Let's look at a couple of karma efficiency items in these two builds though, perhaps:
ELF TROLL delta karma
BOD 5 5 0
AGI 7 3 -110 (Troll can't even actually ever get this high)
STR 1 5 +70 (but you're not actually even using this STR for anything, so what is that +70 actually worth?)
REA 5 2 -60
CHA 7 4 -90 (Troll can't even actually ever get this high)
INT 5 3 -45
LOG 2 2 0
WIL 3 5 +45
EDG 3 2 -15
So, the Elf (whom in this example I dropped the LOG by 2, as it's no longer used in physical B&E, and put those in BOD) comes out ahead by 205 in-game earned karma (and really more like 275 as STR is not important to this type of PC, slap a Shock glove on call it good).
Elf, skills C (28/2) Troll, skills B (36/5)
181 karma worth of Skills 283 Karma worth of Skills
So, the Troll build comes out ahead by 102 in-game earned karma for skills.
Elf, Y E (6K) Troll, Y C (140K)
If 2KY = 1 karma, then the Troll is ahead here by 67 in-game earned cash-for-karma. Although it's worth noting, in-game earned karma may be worth more to an Adept, as he needs a lot of karma to keep pumping that MAG/Initiating along with skills, than in-game earned cash. i.e. the Adept generally has more places to put karma than cash.
+205-275 vs. +169 is quite the disparity depending on how you reckon it, i.e. how much value you put in the unused STR (~36-106 karma), and when we look at the overall potential of the Elf in this PCs primary skills (Facing - CHA, Shooting+Sneaking - AGI, and Limits), he pulls away even more. I guess this is what I was talking about with the long game. Do you want to front-load the PC, and have the pools at the outset, or do you want to have the growth potential?
You could take the same Qualities with the Elf, for the same skill bonus'.