FUN WITH NUMBERS
Let me show you some numbers. (My numbers, let me show you them.)
Total Melee sales in Japan: 1.44m
Melee sales after one week: 360,070
M+2: 423,913 (+63,843) (-82.27% S) (+17.73% T)
M+3: 478,667 (+54,754) (-14.24% S) (+12.92% T)
M+4: 561,578 (+82,911) (+51.42% S) (+17.32% T)
M+5: 715,125 (+153,647) (+85.32% S) (+27.36% T)
M+6: 835,839 (+120,714) (-21.43% S) (+16.88% T)
M+7: 950,242 (+114,403) (-05.23% S) (+13.69% T)
M+8: 976,475 (+26,233) (-77.07% S) (+02.76% T)
M+9: 996,675 (+20,200) (-23.00% S) (+02.07% T)
M+10: 1,010,609 (+13,934) (-31.02% S) (+01.38% T)
M+11: 1,023,775 (+13,166) (-05.97% S) (+01.29% T)
M+12: 1,034,573 (+10,798) (-17.99% S) (+01.04% T)
Obviously Christmas / New Year's has massively affected this data but let's assume Brawl will see a drop of 83% for second week and 15% per week thereafter. Applying that model to SSBM first week sales gives us 12th week sales of 10243. Pretty damn close -- and on the low end, just like you like it, eh?
Total Brawl sales in Japan after one week: 840,000
B+2: 982,800 (+142,800)
B+3: 1104180 (+121380)
B+4: 1207353 (+103173)
B+5: 1295050 (+87697)
B+6: 1369592 (+74542)
B+7: 1432953 (+63361)
B+8: 1486810 (+53857)
B+9: 1532588 (+45778)
B+10: 1571500 (+38912)
In this model Brawl just barely misses outselling Melee LTD in 7 weeks.
Keep in mind wide reports of supply constraint for Brawl in Japan, so these numbers may be higher if Nintendo can pump them out fast enough.
(To those in the know: Am I overlooking something? Is my model just stupid? Do higher-selling games drop off faster, or does it depend more on the game?)