All Possible Ability Scores for 22 Points

[silentounce]

Given the spread of points added as you level up, I think the total number of stat points is relevant too. On that front, both 14 14 13 13 13 11 and 15 13 13 13 13 11 give 78 stat points, the former giving a pretty decent +7 total start. The worst stat arrays give just 72 points.

I wouldn't be too concerned with that considering that most people would use the two +1s on their two main abilities, at least until the last set when that may given them an odd number. Also, you could squeeze out 79 if you wanted to. 14, 13, 13, 13, 13, 13

With the +1 to two different abilities every four levels and the blanket +1s at each tier, you could end up with a 28th level character with the following: 17, 17, 17, 17, 17, 17 before racial adjustment. (that's using my suggested +8 build from earlier)

Here, I'll list the most economical builds, the ones with the most total stat points and that have a +8 total bonus. I know you mentioned earlier their were more variations on this, but I couldn't find any others that give +8. I, honestly, haven't looked at the spread sheet either.

14, 14, 14, 14, 10, 10 - 76
14, 14, 14, 14, 12, 8 - 76

Considering most classes have a primary with two secondaries, I think I'd go with the following.

16, 14, 14, 13, 10, 8 - 75 with a +7
or
16, 16, 12, 12, 10, 8 - 74 with a +7

 

[rrornyll]

I'm not sure where you get your "pluses" from. That first group gives a +4, +2, and -1 for +5 total. The second group gives +2, +2, +1, +1, and +1, for a total of +7.

Careless. Inept. Inattentive. Pick two.

Seriously, I created the spreadsheet and posted it in a ten minute break at work. I herewith apologize sincerely for not perfecting my first contribution before posting, as is the norm in these forums, aka fora.

/snippy

Can you confirm that 8 gives a penalty of -1? I thought there were no ability score-based penalties in 4e? If there are, the "pluses" column of the spreadsheet will still be incorrect.

 

[Revinor]

Can you confirm that 8 gives a penalty of -1? I thought there were no ability score-based penalties in 4e? If there are, the "pluses" column of the spreadsheet will still be incorrect.

I have also heard it long time ago - but I think that what was meant back then is that no PC races will have negative modifier to starting stats. This means we get +2/+2 for most races, as opposed to +2/-2 (or even +2/-2/-2) know from 3e.

Thanks to that, they open up possibility of having slightly more powerful (from point of extra powers) races in future at cost of no ability bonuses.

 

[Bolongo]

Can you confirm that 8 gives a penalty of -1?

Yes. It does.

 

[Mapache]

It looks like the original list of possible combinations is, in fact, incomplete. (For instance, 14 14 14 14 10 10 isn't there.) I think the problem is with this assumption:

I assume the base 8 score will not be increased more than any of the base 10 scores.

You seem to have computed all cases where you don't spend any more points on the 8 than on any of the 10s, which leaves out some possibilities. Instead, you should compute all cases where the resulting value of the score that started at 8 isn't any higher than that of any of the scores that started at 10.

In any case, I have attached my own work, with a little zip file holding a tiny Python script to generate all possible outputs and the resulting CSV file with all 121 possible sets.

 

[Burr]

16 16 12 12 10 8 looks especially worthwhile, if we weight the primary modifier with 2, the secondary modifier with 1.33333, the next two modifiers with 1 each, the last two modifiers with 0.

Scoring it that way, 16 16 12 12 10 8 wins levels 1-3 by a little, levels 1-7 by landslide, and sticks close the top for levels 1-10, 1-13, and 1-17 to ultimately win the average of all of those groupings (1-3, 1-7, 1-10, 1-13, and 1-17).

Edit: Oh, whenever I added level bonuses to stats, I always increased the primary stat. If any of the next three stats were odd, I increased the highest odd stat. Otherwise, I increased the secondary stat.

 

[silentounce]

You seem to have computed all cases where you don't spend any more points on the 8 than on any of the 10s, which leaves out some possibilities. Instead, you should compute all cases where the resulting value of the score that started at 8 isn't any higher than that of any of the scores that started at 10.

In any case, I have attached my own work, with a little zip file holding a tiny Python script to generate all possible outputs and the resulting CSV file with all 121 possible sets.

I added a formula to column G that gives you the total of the ability modifiers. Enjoy the sorting!

16 16 12 12 10 10 8 looks especially worthwhile, if we weight the primary modifier with 2, the secondary modifier with 1.33333, the next two modifiers with 1 each, the last two modifiers with 0.

It wins levels 1-3 by a little, levels 1-7 by landslide, and sticks close the top for levels 1-10, 1-13, and 1-17 to ultimately win the average of all of those groupings (1-3, 1-7, 1-10, 1-13, and 1-17).

You meant 16 16 12 12 10 8, right? And I'm not sure what you mean by wins levels.

 

[Burr]

You meant 16 16 12 12 10 8, right? And I'm not sure what you mean by wins levels.

That I did. Fixed.

And I mean the highest average sum of the weighted modifiers over a grouping of levels. (The weights I used are listed in the previous reply. YMMV.) For levels 1-3, it's just the sum of the weighted modifiers. But the stat bonuses don't come linearly, so for levels 1-7, you have (3*(the sum of the weighted modifiers for levels 1-3)+4*(the sum of the weighted modifiers for levels 4-7))/7. And so on...

Ultimately, I also looked at all of those groupings averaged together, to somewhat account for the idea that you are more likely to TPK or just start a new campaign the further you progress, so there will be more 1-3 campaigns than 1-7 campaigns and so on.

Addendum: If I remember correctly, 17 14 13 12 10 8 won levels 1-10, and 18 12 12 12 10 8 won levels 1-17. So there seems to be a trend towards the extremes for starting higher-level campaigns.

Unfortunately, I stupidly closed the excel file without saving. But then, anyone who would go to the trouble of perusing it would probably want to use their own weights anyway. Edit: *looks below* Ah, guess it doesn't matter then.

 

[silentounce]

Sorry for the double post, but I added more statty goodness. Columns that give the modifier for each stat. A column giving the total mod for a given array. And finally, a column using a formula similar to the one Burr mentioned above. I weighted the secondary stat with a 1.5, but if you are at all familiar with spreadsheets, you can easily change that formula around with different weights.

So, in other words, don't download the one I posted above, this one is better.

 

[mattdm]

16 16 12 12 10 8 looks especially worthwhile [...]

It might not matter so much with +1 to two stats every 4 levels/tier, but 13 is a lot more desirable than 12 for feat access. So, 16 16 13 11 10 8 might be better from a practical point of view — you can then concentrate on putting your every-4 plusses into your top two stats, and the +1-to-all bump at 11 carries the two odd numbers into even.