What can video game console sales tell us about the perils of linear extrapolation?

I enjoy when I can combine more than one great love. For me, as a kid born in 1986, video games have been a major part of my life, so much so that I would say I love them. Also, growing up during the 1990s I got bitten by the stock market bug. Those are two loves. Another love is criticizing bad quantitative analysis. Today is my lucky day because I can talk about all three at once. I feel like George Costanza in Seinfeld when he discovered how to combine sex and food.

One of the greatest follies in quantitative analysis is the great temptation to employ linear extrapolation and rely on the observed "trend". Indeed, this is probably one of the single greatest follies of modern man. Now, in fairness, this seems to work most of the time. There seems to be a deterministic motion to the movement of most things over any given length of time. However, what about when that trend breaks? What if it breaks badly? Not sure what I'm talking about? Well, let's talk about video games. Before you lampoon me, let me first say that video games are big big business. While it is off its highs, Nintendo (NTDOY) was, at one point, worth nearly $100 billion a few years ago during the peak sales of the Wii. That's just Nintendo too.

Speaking of Nintendo, prior to the release of the Wii, you might be forgiven for assuming that Shigeru Miyamoto's proud gaming empire had seen its best days and was in a Sega-like death spiral. Indeed, look at this chart as of the end of the last generation of consoles:

My word that isn't a pretty picture. Also, this is with the backdrop of increasing industry-wide console sales so Nintendo was absolutely hemorrhaging market share. I'm sure many marketing degree-bearing "analysts" at the time would have told me that Nintendo had so narrowed its brand focus to the under 16 crowd that this death spiral was intractable. Something something customer base something something brand confusion something blah blah synergy something. Why, the pretty red trend line that Excel drew said so. Many analysts did say just that very thing and suggested that, with the unwavering dominance of the Playstation 2, Nintendo might be relegated to only the handheld market.

Well, let's update this picture with generation-to-date Wii sales.

My... I broke the trend line. I had to torture the damn thing to make it fit this one. Now, to be similarly idiotic I would say "Golly, if you just take that line and extrapolate it forward.... Nintendo will sell 200,000,000 consoles in its next generation!" That would be, unless Sony and Microsoft produce absolutely nothing, a bunch of crap. However, it is quite likely that Nintendo will breach 100 million units in this generation, which is a four-fold increase from the Gamecube.

Now, how might this have impacted your investment decisions? Well, let's compare Nintendo to Sony (SNE) from the start of the ramp up to the 7th generation consoles through the 2007 Christmas season, which marked the height of 7th generation sales. 

Sony, to use a technical term, got its ass handed to it. Granted there are other problems at Sony including its terribly unprofitable film business and, more importantly, its ailing market share in televisions and portable music players. However, plunging from 75% market share in video game consoles to maybe 20%-25% hurt badly as well. That was a profit center for the company and they lost it. 

Friends of mine can attest that I called Nintendo as the 7th generation winner long before the sales figures started coming in. Why did I do such a thing? Because I knew how kids bought video games and they couldn't pay $500 or $600 for a Playstation 3. Wall Street analysts didn't realize this either because they spoil their kids too much to know that those sorts of prices are well outside a family's birthday or Christmas budget. The Wii, at $250, was right in line for a combined present for the kids, or, for slightly more well off families, for a single kid. The fact that it was $100 less than the XBox 360 also helped. Granted, the "gimmick" of motion sensing controllers helped as well, but price was the key determinant. The proof? As Sony and Microsoft lowered their prices, the relative sales disparity improved. However, Nintendo got the upper hand and has not surrendered its edge. Frankly, it isn't likely to in this generation. Probably 2/3s of all the systems that will be sold in this generation already have been if not more, meaning that Nintendo would almost have to get skunked in order to lose its lead.

Long story short, this is to caution against bad analysis. I see it far too often in industry analyses and it leads to a lot of lost money. This is the simplest quantitative method people use and it is also the worst. As I said, it can be right a lot of the time, but when it is wrong, it is very wrong. When doing your analysis, consider a bigger picture than what an Excel trend line shows you. 

Any source