IPB

Welcome Guest ( Log In | Register )

 
Reply to this topicStart new topic
> For those wanting skills to matter more
Shadow_Wanderer
post Aug 31 2005, 04:30 AM
Post #1


Target
*

Group: Members
Posts: 9
Joined: 29-August 05
Member No.: 7,642



I had noticed while reading through the forum that a few people did not like the fact that in SR4 a character with a high attribute stat and low skill stat was just as good as a char with a low attribute and high skill (following the thought that skill should beat out raw talent). So in my utter boredom i wrote a program to toy around with different formulas for rolling to make skill more important and i came up with the following.

Be warned, this kinda throws out the whole fixed TN thing...

TN = 5 - (Skill / 2 rounded down)
or
TN = 5 - (Skill / 3 rounded down)

I'll show you the difference it makes below:
I calculated the chances for success using a total of 10,000 rolls for each test, first i'll show you some samples using the normal rules, then sets for the house rule. (keep in mind the percentages are kinda rough since i didn't feel like leaving the program to run through a million rolls for each sample)

The tables show how many times in the 10,000 rolls that that many hits were achieved.

SR4 DEFAULT
****
attribute = 1, skill = 1
0 hits = 4453 (%44.53)
1 hits = 4417 (%44.17)
2 hits = 1130 (%11.30)

attribute = 2, skill = 2
0 hits = 1981 (%19.81)
1 hits = 3884 (%38.84)
2 hits = 2987 (%29.87)
3 hits = 1024 (%10.24)
4 hits = 124 (%1.24)

attribute = 3, skill = 3
0 hits = 887 (%8.87)
1 hits = 2643 (%26.43)
2 hits = 3314 (%33.14)
3 hits = 2173 (%21.73)
4 hits = 806 (%8.06)
5 hits = 163 (%1.63)
6 hits = 14 (%0.14)

attribute = 4, skill = 4
0 hits = 385 (%3.85)
1 hits = 1517 (%15.17)
2 hits = 2704 (%27.04)
3 hits = 2796 (%27.96)
4 hits = 1722 (%17.22)
5 hits = 671 (%6.71)
6 hits = 178 (%17.80)
7 hits = 26 (%0.26)
8 hits = 1 (%0.01)

attribute = 5, skill = 5
0 hits = 205 (%2.05)
1 hits = 895 (%8.95)
2 hits = 1973 (%19.73)
3 hits = 2596 (%25.96)
4 hits = 2175 (%21.75)
5 hits = 1376 (%13.76)
6 hits = 552 (%5.52)
7 hits = 183 (%1.83)
8 hits = 40 (%0.40)
9 hits = 5 (%0.05)
10 hits = 0 (%0)

attribute = 6, skill = 6
0 hits = 86 (%0.86)
1 hits = 477 (%4.77)
2 hits = 1253 (%12.53)
3 hits = 2120 (%21.20)
4 hits = 2366 (%23.66)
5 hits = 2000 (%20.00)
6 hits = 1090 (%10.90)
7 hits = 449 (%4.49)
8 hits = 130 (%1.30)
9 hits = 19 (%0.19)
10 hits = 9 (%0.09)
11 hits = 1 (%0.01)
12 hits = 0 (%0)

---------------------------------------
skill / 2 rounded down reduction for TN
****
attribute = 1, skill = 1
0 hits = 4453 (%44.53)
1 hits = 4417 (%44.17)
2 hits = 1130 (%11.30)

attribute = 2, skill = 2, TN = 4
0 hits = 637 (%6.37)
1 hits = 2509 (%25.09)
2 hits = 3743 (%37.43)
3 hits = 2501 (%25.01)
4 hits = 610 (%6.10)

attribute = 3, skill = 3, TN = 4
0 hits = 150 (%1.50)
1 hits = 1000 (%10.00)
2 hits = 2269 (%22.69)
3 hits = 3143 (%31.43)
4 hits = 2351 (%23.51)
5 hits = 949 (%9.49)
6 hits = 138 (%1.38)

attribute = 4, skill = 4, TN = 3
0 hits = 1 (%0.01)
1 hits = 31 (%0.31)
2 hits = 181 (%1.81)
3 hits = 723 (%7.23)
4 hits = 1719 (%17.19)
5 hits = 2740 (%27.40)
6 hits = 2691 (%26.91)
7 hits = 1524 (%15.24)
8 hits = 390 (%3.90)

attribute = 5, skill = 5, TN = 3
0 hits = 2 (%0.02)
1 hits = 4 (%0.04)
2 hits = 39 (%0.39)
3 hits = 190 (%1.90)
4 hits = 583 (%5.83)
5 hits = 1385 (%13.85)
6 hits = 2267 (%22.67)
7 hits = 2596 (%25.96)
8 hits = 1881 (%18.81)
9 hits = 882 (%8.82)
10 hits = 171 (%1.71)

attribute = 6, skill = 6, TN = 2
0 hits = 0 (%0)
1 hits = 0 (%0)
2 hits = 0 (%0)
3 hits = 0 (%0)
4 hits = 0 (%0)
5 hits = 9 (%0.09)
6 hits = 67 (%0.67)
7 hits = 279 (%2.79)
8 hits = 911 (%9.11)
9 hits = 1973 (%19.73)
10 hits = 3005 (%30.05)
11 hits = 2648 (%26.48)
12 hits = 1108 (%11.08)

---------------------------------------
skill / 3 rounded down reduction for TN
****
attribute = 1, skill = 1
0 hits = 4453 (%44.53)
1 hits = 4417 (%44.17)
2 hits = 1130 (%11.30)

attribute = 2, skill = 2
0 hits = 1981 (%19.81)
1 hits = 3884 (%38.84)
2 hits = 2987 (%29.87)
3 hits = 1024 (%10.24)
4 hits = 124 (%1.24)

attribute = 3, skill = 3, TN = 4
0 hits = 150 (%1.50)
1 hits = 1000 (%10.00)
2 hits = 2269 (%22.69)
3 hits = 3143 (%31.43)
4 hits = 2351 (%23.51)
5 hits = 949 (%9.49)
6 hits = 138 (%1.38)

attribute = 4, skill = 4, TN = 4
0 hits = 39 (%0.39)
1 hits = 325 (%3.25)
2 hits = 1121 (%11.21)
3 hits = 2189 (%21.89)
4 hits = 2730 (%27.30)
5 hits = 2185 (%21.85)
6 hits = 1084 (%10.84)
7 hits = 295 (%2.95)
8 hits = 32 (%0.32)

attribute = 5, skill = 5, TN = 4
0 hits = 0 (%0)
1 hits = 107 (%1.07)
2 hits = 420 (%4.20)
3 hits = 1140 (%11.40)
4 hits = 2057 (%20.57)
5 hits = 2472 (%24.72)
6 hits = 2102 (%21.02)
7 hits = 1160 (%11.60)
8 hits = 426 (%4.26)
9 hits = 94 (%0.94)
10 hits = 14 (%0.14)

attribute = 6, skill = 6, TN = 3
0 hits = 0 (%0)
1 hits = 0 (%0)
2 hits = 9 (%0.09)
3 hits = 25 (%0.25)
4 hits = 173 (%1.73)
5 hits = 455 (%4.55)
6 hits = 1108 (%11.08)
7 hits = 1943 (%19.43)
8 hits = 2348 (%23.48)
9 hits = 2134 (%21.34)
10 hits = 1259 (%12.59)
11 hits = 447 (%4.47)
12 hits = 99 (%0.99)

Some Examples:

Skill / 2 round down option
~~~~
Using this system -
An attribute 3, skill 6 (TN 2) character will beat a attribute 6, skill 3(TN 4) character about 90% of the time, lose 5%, and tie 5%.

Skill / 3 round down option
~~~~
Using this system -
An attribute 3, skill 6 (TN 3) character will beat a attribute 6, skill 3(TN 4) character about 70% of the time, lose 15%, and tie 15%.

thats a nice improvement over the old 50% chance to beat each other

Also, with both systems a Skilled Weakling(attribute 1, skill 3, TN4) will be able to beat an unskilled average character(attribute 3, skill 1, TN5) about 55% of the time, lose 20%, and tie 25%.

Skill / 3 looks to be a much better choice if you use this rule to keep high skill characters from being unstoppable monsters.

well, thats enough outta me... tell me what you guys think..
Go to the top of the page
 
+Quote Post
Hell Hound
post Aug 31 2005, 05:12 AM
Post #2


Moving Target
**

Group: Members
Posts: 109
Joined: 26-April 05
Member No.: 7,360



Well, my first thought was "Wow, long post". :)

It certainly seems to be a decent fix to make skills more important than attributes but probably not one I would use in my games (when I finally get ahold of SR4). Personally I would like to leave altering the Fixed Tn 5 rule to an absolute last option. The entire system is built on a fixed target number, take it away and the system is sure to misbehave in a lot of places.

One possible idea that has occured to me is to make the skill rating set maximum hits. A character cannot have more hits (total not net) on a skill test than their skill rating+1, exploded dice (via Edge) count as 1 hit rather than 2 towards this maximum.

So a character with attribute 6 skill 1 cannot exceed 2 hits total on their skill test, which is about their average result, without using edge. With edge they can get 4 hits since they can explode two dice which only count as 1 hit each for determining maximum (characters get to keep exploded dice over unexploded ones when eliminating extra hits beyond their limit). However a character with attribute 3 skill 3 can get up to 4 hits normally and 8 with edge. Their average number of hits is still 2 however, they won't perform better than the attribute 6 character all the time but they can surpass them because of their superior training.

There are of course a few problems with this idea. Firstly there are apparently some tests with no skill or multiple skills (A Dodge test can be just reaction, or it can be reaction plus dodge skill AND melee skill in close combat) Setting a threshold on these might cause problems. Plus, setting this threshold means that most characters can never roll all successes without using edge. You may be rolling 6 dice but unless that is skill 5 and attribute 1 you can't get 6 hits.
Go to the top of the page
 
+Quote Post
Shadow_Wanderer
post Aug 31 2005, 05:26 AM
Post #3


Target
*

Group: Members
Posts: 9
Joined: 29-August 05
Member No.: 7,642



long post? ya... i guess i went alittle overboard :D

Just wanted to make sure all the information was out there so people could see the difference and decide if it was worth it.
Go to the top of the page
 
+Quote Post
Crusher Bob
post Aug 31 2005, 05:29 AM
Post #4


Runner
******

Group: Members
Posts: 2,598
Joined: 15-March 03
From: Hong Kong
Member No.: 4,253



Here's some sample probabilities for limiting successed to skill +1

(the numbers given are esentailly how likely you are to be success capped by your skill level)

6 dice, skill 3 (more than 4 successes): ~.02
6 dice, skill 1 (more than 2 successes): ~.32

8 dice, Skill 4 (probability of more than 5 successes): ~.02
8 dice, skill 2 (probability of more than 3 successes): ~.26

10 dice, skill 2: ~.70
10 dice, skill 3: ~.44
10 dice, skill 4: ~.21
10 dice, skill 5: ~.08

for 24 dice at skill 7, the chance of being capped is ~.58

I've considered this alteration myself, but wanted to get the book before suggesting it, as I can't be sure exactly how this limit would play out in the game. The math looks ok.

I'd make the following additions:

Total number of successes is generally limited to skill +1.
Defensive rolls (dodge, spell resistance, parry) are not covered by this rule.
Using edge:
Not decided yet, options are
add edge to max successes
don't conut exploding dice toward max successes
using edge removes success cap altogether
using edge doubles success cap

I'm not sure how prevalent the use of edge will be (since I'm not exactly sure how often you can use it).
Go to the top of the page
 
+Quote Post
Shadow_Wanderer
post Aug 31 2005, 05:37 AM
Post #5


Target
*

Group: Members
Posts: 9
Joined: 29-August 05
Member No.: 7,642



Major problem with limiting the number of successes is that you kill the chance for those "lucky hits" or the ability of a weak character to "beat the odds". of course if thats what your going for, more power to ya then.
Go to the top of the page
 
+Quote Post
blakkie
post Aug 31 2005, 05:41 AM
Post #6


Dragon
********

Group: Members
Posts: 4,718
Joined: 14-September 02
Member No.: 3,263



Changing around the TN seems like bad mojo.

Besides if you are going to do all that bean accounting for each skill why not just work out a formula for what you want to do. Like i tossed out the suggestion in a different thread that what people have a problem with is the idea that you just add the Skill + Attribute, but the way we think of ranking is that we actually want more like Skill * Attribute. But 6*6=36 dice, which is a bit excessive. Plus it is rather a steep rise going from Skill 1 / Attr 1 = 1 die. So instead you take sqrroot( Skill*Attr)*2, rounding off. It is a bit wierd that Skill 1 Attr 1 would be 2 dice, Skill 2 Attr 1, Skill 1 Attr 2 come out to 3 dice. As does Attr 3 Skill 1 and Skill 3 Attr 1, while Skill 2 Attr 2 is 4 dice.

Basically if your Skill gets too far out front of your Attr, or visa versa, the total ability is pulled closer to the lower one. If you know the moves, but are weak or are strong but know jack-squat, the synergy isn't there and your weakness pulls you down.

Note: For untrained you'd actually have say Skill 0.5 instead of 0 for the purpose of calculating, and Skill 0 for unknown (meaning you can't Default). With this you shouldn't need a Defaulting penalty outside of what the formula provides.

EDIT: Oops, fixed up some calculations here. Forgot to *2. So the progression for untrained is 0.5 :
CODE

1 1
2 2
3 2
4 3
5 3
6 3
7 4
8 4
9 4
10 4
11 5
12 5


To be a bit harsher for Defaulting you could drop the Skill value for untrained to 1/3 or something.
Go to the top of the page
 
+Quote Post
Hell Hound
post Aug 31 2005, 05:54 AM
Post #7


Moving Target
**

Group: Members
Posts: 109
Joined: 26-April 05
Member No.: 7,360



QUOTE (Shadow_Wanderer)
Major problem with limiting the number of successes is that you kill the chance for those "lucky hits" or the ability of a weak character to "beat the odds". of course if thats what your going for, more power to ya then.


True. It would be nice to come up with a houserule that does not completely remove the possibilty for lucky hits. It would be particularly hypocritical of me to remove that possibility since I was one of the people back before the release of SR4 that was complaining how the new system looked like it would take away the 'long shot' pure luck skill test success.

Edge may be the means to alleviate that problem with the houserule, at least partially.
Go to the top of the page
 
+Quote Post
Shadow_Wanderer
post Aug 31 2005, 06:07 AM
Post #8


Target
*

Group: Members
Posts: 9
Joined: 29-August 05
Member No.: 7,642



At first i thought the changing the TN would be bad, but after trying it out in the program it seemed to work pretty good at making sure the higher skill won out but still gave the talent a decent chance at success. but thats all theory... wont really know till i have a chance to try it in game.
Go to the top of the page
 
+Quote Post
Hell Hound
post Aug 31 2005, 06:27 AM
Post #9


Moving Target
**

Group: Members
Posts: 109
Joined: 26-April 05
Member No.: 7,360



QUOTE (blakkie)
Changing around the TN seems like bad mojo.

... i tossed out the suggestion in a different thread that what people have a problem with is the idea that you just add the Skill + Attribute, but the way we think of ranking is that we actually want more like Skill * Attribute. But 6*6=36 dice, which is a bit excessive...

You're certainly right about 6*6 being excessive. I would be hesitant of any houserule that can create a larger dice pool than the official rules, same as I am hesitant of changing the target number. Make either of those changes and you will almost certainly need to alter a lot of the rest of the system as well. My personal preference in houserules is something that adds on to existing rules rather than changing them, that's why I suggested a cap on successes based on the skill rating.

The 'Square Root of (Att*Skill) doubled' idea keeps dice pools around the same level as the official rules but I'm still inclined to go for a solution that doesn't involve getting out a calculator (my mental maths isn't great on squares) and involves as little change to the official rules as possible.
Go to the top of the page
 
+Quote Post
blakkie
post Aug 31 2005, 06:34 AM
Post #10


Dragon
********

Group: Members
Posts: 4,718
Joined: 14-September 02
Member No.: 3,263



QUOTE (Shadow_Wanderer)
At first i thought the changing the TN would be bad, but after trying it out in the program it seemed to work pretty good at making sure the higher skill won out but still gave the talent a decent chance at success. but thats all theory... wont really know till i have a chance to try it in game.

The integrating it into the game part. At the low end not so much a problem, but as your skill rises the existing scaling goes all out of wack. Especially subtracting TN/2.

Incidentally what would you do for the defender pool in opposed Combat rolls? Without something there (and you don't always have two dice sources, you have sometimes three and sometimes one). Unless you do something to bring up the defender's number of Hits the higher the Skill the more skewed it is to favour the attacker.
Go to the top of the page
 
+Quote Post
blakkie
post Aug 31 2005, 06:42 AM
Post #11


Dragon
********

Group: Members
Posts: 4,718
Joined: 14-September 02
Member No.: 3,263



QUOTE (Hell Hound)
The 'Square Root of (Att*Skill) doubled' idea keeps dice pools around the same level as the official rules but I'm still inclined to go for a solution that doesn't involve getting out a calculator (my mental maths isn't great on squares) and involves as little change to the official rules as possible.

But with the moving TN you have always be watching the rolls for which TN you are using on that roll.

I don't see any reason i'd play with my suggestion, but then i don't see any reason for me to change the standard SR4 rules on this. But i wouldn't expect that the ability calculation would need be done that often. Even with standard SR4 i expect to normally only do the math for a skill once when i raise it or the linked attribute. For sqrt(S*A)*2 you'd definately want to build and even store the characters in an electronic form that does the calculations, like a spreadsheet, between printing it out.

P.S. Having a TN 2 is just plain silly. :P
Go to the top of the page
 
+Quote Post
Synner
post Aug 31 2005, 06:50 AM
Post #12


Runner
******

Group: Members
Posts: 3,314
Joined: 26-February 02
From: Lisbon, Cidade do Pecado
Member No.: 185



A simple fix you might want to try, one which won't require changing the rules to any great extent, is tying glitch triggering to Skill rating. This means that a high attribute/low skill character will be more likely to experience glitches than a high-skill/low attribute character even though they might have the same number of dice.
Go to the top of the page
 
+Quote Post
Hell Hound
post Aug 31 2005, 07:06 AM
Post #13


Moving Target
**

Group: Members
Posts: 109
Joined: 26-April 05
Member No.: 7,360



QUOTE (blakkie)
QUOTE (Hell Hound @ Aug 31 2005, 12:27 AM)
The 'Square Root of (Att*Skill) doubled' idea keeps dice pools around the same level as the official rules but I'm still inclined to go for a solution that doesn't involve getting out a calculator (my mental maths isn't great on squares) and involves as little change to the official rules as possible.

But with the moving TN you have always be watching the rolls for which TN you are using on that roll.

I did say in that post that I'd also be hesitant about using any houserule that changes the fixed target number (but for a different reason than the repeated recalculation of target numbers). This is why my original suggestion was a cap on maximum hits, the dice pools and target numbers remain unchanged.

QUOTE (Synner)
A simple fix you might want to try, one which won't require changing the rules to any great extent, is tying glitch triggering to Skill rating. This means that a high attribute/low skill character will be more likely to experience glitches than a high-skill/low attribute character even though they might have the same number of dice.

Exactly what impact do glitches and critical glitches have on success tests? I want high attribute low skill characters to be less capable than average attribute average skill characters but I don't want them tripping over their toes every time they go to use their low rated skills.
Go to the top of the page
 
+Quote Post
Ellery
post Aug 31 2005, 08:31 AM
Post #14


Moving Target
**

Group: Members
Posts: 778
Joined: 6-April 05
Member No.: 7,298



Tying glitches to skill (rather than, say, skill+attr/2) means that a high-attribute, low skill character is more likely to glitch that a low-attribute, low-skill character.

That's pretty weird.
Go to the top of the page
 
+Quote Post
blakkie
post Aug 31 2005, 08:36 AM
Post #15


Dragon
********

Group: Members
Posts: 4,718
Joined: 14-September 02
Member No.: 3,263



Think REALLY BIG ENGINE, tiny-weeny steering wheel. Like a drag car, and we've all seen what happens with them. :)

P.S. I don't think you'd use just straight Skill for the value though, because you'd be back to SR3 where nobody in their right mind would use a Skill(1). Also there is the issue of Defaulting with Skill (0).
Go to the top of the page
 
+Quote Post
Synner
post Aug 31 2005, 10:02 AM
Post #16


Runner
******

Group: Members
Posts: 3,314
Joined: 26-February 02
From: Lisbon, Cidade do Pecado
Member No.: 185



QUOTE (Ellery @ Aug 31 2005, 08:31 AM)
Tying glitches to skill (rather than, say, skill+attr/2) means that a high-attribute, low skill character is more likely to glitch that a low-attribute, low-skill character.

That's pretty weird.

Correct. Which is why Skill+Att/2 (ie. Dice pool/2) is the formula to trigger glitches and the one I use.

The suggested alternative was something I tried out in playtesting and emphasizes skills regardless of Atts, period. I believe I tried both Skill and Skill*2.

You are correct that it makes that high Atts more probable to glitch than low atts given the same skill level, which is why I don't suggest it's the solution for everyone. However, I found that in practice - taking into consideration the typical Att. ranges in SR4 - it doesn't skew the current balance significantly and helps partially redress the balance for anyone who doesn't want Atts to dominate the equation and promotes skill investment.
Go to the top of the page
 
+Quote Post
Ellery
post Aug 31 2005, 11:44 AM
Post #17


Moving Target
**

Group: Members
Posts: 778
Joined: 6-April 05
Member No.: 7,298



I meant skill+(att/2), not (skill+att)/2. Division is higher precedence than addition, so skill+att/2 should normally be read as the former.

Anyway, skill + (att/2) will reduce the glitch rate overall, but it will also favor those with high skills.
Go to the top of the page
 
+Quote Post
Shadow_Wanderer
post Aug 31 2005, 03:38 PM
Post #18


Target
*

Group: Members
Posts: 9
Joined: 29-August 05
Member No.: 7,642



Blakkie
QUOTE
The integrating it into the game part. At the low end not so much a problem, but as your skill rises the existing scaling goes all out of wack. Especially subtracting TN/2.


I agree that that TN/2 does throw it out of whack, but its up there for comparision and those who want (i wouldnt use it).

QUOTE
Incidentally what would you do for the defender pool in opposed Combat rolls? Without something there (and you don't always have two dice sources, you have sometimes three and sometimes one). Unless you do something to bring up the defender's number of Hits the higher the Skill the more skewed it is to favour the attacker.


I wouldnt do anything for the defender. It makes sense to me that a skilled attacker should beat someone with no skill and little to average natural talent at dodging/blocking attacks. And the the rule still allows for those with alot of natural talent to compensate to a point. Heres another example:

Using the Skill/3 method
A attribute 1, skill 3 character (4 dice, TN 4) will only beat a attribute 5, skill 1 (6 dice, TN 5) 38% of the time, while tieing at 25% and lossing at around 36%.

Go to the top of the page
 
+Quote Post

Reply to this topicStart new topic

 



RSS Lo-Fi Version Time is now: 12th July 2024 - 08:27 PM

Topps, Inc has sole ownership of the names, logo, artwork, marks, photographs, sounds, audio, video and/or any proprietary material used in connection with the game Shadowrun. Topps, Inc has granted permission to the Dumpshock Forums to use such names, logos, artwork, marks and/or any proprietary materials for promotional and informational purposes on its website but does not endorse, and is not affiliated with the Dumpshock Forums in any official capacity whatsoever.