Jump to content



Photo

Task Difficulty


  • Please log in to reply
11 replies to this topic

#1 Yepesnopes

Yepesnopes

    Member

  • Members
  • 1,251 posts

Posted 15 March 2013 - 07:59 AM

Hello all,

I have seen that the task difficulty for opposed checks in EotE have been fixed as compared to its predecesor, Warhammer 3, where it was a disaster matematically speakinga. In EotE equally prepared and trained contendants have around a fair 50% of success rate. Good!

On the other hand, it seems to me they have done a rather worse job regarding the non opposed task difficulties. I will cite the book

For a typicak character I have picked up someone with a stat of 3 and skill training rank 1.

Easy tasks: A typical character with the proper training, resources, and the right tools for the situation should reasonably expect to succeed at most easy tasks he attempts. By puting this in the probability calculator it gives a success ratio of 78.1%. Sounds good to me.

 

Average tasks: A typical character with the proper training, resources, and the right approach to the situation should reasonably expect to succeed at average tasks slightly more often than he fails. By puting this in the probability calculator it gives a success ratio of 65.1%. Sounds good to me.
 
Hard tasks: A typical character with the proper training, resources, and the right tools for the situation should accept that he might fail at hard tasks more often than he succeeds. By puting this in the probability calculator it gives a success ratio of 53%. Something starts to go wrong between definition of Hard and probabilities.
 
Daunting tasks: A typical character with the proper training, resources, and the right tools for the situation will likely fail more often than he succeeds at daunting tasks. By puting this in the probability calculator it gives a success ratio of 42.2%. Since "will likely fail more often" is vague, we can argue if 42% is the right number or not. I personally think it is a bit too high.
 
Formidable tasks: Typical characters almost always fail formfdable tasks. Even trained veterans fail formidable tasks more often then they succeed. By puting this in the probability calculator it gives a success ratio of 33% for the typicall character, quite high as compared to the "almost always fail" definition. On the other hand, if we do the numbers for the "trained veteran" e.g. Stat 4 and skill rank 3, I obtain a fantastic 53.4% of success ration, totally at odds with the definition.
 
The numbers feel at odds with the tasks difficulty definition given in the book. In warhammer 3 they did a similar "mistake" which they "fixed" in a supplement later on by introducing an optional rule which rises the difficulty of Daunting and Formidable tasks. Therefore, I was wondering to do something similar in the EotE. For example, by adding a set back dice by default in all Hard tasks the probability of success decreases down to 45.3%, more in line with the definition. Similarly, by upgrading one difficulty die by default in all Daunting tasks, the success ratio decrease to 36.7%, again more in line with the definition. Finally, by upgrading two difficulty dice by default in all Formidable tasks the success ratio decrease to 24.2% for the typical character and down to 43.7% for the trained veteran.
 
I would like to know which is the experience of "veteran" GMs regarding the non-opposed tasks difficulties.
 
Cheers,
Yepes

The Book of the Asur - High Elf fan supplement

The Dark Side - Witches, Warlocks, Dark Magic and more

Secrets of the Anvil - Advanced Dwarf careers and runes

Dice statistics calculator for SW EotE


#2 aljovin

aljovin

    Member

  • Members
  • 171 posts

Posted 15 March 2013 - 09:33 AM

I think that your premise is wrong. You say an average character has an ability of 3 and a rank of 1, right?

This actually means that the person is above average on his ability (average being 2) and trained, thus being considered proficient. I think the average values should be an ability of 2 and a rank of 0.

in my games, not everyone is trained in ranged combat, but everyone can try shooting with the Blaster (think of Leia in SWIV, when they try to leave the detention level, she's having a hard time holding the blaster (Agility 2, Rank 0), but fast forward to  SWVI, she's much more comfortable with personal combat (Ability 2, but Rank 3?)

Don't you think?



#3 aljovin

aljovin

    Member

  • Members
  • 171 posts

Posted 15 March 2013 - 09:36 AM

I've reread your post, and I'm not sure my answer makes sense anymore…



#4 Shakespearian_Soldier

Shakespearian_Soldier

    Member

  • Members
  • 889 posts

Posted 15 March 2013 - 09:37 AM

aljovin said:

I think that your premise is wrong. You say an average character has an ability of 3 and a rank of 1, right?

This actually means that the person is above average on his ability (average being 2) and trained, thus being considered proficient. I think the average values should be an ability of 2 and a rank of 0.

in my games, not everyone is trained in ranged combat, but everyone can try shooting with the Blaster (think of Leia in SWIV, when they try to leave the detention level, she's having a hard time holding the blaster (Agility 2, Rank 0), but fast forward to  SWVI, she's much more comfortable with personal combat (Ability 2, but Rank 3?)

Don't you think?

This. :)


"Beg for your life. No, doing so won't save you - but it will make your death more amusing to watch."
- Vago the Hutt; Star Wars: Edge of the Empire


#5 Yepesnopes

Yepesnopes

    Member

  • Members
  • 1,251 posts

Posted 15 March 2013 - 11:16 AM

Not sure I agree with your vision. The book states two things

1) Proper training. I guess if they say so, it implies at least rank 1. Otherwise, if proper training implies rank 0, why to bother at all with the sentence "proper training". Moreover, if proper training is rank 0, what would be "no training" or an "improper training"? it does not exist!

2) It says average character. I don't know your characters, but mine have easely 3 out of 6 stats at 3 when they start, plus several skills trained (unless I am doing something wrong with the rules).

any more thoughts about it?

 

Cheers,

Yepes


The Book of the Asur - High Elf fan supplement

The Dark Side - Witches, Warlocks, Dark Magic and more

Secrets of the Anvil - Advanced Dwarf careers and runes

Dice statistics calculator for SW EotE


#6 aljovin

aljovin

    Member

  • Members
  • 171 posts

Posted 15 March 2013 - 05:27 PM

Actually, a Characteristic of 2 is "average" for that characteristic, not an average of all the characteristics of the PC.

I understand your point about the ranks though. But in this case, I would be tempted to understand that you're getting into the semantics of the description, more than the intent. Remember that we are using a beta version of the rules.

It's probably a case where the beta should be changed in the wording of what is an "Average" check. According to p.70 of the Beta, skill 0 is "unskilled".

But a rank 1 is above average. Thus why I think that we should compare with a Rank 0, not 1.

 

The description of "Average" is a bit misleading in that sense (if my understanding of "average" is correct), as it is described "with proper training". "Proper training" does not mean "unskilled", but I think that rank 1 should be considered above "proper training".

Here is a comparaison: Driving a car

Rank 0 : Does not drive or occasional driver

Rank 1 : Taxi Driver (though it may be argueed!) - Professional driver (Drives car for a living)

Rank 2: Stock car driver

Rank 3: Nascar Driver

Rank 4: Formula 1 Driver

Rank 5: Sonic-speed-car driver

 

Again, maybe we're just trying to overthink it.



#7 Diggles

Diggles

    Member

  • Members
  • 324 posts

Posted 15 March 2013 - 07:43 PM

I get a little hung up on the choice of wording for the levels.

 

Who actually uses formiable in their vocab as meaning difficult than DAUNTING.

To me formiable means its going to be a challenge, possible with good performance but plenty of chance for failure.

Daunting is MUCH bigger, with high likelyhood of failure.



#8 Kallabecca

Kallabecca

    Member

  • Members
  • 685 posts

Posted 16 March 2013 - 06:43 AM

aljovin said:

Actually, a Characteristic of 2 is "average" for that characteristic, not an average of all the characteristics of the PC.

I understand your point about the ranks though. But in this case, I would be tempted to understand that you're getting into the semantics of the description, more than the intent. Remember that we are using a beta version of the rules.

It's probably a case where the beta should be changed in the wording of what is an "Average" check. According to p.70 of the Beta, skill 0 is "unskilled".

But a rank 1 is above average. Thus why I think that we should compare with a Rank 0, not 1.

 

The description of "Average" is a bit misleading in that sense (if my understanding of "average" is correct), as it is described "with proper training". "Proper training" does not mean "unskilled", but I think that rank 1 should be considered above "proper training".

Here is a comparaison: Driving a car

Rank 0 : Does not drive or occasional driver

Rank 1 : Taxi Driver (though it may be argueed!) - Professional driver (Drives car for a living)

Rank 2: Stock car driver

Rank 3: Nascar Driver

Rank 4: Formula 1 Driver

Rank 5: Sonic-speed-car driver

 

Again, maybe we're just trying to overthink it.

I wouldn't say that there is that much of a difference between the three different racing drivers. They're driving cars on tracks designed for the speeds they will be going. And the sonic-speed car driver is just holding the vehicle straight on a really long, flat course. Those higher skills would more likely represent things like combat driving/flying where you're trying to do some really tough maneuvers to get out of a deadly situation (like the Matrix: Reloaded freeway chase).



#9 Yepesnopes

Yepesnopes

    Member

  • Members
  • 1,251 posts

Posted 16 March 2013 - 01:16 PM

Obviously the difficulty of a game is a matter of taste for GM and players. The last rpg I have been regularly playing was Warhammer 3, and I ended up totally feed up with the ridiculous challenge that tasks posed to player characters (even new ones). This means that I might be a bit biased right now.

It is just that after doing some numbers, I had the fear that EotE may go the same direction, but indeed its true that starting characters start typically with stats of 2 and 3, and skill ranks between 1 and 2 in they area of expertise. Meaning that in they area of expertise they have very good chances of succeeding at Hard tasks and around 47% for Daunting tasks.

I have to see a bit more if this is what I want in my games, or I prefer a bit more of long run characters.

 

Cheers,

Yepes


The Book of the Asur - High Elf fan supplement

The Dark Side - Witches, Warlocks, Dark Magic and more

Secrets of the Anvil - Advanced Dwarf careers and runes

Dice statistics calculator for SW EotE


#10 LethalDose

LethalDose

    Member

  • Members
  • 747 posts

Posted 17 March 2013 - 09:46 AM

So, I'm going to end up eating some of my words from earlier posts, but screw it.  Such is education and experience.

First off, I do agree that the probabilities of success for starting characters attempting crazy hard tasks is pretty high, and I would argue too high.  I think that was more or less the OP's original point.  We can quibble about what "average" means, and it does make some difference, but not a lot.  Different argument for a different thread.

Now for the main course:  While the success or failure of an action is what moves an encounter forward, and therefore arguably the most important axis of resolution on a dice roll…

sigh

… you can't only look at the results along any single axis.

 

There, I said it.  

 

The idea behind the narrative dice system is that success really isn't a binary outcome, or even an ordinal outcome on a single axis; it's mutlidimensional.  There have been other threads discussing this, but they never really addressed the associated probabilities.  Getting back to this, the idea is that, even though you succeeded on a roll, nasty $#!t can still happen.  Sometimes, even so bad that you wish you had never attempted the task to begin with.

We can demonstrate this using the examples above.  Below are the results of two simulations from my R script EotE simulator (available on request).  Simulation paramters are described in the simulations below.  Both simulations are based on 100K iterations.  Note percentages are reported to the nearest tenth of a percent, and some round-off error may be present.

Simulation 1

If we have freshly minted character with an attribue of 3 and a single skill rank attempt a "Hard" task, he rolls GGYPPP, and he has a ~ 52.6% probability of success .  Okay, super.  BUT, given that the character was successful, there's about a 65% chance that he generates threat on the roll.  The probabilities associated with the outcomes (only on the success-fail and adv-threat axes) look something like:

  Net Advantage 0 Net Threat/Adv Net Threat
Success 7.6% 10.9% 34.1%
Failure 23.1% 11.7% 12.6%


Pr(Threat > 0|Success) = 34.1/(7.6+10.9+34.1) = 64.9%

So even though the character may succeed at the task, there's a good chance that something nasty happens.

 

Simulation 2

The same character above attempts the same skill check, but now with a "Formidiable" difficulty.  There is now a  33.1% probability of success.  Still pretty high, but far from reliable.  And now theres about a 75% chance of threat being produced.  The result matrix now looks like:

 

  Net Advantage 0 Net Threat/Adv Net Threat
Success 0.6% 1.7% 30.8%
Failure 10.6% 11.5% 44.8% 

 

Pr(Threat > 0|Success) = 30.8/(0.6+1.7+30.8) = 93.2%

We see a strong shift to the right, especially in the successful attempts: 93% of the successes also produce threat, and, on average, just under 3 threat are produced on each success. 

So, while the increased difficulty does have a minor to moderate effect on the probability of success, it has a more substantial effect on the likelihood of producing threat.  At least in these examples.

On the math: The calculations and notations I've used above are from a concept known as conditional probability.  In brief, conditional probability refers to the probability of an event given that something else has/will/may have happened.  It is of primary interest when the two different outcomes (e.g. success and net adv/threat) are not independent.

-WJL

PS: @Yepesnopes, did you use the simulator script I sent you for the results you provided in your posts?

 



#11 Yepesnopes

Yepesnopes

    Member

  • Members
  • 1,251 posts

Posted 17 March 2013 - 09:35 PM

LethalDose said:

PS: @Yepesnopes, did you use the simulator script I sent you for the results you provided in your posts?

I used a probability calcutor I addapted for EotE. You can find it here.

Cheers,

Yepes


The Book of the Asur - High Elf fan supplement

The Dark Side - Witches, Warlocks, Dark Magic and more

Secrets of the Anvil - Advanced Dwarf careers and runes

Dice statistics calculator for SW EotE


#12 FuzzyLog1cZA

FuzzyLog1cZA

    Member

  • Members
  • 110 posts

Posted 18 March 2013 - 12:15 AM

 

I played around with your stat calculator, very nice!

However I think your issue is that you are giving people way too many ranks in their ability. 2 is average, not 3, and while giving your Veteran 3 ranks in his skill makes sense, giving him 4 ranks in an ability does not make sense. Just because your characters will average 3 ranks in an ability doesn’t make them an average person, they are beyond average which is the whole cinematic point and are therefore more likely to succeed then the average person.

Those definitions are clearly built on using an average person that has no training in a skill and 2 in their ability. A veteran person will still have only 2 ranks in their ability but may have 3 in a skill. That gives me a success rate of 41% with a formidable task meaning he will fail more often than he succeeds.






© 2013 Fantasy Flight Publishing, Inc. Fantasy Flight Games and the FFG logo are ® of Fantasy Flight Publishing, Inc.  All rights reserved.
Privacy Policy | Terms of Use | Contact | User Support | Rules Questions | Help | RSS