Your working make no sense
you can't say this; "Without Halo 3, Xbox 360 still outsold (should be out-revenued) PS3 2:1, which means that if PS3 accounted for 9% of total revenue than Xbox 360 without Halo 3 accounts for 18%."
It makes no sense, this is percentages we are talking about, PS3 software accounted for 9% of software revenue with halo 3, If you take Halo out the PS3 percentage will be higher because percentage adds to 100 whatever information is excluded.
If anything, because Halo3 accounts for slightly over half of total revenue itself
(56% in fact) the PS3 percentage of total without revenue should be more than double (up to 20%) with Wii also being more than double (18% up to 40%) which means that 360 had very similar software REVENUE to Wii (both about 40%)
here are my workings;
total software revenue- PS3 9%, 360 is 8:1 over PS3 making 360 72%, Wii is 1:4 of 360 meaning Wii 18%....adds up to 99% (allow for innacuracy of ratios)
If you use the 2:1 ratio of 360-Halo:PS3, that means Halo made up 3/4 of all 360 software revenue;
so 72% /4 = 18%.....18% X3 = 56%
That means you have taken 56% of software out of the equation yet still used PS3s 9% of total as a base, which is why you should be multiplying both PS3s and Wiis percentages by 2.3 (100/44% = 2.2727....)
EDIT; wait, i think i see what you have done with the table now.....i'm not totally sure but i may be wrong in my accusations, but like everyone else says, you have assumed price points (average 360 game $75?) and have used VGC data which has overtracked PS3 hardware this past month so it's safe to assume it overtracked software too.
