Jeffrey correctly pointed out that some of the commenters may be missing the point of the puzzle. Bean is correct that charging a lower console price will cause more people to want to buy the console. If the company can sell all the units it wants, this low console price may be made up for by sales of peripherals, accessories, and games, which may be more profitable. This is basically a form of price discrimination. As benh57 said, the printer/ink model is another example of this--sell the printer for next to nothing, then charge a lot for the ink. This means that customers with the highest willingness-to-pay (those who will use the printer or game console a lot) end up paying more than those with a low willingness-to-pay.
But as Jeff said, this only makes sense if console makers can sell all the consoles they want at this low price. If there is a limit to how many consoles they can sell, and they still charge a low price--a price so low that there is excess demand--this is a puzzle. To put it another way, suppose Nintendo could sell just as many Wiis, and therefore just as many accessories, peripherals, and games, if they were to raise the price by $25. Why wouldn't they just raise the price? Even if they're trying to price discriminate, they're not following the optimal price discrimination scheme. Kudos to Jeff for seeing the deeper question.
Suppose Nintendo had initially sold the Wii for $400, and then lowered the price to $250 after making all possible sales at that price. Jeff proposes that bad will from quickly lowering the price could be important--buyers could be angered, and this might affect future sales. I doubt this, though. People were initially angry about the iPhone's rapid drop in price, but they got over it quickly. He has a point regarding an upgraded product, of course (he mentions buyers waiting for a new DS to come out, rather than buying the current one), but we're not exactly talking about that. We're just talking about a lower price for the existing product.
Pup mentions that SKUs never go up in price; that's probably true, although there are ways around that, and it still begs the question of why all the console makers seem to get the price too low in the first place. The prices of electronics probably have a hard time rising because of contracts with distributors and retailers. One way to get around this problem is to simply release a new SKU (Say, a new Wii in black with stronger wrist straps) and charge significantly more for it. It's true, this might make them look greedy, but for a couple hundred million dollars, couldn't they live with this? They've been accused of it before, for other reasons, and survived. Pup also mentions that shortages create news and buzz, but so does selling out--and what good is additional buzz if there are no consoles to sell? The point of generating buzz is to sell more product when there is more product to sell. If a company is already selling all it can, buzz doesn't do much good.
Ted Vessenes suggests (I think) that the idea of selling online at a higher price might anger retailers, who might refuse to stock the product. This is possible, and if retailers are organized it might succeed. There's a prisoner's dilemma problem here, though. Suppose I am one of ten retailers selling Wiis. If the other retailers decide not to carry it, it's in my interest to continue carrying it, as I'll get all the sales of the remaining retail Wiis, regardless of what Nintendo does online. Again, this might work if the retailers are sufficiently organized. He mentions the case of labor unions, which are very organized.
Ted has a very good point regarding specialization and economies of scale. It may cost Nintendo so much to set up a mechanism for selling Wiis online (on eBay or whatever) that it's not worth it; it's better to just let Best Buy and Target and the other stores do the selling, since that's what they're good at.
I'm sad to say that the explanations provided on my Econ 265 exams were mostly disappointing. One interesting explanation was that Nintendo (which, if i remember correctly, started selling the Wii in Japan before the U.S.) made a mistake when bringing the Wii to the U.S., relying too heavily on sales information from Japan (and on Japan's anticipated demand for the PS3). A couple students suggested the price discrimination scheme, but they did not see the problem with that argument. It's a difficult problem, and hard to think through during a stressful exam (especially since the exam was on Law and Economics, and this was an extra credit general economics question).