To augment or not to augment

[Kloster]

4 hours ago, French Desperate WindChild said:

in fact with 90% at base skill (without augment but maybe modified in a combat with the reduction of 100%+ rules if i can call it like that) :

you may have 140% but you will fail at 96+, you may have 140% but you will fumble at 00

so the success chance is only raised by 5 when the augment skill roll is succeed (critical, special or normal)

in the other hand, you really reduce your success chance when you failed your augment skill roll

so the average success is

(90+5) * 0.55 + (90-20) * 0.43 + (90-50) * 0.02 = 83.15 < your 90%

the only gain is the probability to have a critical success (20% of 140, 120, 110, there is no "cap" there)

You're right. My calculation is wrong here.

[Baron Wulfraed]

10 hours ago, Bren said:

For example, finding the rules for calculating the physical and spirit combat damage bonuses in the middle of the long character creation section is awkward for me as a GM as I need to reference those calculations whenever I create a new NPC or creature.

Spirit Combat Damage is on the fourth panel of the GM Screen. Regular damage bonus is on the second panel of the GM Screen, along with Strike Rank Modifier(s). Might be worth considering at least the PDF version.

[Baron Wulfraed]

 

Duplicate posting

Edited September 23, 2021 by Baron Wulfraed

[Bill the barbarian]

7 minutes ago, Baron Wulfraed said:

Spirit Combat Damage is on the fourth panel of the GM Screen. Regular damage bonus is on the second panel of the GM Screen, along with Strike Rank Modifier(s). Might be worth considering at least the PDF version.

I should not speak for others, but I am reasonably sure that the point was not that it could not be found but the that it was not easily found. Having to leave the core book  Char Gen entirely to create an NPC or critter does not seem optimum.

[Kloster]

On 9/22/2021 at 9:43 PM, PhilHibbs said:

No, I really can't. I just crunched the numbers and reported the result.

Here it is, if you want to have a look. Maybe I made a mistake.

For a 60 ability augmenting a 90 skill, out of 10,000 rolls 1,958 get a better result and 2,439 get a worse result.

Ok, after taking in account Windchild's remarks, here are the calculations for a 90% skill, possibly augmented by a 55%, 65% and 75% skill:

Without augment:
Critical 5%
Special (18-5)=13%
Success (90-18)=72%
Failure (10-1)=9%
Fumble 1%

With 55% skill augment:
Critical ((7*3)+(6*8)+(6*44)+(4*44)+(2*2))/100=5.13%
Special (((28-7)*3)+((24-6)*8)+((21-6)*44)+((14-3)*44)+(3*2))/100=13.57%
Success (((95-28)*3)+((95-24)*8+((95-21)*44)+((70-2)*44)+(2*2))/100=70.21%
Failure=(100-(1.49+70.21+13.57+5.13))=9.6%
Fumble ((1*3)+(1*8)+(1*44)+(2*44)+(3*2))/100=1.49%

With 65% skill augment:
Critical ((7*3)+(6*10)+(6*52)+(4*33)+(2*2))/100=5.29%
Special (((28-7)*3)+((24-6)*10)+((21-6)*52)+((14-3)*33)+(3*2))/100=13.92%
Success (((95-28)*3)+((95-24)*10+((95-21)*52)+((70-2)*33)+(2*2))/100=70.07%
Failure=(100-(1.37+70.07+13.92+5.29))=9.35%
Fumble ((1*3)+(1*10)+(1*52)+(2*33)+(3*2))/100=1.37%

With 75% skill augment:
Critical ((7*4)+(6*11)+(6*60)+(4*24)+(2*1))/100=5.54%
Special (((28-7)*4)+((24-6)*11)+((21-6)*60)+((14-3)*24)+(3*1))/100=14.49%
Success (((95-28)*4)+((95-24)*11+((95-21)*60)+((70-2)*24)+(2*1))/100=71.23%
Failure=(100-(1.26+71.23+14.49+5.54))=7.48%
Fumble ((1*4)+(1*11)+(1*60)+(2*24)+(3*1))/100=1.26%

This is now more in line with what you have experimentally got. And you are right, the breakpoint is caused by the fumble % going from 99-100% to 100%.

[Bren]

7 hours ago, Kloster said:

Ok, after taking in account Windchild's remarks, here are the calculations for a 90% skill, possibly augmented by a 55%, 65% and 75% skill:

Thanks Kloster for posting this and for partially showing your work. While I don't need the exact numbers in play, having some sense of where the numbers are for a rational optimization choice vs. a do-you-feel-lucky choice.

Coincidentally, in tonight's session, all three PCs failed their Runic Augmentation in the same short scene. But, as has been mentioned, that didn't impact the skill they were trying to improve, though it did cause some complications for one PC when later that day he wanted to cast some Rune Magic. This thread helped me decide to (rightly) decide to roll my second best Runic affinity, so that I could save my best affinity for casting Rune Magic in climactic combat that came later.

 

[Kloster]

Here are the full calculation I've found for every value of skill and augmenting skill below 490% (above that, Special% should be capped to 95%).
It is easy to see that to get a good benefit of the augment, the skill should not be above 65%, and if it is above 75%, the gain will be even lower because the max probability of success is 95%.

Critical: ((Skill+50Crit%*AugmentCrit%)+(Skill+30Crit%*(AugmentSpecial%-AugmentCrit%))+(Skill+20Crit%*((Min(95;Augment%)-AugmentSpecial%))+(Skill-20Crit%*(100-Augment%-(101-AugmentFumble%)))+(Skill-50Crit%*(101-AugmentFumble%)))/100

Special: (((Skill+50Special%-Skill+50Crit%)*AugmentCrit%)+((Skill+30Special%-Skill+30Crit%)*(AugmentSpecial%-AugmentCrit%))+((Skill+20Special%-Skill+20Crit%)*((Min(95;Augment%)%-AugmentSpecial%))+((Skill-20Special%-Skill-20Crit%)*(100-Augment%-(101-AugmentFumble%)))+((Skill-50Special%-Skill-50Crit%)*(101-AugmentFumble%)))/100

Success: (((Min(95;Skill+50%)-Skill+50Special%)*AugmentCrit%)+((Min(95;Skill+30%)-Skill+30Special%)*(AugmentSpecial%-AugmentCrit%))+((Min(95;Skill+20%)-Skill+20Special%)*(Min(95;Augment%)-AugmentSpecial%))+((Min(95;Skill-20%)-Skill-20Special%)*(100-Min(95;Augment%)-(101-AugmentFumble%)))+((Min(95;Skill-50%)-Skill-50Special%)*(101-AugmentFumble%)))/100

Failure: (((100-Min(95;Skill+50%)-(101-Skill+50Fumble%))*AugmentCrit%)+((100-Min(95;Skill+30%)-(101-Skill+30Fumble%))*(AugmentSpecial%-AugmentCrit%))+((100-Min(95;Skill+20%)-(101-Skill+20Fumble%))*(Min(95;Augment%)-AugmentSpecial%))+((100-Min(95;Skill-20%)-(101-Skill-20Fumble%))*(100-Min(95;Augment%)-(101-AugmentFumble%)))+((100-Min(95;Skill-50%)-(101-Skill-50Fumble%))*(101-AugmentFumble%)))/100

Fumble: ((Skill+50Fumble%*AugmentCrit%)+(Skill+30Fumble%*(AugmentSpecial%-AugmentCrit%))+(Skill+20Fumble%*(Min(95;Augment%)-AugmentSpecial%))+(Skill-20Fumble%*(100-Min(95;Augment%)-(101-AugmentFumble%)))+(Skill-50Fumble%**(101-AugmentFumble%)))/100

 

[Bren]

3 hours ago, Kloster said:

Here are the full calculation I've found for every value of skill and augmenting skill below 490% (above that, Special% should be capped to 95%).
It is easy to see that to get a good benefit of the augment, the skill should not be above 65%, and if it is above 75%, the gain will be even lower because the max probability of success is 95%.

Critical: ((Skill+50Crit%*AugmentCrit%)+(Skill+30Crit%*(AugmentSpecial%-AugmentCrit%))+(Skill+20Crit%*((Min(95;Augment%)-AugmentSpecial%))+(Skill-20Crit%*(100-Augment%-(101-AugmentFumble%)))+(Skill-50Crit%*(101-AugmentFumble%)))/100

Special: (((Skill+50Special%-Skill+50Crit%)*AugmentCrit%)+((Skill+30Special%-Skill+30Crit%)*(AugmentSpecial%-AugmentCrit%))+((Skill+20Special%-Skill+20Crit%)*((Min(95;Augment%)%-AugmentSpecial%))+((Skill-20Special%-Skill-20Crit%)*(100-Augment%-(101-AugmentFumble%)))+((Skill-50Special%-Skill-50Crit%)*(101-AugmentFumble%)))/100

Success: (((Min(95;Skill+50%)-Skill+50Special%)*AugmentCrit%)+((Min(95;Skill+30%)-Skill+30Special%)*(AugmentSpecial%-AugmentCrit%))+((Min(95;Skill+20%)-Skill+20Special%)*(Min(95;Augment%)-AugmentSpecial%))+((Min(95;Skill-20%)-Skill-20Special%)*(100-Min(95;Augment%)-(101-AugmentFumble%)))+((Min(95;Skill-50%)-Skill-50Special%)*(101-AugmentFumble%)))/100

Failure: (((100-Min(95;Skill+50%)-(101-Skill+50Fumble%))*AugmentCrit%)+((100-Min(95;Skill+30%)-(101-Skill+30Fumble%))*(AugmentSpecial%-AugmentCrit%))+((100-Min(95;Skill+20%)-(101-Skill+20Fumble%))*(Min(95;Augment%)-AugmentSpecial%))+((100-Min(95;Skill-20%)-(101-Skill-20Fumble%))*(100-Min(95;Augment%)-(101-AugmentFumble%)))+((100-Min(95;Skill-50%)-(101-Skill-50Fumble%))*(101-AugmentFumble%)))/100

Fumble: ((Skill+50Fumble%*AugmentCrit%)+(Skill+30Fumble%*(AugmentSpecial%-AugmentCrit%))+(Skill+20Fumble%*(Min(95;Augment%)-AugmentSpecial%))+(Skill-20Fumble%*(100-Min(95;Augment%)-(101-AugmentFumble%)))+(Skill-50Fumble%**(101-AugmentFumble%)))/100

 

😀 Thanks again!

[Arcadiagt5]

I've done a similar analysis, but took a different approach. Essentially I only looked at the ability being used to provide the augment.

1. I drew up a table from 1 - 100, and for each value calculated the chance of:
   • A critical success (for a  +50 augment)
   • A special success (for a +30) (incidentally an ability has to be >= 8 for there to be separate critical and special success chances)
   • A success (for a +20)
   • A fail (for varying penalties)
   • Fumble (for varying penalties)
2. I then calculated the average return at each augmenting level as equal to: Chance of a Critical x 50 + Chance of a Special x 30 + Chance of a Success x 20 + Chance of a Fail x Penalty + Chance of a Fumble x Penalty (assessed at -100 for Runes/Passions)

The breakeven point for augmenting with a skill is about 49, 51 for a Rune, and 60 for a passion. The latter assumes that the -10 to everything is worth about -30 in the calc.

The key difference in my analysis is that I don't think the value of the skill being augmented matters - it's the source of the augment that determines the average return (which rises to about 20 across the board when the source ability reaches 100). 

And here it is as a chart:

Edited September 25, 2021 by Arcadiagt5

[French Desperate WindChild]

6 hours ago, Arcadiagt5 said:

calc.

The key difference in my analysis is that I don't think the value of the skill being augmented matters

But it matters

and the situation matters too

for example

Brutus has climb 90%
 

he has to climb the wall to escape

he can only try one time : if he succeeds he is free, if he fails he is catched.

The best augment he has is 70%
should he take the risk to lose 20% or worst to escape or just roll against his only climb et 90% ?

now Alienor is in the same situation but she has only climb 45% (the augment best roll is 70 too) in my opinion she should try the augment to save her Life.

then back to our Brutus

he is participating in a spring contest and one of the test is … to climb this tree and take an apple 

depending on the success he can gain more or less points:

critical 5points,special 3 points, success 1 and fumble -30 in any social roll for the party this evening

if he wants to gain the contest I suggest to try to augment his climb skill (more opportunities to crit / spécial)

if he wants to charm someone tonight maybe keep his 90

 

 

[Arcadiagt5]

12 hours ago, French Desperate WindChild said:

But it matters

and the situation matters too

for example

<<SNIP>>

That's true, but not quite what I was getting at. The point I was trying to make was that the expected return from an Augment in terms of average benefit/penalty is independent of the ability being augmented. I was also trying to illustrate where, from that perspective, the break even points are. 

Yes, the situation matters, and yes depending on the target ability, the tactical effect of an augment can vary wildly. 

[soltakss]

On 9/22/2021 at 9:00 PM, PhilHibbs said:

Unless you're using a dice roller app that uses a deterministic pseudorandom algorithm with no stochastic interference.

We used Roll20, which I am sure increases the number of Fumbles just for a laugh.

On 9/22/2021 at 9:14 PM, PhilHibbs said:

In other words, like @soltakss said, "I would have made it if the first die roll was the skill, but the last one was a fumble so I was robbed".

Yes, I agree, I was robbed!

My argument is not against Augments, I think Augments are great. I just hate rolling for rolling sake, and for me rolling for Augments is exactly that. Just give me a straight +10 for an augment and I would be happy.

 

[Kloster]

On 9/25/2021 at 8:27 AM, Arcadiagt5 said:

The key difference in my analysis is that I don't think the value of the skill being augmented matters - it's the source of the augment that determines the average return (which rises to about 20 across the board when the source ability reaches 100). 

In fact (and this is the mistake I did originally), the skill value being augmented does matter because the chance of success is capped to 95% (in fact, the formula I gave are wrong because I didn't take in account the fact that it can't goes below 05%), so if it is above 45%, you will benefit of the 50% bonus (for a critical) on your crit and special chance, but not fully on your success chance. If you are above 75% (as in PhilHibbs original test with a 90% skill), you don't even fully benefit of a standard success augment. On the contrary, if you are below 20%, you are not fully affected by the -50% of a fumbled augment. This is one of the cause of the breakpoints, the other being the granularity of the crit, special and fumble table for the augmenting skill.