Sure, sure.

How about fumbling a dribble? You never once had he ball bounce in a way you hadn't expected, or put slightly the wrong spin on it? You've never tried to steal the ball from someone, only to see it go out-of-bounds with you having been the last to touch it? You've never tried to recover from having the ball stripped from YOU, only to see it go out of bounds?

Because if you haven't had little things like that happen to you, in the course of fifteen years? Why hasn't the NBA recruited you, yet? O_O Or your country's Olympic team, for that matter.



No, trying to throw a kick and slipping on the mat, is not simply a failed roll.


And I'm arguing that you're discounting too many things that should be "glitches", as being of no consequence at all.

mmm

There's something I fail to get. In so far, I've read there that a glitch was a complication occuring in your task. I've fumbled dribbles, airballed, made passes where I knew while making it that they would be intercepted.

But to me, those are not complications. I mean they are just bigger failures.



We've got a difference in how we interprete a "glitch".

The fumbbled pass, the kick that ends in falling on the mat just fits in a system like : "roll dice pool, if more ones than successes, you have a glitch" as they are skill related.

If the glitch is a complication unrelated to skill level (like roll 2d6, double 1=glitch) and happens often (maybe as a story-enhancer), I'm not for it, I handle them better off-skill roll.

See, I consider a 'glitch' to be a complication.
So, failing a pass or getting an airball might be a failure. But slipping on the court, getting a sprain, or passing right to an opponent (interception!) could be a glitch. It isn't that 'you failed', it's that your life just took a step back, making success that much harder. A critical glitch, of course, would be a 'you failed and it wound up horrible for you'. You pass the ball, someone intercepts it and gets the point. Or, in hockey, you scored on your own net (seen it happen when Wayne Gretzky accidentally scored on his own net during the Stanley Cup playoffs some time ago. Wasn't that something... he cost his team the cup).

In MMA, whiffing on a kick? That's a failure. Landing on your ass because of it? That's a glitch... you're now prone, and your opponent has a significant advantage. If you kick your opponent, and shatter your foot? That's a critical glitch. Now, does anyone want to sit down and math how often this happens? Probably not - but you can check medical reports for injuries ... training injuries in MMA are really... significant. Injuries gathered in a game of hockey or football? I'd say a number of those are from glitches coming up over the course of a game. These games aren't meant to cause injuries - but they're contact sports, and injuries happen. (Except MMA, where injuries are the intent...)

I'd say that most of these injuries are glitches - in hockey / football, I'd say that most of the damage is 'stun' damage, but a glitch shifts the injury to 'wound'. And in a hockey game, or even during training, you'll see these glitches come up... even once a game. In football, there's a lot more involved, so there's a higher chance of sucking back a glitch, thus getting a few wound boxes filled. And in MMA...

And definitely, a glitch doesn't have to be horrible. In a gunfight, you get the shot off and your weapon jams, requiring one IP to unjam it (glitch), or you wind up shooting through the fruit stand, giving the guy an extra 2 points of soak (glitch), or you have to hesitate because just as you were about to pull the trigger, a group of civilians comes out of the alley in a panic, trying to get out of the way and completely spoiling your shot (critical glitch) - or worse, one got IN the way, and died (critical glitch).

Glitches make for a good narrative tool, and don't always have to be 'hose the player'. A slip in the MMA ring isn't going to necessarily screw the entire fight, and having to spend an IP fixing your gun isn't necessarily going to screw the entire firefight. It gives the player or the game master a nice twist to throw into the scene, without having to upend the entire scene as a whole.

Hell, I had a player with gremlins completely screw his commlink. Rather than having it destroyed, I had the language filter switch to the ancient dwarvish language. Once the PCs figured out that it was an actual language, they were able to sell it to the Atlantian Foundation. The dwarvish language is slowly showing up in my Shadowrun campaign now.

Now, the second time, I had the perfectly-sealed custom-built commlink's internal components leak blue ooze out the sides. This made the hacker who had designed it for him perplexed. "How did it even DO that?!"

Sorry to derail the glitch discussion for a minute...



Hm, might give some better granularity for the description of proficiency levels, but I dont like the increased number of dice and the different scale between attributes and skills - i had some other house rules in place, to make sure skills had more importance in a roll then attributes. i.e. I limited the maximum number of rolls in extended tests to skill*2 (1 roll only for unskilled) and glitches where based on skill instead of pool/2 (see below).



I used that for a while and it worked ok, we only had a minimum ammount of hacking though.



So they peaked into my houserules
I dont think its a far way once you have contemplated the matrix mechanics. I used it for most rolls (vehicle, matrix, combat, magic, healing (limit based on lifestyle)) and it worked quiet well. I only used soft limits thought (i beleave there was an optional rule suggestion for spellcasting like that) in that hits above the limit would only count 1/2. Unarmed combat (except for ppl w martial arts) was 0, i.e., I like armed combat to be inherently more dangerous then unarmed. I also used it for damage resistance (I roll body+strength, limited by armor (so w.o. armor all your hits count half)*. This worked really well for me, allthough my players needed getting used to it.

*armor would still change phys to stun normaly (actually all physical attacs also do the same amount of stun in my game, getting rid of the whole when you are tired its better not to use your armor nonsense), but it would not give extra dice to resist.

back on topic... glitches and edge

For the record, I love the edge mechanic. Its way better then karma pool and also i interpret it as not pure luck, but also how much a character can push him/her/it-self if just a bit more is needed - and how often. So I think its also cool thematically. The uncybered mundane can spend the points safed on edge, which fits because you need to be abled to push yourself more, if you dont have any other boosts.

With the soft limits I used edge was not necessary as a limit break though. I think refreashing compleatly after a run/w a lot of downtime and sometimes a single point for big acheavements or some good rest or an exceptional & called trickshot (in other words wasting hits and risking a miss just for dramatic effect - but only if the risk is high enough relative to the dicepool, importance and dificulty) or for suffering a crit glitch w/o edging out.

As stated above, I used skill not pool/2 to decide glitch or no glitch. I thought thats cool, because a person can be immensely capable (high attribute/bonus dice) but if the person is not highly skilled at the same time, the chance to break your knuckles while breaking s.o.'s face get higher. with unskilled rolls, 2s also count for glitches in my game (but i do not reduce the pool by 1 for defaulting), so when defaulting glitches get quiet likely.

Now normal glitches (as opposed to crictical once) also are not that tough in my games. might have to unjam for i action, or get off the ground, or pick up your knife, or make some unintended noise (for infiltration that would be more a critical glitch), etc.. If nothing else fits, I would give 'stress damage' to gear or attributes (usually 1d3, for every 3 points the gear or attribute would suffer a -1 dice pool mod - that might be 'stun'/temporary for regular glitches, which is easily fixed or healed quickly, or permanent/'physical' for critical glitches, which would need some money and/or time). This worked really fast and well. Even mental attributes could accumulate this, which dosnt necessarily reflect braindamage - could just be confusion or distraction (intuition), some missremembered facts (logic), bad mood or spinach between your teeth (charisma) or plain tiredness or demotivation (willpower).

The glitches were farely freequent in situations w many rolls, like combat, but since the effects were minor and npcs would glitch just as much, it made for interesting twists and turns.

No, but neither do they occur at a fixed percentage, regardless of skill. Saying that the best-skilled, best-equipped person has the same chance of getting it wrong as an untrained, unprepared amateur is more than a little silly.

The odds of a complication need to scale with skill levels. A flat one-in-six chance of something going wrong, regardless of how well you've prepared or how skilled you are, just doesn't work.

Cain; but the current system doesn't work well either. Once you go past 10 dice the chance practically disappears to the point even a normal glitch becomes a freak occurrence, let alone a critical glitch... someone else got it dead right I think... it was designed for SR3 where the dice rolled rarely entered the teens.

It's the nature of the beast... how high should the glitch chance be on small pool? How much should it diminish as the pool gets larger and larger? My own take is that it shouldn't fall below 1% and not really go higher than say 10%. The hard part is coming up with some scheme which isn't overly complicated based on the d6 mechanic.

The best idea I could come with was was flawed by that at the low end, though it did scale nicely.

At what point do you just throw up your hands and say a glitch is when you roll a 1 on a d20 (5%).. or snakeeyes on two separate 2d6 (2.8%) which at least keeps with SR's d6 motif.

I've tried statistically:

more ones than successes=>numbers too high, see previous posts.
more one than half the die pool=>drops under 1% over 16 dices.
more ones then attribute would be inconcistant as you had more chances to get a glitch as your skill improves.

The only possibility left is to pick a colored die (like Star Wars d6) per 5 or 6 dice pool for exemple and if all those shows ones, it's a glitch.

kinda inelegant but it's the only statiscal solution I see that doesn't requiere a throw of additionnal dices to your dice pool.


(that is, I can live with a low glitch probability system)

What do you think about ..

I have no time to calculate the propabilities right now, but at least it fits w the notion, that a skilled person is less likely to glitch... the only odd thing is, that as you get negative dice mods, its also getting less likely to glitch. Mayb be less likely with less dice mods in SR5. You could even subtract dice mods from glitch threshold, as well (may be feasable w higher skill range and less dice mods in SR5). That way, you have a system thats less likely to glitch for with increased skill and very skilled ppl are unlikely to glitch at all, unless there are adverse conditions, which i think makes sense, too.

Well but in this case, the higher the attribute, the higher your chance to glitch. Ofc it's feasible, though somewhat weird that skill 3+ atrribute 3 has less chance to see a glitch occuring than a skill 3 attribute 6 character.

That is, you could argue that the raw potential of the S3/A6 character makes him succeed more often, but his success are less refined (hence with glitch)?



rough stats of your proposal

I'm not quite sure, what the table is supposed to signify, but I think your math is off when comparing a 3+3 roll to a 3+6 roll.
Uneven numbers have a lower chance to glitch than even numbers, but even so, the glitch percentage would be lower on a 2+3 roll than on a 3+3 roll, and even less on a 3+6 roll.

horizontaly: Attribute
Vertically: Skill

I am assuming, you're calculating the regular glitch chance, not the critical, then?

Edit: And in the system you proposed earlier?

I've used a random function with 10 000 trials for each value. Also I've got a stat diploma, it's easier that way (so it may differ but nothing over 0,3-0,5% I guess)

and yes, it's just a regular glitch into the table.

Well, I'm hoping that you know with your diploma that there is a lot of rounding imprecision going on there, with just one point after the comma
Anyways, I was under the false impression that this would be the calculation for the regular glitch mechanic from the books, so disregard my comments as a mere question for clarification

For my info, critical glitch occurs when you don't have the requiered success + glitch occurs, right?

If you need 3 successes, that would be the table of a critical glitch:



If you need 2 successes:

No, a critical glitch occurs if you have NO successes (at all) AND more than half of your pooldice come up as 1s.
A friend of mine wrote a script recently for the regular mechanic, but that goes deep into mathhammer territory

Which one?

I have no idea anymore at this moment
A lot of systems have been thrown around in the last posts and the only thing I am sure of is that it can't be the regular mechanic.

I hope.

ok so no succes at all with number of ones = skill dice.



no success at all is kinda harsh. It happens very seldomly.

Ah, thanks for the clarification


Yes, which is why I am perfectly okay with the current glitch system *g*

As a mathematician and programmer, I salute the Monte Carlo simulation.

I don't agree with Monte Carlo. I prefer to just outright calculate exact odds and distributions.

It's not that hard to do in these cases.

Also... 10,000 sounds like a lot, but it isn't a sufficient population size when your actual population size of results is in the trillions. (16 dice, 2.82 E12 possibilities of merely calculating the combinatorial distributions of 0-16 1's. 1:5 odds weighted coinflips).

What's that old saying about formulas and reader count again?

Not quite. There are only 245157 distinct combinations of 16 dice. The 2.82E12 number includes all permutations -- that is, where 1 2 3 is counted separately from 3 2 1. We don't really care about the functional duplicates here, since dice order isn't important.

EDIT: Oh, your general formula for n dice is (7+n)!/(7! n!)

The hell does the exclamation mark mean?

Factorial.

3! = 3*2*1
7! = 7*6*5*4*3*2*1

And so on.

EDIT: And 0! = 1, strangely enough.

Not really. A factorial is a set of numbers which are multiplied. Specifically n! = Π(k=1 -> n).

Π = product the same way Σ = sum.

So 0! follows the rules of the Empty Product.

Your math is wrong...... I can instantly prove you wrong as well.
n=1 in your formula. Produces a value of 8. Meaning 1 d6 can roll to 8 different results.

The actual number of combinations without respect to order of 16 d6 dice is in nCr notation... 21 C 5 == 20349.

But this number is meaningless to us as this only tells us how many total distinct combination bins there are to count results into. It does not tell us how many results fall into each bin or their probability density. So if you roll toss out 10,000 results at random... you fill less than half the 'bins' even if no bin gets more than 1 result (more likely some bins get multiple balls, and others get none like some monstrous 'plinko' machine).


And monte carlo won't tell you the odds of rolling 16 1's.... it can't give you even a close guess on it until you've rolled a good order of magnitude over that 3 trillion figure I just gave you. At which point it may have dropped 1 ball into the 16'1s bin, or 20... depending on how fickle lady luck is feeling.

For the record the odds of rolling 16 1's on 16 dice are exactly
16: 1 in (6^16)!~= 2.82111... trillion
15: 80 in
14: 3000 in
13: 70000 in...
....
0 1's: 1.52588E11 out of this. Roughly 5.4% of the results have no 1's at all.

At the end of the day each of the sub categories summed together must result in precisely 6^16 power to cover 100% of the probability distribution.


That's why I said the monte carlo numbers aren't reliable. You're not doing it enough to invoke the law of large
numbers. You're not even doing it enough to cover all your 'without respect to order' bins.

Or more descriptively, n! is the number of different sequences which can be produced from a set containing n different elements. How many (distinguishable) ways are there to arrange nothing? One obviously

Hmm.. you're right, but what is it missing? Ah well, I was going from memory, which is painfully unreliable. So on 1 die you've got 7 ticks and 1 star:

| | | | | | | *

Ah, now I see, the 8 was coming from allowing the star outside the ticks, you want to go by the whitespace (bins). This would rather be written as...

(5 + n)! / (5! n!)

which for 16 is 20349, and for 1 is 6.



It's not meaningless, but there are a few trillion fewer bins than you made it seem. And Monte Carlo will get you the odds, but you'll have to increase the sample size by a lot -- Monte Carlo is useful when computational cycles are cheaper than actual figuring. In this instance you still get the practical answer: the odds of rolling 1s on all 16 dice is so low that it may as well be counted as zero. We're not launching a rocket to Alpha Centauri, after all.

I don't actually think it's strange, but it certainly doesn't follow by the short (read: awful) way I presented it.

Since you seem lost the general form for rolling any arbitrary number of dX's is.
n= number of sides, r=number of dice
(n+r-1)!/(r!(n-1)!)


But no you completely missed the point. People have posted that the 'glitch' rate going up to 16 dice is about 1% based on flawed monte carlo assumptions. Because they didn't do the basics to determine the 'confidence'... generally you need to have far more than the expected rate of occurance for the law of large numbers to start working. Whoever did the monte carlo did not calculate the confidence interval and realize that his figures were wholly innaccurate.

https://docs.google.com/viewer?a=v&q=ca...MTo4KL0U7-IR5Dg

Good example of craps... to get to 0.1% confidence interval on a nearly 50:50 event... nearly 1million!!! results need to be run. On an event with a far lower expectation rate (below 1%) far more than even that need to be run. Note the line graph on page 2 of the same program run 100x in a row and notice how often a single run 'spikes' outside the expected error showing as high as 70% win odds or as low as 30%.


How do I know this because I know the *EXACT* non-random guestimation is ~0.21% (6.06E9/2.82E12). Because I can enumerate all these without enumerating all them, simply by enumerating 16 coinflips probability distribution and then assigning weight to each side of the coin (a coin 5x more likely to flip heads than tails) to calculate precise odds.

Even at 11 dice... the odds of 6 or more 1's is a paltry 0.46%


So anyone claiming 1% on 16 dice is off by a factor of *5*. That would get them an 'F' in most prob and stat courses. Monte carlo would generally not be used for this at all because an exact mathematical derivation is easily obtained.

Mmmno.



Mmmalsono. Since the percentage drops below 1% somewhere around the 10 dice mark, as I've stated previously, it's safe to assume that at the 16 dice mark it's far less. It's great that you calculated the actual percentages, but there's no reason to get increasingly anal about it.



Easy, but a little tedious. Since this isn't a prob and stat course, you probably shouldn't fault a programmer for coming up with a programming solution, especially one you can write in a minute and run a few hundred thousand iterations of in the time it takes to make a cup of coffee.

Exactely.

I mean, I'm ok to spend 5 mins for a result I find accurate enough.

Now I've could have calculate the exact numbers but it would have taken me time and really, really, it doesn't bring a lot to the thread.

By the way, this is misleading:



Where the hell would I need a 0,1% Interval?


You (Falconer) come here with a very good mathemacial point of view (nice) but a very narrowed mental state (EDIT: badly written, too mathematical straight, too cartesian, too scholar. I don't know how to explain rightly what I mean. In french: "trop droit"). To me, if my 16,3% glitch chance through 10 000 trials is enough to be at 1% interval confidence at a probability à 80% (standartly, most tests use 95% but I don't need that accurracy just to discuss of an RPG glitch-system proposal). -ps: still not wanting to go in exact calculations-.


Anyway, just calculate them. I didn't feel like doing it. Just do it.

To me it was enough to find out that glitches just falls off with high pools and under which dice ranges they're still likely to occur.



ps:
(for non math users, in the case of craps: you flip a coin and you tell yourself: I'll look how many times I'll win. How many times should I flip the coin to have my calculated value at 0,1% difference from the real win probability (with a likeliness of 95%). That's 1 million.

But there's two factors you can play with: the interval-the confidence probability.

I made a test on computer:



My point is:
5056/10000=50,56% is enough for me to get the feeling of a rule.

But I welcome the calculation of the real values.

Which is a) overkill for the purposes of a discussion like this and b) generating one million random numbers plus checking how many of them are successes still happens while fetching coffee, even without CUDA tricks.

Sk8 and Sengir make good points. For the purposes of idly discussing mechanic, "back of the napkin" accuracy is really all we need in order to get a broad feel for the mechanic in question. Most people playing the game around their dining room table aren't going to be more precise than that, in deciding whether or not they like a system, after all.

There comes a time where too much accuracy (or rather, the effort needed to achieve it) is actually more detrimental than not enough of it.

As an example, my simulation of hacking probabilities used 10k samples for each dice pool. The resulting graphs show some deviations, but they are easily recognized by just looking at them and do not diminish the information value of the graphs one bit. Burning more CPU cycles on smoother curves would effectively just be eye-candy.

Also, the simulation code was written on the back of a letter before going to bed, whereas I spend a good week trying to figure out a closed formula for it.