[Sphinx]

The Master Character Creation Table in First Edition (SR1 p.53), Second Edition (SR2 p.47), and Third Edition (SR3 p.54) all awarded the same number of attribute points at five priority levels: A30, B24, C20, D17, E15. These points were distributed among six attributes (three physical, three mental) on the same six-point scale we use now. A human character with attributes at Priority A (30 points) could have a rating 5 (one below maximum) in all six attributes. Priority B (24 points) was enough for rating 4 in all attributes. Even at Priority C (20 points), characters could do a little better than average with rating 4 in two attributes and 3 in the rest.

Fourth Edition dropped the priority system in favor of build points, and increased the number of attributes from six to eight (four physical, four mental). When Fifth Edition brought back the Priority Table (SR5 p.65), it assigned the following attribute priorities: A24, B20, C16, D14, E12. Note: earlier editions allocated 1 to 6 points to each attribute and then added modifiers for metatype, while Fifth Edition attributes start at a minimum set by metatype and then allocate 0 to 5 points from the priority table. Even so, beginning attributes are significantly lower now than they were before. In Fifth Edition, a human character with attributes as their highest priority can manage a rating 4 in each, a full step behind the first three generations of the game. Three of the five priorities leave you with average stats or worse. It almost makes you wonder if they forgot there were eight attributes instead of six. I might not mind, except that NPCs clearly don’t have the same handicap—it’s not unusual for ordinary grunts to have attributes equivalent to the most elite PCs.

For Fifth Edition attributes equivalent to classic Shadowrun, the priority table should be: A32, B24, C20, D16, E12. This would give (human) characters an average rating 5 for priority A, 4 for B, 3.5 for C, 3 for D, and 2.5 for E. Honestly, that feels overpowered to me after a dozen years playing Fourth and Fifth Editions. But I still think the priority table needs an adjustment, so I suggest the following attribute spread, strictly as a house rule:

A28 (add 4), B23 (add 3), C18 (add 2), D15 (add 1), E12 (no change).

Thoughts?

[ShadowcatX]

Other than "this isn't how they did it in previous editions" what benefit do you hope to see from this? Attributes is already one of the best places to assign a high priority, so it isn't like you are changing how people allocate priorities much. And sure, an "average grunt" might have mental attributes equal to your street samurai, but I doubt they will match his 9 agility...

To me, the attributes are supposed to be give and take. You're not making super humans, that's the realm of nuyen / magic slots. Attributes is where you begin introducing flaws to your character because you probably won't get all the points you want. And flaws, again IMO, are cyberpunk.

[adzling]

I see no reason for this change.

The current points are just fine imho.

[Kythandra]

In a way, you're right, attributes for high priorities tend to be lower in 5th edition than it was in 1st to 3rd. But, you also have to consider the way things are balanced in the game. You compared the average stats of a high-ability scored human, but did you make the calculations for every other metatype ?

You're trying to compare with previous editions, which is, in theory, not a bad way to consider things. Though, you have to consider the priority table as a whole, not just a single column. Let's have a detailed look at what was done in 3rd (the edition I know best besides 5th), and why it explains a lot of things that were done around total points.

Ability scores
A - 30 (av. 5)
B - 27 (av. 4.5)
C - 24 (av. 4)
D - 21 (av. 3.5)
E - 18 (av. 3)

In 3rd, all of your points had to be bought, and you only had 6 attributes. Which equates to a low-end average of 3, and a high-end average of 5 for humans. Plus, there's a good reason in the next part that explains why.
In 5th, you have 8 attributes, and less points, thus you have more choices to make, and you have to leave more flaws. That's part of the design principle : attributes shouldn't always be top-priority for most characters. We'll keep in mind the average low-end of 3 (1 free point in every attribute, 16/8 = 2 additional points) and the high-end of 4 (free, 24/8 = 3 additional) for humans.

Metatypes
A - Nothing
B - Nothing
C - Elf, Troll
D - Dwarf, Ork
E - Human

Well, that would call a rework for the metatype column, considering Special points.
Second thing, races used to be different : their bonuses to stats were real bonuses, but they also had penalties, you had to buy off. (Yeah, the troll had to invest at least 3 attribute points in both intelligence and charisma to have a 1 in both)
If we consider a troll in both editions,

• 3rd Troll

Attributes E : B 8 ; Q 2; S 7 ; I 1 ; C 1 ; W 3 (av. 2,66)
Attributes A : B 10 ; Q 4 ; S 9 ; I 3 ; C 3 ; W 5 (av. 4,66)

• 5th Troll

Attributes E : B 7 ; A 3 ; R 3 ; S 7 ; W 3 ; L 3 ; I 3 ; C 3 (av. 4)
Attributes A : B 8 ; A 4 ; R 4 ; S 8 ; W 4 ; L 4 ; I 4 ; C 4 (av. 5)
[/list]
Here, you can see that a low-attribute troll in 3rd has huge flaws (I 1 and C 1) making it unplayable, on average. In 5th, a low-end troll, even if not excellent, is still playable. This reasoning applies to every metatype that used to have negative modifiers, that don't exist in 5th. And high-attribute trolls are also better on average than in 3rd. (which totally explains their being a B+ priority)


Magic
A - Full magician
B - Aspected magician, adept
C - Nothing
D - Nothing
E - Nothing

Basically, you want magic, you have to expend a high priority on it. However, you have a Magic attribute equal to your rounded down Essence. That also meant that if you wanted to play a human character without magic, you had NO benefit for putting either at D.
In 5th, you can be aspected from D to B, and full magician from C to A, due to the expansion of the metatype table to include special attributes.

Skills
5th equivalent values are between parentheses, skill group points are worth their 3 skill points minimum value (not considering worthiness of either skill, that's another debate)
A - 50 (46 + 10 Group = 76)
B - 40 (36 + 5 Group = 51)
C - 34 (28 + 2 Group = 34)
D - 30 (22)
E - 27 (18)

In terms of equivalency, the average stays the same : 34 in 3rd, 28+2G=34 in 5th.
But it's not the only value to compare :

• the ratio between high and low values is completely different : in 3rd, you have a roughly 2:1 discrepancy between A and E. In 5th, you have a 4:1 ratio, making a low-skill choice much more painful.
• plus, as it average stays the same, it means either or both low and high reach more extreme values. In 5th, it's both.
• still, it's compensated for Magic characters, who get free skill points in their dedicated skill groups.

Specializations also worked differently : more specific (no SemiAuto spec), free at character creation, but they also imposed a -1 to the base skill rating, creating a difference of 2 points between base skill and spec. Thus, Pistols 5 becomes Pistols 4 (Ares Predator 6). It also means you couldn't have a skill at 6 and a spec : you had to choose between a skill at 6 and a skill at 5 with a spec at 7.
The second thing here, is that specializations are improved separately from the base skill during gameplay. You could evolve to Pistols 4 (Ares Predator .

To summarize
As such, 3rd characters are more ability score-focused, where 5th are more skill-focused. That's where the game paradigm shifted, to create more interest in the way the priority system could be used.  So, no. There's no real reason to change the column for attributes, without making many other changes to the table.

You can also, lore-wise, attribute that fact to the progressive increase in pollution levels, and other supernatural phenomena, that had an impact on attribute development, but as you could see, it did, surprisingly, only apply to some races.

[Tecumseh]

If we're going to look at previous chargen rules for guidance on the appropriate number of attribute points, then we also need to look at the previous rule systems as a whole to understand what those points meant and what they were used for in the context of the entire system.

I suspect Sphinx already knows this, but attributes are far more important in SR5 than they were in SR1-3. The current Attribute+Skill dice pool mechanic means that attributes are the foundation of everything you do, and thus essential to success in SR5. In SR1-3 you could take Attributes A and still end up as Sir Derps-a-Lot because your base dice pool was equal to you skill rating only.

Attributes are doubly important in SR5 because they also set Limits. Your final tally of hits is not only dependent on your base dice pool but also your Limits. Attributes are a key factor in both (dice pool and Limits), which means they're arguably the most important aspect of a character, crunch-wise. Adding a significant number of attribute points at chargen would only reinforce that.

I know attributes did useful things in SR1-3 - like establish your specialized pools (combat / hacking / control, etc.), determine your karma costs for advancing linked skills, and serving as a default dice pool if you were untrained in a skill - but I still think it's fair to say that attributes are much more important in SR5 than they used to be. If we agree that's true, then the logical conclusion is that the values of the current chargen system don't need to mirror the values of the old one.

All of this is being said to address the "this revision is based on SR1-3 values" argument. If a GM wants to run a super-prime campaign composed of Sioux Wildcats, then yes, by all means, set Attributes A at 32. The professional rating 6 grunts on p. 384 are at 33 and 35 respectively, I believe.

For your typical campaign, I think the current Attributes A is fine in terms of establishing a valid upper end of the spectrum. I wouldn't be opposed to a slightly revised distribution of 24/21/18/15/12 for a regular game, or even  24/22/20/18/16 for a prime runner campaign, but I don't see a default need to go above 24 points for Attributes A.

[Carmody]

There is another fundamental difference between SR1-2-3 and SR4-5 in the way attributes are used:

In SR1-2-3 attributs were used for dice pools, with formulas like (quickness + Intellect + Willpower)/2, and the dice pool were only refreshing once per turn (if I remember properly), which means roughly 1.5 attribute value spread among all the combats test of a given turn

In SR4-5 attributes are used at their full value on every single test, which means that each attribute point represent much more die than in SR1-2-3

To me, this means that SR1-2-3 and SR4-5 attributes are different things, and such a comparison makes little sense.

[Kiirnodel]

Sphinx, take a look at the conversion document for converting from SR3 to SR4. You'll note that it specifically calls out that attributes (and skills) should be lowered by approximately 2/3rds. So an Attribute in SR3 of 5 is only worth an attribute of 4 in SR4/5.

Your analysis showing that humans maximized for attributes getting a 5 in every attribute with SR3 compared to only getting a 4 in SR5 is exactly the result you would get if you converted that character. The attributes seem on point in that regard.