once per refresh period and you'll pretty much need everything that's left for emergencies.

(The overestimation involves the fact that I assumed the player could just roll 90 dice, when actually you lose a die every time you get a success—actual odds will be somewhat worse)

Actually, I guess that's another important question—how much action typically goes down between karma pool refreshes for you? If KP refreshes frequently, I could see that causing issues as there's a significantly reduced incentive to keep a sizable pool on hand for emergencies and an increased incentive to spend liberally ("use it or lose it").


This is actually a perfect example of the kind of stuff you can't reliably pull with karma pool if they're "skydiving" from 21 meters, they need to resist 10D Physical on impact and get 8 successes to come out uninjured, and they're throwing 10 dice, the entire 36-KP-8-rerolls still doesn't get them above 47.8% odds, and with 8 successes needed my simplifying assumption of 90 dice is now a massive overestimation. From 40 meters, my probability calculator doesn't even display a non-zero digit at default precision, so it's under 0.0001% odds of success.

Granted, most of my high-KP experience thus far has been for a team with KPs in the low 20s, but thus far I've mostly found that it expands the range of little things they're willing to do a single reroll for—making it more possible to take risks or try hard tasks with consequences for failure. I think large karma pools are critical to high-powered games, but I'm trying to make sure there aren't reefs out there waiting for me.



Well, yes, in the sense that you can no longer drop a Rating 10 maglock with a Rating 4 anti-tamper system and have it be a "you must find a different way to get in" moment, but yes, the it does become more of a strategic challenge—not "can we open this door without sounding the alarm", but rather "can we open this door without sounding the alarm and without spending too much karma pool"—whereas KP is mostly a panic button when it's small, it becomes a real resource to be managed when it gets large, insofar as it needs to last long enough to get the team to the objective.


This probably gets at the heart of the issue. It's not intrinsically a problem that, if the team decides to spend the resources, they can reliably punch right through a Rating 10 maglock with Rating 4 anti-tamper system. The issue comes when a GM drops that in front of the team thinking that this will make them go find the guard with the key or something.


Really, I think the variable-TN system is mostly sufficient here (modulo a few edge cases where characters can pile up gigantic stacks of mods—social skills are the obvious place, especially if you allow SotA:64's social adept powers). If the players can calculate the odds better than the GM can, though, I can see that becoming a problem.

~J

Yeah. That was mostly a preface to issue #2, though.

In terms of the disappearing advantages, though, it's seems less easy to hard-cap your attributes, because even though you can spend karma on them, the caps are so much higher that you're effectively going to top out via ware limits earlier. Admittedly, we didn't have any trolls, but we had an Elf and an Otaku, and even though they were monstrously adept in their respective areas of expertise, the skill level disparity between them and my mage and the physad weren't too big; simple ware had boosted most of our attributes to where we'd never push past them with the skills they were linked to - Why does my mage need a Pistols above 6, or the Physad need a Magical Threats skill above 6?

While attribues were valuable, I never felt they were *as* valuable in 3rd as they are in 4th, since they don't directly contribute to diepools except as how they interact with Combat, Hacking, and Magic pool. Body and Strength are outliers here, because they DO directly factor into stuff that comes up every game, though strength to a lesser degree than Body. But the functional difference between the Otaku's Intelligence of 11 and my Mage's Int of 9 wasn't particularly large. Or at the very least, it never *felt* particularly large. And I think a lot of adjudicating imbalance comes down to feelings of disparity rather than actual disparity. For instance, while everybody knows Orks provide the best BP-to-bonus ratio in 4th edition, how useful that bonus is directly depends on how prevalent racism is, and how often Charsima will be needed in the game. For the game we were playing, the disparity in attributes after 400 karms and maybe 1,500,000 nuyen between us just wasn't very big.



Generally it was a "pool refreshes after every 'story'", to use a White Wolf term. Usually this equates to one run, but in the case of, say, Brainscan, it didn't refresh for a *large* chuck of the main action.

In terms of TN and diepool, yes - it's really hard to hit a TN 20 task, but functionally it's rare to get TNs above 9 outside hacking and extemely odd B&E situations. Smilarly, we'd housruled out as many open-ended tests as possible for the same reasons SR3R has tried to do away with them: they're way too erratic. In terms of *contested* rolls, or rolls with listed difficulties, it often didn't matter if we didn't have tools, or were high on crack, or had our heads stolen by ninjas, because with enough brute force it seemed like we'd just roll on through stuff. It was less about re-rolling one critical test multiple times, and more like, "I can re-roll 40 tests this run at least one time." It tended to screw with the results of basic tests a little, to the point where if we were merely *annoyed* with something we would toss a karma pool down the hole, because hey, I still have 32 left. That was the sort of feeling the GM wanted to avoid.

The "skydiving without a parachute" example was a bit of hyperbole, I admit. But here's an example: the elf was in the back of a van. Some dudes opened the back of the van, and she dropped two of the dudes. The remaining dudes - about six of them, if memory served - opened up with fully automatic gunfire on her with die pools of about 8. She dodged all six of them, and then killed all six of them. She burned some karma pool, make no mistake, but she only burned (I THINK) 5 of it, pretty much one per dodge roll except for one. Which means that in the combat equivalent of a pants-shitting-oh-fuck situation that would have totally annihilated a merely "competent" runner unless if they spent 50% of their karma pool, she only exhausted about 15% of her total karma pool for the run.

The question is, is being 4 times as experienced as the competent runner (400 karma vs 100 karma) a valid reason to have that task use up so much less of their available pool?



I like the resource management aspect of Karma Pool, don't get me wrong. It was just that 40 of the resource was a bit too much, in the GM's opinion. Obviously, if the game is not currently having a problem with it, then there's no reason to adjust the pools. Capping the pool at 10, 20, 30 - whatever your game's break point seems to be - seems like another totally fine option, too.

Well, for the Pistols you get several advantages—you can throw the same number of dice while spending less Combat Pool, or if you're going all-out you can dig deeper into the Combat Pool. It's not a mandate for a Mage like it would be for a physical combat character, but it can be handy. That said, the basic point of "you won't always want to bump linked skills past human levels" stands.


The Otaku situation I think is atypical, because low Quickness offsets most of the gains of the big Intelligence. For a character that isn't an Otaku (EA:Int, say), that's two dice on Perception, a point of Combat Pool, two-thirds of a point of Hacking Pool or Spell Pool, and a full point of Reaction right there—not a towering insurmountable advantage, but a clear and visible one. But I see your point about it not feeling like a big disparity in a lot of cases.


Mm. Yeah, that's pretty much how I play it, but I guess that pushes the question over to how long and involved a typical run is?


Depends on the GM and the player loadouts; for a character with a Smartlinked (-2) Heavy Pistol firing a shot at a second target (+2) at Long Range (21-40 meters, base TN 6) from 50% Cover (+2) in minimal light they're going to need to be packing Low-Light with Eye-Lights to stay below TN 10. Or if you think that's an unusual scenario, move the range to Short (now the target's within 5 meters!) and make the lighting perfect (+0), but put the target behind 50% cover (+4).

Outside of combat, there are high-rating keypads with anti-tamper systems. Detecting monowire obstacles starts at TN 8, but that's before the mods for being obscured by brush or for being distracted or for lighting conditions. Noticing a very small object is a +6 TN mod, before visibility modifiers. Object Resistance for highly-processed objects starts at 10. Signature for spyblimps is often 10 or above. First Aid for Deadly Wounds is base TN 10, as is the autostabilization test. A number of Availability tests will have TNs over 10, even for highly-optimized Faces. Scaling a 3-meter brick wall without climbing gear. For a non-Face (Good Reputation 2, no other social TN mods), trying to fast-talk a reasonably smart opponent (Int 5) who is Suspicious (+2) into doing something that could be Disastrous to them (like let them go, or allow them into the secure facility, or whatever).

That's not to mention what can happen when autofire and AP rounds start getting mixed.

I think part of the reason you often don't see those numbers is (aside from the fact that I've noticed that combat lighting conditions and ranges are often implicitly assumed to be very favourable—nasty habit of mine, that) that until you have a big karma pool most things with TNs that high are simply not worth attempting. In many cases, there's a cost for trying and failing (whether setting off an alarm or otherwise tipping one's hand, or the opportunity cost of a difficult attack when there's something else that could be done), so it never even gets attempted.


To some extent this is what I like, because it means that players can trade some of a finite resource in exchange for even more consistency, which allows more complex plans to be put together successfully because there's more ability to prevent a single freak piece of bad luck midway through sinking the plan. However, maybe the point here is that more GM guidance is necessary—if you generally weren't encountering TNs in the >9 range, you'd never need to maintain a reserve of KP to ensure that those tests went successfully when they needed to, so you'd need really absurd times between KP refreshes in order to drain the pool.


We're talking five points permanently burned here, and an unknown additional amount probably spent. That's a very high cost, and this only for one confrontation, so the spent KP is also a real cost.

More to the point, though, I think it is a valid reason—that's basically the idea behind Karma Pool, that with enough experience you can simply get into situations that less-skilled runners could not survive, and still be able to get back out of them. You still can't kill demigods, but you can plan to do something difficult and actually have a way to know you'll succeed—and because it has a cost, becoming more experienced means you can do this more times per run instead of meaning that you can do it an infinite number of times at a higher difficulty level.

~J

When I say burned, I meant spent, in that last example. There wasn't any permanent burning. Using the wrong terminology there. I'm also not advocating using Orient's system, necessarily - I'm simply saying that it seemed to work pretty well if huge pools are percieved to be a problem. I think there's a lot to be said for huge pools, but I also think there comes a tipping point. Is the return for "having a lot of karma" good enough just in terms of 'things purchased'? In many cases, having redonkulous stats and sklls seems like a good enough reward in and of itself.

In terms of regular TNs we faced:
-We rapidly ran into "huge TNs due to range", so most of us started packing assault rifles & SMGs - which 90% of the time meant we were in short-to-medium territory. Enemies were smart about cover, but we all had stupid vision enhancements and whatnot, so lighting was generally 0 to +2 tn, only higher very rarely. So combat TNs were generally 4, -2 for smartlink or other enhancements, +1-4 for range & vision, +2 for 50% cover. So we're looking at between 5 and 9 TN for the "usual" sort of encounter. If we were in some kind of situation with higher numbers, generally we'd reposition or fall back to a more advantagous position, because hey, why not, right? A couple of times that didn't work, so we just used airburst linked grenades to demolish things. Which, looking back on it, seems kinda like what they're for...

Things would get a bit higher with B&E. Rating 10 maglock, 6 anti-tamper, but I don't remember exactly how those rules worked, but they were... hard but not impossible?

So it wasn't a cakewalk, but things seemed reasonable. One of those situations where, "Okay, not EVERY major battle can take place in smoke, at night, against ninjas in max'd ruthenium at extreme range" was Orient's basic take on versimilitude. To be fair, though, we were all big, big planners who took a lot of OOC time to maximize our advantages; our KP spending was primarily on freak rolls or the inevitable double crosses & ambushes.

I'd like to share one reason I did love having a huge Karma pool. We were doing whichever mission it was in Survival of the Fittest where you have to bring the Elemental Scrolls of Ak'le'ar to south america, and right at the climax, MCT or whoever jumps the PCs from helicopters with black ops ninjas and OMG death from above. Pretty high-powered shit. I recall that on the first round, one of the attack helicopters launched a frag missile at us - typical 18D type stuff, I don't recall exactly what kind of crazy-ass code it had. And everybody rolls to dodge. I can't recall the numbers, but everybody tossed their whole combat pool into dodge but me, because I realized it would be more statistically effective for me to roll soak with combat pool, since I had a stupid amount of armor between spells and worn gear. It was something like TN 6 to dodge because of confined terrain, with a huge number required to reduce damage to zero... but my soak TN was 3, or something.

So I rolled soak, and took no damage. Everybody else was like, M, S damage or something. That was a pretty heavy "Karma Pool makes you super bad-ass" moment. It's not every day a human takes a missile to the chest and brushes it off.

our gm just made it burn for a success or buy extra die only. We were all in the 500+ karma range by the end.

During the days of the RPGA's Virtual Seattle, complaints arose about the 400⁺ karma pools that blew away the modules, even with graduated opposition.

I pride myself on writing one module that was successfully completed by a number of rookie teams, yet TPK'd a 400⁺ KP team, at Gencon I seem to remember. Arrogance and failure to think killed them. The opposition was merely the tool.

This seems like a danger to me with rising karma pools. Players can get lazy, complacent and stupid.

EDIT: To clarify, it was 400⁺ karma, giving a karma pool of 40⁺

Ah, I see. I'd considered that, but without tactical errors by the attackers I'd figured it was a sufficiently nasty situation to cause pool burning.


Well, the thing is that redonkulous stats and skills can't make up for high TNs—even at TN 10 you're only looking at a 72% chance of success, which is probably about right for the general case (that is, with no resource expenditure) but which is too much of a risk for a lot of actions (like the ones with alarms, or that involve trying to fast-talk people).


Mm, granted, if you pack bigger guns the higher range mods become rarer, especially for ARs. 41-80 meters doesn't seem like it should be that rare, though, which is Long for an SMG. Our group packs smaller guns a lot of the time, but especially if you've got a Rigger to do close-range dropoffs and pickups while still getting away cleanly, I guess you can afford to move up the size/power/range scale.


By canon anti-tamper tops out at 4, I think; it's TN of Rating+Anti-Tamper Rating to crack the case and failure sounds an alarm (if Anti-Tamper is present), followed by an additional test with TN of Rating with the same consequences for failure (regardless of Anti-Tamper). Without KP they're hard but not impossible if and only if setting off the alarm is acceptable.


Sure, but honestly, if the conditions aren't bad and the opposition isn't either well-equipped or well-prepared, the combat is mostly a narrative road-bump.


Heh, nice. Of course, the armor was the real secret sauce


I think the key there might be that the use of karma pool never became expected. Once you get up to the 10, 15+ karma pool range, let alone the 20-40 range, KP use becomes part of the basic capabilities of the character, and a run designed to be accomplished entirely without it probably becomes excessively easy.

~J

Our SR3 group has an intesting system for handling Karma pool that works very nicely.
This is cut straight off of our cast forums.

--------------------------------------------------------------------------------

Scale Rules
All characters are classified as either being On-the-Scale or Off-the-Scale, by player choice. All characters are classfied as Off-the-Scale at character generation, and will remain so until the player formerly announces that said character is officially On-the-Scale from this day forth. Once On-the-Scale, a player may never go back to the Off-the-Scale status.

Off-the-Scale - While in the Off-the-Scale status, players may spend any earned Karma points earned on any skills or attributes less than Rating-4 with no regard to training time (per SR3 Main Rulebook P.245 and Shadowrun Companion P.48). On the downside, Off-the-Scale characters may not have a Karma Pool statistic (SR3 p.242), which makes things MUCH more dangerous for gaming (no re-rolls, etc). While in the Off-the-Scale status, characters may be fully developed at a much higher risk during gameplay. Everything "learned" during this time is assumed to have been known all along (which is why there is no learning time), helping the PC to flesh out previous training, etc. Nothing is considered actually learned during this process. GM's may even allow Karma to be used for things outside the Attribute/Skills area (contacts, etc), but these have to be publically declared.

On-the-Scale - Once it has been declared that a specific character is On-the-Scale, he immediately gets a Karma Pool Rating of a 1, and must then start maintaining a Earned Karma (from that point forward) tally. That tally represents all EARNED Karma from that point forward. The Karma Pool Rating will increase as the Earned Karma tally reaches specific totals, as per the table below. Note that Earned Karma is a value that represents all of the karma gained during game play. It does not include bonuses for things like background story bonuses, character questionairres, etc. Also note that this value can never be lowered, it can only be increased.

Burning Karma - Karma Points may be burned as per current SR3 rules. All On-the-Scale characters must keep a new stat on record identified as Burned Karma. Any Karma Pool points that are burned by the player will increase this rating. A player's Karma Pool rating will equal to the value earned per their Earned Karma rating, referencing the chart below, MINUS their Burned Karma rating. So, if in your character's career, he has burned 2 Karma Pool (Burned Karma: 2), and has 135 Earned Karma, then he would have 8 Karma Pool (10 from the chart below minus 2 Burned Karma)

For rules on how Karma Pool works, refer to page 244 in the main rulebook.
It has been decided by the GM team that race will play no role in the new Karma Pool system.


Earned Karma Karma Pool
0 1
3 2
9 3
18 4
30 5
45 6
63 7
84 8
108 9
135 10
165 11
198 12
234 13
273 14
315 15
360 16
408 17
459 18
513 19
570 20
630 21
693 22
759 23
828 24
900 25

I found that my group got pretty conservative as their Karma Pool went up. It let them regularly succeed at things that used to be challenging and at least gave them the option of trying the crazy stuff. By that time they were so attached to their character, and so paranoid that they were always expecting worse things ahead so unless they knew for sure that succeeding at that one roll would keep them alive and let them escape to safety they wouldn't invest too much KP in that one roll. Most of them were seriously looking at retiring their characters soon ("I'm too old for this shit!") so their goal was to "win" Shadowrun, not necessarily succeed at the current job.

Led to some pretty cool role-playing as the new runners that joined them were still all full of piss and vinegar and the jaded old bastards were constantly telling them to GTFO. Gave it a war movie feel.

My group has changed it to:
Humans start needing 10, metas 20 like normal, but doubling the karma needed amount.
so a human would need:
10: +1 pool
20: +2 pool
40: +3 pool
80: +4 pool
...

In the end the metas are 1 pool behind and you don't end up with crazy pools.

So you also believe that the pools are inherently crazy when they get large, independent of in-group disparity? Is this based on experience?

~J

I'd say its more of the groups play style. I don't see a problem with the default rules, it really doesn't change a whole bunch to me. We based our rules from the companion p. 82 where it talks about limiting karma pool.

We'd encounter situations where every turn someone was using a pool to do things even if it wasn't required. It'd often slow down the game and wouldn't be used for what the group felt was dramatic effect. My group decided to reduce the points gained due to similarly bad experiences with large pool sizes and have since enjoyed the change.