Steve Jackson Games - Site Navigation
Home General Info Follow Us Search Illuminator Store Forums What's New Other Games Ogre GURPS Munchkin Our Games: Home

Go Back   Steve Jackson Games Forums > Roleplaying > GURPS

 
 
Thread Tools Display Modes
Prev Previous Post   Next Post Next
Old 05-01-2007, 06:30 PM   #1
Luther
Grim Reaper
 
Luther's Avatar
 
Join Date: Aug 2004
Location: Italy
Default Stat Normalization -- A Summary

Stat normalization is something I've advocated since the dawn of the forums, well I'm the self proclaimed God of Stat Normalization (see signature). The 4th edition somewhat helps with this. We have discussed the topic at lenght and I'm giving here a little recap for all the new guys and gals. Keep in mind that the following is for realistic humans.


The Short Answer

I've put A LOT of thought into the subject. I will illustrate here my conclusions, but I'll make little effort to justify them. If you are interested you can read the threads linked below, just keep in mind that this is a game and that my assumptions and reasoning have to be used in a game, not for a scientific publication. I cap attributes for realistic humans, and assume a normal, simmetric distribution of them -- I feel that's good enough for gaming purpouse.
  • IQ, DX, and HT cannot exceed 15.
  • ST cannot exceed 20
  • 50% has attribute 10
  • 95% has attribute in 9-11 range
  • +99% has attribute in 8-12 range


The Long Answer

Strength

Lately I changed my mind on the subject. According to Basic Lift, and energy consideration, I think the maximum realistic human strength in GURPS is about 15. However, due to what I consider an imperfect mechanic, you can buy up to Lifting ST 20. In fact I require realistic characters to buy (ST-10) levels of Lifting ST.

ST: up to 15.
Lifting ST: up to 20.
Striking ST: not allowed.


Intelligence

Using a normal distribution for the attribute (in a way similar to I.Q., even if the two are different things), according to the attribute-skill relationship, and to studies about human learning abilities, I've developed a consistent (with its premises) model, which gives fairly believable results. I cap IQ at 15, and I think this is really stretching . . . see IQ and DX human limits.

IQ: up to 15.

Before you ask here is a rough, imperfect, approssimative formula that correlates real world IQ with GURPS IQ. This is just to get an idea, because GURPS IQ is such a broad attribute:

GURPS_IQ = 10 + (Real_World_IQ-100)/20


Dexterity

This attribute is the basis for a lot of skills, which it helps learning -- just like IQ. DX obviously includes visual/spatial and bodily/kinesthetic intelligence. On that ground, and in order to have more credible defaults, I use the same model as IQ, capping DX at 15.

DX: up to 15.


Health

Health has a lot of applications, which makes it very difficult to find a satisfying value. If you think about the bleeding mechanic then perhaps you should limit HT to 13. Talking about skills and defaults maybe 14-15 is ok. Resisting other hazards could justify 16+. Given that I think bleeding and death checks are not so well modeled under the system, I'm inclined to use a believable, easy to remember, coherent cap: 15. See HT human limit.

HT: up to 15.


Hit Points

Mass consideration and various tests lead me to believe that HP 15 is more than enough. See HP human limit

HP: up to 15.


Conclusion

If I were to enforce strictly realistic limits, I would probably lower the above numbers a bit, something like ST 14 (about 20 for lifting), IQ 14, DX 14, HT 13 (14 for skills). However this would be overkill (the system works on approximations), would reduce the usable range too much, and wouldn't be nice and easy. In the end my players and I are quite happy with 15 as a cap.

If you are interested in the subject, there was quite a lot of discussion on the boards, here are some threads that are worth reading, from older to newer:
Realistic human ST?
Realistic Limitation for Strength
IQ: Probably the hundreth or so thread on the subject



Other

For an application of the above, and also an interesting thread about the population of a typical village see: The Village (Stat Levels)

Interesting discussion about Appearance Distribution, complete with percentages.

There are other interesting and consistent takes on the subject. For example Bookman proposed a lognormal distribution of IQ (because of some studies), which results in IQ limits about 17-18. Douglas Cole solved elegantly the default problem using Attribute/2 instead that Attribute-5; of course he elaborated much more to include skill difficulties and learning.

Maybe these good guys and others will want to contribute posting their structured, well thought out ideas. Maybe this thread will turn in a worthless mess of incoherent, "cool fixes". Maybe, and that's even worse, people will start quoting the Basic Set. ;-)
__________________
bye!
-- Lut

God of the Cult of Stat Normalization

Last edited by Luther; 05-01-2007 at 06:44 PM.
Luther is offline   Reply With Quote
 


Posting Rules
You may not post new threads
You may not post replies
You may not post attachments
You may not edit your posts

BB code is On
Fnords are Off
[IMG] code is Off
HTML code is Off

Forum Jump


All times are GMT -6. The time now is 01:06 PM.


Powered by vBulletin® Version 3.8.9
Copyright ©2000 - 2024, vBulletin Solutions, Inc.