Success Test Thresholds, Or, Success to Extended Test Conversion Options

[Dashifen]

I was reading stuff over today following my RL game and I noticed that the book seems to indicate that the hardest success test threshold is 4. However, there are some situations, e.g. detecting a stealthy individual who gets 5+ hits on Agility + Infiltration, where it's very easy for there to be a higher threshold. I've always just used the threshold regardless of it's magnitude and had opponents attempt to beat it. Sometimes they do, often they do not.

However, I had an insight today: if the threshold of a success test exceeds 4, it could be "converted" into an extended test. It'd probably be up to the GM as to the interval, but it should probably not exceed 1 minute and the threshold would remain the same. In other words, let's say that after all modifiers, the threshold for someone's success test is 6. It could be very difficult to impossible for the average mook to meed such a threshold. What do you all think about, instead, allowing the mook to re-roll their dice pool as in an Extended test. Obviously, the situation which allows them to roll the test in the first place must continue to be available to them as they roll again, but it might ratchet up tension levels as players know they can't give the opposition too much time to work on a problem or they'll likely figure it out.

For example, let's say Mr. Ninja is sneaking across the open area between a public road and a private office facility. Mr. Ninja rolls his Agility + Infiltration. Let's make him an elf with Exception Attribute: Agility (for a total of 8 agility) plus muscle replacement 2 (10 agility) and an Infiltration skill of 6 with a reflex recorder. That makes his dice pool 17. And, there's a variety of ways to get the DP higher, but those are reasonably easy to acquire and I'm sleepy so I don't feel like maxing it all the way. He rolls, and gets 6 hits.

The guards in the facility are less Ninja and more Mook. They have only a dice pool of 8 to see him. It's possible, though unlikely that they'll reach the threshold of 6 to do so. However, since Mr. Ninja has to cover not an insignificant amount of ground, the guards roll and get 3 hits. The interval of a minute goes by and they get to roll again and get another 3 hits. They spot Mr. Ninja.

Or, we could say that Mr. Ninja covers the ground in the minute of the extended test interval and, thus, the guards don't get their second roll and, thus, don't spot Mr. Ninja. The downside is that most Extended tests succeed if there's enough time in which to take them. Thus, it's it might be unlikely that the PCs ever get away with their actions, which concerns me. But, the flip side is that sometimes it seems like the PCs *always* get away with their actions, which also concerns me.

Anyway, there's the idea, I'm interested in reactions so fire away.

[kzt]

Making it some that you don't need any real skill to succeed at foiling someone who has invested hugely to be insanely expert doesn't seem like a good plan. As the key to successful infiltration is to move EXTREMELY slowly, this forcing people to run is kind of nuts. If the security system depends on idiots seeing someone sneaking across the ground, then yeah they are screwed. Life is hard. NVA & VC Sappers demonstrated how it's perfectly possible for someone to sneak into a heavily guarded location protected by layers of concertina and lots of alert armed guards if they know what they are doing, are extremely patient and have a good plan.

"This particular action involved an apparent mix of VC and NVA elements that destroyed 9 Chinook double-rotor heavy lift helicopters, damaged 3 more and blew up an ammo dump.[64] VC sappers by some reports led the assault, with NVA providing follow-on ground or fire-support attacks. . . . Penetration teams achieved almost complete surprise, with the sappers cutting 10 barbed wire fences, and advancing without being detected by sentries, obstacles or patrols."

[Aaron]

Arguably, every test could be an Extended Test. I mean, it doesn't have to be whether you hit that go-ganger Right Now, it could be a question of how long it takes you to hit him. It's kinda like how a roll could indicate the longest range at which your shot is accurate, rather than using the range to determine how hard it is to hit.

Algebra rocks.

[Method]

At first glance I like the idea, but like kzt said I'd be careful how you use it. Conceptually the whole point of an extended test is to speed up actions that occur over time that would otherwise require a monotonous stream of success tests. In that regard I don't think this really diverges from the core mechanics all that much.

I think the really tricky part is to get the intervals right, so Mr. Ninja doesn't feel like he's getting shafted. Maybe an alternate mechanic would be that if the Mook gets 4 hits on the initial test (the theoretical limit on a success test) his suspicions are aroused and he takes a second look (the extended test).

[ArkonC]

If you allow Mr. Guard this extended test, than Mr. Ninja should get it too...
So the only thing you did was limit the number of hits to 4 per round...

I might see merit to this method if it wasn't someone actively hiding, but, for example an object hidden...

[Hank]

I dunno...this seems like a good idea, but maybe not one you can put into hard rules. I agree with Method...you could really irritate a player who "heavily invested" into one trick to make his character survive.

On the other hand, if a character invests enough he can easily turn 6 successes or more on a stealth test, and how much fun is that game? If the characters are completely invulnerable, undetectable, I don't think it's a fun game, anyway.

[Cardul]

*hmms* I agree with Hank..It is not much fun to not be challenged. I also think that in Shadowrun, often times the game is faster paced then in RL. In RL, a sniper can take 2 hours to move a foot, and that gives him an incredible bonus to not being noticed. However, in Shadowrun, you are moving between buildings, or into patrolled buildings, things that seem like they would be faster paced. This idea I like because it gives the mooks a chance, and gives the team a reason not to loiter. However, there ARE ways to defeat it, if you think about it...

First: it is easy to get caught in one or two rolls if you just cross through an open area, even if you are careful with how you move. However, if you move from place of concealment to place of concealment, effectively having to re-roll infilitration each time, then you also force the guards to have to start over.

Second: Me, personally, I would say that if you make a perception test against the guards initial successes, you can notice that the guard is starting too look actively, and..can go immobile, and begin making an extended test against the guard..(I would limit the number of time a Guard can do the extended test in a row to his/her Perception Skill, at which point they conclude whatever triggered them was "nothing").

As Method pointed out, I would say that there is a certain threshold that needs to be met on the first rolll, representing the guards suspicions being arroused. I don't think it should, necessarily, be 4, though..I can see it more as a sliding scale based on perception skill. 4 hits is needed for peception 1&2, 3 hits for Perception 3&4, 2 hits for perception 5&6, 1 hit for Perception 7(and if you are going up against someone with Perception 7...heaven help you, because your GM certainly won't!)

[Ryu]

Any extended test does not provide a chance of failure. I´d be more in favour of a mechanic that gives the guards bonus dice after a certain time has passed. In combination with those situation were a stealth roll is not even permitted, it should still give a chance of detection. If it doesn´t, fine. Flush them out with radar and spirit patrols.

[Abbandon]

.......modifiers.........suck on my loud crunchy gravel walk ways, ultrasound emitters, and glow grass and my ten spy drones you stupid ninja!!!

Bob, "What was that?"
Fred, "What was what?"
Bobm "Turn your audio enhancers up you dumb ass!"

[Aaron]

Any extended test does not provide a chance of failure.

That's a good point. However, if you really wanted to make everything Extended Tests, you could run parallel tests, one for the action and one for the failure state, e.g. the sneaking runner and the vigilant guard (respectively).

I'm not advocating this, of course. But it's possible.

[Hank]

Any extended test does not provide a chance of failure.

This is only true on an infinite timescale.

[Larme]

For example, let's say Mr. Ninja is sneaking across the open area between a public road and a private office facility. Mr. Ninja rolls his Agility + Infiltration. Let's make him an elf with Exception Attribute: Agility (for a total of 8 agility) plus muscle replacement 2 (10 agility) and an Infiltration skill of 6 with a reflex recorder. That makes his dice pool 17. And, there's a variety of ways to get the DP higher, but those are reasonably easy to acquire and I'm sleepy so I don't feel like maxing it all the way. He rolls, and gets 6 hits.

The guards in the facility are less Ninja and more Mook. They have only a dice pool of 8 to see him. It's possible, though unlikely that they'll reach the threshold of 6 to do so. However, since Mr. Ninja

Nonono. Stealth is an opposed test. You're saying that Mr. Ninja has to roll once for stealth, and then his opponents get to roll as many times as they want, adding their successes each time? That would nerf stealth. You want to encourage players to use stealth and be smart. If you nerf inflitration, your players will just go in guns blazing every time and you'll have to deal with the escalation that entails.

Now, at the GM's discretion, I'd say the guards can make multiple tests to notice Mr. Ninja. Like if he's sneaking out there for 10 minutes, and they're watching carefully the whole time, give them another shot every few minutes. But if, say, they had 4 dice to notice Mr. Ninja, and he had 5 hits on his infiltration test, they are SoL. They are just not perceptive enough to notice him. That's how opposed tests work.

So you picked a bad example. You're talking about success tests, but you chose stealth, which is an opposed test.

For success tests, you shouldn't make a rule that anything over threshold 4 requires an extended test. Threshold 4 represents "extreme" difficulty - is that, like, as high as difficulty can get? What about "almost impossible" or "requiring a miracle?" The threshold for a success test can go as high as the GM wants it to. Success tests are, by definition, things that are accomplished all at once, they have no time interval. The thing that seperates success tests from extended tests is that one has an interval and the other doesn't. And, of course, because of their nature, extended test thresholds typically go a lot higher. But it isn't a matter of deciding whether something is too hard to be a success tests. Rather, you must decide whether something is doable no matter how long it takes (i.e. you could keep picking a standard lock for an hour and still succeed) or whether it's all or nothing (you have to jump four meters right now or you die). There isn't any interplay between the two.

[Blade]

I consider that Mr Ninja who's got such a high dice pool, and manages to get that many hits, isn't just someone very stealthy. He's superhumanely stealthy. I consider the 4 threshold to be "the max" for unaugmented characters. A treshold 4 test is very difficult yet possible for someone like this. But with adept powers or cyberware, all bets are off.

But it doesn't mean that the guards don't stand a chance. Security companies know that there are people able to pull off superhuman feats. That's why they'll also use tech and magic (in that case drones, spirits (or astral perception) and vision/audio enhancements) so that the guards will be able to do something (at least where it's worth the cost).

Still, Mr Ninja probably won't be noticed. But it's fair: if I designed my character to be the stealthiest ninja in the world, I'd be really disappointed if I got caught every time I tried to sneak around. But if I'm that good, I wouldn't go for low security places. A runner as stealthy as Mr Ninja will probably be sent to B&E places with extra-high security, where he'll face a challenge worthy of his skills.

[Dashifen]

So you picked a bad example. You're talking about success tests, but you chose stealth, which is an opposed test.

For success tests, you shouldn't make a rule that anything over threshold 4 requires an extended test. Threshold 4 represents "extreme" difficulty - is that, like, as high as difficulty can get? What about "almost impossible" or "requiring a miracle?" The threshold for a success test can go as high as the GM wants it to. Success tests are, by definition, things that are accomplished all at once, they have no time interval. The thing that seperates success tests from extended tests is that one has an interval and the other doesn't. And, of course, because of their nature, extended test thresholds typically go a lot higher. But it isn't a matter of deciding whether something is too hard to be a success tests. Rather, you must decide whether something is doable no matter how long it takes (i.e. you could keep picking a standard lock for an hour and still succeed) or whether it's all or nothing (you have to jump four meters right now or you die). There isn't any interplay between the two.

No, I picked a very specific example intentionally. The only test in the game, IMO, is the Success test. Some have an arbitrary threshold between 1 and 4 (which the book refers to as a Success test) others have a threshold set by a dice pool representing someone else's skill (which the book refers to as opposed tests) and others have even higher threshold arbitrarily set between 4 and 16+ and take longer to complete (which the book refers to as extended tests).

Regardless, all of these rolls come down to the same thing: you roll dice and beat a threshold. The problem I've had in SR4 for a few years running now is that outside of combat, the thresholds in opposed tests -- that is thresholds set by rolling dice representing one character's facility in a skill -- frequently end up higher than 4, which the book lists as extremely difficult to impossible.

Thus, I'm faced with a problem in that the book indicates that a threshold of 4 is the highest that any character can expect to accomplish quickly. But, I see 5+ thresholds often at the table from mooks to prime runners to 50+ karma PCs. I've frequently use the Astral Patrol rules from Street Magic and found that my players have to hit a threshold of 7-10 to not be found out by spirits. That makes the game not fun for anyone because Astral reconnaissance is out. Now with some of the Sensor software available in Arsenal, even meat-body reconnaissance might be more problematic.

It just feels to me like the old open-test problem of SR3 (where someone's target number was the highest roll of someone else's skill) seems to creep its way back into SR4. I suppose the answer might just be Edge, but I hate having mooks use Edge.

[Ryu]

No, I picked a very specific example intentionally. The only test in the game, IMO, is the Success test. Some have an arbitrary threshold between 1 and 4 (which the book refers to as a Success test) others have a threshold set by a dice pool representing someone else's skill (which the book refers to as opposed tests) and others have even higher threshold arbitrarily set between 4 and 16+ and take longer to complete (which the book refers to as extended tests).

Regardless, all of these rolls come down to the same thing: you roll dice and beat a threshold. The problem I've had in SR4 for a few years running now is that outside of combat, the thresholds in opposed tests -- that is thresholds set by rolling dice representing one character's facility in a skill -- frequently end up higher than 4, which the book lists as extremely difficult to impossible.

Thus, I'm faced with a problem in that the book indicates that a threshold of 4 is the highest that any character can expect to accomplish quickly. But, I see 5+ thresholds often at the table from mooks to prime runners to 50+ karma PCs. I've frequently use the Astral Patrol rules from Street Magic and found that my players have to hit a threshold of 7-10 to not be found out by spirits. That makes the game not fun for anyone because Astral reconnaissance is out. Now with some of the Sensor software available in Arsenal, even meat-body reconnaissance might be more problematic.

It just feels to me like the old open-test problem of SR3 (where someone's target number was the highest roll of someone else's skill) seems to creep its way back into SR4. I suppose the answer might just be Edge, but I hate having mooks use Edge.

Sometimes the "threshold" on opposed tests is 1, sometimes it is 9. That is just a veil. The key is that the probability of success depends on the relative DP sizes, and that the variance of results still decreases with absolute DP size. The mook will rather constantly loose, even if the player rolls less dice. Increase the mooks Perception Pool, Audio Enhancers 3 cost 300¥. You should by all means assign DP mods to carried gear and environment. The most important trick is having the players re-roll stealth occasionally (for different sections of the facility), and having the guards reroll in certain intervalls. The guards will roll well once upon a while, and one of the players is bound to have a mediocre roll.



Astral Patrol: May I ask what kind of spirit you send on patrol? Or better, how many? Half a dozen force 4 spirits have a base DP size of 13 for teamwork tests, and are quite lucky if the sum of DP mods is not negative.

[Larme]

You're saying that the players are too powerful, so you should alter the game system to make it harder for them. If they're down with that, then go for it. But I don't think it's good GMing. There are ways to increase difficulty without monkeying with the dice system. And remember that altering the system boosts the NPCs as much as the PCs. Though you're trying to give the NPCs an edge over the players, the players can and will use it against you at another time, so you won't have changed anything. Try adding some security systems. Give your PCs situations that they can't succeed just by throwing dice at them. Creativity is the answer.

[Dashifen]

I do think my players are challenged, that's not the issue. I think the issue is that I don't feel challenged. I feel like I can sit back, relax, let the players come up with a plan, throw some dice, and then see what happens. Due to the nature of the system, it seems to me like I can very likely predict what's going to happen.

Probably doesn't help that I glitch 5+ times in a four hour session on average even with 10+ dice pools. And I switch dice so I'm just cursed.

[Larme]

That's the sad part about being god, you always know what will happen before it happens, because you control all the variables... But I don't see how your proposed system change will make the system less predictable. I guess there's a new dynamic - the NPCs can always theoretically succeed on an opposed test, but you don't know exactly when they will. Except statistically, that's just as predictable as any single throw of the dice.

I still think the answer lies in giving your NPCs the advantages that come with higher resources. They're not limited to just having mooks with low dice pools. They can have thermal sensors that essentially cannot be spoofed. Or laser motion detectors that are completely foolproof. They can have scary underground facilities, or undersea bases, or mile high towers. And trained critters who can smell the characters, even if they're functionally invisible. If you're bored with how your characters don't even give mooks a chance, throw in something other than mooks. Converting opposed tests into extended tests is just a minor tweak that doesn't seem to get you what you want.

[Dashifen]

Larme ... I think you're missing it when I say that this isn't just for me. It's for PCs and NPCs alike and it's not that my NPCs are all mooks, it's that at any level of the game, it is possible to create a threshold for a test that is greater than 4, which is what I feel the book indicates as the highest threshold for actions that are completed instantaneously (anything that takes longer is an extended test). Thus, for such tests, I considered turning the success test into an extended test. It has nothing to do with NPCs that can't act in a tactical way or PCs that are overpowered with super dice pools. It has everything to do with the fact that the by-the-book idea of a threshold seems to fall apart, IMO, for opposed tests outside of combat.

I just realized, though, that it could be because of the way I roll opposed tests so, since that's a vastly different concept than I was trying to get at here, I started a new thread.

[deek]

I'm going to check in at the other thread, because I don't have an issue with the threshold of 4 being the highest threshold. Granted, I do enforce skill caps on all tests, so maybe that is why I don't have much of a problem with the hard cap of 4.

[Fortune]

There are Thresholds over 4 listed in the SR4 Core book.

Assensing Table
5+ Any other implants.
The general cause of any emotional impression (a murder, a riot, a religious ceremony, and so on).
The general cause of any astral signature (combat spell, hearth spirit, and so on).
The fact that a subject is a technomancer.

I don't really see the need to alter tests with larger Thresholds into Extended tests.

[Larme]

Larme ... I think you're missing it when I say that this isn't just for me. It's for PCs and NPCs alike and it's not that my NPCs are all mooks, it's that at any level of the game, it is possible to create a threshold for a test that is greater than 4, which is what I feel the book indicates as the highest threshold for actions that are completed instantaneously (anything that takes longer is an extended test). Thus, for such tests, I considered turning the success test into an extended test. It has nothing to do with NPCs that can't act in a tactical way or PCs that are overpowered with super dice pools. It has everything to do with the fact that the by-the-book idea of a threshold seems to fall apart, IMO, for opposed tests outside of combat.

I just realized, though, that it could be because of the way I roll opposed tests so, since that's a vastly different concept than I was trying to get at here, I started a new thread.

It seems to me that you're giving entirely too much weight to that silly table. 4 might be 'extreme' difficulty. But that's clearly not as high as it goes. Like if you want to do a standing jump of 10 meters, you'll need to hit a threshold of 10 on a success test. Does that table forbid you from trying to jump 10 meters because 10 > 4? No, it doesn't. The table, as far as I can tell, is only relevant to arbitrary difficulty levels set by the GM. If the player says "I want to do a gymnastics trick, how hard is it?" the GM has to think about it, decide what difficulty level, 1 through 4, to assign, and then give that as a threshold. The table does not forbid thresholds higher than 4, as shown by the jumping rules. The only thing it does is give some general guidance for a GM to make a quick decision about how hard something is. If 4 were the highest difficulty allowed by the RAW, you'd also expect the RAW to limit all dice pools to 12 or so. But the fact that players can fairly easily blow a dice pool of 12 completely out of the water, and hit thresholds of 4 in their sleep, shows that the table is really not meant to control all aspects of SR4's dice system.

[nathanross]

For example, let's say Mr. Ninja is sneaking across the open area between a public road and a private office facility. Mr. Ninja rolls his Agility + Infiltration. Let's make him an elf with Exception Attribute: Agility (for a total of 8 agility) plus muscle replacement 2 (10 agility) and an Infiltration skill of 6 with a reflex recorder. That makes his dice pool 17. And, there's a variety of ways to get the DP higher, but those are reasonably easy to acquire and I'm sleepy so I don't feel like maxing it all the way. He rolls, and gets 6 hits.

First off, I really disagree with your idea of a Perception extended test. You are saying that a player made a character to do something well (avoid getting spotted by guards) and you are going to pull the rug out from under them by allow said guards a way to spot him given enough time.

That is bullshit.

You are completely forgetting modifiers. Why on hell is this moron walking across the street well in view of the guards. Is the street well lit? Do you let your players Infiltrate through the front gate? You need to provide modifiers based on the situation. Not to mention that Perception has some of the cheapest DP inflators in the game. Use them!

I do think my players are challenged, that's not the issue. I think the issue is that I don't feel challenged. I feel like I can sit back, relax, let the players come up with a plan, throw some dice, and then see what happens. Due to the nature of the system, it seems to me like I can very likely predict what's going to happen.

Probably doesn't help that I glitch 5+ times in a four hour session on average even with 10+ dice pools. And I switch dice so I'm just cursed.

Can't help you with bad luck, however, boring games can easily be turned around. First, change things up. Guards in my world don't carry guns, they carry radios and biomonitors/doc-wagon bracelets. Why would I want my grunts damaging my facility? Maybe tasers/hold-outs with stick-n-shock but nothing more for sure.

As a runner, I am always more scared of the guys that are on call that the facility personnel. I say five to ten minutes and in comes the black ops to mop up the problem. Also, try letting them get away with a few easy runs to relax their guard and then throwing something in their face. Think deep into things, and don't let your dudes be target practice. They are characters too, with lives that they don't want to give up. Play them that way. Use edge!

[kzt]

You as a GM are going to be able to predict how smart players are usally going to approach a problem. Because they will choose the approach that offers the best chance of success. And after you as a player have played a few times with the group and the GM you figure this out. There are only three basic approaches that can be used to get into a secure site: Blitz, Stealth and false flag.

This is also what the expensive consulting companies that get hired to design security architecture/systems understand and design against. And they have lots more time than the PCs, very high skill levels in both designing threats and designing security systems against them, full plans, custom designed software to analyze the site and the ability to redesign the landscape to help them out. Depending on the constraints (like "It has to look good", or "we can only spend x", "The threat model doesn't include use of mortars and ATGMs") and the skill of the designer it will be better or worse at standing up to the attack.

[Method]

nathanross: I disagree. If a guard has an extended length of time to possibly observe an infiltrating PC the probability of the guard detecting the PC *should* increase in proportion to the time. You could say that getting multiple chances to pass a opposed success test gives him a greater probability of success, and you'd be right. Dashifen's mechanic just increase the probability *much faster* in proportion to the time because successes carry over, making the guard continually more likely to succeed.

As someone mentioned above, yes there are snipers that infiltrate enemy teritory by moving inches per hour but the SR game mechanics aren't very well suited to that kind of situation anymore than the combat system is suited to countering sniper tactics (i.e.- if you are surprised by a sniper you're pretty much dead). Neither of those scenarios are particularly fun to play out IMHO. And the core mechanic in SR for things that take place on long time scales is.... the extended test!! (IMG:style_emoticons/default/eek.gif) (IMG:style_emoticons/default/wink.gif)

Anyway, I think that the larger problem that Dashifen is alluding to (please correct me if I'm wrong) is that all too often PCs can come up with a plan and execute it without any complications. Why is that bad you ask? Because IMHO a game without complications gets dull. The whole point of rolling dice is to inject the possibility of failure. The alternative is to come up with ways that force complications on the players, but then you are treading dangerously close to the slippery slope of railroading.

So the question becomes, *if a GM and his players desire more complications* (and there's nothing wrong with that) what can they do to increase dice roll failures without railroading? Situational modifiers are a good start. Smaller dice pools work. The optional rules for grittier game play and limited edge are good too. I think Dashifen's mechanic could be useful in some situations as well. It all depends on how you want your game to work.