Spectacular Failure is broken

[mvdwege]

Ok, I've read and reread the T5 book, and even started prepping my first adventure to run with it. To start with: I like it. The overall playing system is simple and elegant; I am not wedded to a roll-over system as I grew up playing ASL, so as opposed to some on these boards, I find a roll-under system fairly normal.

Character creation has a few minor warts, and is not something I would give to a completely uninitiated player, but then neither would such a player be able to easily build a character in systems of comparable complexity; I DM a D&D3.5 campaign, and I have at least two players who cannot maintain their own characters' experience progression without help from the rest of the table.

That being said, I have found at least one mechanic which is broken in its very design: the Spectacular Failure rule. The simpler formulation: roll three 1s on any task requiring three or more dice is a spectacular failure.

For an author that has spent so much time on getting the statistical underpinnings right, I am rather shocked that Marc Miller made the same mistake as Mark Rein*hagen did in Vampire: The Masquerade: the higher your proficiency at a task, the higher the chance that you fail spectacularly.

I cannot think of any justification for such a completely broken concept. The warts in T5 such as the messy organisation I can live with, but my first houserule will be to discard this particular rule without prejudice.

 

[cuTTer]

For an author that has spent so much time on getting the statistical underpinnings right, I am rather shocked that Marc Miller made the same mistake as Mark Rein*hagen did in Vampire: The Masquerade: the higher your proficiency at a task, the higher the chance that you fail spectacularly.

I cannot think of any justification for such a completely broken concept. The warts in T5 such as the messy organisation I can live with, but my first houserule will be to discard this particular rule without prejudice.

Can you give an example? Surely the chance only increases as dice are added, and dice are only added with difficulty...

To put it another way, if my stat is 5 and my skill is 1 and a task requires I roll 3d6, boosting my stat to 7 and my skill to 7 doesn't make a Spectacular Failure any more likely on the same 3d6.

Does it?

 

[mvdwege]

The difficulty increase gives you more dice to roll. The higher skill level gives you the opportunity to roll for the task.

Statistically, even with climbing difficulty, you are more likely to succeed the higher your Assets; however, with that higher probability to succeed also comes a higher probability of Spectacular Failure.

I fail to see the justification in your argument. The basic statistics still hold up: the higher the amount of dice you roll, the higher the chances of spectacular failure. And worse, the probability of total failure is completely decoupled from your skill level. If you're a highly trained surgeon in a modern operating theatre, you have a higher chance of killing your patient than a soldier attempting to do a hasty battlefield stabilisation of his injured comrade prior to having him moved out by the medics.

Like the 'botch' mechanic in the original Storyteller, it is objectively, measurably, broken.

 

[McPerth]

For an author that has spent so much time on getting the statistical underpinnings right, I am rather shocked that Marc Miller made the same mistake as Mark Rein*hagen did in Vampire: The Masquerade: the higher your proficiency at a task, the higher the chance that you fail spectacularly.

It was time ago I played Vampire, but, IIRC, there you rolled as many dice (ten sided) as stat + skill you had, each against a target number. Each 1 was a botch, that nullified one success, and if more botches than successes were achieved, it was a fumble.

If you rolled 4 dice against a difficulty of 7, you were more likely to roll a 7+ tan a 1, so a botch was posible, but unlikely. If you rolled 10 dice, at average you'll roll 4 successes and one botch, giving you a 3 successes result and a botch very unlikely.

If your target is 10, at average you'll roll as many successes as botches.

So, I disagree with you as possibility to roll a fumble was higher the more dice you rolled, as, except at target 10, even if you were more likely to roll a botch, you were even more likely to roll more successes, so nullifying the botch and even giving you more successes on the average.

About T5, I have not read it, but, for what I've read in this thread (an I guess DonM can be trusted), the possibility to roll a spectacular failure is dice dependent, not target number dependent, and so it's difficulty dependent, not stat/skill dependent.

Pecisely, the problem is the opposite that you say: spectacular success is imposible for easy or average tasks, very difficult with difficult task and the possibility increases at each die (difficulty level) you add.

EDIT: And I don't know if tha game specifies what happens if you roll tree ones and tree sixes on a hopeless or above task roll :devil:. END EDIT

Of course I might be whorng in what I've undertood from what I've read here...

 

[mvdwege]

It was time ago I played Vampire, but, IIRC, there you rolled as many dice (ten sided) as stat + skill you had, each against a target number. Each 1 was a botch, that nullified one success, and if more botches than successes were achieved, it was a fumble.

If you rolled 4 dice against a difficulty of 7, you were more likely to roll a 7+ tan a 1, so a botch was posible, but unlikely. If you rolled 10 dice, at average you'll roll 4 successes and one botch, giving you a 3 successes result and a botch very unlikely.

If your target is 10, at average you'll roll as many successes as botches.

So, I disagree with you as possibility to roll a fumble was higher the more dice you rolled,

I still disagree. It is simple, fundamental, probability mechanics: the larger your dice pool, the higher the average number of 1's rolled. Even if you set a low difficulty (in Storyteller terms), the chances of rolling enough ones to cancel out your successes increase with the number of dice rolled. In other words, the better you are, the higher your chance of failure.

T5 does it slightly better in that a larger dice pool means that the task is more difficult, so the relation to skill level is indirect (only a high skill level would attempt a more difficult roll), but the basic problem remains: if you have the skill levels and the Mods to make a large dice pool irrelevant to your chance of normal failure, you still increase your chance of Spectacular Failure, which is in my view complete nonsense.

 

[whulorigan]

That being said, I have found at least one mechanic which is broken in its very design: the Spectacular Failure rule. The simpler formulation: roll three 1s on any task requiring three or more dice is a spectacular failure.

For others who may be reading this thread, the Spectacuar Results Rules are on p.136-137 of the T5 Core Rules. That being said, the following are the possible Spectacular Results:

1) Spectacular Success Three Ones. If the actual dice roll includes 3 ones (but not possible on 1D or 2D) the result is a Spectacular Success (even if the result would otherwise be a failure). The task succeeds. The task produces the results desired and positive consequences as well.

2) Spectacular Failure Three Sixes. If the actual dice roll includes 3 sixes (not possible on 1D or 2D), the result is a Spectacular Failure. The task fails to produce the results desired, and it produces negative consequences.

3) Spectacularly Interesting It is possible (if the task calls for 6 or more dice) to roll both 6-6-6 and 1-1-1. The result is a Spectacularly Interesting situation involving both Spectacular Success and Spectacular Failure (and a sign that the referee should make situation a rousing, interesting event for all concerned).

4) Spectacularly Stupid If C+S is less than the number of dice being rolled, the task cannot ordinarily be successful. Some characters will desperately try such a task in hope of Spectacular Success. There is a chance (although vanishingly slight) that the result will indeed be spectacular.

​

I think part of the problem may in fact not be a "broken" rule, but rather that parts of it are examples of bona fide errata. Look at the comparable section in T4, p. 50 (from which T5 evolved):

Spectacular Results
Occasionally the outcome of a task can be spectacular in either a positive or negative way.

Spectacular Success: Sometimes when a person tries something, things just fall perfectly into place, and the result is so amazing that it seems almost like magic. In game terms, whenever a player rolls the minimum possible result for a task (a 2 on 2D, a 3 on 2.5D or 3D, or a 4 on 3.5D or 4D), that attempt succeeds spectacularly. The referee will decide the exact effects of this success (but the player can certainly make suggestion). .
.
.
.
Spectacular Failure. On the other hand, no matter how hard a person tries, no matter how well prepared he is, sometimes the universe just seems set against him, and his action fails dramatically, against all odds. In game terms, whenever a player rolls two sixes on a task roll, the attempt fails in some particularly awful way.

​

It seems that what was intended for T5 was that whenever a roll of all 1's is made (regardless of # of dice), a spectacular success occurs, and whenever a roll that includes three 6's is made (regardless of # of dice), a spectacular failure is the result. This will of course eliminate the possibilty (above) for a "Spectacularly Interesting" Result.

This allows for Average and Easy tasks to get Spectacular Success, while at the same time not increasing the chance for Spectacular Success as the difficulty level increases. Likewise, Spectacular Failure probability increases as the difficulty increases.

 

[cuTTer]

T5 does it slightly better in that a larger dice pool means that the task is more difficult, so the relation to skill level is indirect (only a high skill level would attempt a more difficult roll), but the basic problem remains: if you have the skill levels and the Mods to make a large dice pool irrelevant to your chance of normal failure, you still increase your chance of Spectacular Failure, which is in my view complete nonsense.

But you have no problem with you chance of Spectacular Success rising with difficulty?

 

[whulorigan]

But you have no problem with you chance of Spectacular Success rising with difficulty?

I believe if you look at the original post, he has Spectacular Success and Spectacular Failure swapped. That's why I have the two undelrined in my citation from T5 Rules.

Spectacular Failure should work fine as is. It is Spectacular Success that needs adjusting.

 

[mvdwege]

Yes, I messed up the numbers. Spectacular Failure is of course 3 or more 6s. I typoed that because I had the Storyteller mechanic playing around in my head, where it is 1s that are to be avoided.

Just read 6 where I put 1, and my objections still stand.

As for Spectacular Success chances rising with rising difficulty, yeah, that bothers me too, although I can at least defend that with the rationalisation that succeeding at a difficult task can at least feel spectacular, and in fiction often is spectacular (like Luke hitting the exhaust port). I still dislike it, and I would probably houserule it out as well.

 

[whulorigan]

As for Spectacular Success chances rising with rising difficulty, yeah, that bothers me too, although I can at least defend that with the rationalisation that succeeding at a difficult task can at least feel spectacular, and in fiction often is spectacular (like Luke hitting the exhaust port). I still dislike it, and I would probably houserule it out as well.

Does my post upthread about how it is easily fixed (which is based on the T4 mechanic from which it evolved) work for you? Do you see additional problems that my post does not address?

 

[McPerth]

I still disagree. It is simple, fundamental, probability mechanics: the larger your dice pool, the higher the average number of 1's rolled. Even if you set a low difficulty (in Storyteller terms), the chances of rolling enough ones to cancel out your successes increase with the number of dice rolled. In other words, the better you are, the higher your chance of failure.

While intuitively I'd say that the more dice, the more botches neutralized (and average results are more successes), I'm not good enough in probabilistic calculations to challenge you in this, so I'll accept your point.

T5 does it slightly better in that a larger dice pool means that the task is more difficult, so the relation to skill level is indirect (only a high skill level would attempt a more difficult roll), but the basic problem remains: if you have the skill levels and the Mods to make a large dice pool irrelevant to your chance of normal failure, you still increase your chance of Spectacular Failure, which is in my view complete nonsense.

If what you mean here is that, as the higher the dificulty, only the best will try and so will have more spectacular failures, I'll agree with you here, but I don't believe this is broken, but quite real.

In real world, only the best, the fool or the deseperate would try too difficult tasks, and so they have more probability to fail spectacularly than the not so good (if they are not fools nor desesperate) people that will not dare to try (so to say, Mr Schumaker is more likely to die in a car crash while driving than myslef, as he takes more risks).

Of course, if you character has stat + skill 16 and mine has 8, I would not even try (unless desesperate) a 3+ dice possibility, while yours will probably try it more often, so my character will never acheve a spectacular failure (nor a spectacular success), while your will sooner or latter.

But if we both try the task, we will have exactly the same possibilites, raising with the difficulty level and regardless our target number difference, to acheve a spectacular failure, so I don't see the rule broken, while I see broken that we have the same possibilities (also raising with the difficulty and independent of the target number) to achieve a spectacular success.

I am not wedded to a roll-over system as I grew up playing ASL, so as opposed to some on these boards, I find a roll-under system fairly normal.

Glad to see I'm not the only one here to have grown up playing SL (and latter ASL) .

 

[esampson]

Ok. A couple of things;

Firstly, the odds of botching do not increase in the Storyteller system as the skill of the character increases. This is a misunderstanding some people have of the fact that if you graph out all the difficulties and dice totals for basic rolls then there are some funny spots that occur where the odds of success decrease slightly. These spots tend to occur right around difficulty 8 and 9 and with fairly large dice totals. At difficulty 10 odds jump all over the board on basic rolls but tasks that are difficulty 10 are always crazy-land difficult.

I know this for a fact because I've worked out the statistics myself and I know the statistics are correct because on lower dice totals (up to around 6 or 7 dice) I've run all combinations through with a computer program and the results line up.

As for the odds of spectacular failure increasing in Traveller as characters get more skilled, they don't. They remain constant for difficulties from 3D to 5D (1 in 216), don't exist for difficulties of 1D or 2D, and decrease slightly at 6D and above (because of the 'something interesting' result). More skilled characters might be more likely to attempt the things that are more difficult, but that doesn't mean that the odds increased because the character is more skilled. The odds occurred purely because the character was trying something more difficult.

I suppose very technically the odds do decrease in some cases since the 'this is hard' rule could push some rolls from 5D to 6D but the decrease in the odds of spectacular failure changes from 6 in 46,656 to 5 in 46,656. The particular dice you are rolling probably have a more significant impact than that.

 

[mvdwege]

Does my post upthread about how it is easily fixed (which is based on the T4 mechanic from which it evolved) work for you? Do you see additional problems that my post does not address?

For Spectacular Success to be all 1's would make it exceedingly rare, and would increase the rarity with difficulty, so yeah, I could live with that.

The Spectacular Failure Rate increase is still something I consider broken though.

 

[McPerth]

The Spectacular Failure Rate increase is still something I consider broken though.

Making the spectacular failure probability higer as difficulty increases does not seem broken to me. Doing the same with spectacular success is enterelly another matter...

I'm not a good driver (to say the least). To assume that the more difficult is my driving (with a not well maintained car, on a slippery road, while raining, on the dark and without lights) will increase my possibility to have a deadly crash (spectacular failure) does seem right to me.

Saying that the same conditions sill increase my possiblities to reach my destination sooner (spectacular success) does not (just to reach my destination alive would be a spectacular success here, given de odds).

 

[mvdwege]

As for the odds of spectacular failure increasing in Traveller as characters get more skilled, they don't. They remain constant for difficulties from 3D to 5D (1 in 216),

Sorry, but no. Your calculation is correct for 3D, the chance of rolling 3 6s is indeed 1/6^3. However, for 4D and 5D, the chance of rolling at least 3 sixes rises, because you no longer straight up multiply the probability but use the binominal distribution.

 

[esampson]

Sorry. You're right. I wasn't thinking about permutations. Even instinctively I should have realized that the odds increased as you roll more dice.

However that does not remove the fact that the increase in odds for a spectacular failure is increased not by the skill of the character but by trying things that are harder (i.e. it does not matter how skillful you are, if you try a 5D task then you have the same odds of spectacular failure whether your C+S is 20 or 30.

This is of course untrue.

A skilled character has a lesser chance of rolling a spectacular failure when trying that 5D roll than an unskilled character. Why? Because thanks to the 'this is hard' rule the unskilled character is rolling 6 dice instead of 5 and so has a higher chance (6.5%) of getting 3 6's than the skilled character (at 3.2%).

 

[mvdwege]

Sorry. You're right. I wasn't thinking about permutations. Even instinctively I should have realized that the odds increased as you roll more dice.

However that does not remove the fact that the increase in odds for a spectacular failure is increased not by the skill of the character but by trying things that are harder (i.e. it does not matter how skillful you are, if you try a 5D task then you have the same odds of spectacular failure whether your C+S is 20 or 30.

This is of course untrue.

A skilled character has a lesser chance of rolling a spectacular failure when trying that 5D roll than an unskilled character. Why? Because thanks to the 'this is hard' rule the unskilled character is rolling 6 dice instead of 5 and so has a higher chance (6.5%) of getting 3 6's than the skilled character (at 3.2%).

Still not quite true. If, thanks to your assets and circumstantial mods you don't have to worry too much about making the 6D roll, you still have to cope with the fact that the probability of Spectacular Failure is higher than someone attempting a 3D roll against a lower difficulty.

I dislike that.

 

[esampson]

Still not quite true. If, thanks to your assets and circumstantial mods you don't have to worry too much about making the 6D roll. . .

Well, yes, if there is something that gives the unskilled person a modifier so they are only rolling 5D then you're right, their odds of spectacular failure drops to around 3.5%. However in order to compare apples to apples the roll for the skilled person would drop to 4D which is only a 1.4% chance of spectacular failure. If you aren't going to give the skilled person the same bonus then what you are really saying is that the skilled person has the same odds of spectacular failure when they don't have the advantage that the unskilled character has. That's hardly a condition unique to the T5 rules.

. . .you still have to cope with the fact that the probability of Spectacular Failure is higher than someone attempting a 3D roll against a lower difficulty.

I dislike that.

3D at a lower difficulty seems redundant. Unless I've missed something (which is entirely possible) 3D is the lower difficulty.

Assuming that's the case what you're saying is that someone making a more difficult roll than someone else has a higher chance of spectacular failure. I don't really see anything wrong with that.

The only thing I see that I might consider broken is that assuming both characters have enough skill to avoid 'this is hard' then a highly skilled character has the same odds of spectacular failure as a character with barely enough skill to avoid 'this is hard'.

Again though, this mechanism is hardly unique to T5. It isn't much more than a slightly more complicated version of 'a roll of 1 on 1d20 always produces a fumble' or 'a roll of 3 on 3d6 always produces a fumble'. The only change in this case is that as you try things more and more difficult the odds of that fumble increase, regardless of character ability.

 

[mvdwege]

Again though, this mechanism is hardly unique to T5. It isn't much more than a slightly more complicated version of 'a roll of 1 on 1d20 always produces a fumble'

Guess which D20 rule I routinely ignore when DMing?

 

[Shonner]

I try not to read examples in the book because they didn't get an update if a rule was updated. But last I heard... rolling three 1s meant spectacular success in Traveller 5. Rolling three 6s meant spectacular failure.

So far, I don't see anything about rolling more than three 1s or 6s to be spectacular. Just three of them is needed.

ADDED:
I see a spectacular interesting 6D roll: 1-1-1-6-6-6