He means highlight the past sales to differentiate them from the future predicted sales.

I still have PS3 reaching LTD 8.75 million units by year end, but I may have to revise that if some games get delayed at Sony's event early next week.

If you look at Japan it has been averaging (roughly) 10,000 units per week and has very few games being released over the next few months; 120,000 would be fairly generous as an estimate for Japan at the current time.

North America has had a few good releases and sales have decreased every month in 2007, expecting much more than 500,000 or 600,000 from North America would be optimistic.

Europe (and other regions) are an unknown at the moment but I suspect that expecting much more than north america would be unrealistic ...

Exactly. Claiming that a game like Lair or Heavenly Sword (played the demo, left very unimpressed) is going to move PS3s is akin to a 360 owner claiming Bioshock or Mass Effect are going to move 360s. While the games might have decent sales (I think Mass Effect will probably do the best out of the four listed) and may make money for their respective developers, those games are not system movers. They're just not, plain and simple.

Games like those are what pad a console's sales and back up gamers' collections beyond the blockbusters like Halo, MGS, FF, etc. They're the type of games that turn a mediocre library (Xbox) into a fantastic library (PS2). 

Very nice extrapolations!!!! Now everyone sees why the Wii outselling the 360 by June or July is near impossible. I have always said September (late). It's pretty close to your extrapolations and estimations. Very good chart. Everyone, pay attention to this man, he is nearing TheSource level's of respect from me.

Anyone know how to post a Microsoft Excel graph in the forum?  It would be nice to his the above prediction graphically.  I think it'd be neat to have everyone's graphical extrapolations in a forum.  In time, we could see whose are the most accurate.  As an average, I think we would come pretty close to actual hardware sell through.

I'd like to show you all the graph I made in late February.  It still looks pretty accurate...and I have graphs of US sales for next gen consoles in the USA (NPD & VG Charts).

I say you should take a look at what kind of bumps a game similar to lair has given the PS3 already. Calculate the amount sold on average before the bump then calculate the sales difference during the bump and how long it lasted. Then add bumps and correspond them to the release dates of each of the major titles. (ie Heavenly Sword, Lair, GRAW2 etc)

PS3 will be lucky if they get over 7.5mil by year end if there is no prices drop.

I think the "guess" here is referring to the supposition that the last month of sales is an accurate predictor of future sales. Of course anyone with a basic grasp of statistics understands the problem with this (and you address this in your post), but these same people should also be able to recognizethis as a "this is the best we have to go on" situation.

One could argue that external factors that affect one console will also affect others - holidays, economic slowdown, etc. Even new game releases (I'm not sure Lair and Heavenly Sword are going to be system sellers) will likely buffer out unless one system has the next GTA3 waiting to suprise us all.

cjpierciiw, I do wonder where you got your other numbers from. I added some numbers I wonder if XBox 360 is really selling around 32k per week in the other region. Here are the numbers I added from VGcharts: