I don't think these numbers work out, at least for our purposes.
The Nielson surveys are dealing with households, which is a bit different than what we're talking about. I'm pretty sure we all mean (at least I did) people who played games primarily on the Wii, since that's really what's relevant if we're trying to figure out what factors are driving the Switch's success.
If we're using households, then a large amount of that ownership is going to be purely incidental. If a person is buying a Switch now, and they lived with their grandma during the Wii days who was one of those octogenarian Wii fans you hear so much about and they never played it themselves, then their Wii ownership didn't really factor into the Switch purchase. Household numbers don't give great insight into what kind of users are buying a console.
And,, I'm not sure about these numbers anyway. If a console owning house generally had 1.38 consoles, That means that there have to be more single console owning houses than double dippers... right? Unless Wii owners are vastly overrepresented among double dipping households, then 62% can't be right.
Which begs the question of whether there was only a 5% overlap between 360 and PS3. Which sounds really sketchy to me. The two systems applied to a similar userbase. I get that they are somewhat redundant, but exclusives and the ability to play with friends who own other consoles would drive it to way higher than 5% imo. And the data we have, if accurate, seems to bear that out. As of 2009 according to NPD, 14% of Wii owners had a PS3 and 26% had an XBox 360. https://www.alistdaily.com/media/most-wii-owners-do-not-own-xbox-360-ps3/ So a maximum of 40% of Wii owners owned one of the other two systems. This number could have drifted up by 2012, but 60% would be really high.
The PS4 data is really suspect to me and wouldn't jive with a 5% PS360 crossover.
Unless I'm missing something, if you add it up you get to about 14 million 7th gen systems among the 6.3 millionish PS4 households in the US which didn't own a 7th gen console. This means either virtually all of them owned two consoles (or alternatively an offsetting number of people owned one and three respectively). That's pretty sus. Of course, these are early adopters though so they're bigger gamers. Still, that number seems crazy high.
But if that number is even remotely accurate, then 5% can't even be close. If there were 5 million households that have a Wii, even if all of those households had a second console, then you're left with about 3.5 million PS3s and 360s among the 1.3 million or so non-Wii owning PS4 households. Even if all 1.3 million of those people owned both systems, there would still be be about 800K left over. Meaning we'd have to have 800K PSWii60 owners which would be about 12% of the PS4 households on their own.
These are early adopters so their habits are going to be different than the entire market, but they can't be THAT far off that a 5% overlap with PS3 and 360 would make sense. So, some figure here is off.
And... according to Nielson the figured were very different in June of 14.
At that point, the number of non gen 7 owners was 17%. Of course, the number could go down over time, but you'd figure the earliest adopters would be the most likely to have been 7th gen gamers, so if anything we'd expect those numbers to be going up and not down.
They also say that 31% of PS4 owners were Xbox 360 or Wii owners that did not own a PS3... If around 60% of PS4 owners in mid 2014 were PS3 owners, and 31% were 360 or Wii owners who did not have a PS3... Then together, that would make about 90%. About 10% of PS4 owners didn't have 7th gen systems... So... that means there would have to be zero crossover between PS3 and Wii/360 owners.
So I dunno. Either these numbers are off, or they're being reported wrong, or the terms are being used in unintuitive manner.
I'd say the best evidence we have if we want to figure out the type of audience is buying which system, the best evidence we have is the games that they're buying, particularly when they're big enough that we're getting figures from the company.
I had my reply mostly finished yesterday morning after working on it a bit Monday night after work, but there was a power outage and I lost everything I had written. I'll try my best to replicate what I had.
That 5% figure was just me spitballing. And I flubbed the numbers slightly, as I was using Gen 7 LTD sales as of March 2012 instead of January. The survey was published in March, but it specified the number of owners "as of January 2012." Bit of oversight on my part. It may be only two months, but given the state of console sales in the U.S. in early 2012 (total 360+PS3 sales were over three times that of Wii sales) it does nudge the figures by several points. Instead of the percentage of Wii owners owning a 360 and/or PS3 being 62.2%, it should be 58.1%. And that may explain why that percentage was at most only 40% in 2009. I know it's just anecdotal, but I bought my 360 in 2007, but I didn't own a Wii until 2009. Waiting until later in a gen to get a second console probably isn't unusual.
In any case, we don't know what the actual overlap is. But let's play with those numbers to see what other results we get. If the percent of 360-owning households in the U.S. that owned a PS3 rises to 10%, then percent of Wii-owning households that own a 360 and/or a PS3 drops to only 53.8%. Even if the former increases to 20%, the latter declines to 45.4%. For the vast majority of Wii-owning households to not own either a PS3 or 360 it would require a very large overlap. But even 20% of 360 owners owning a PS3 seems kind of high considering the redundancy between the two. It's not impossible, but more unlikely than lower figures. It is probably higher than 5%, but I doubt it's something like 20+%, either.
And once we figure in Nintendo loyalists who buy every Nintendo system and only buy Nintendo systems, that's likely to still propel the percentage of Wii owners that were core gamers to over 50%. And of course this is all assuming that Nielsen is only counting just the Wii, 360, and PS3. It's not clear if they were counting the DS and PSP as "consoles," but if they were then the Wii↔PS360 overlap may be even higher.
While you're right that households aren't the same thing as individual owners, I imagine it doesn't make a huge difference. There probably wasn't a huge percentage of Wii buyers getting one for, say, live-at-home grandparents. While non-gamers in the household may have gotten into playing the Wii, I still think there's a good argument to be made that most people that owned a Wii were core gamers. Even Nintendo stated that the average age of Wii players was 29, about the norm for the average gamer. Granted, that was a year after the Wii was released and that could have changed, but I doubt it shot up so high as to indicate large swaths of octogenarians getting in on the craze. There was another Nielsen survey measuring time spent playing on particular consoles by age and gender (download the PDF further down the page for details), and while it implied that the Wii was more popular among players aged 35+ and among women based on the metric being used, there was nothing to indicate that the Wii's demographics were drastically different from that of the PS3 & 360.
Since you brought up the kind of games people bought on the Wii, while the Wii series was very popular, it's worth pointing out that several of them benefited from heavily bundling. Wii Sports owed nearly all of its sales to bundling (its global attach rate was 81.6%, but in Japan where it wasn't bundled the attach rate was only 29.3%). Wii Sports Resort also benefited from heavy bundling as it became the new pack-in in May 2010, less than a year after its release. Wii Play was essentially a $10 mini-game collection bundled with a Wii Remote. Why not spend $60 instead of $50 and get some extra games along with a second controller? Wii Fit succeeded on its own merits, though, as it was bundled with a peripheral made just for it rather than the system or a general purpose controller. If we assume zero overlap between people who bought Wii Fit and Wii Fit plus, 43% of Wii owners had one or the other; it's likely a rather high overlap, but even if it was 100% overlap that still means 22.3% of Wii owners had Wii Fit, and I imagine non-gamers were a decent chunk of Wii Fit buyers. Finally, Wii Party didn't have nearly the attach rate because it wasn't bundled with anything.
But what about other games? Outside the Wii Series, the Top 10 best-selling games on the Wii were Mario Kart, New SMB, Smash, Mario Galaxy, Just Dance 3, Mario Party 8, Twilight Princess, Galaxy 2, Mario & Sonic at the Olympics, and either DKC Returns or Mario & Sonic at the Winter Games. One third-party party game, one or two Sega-made Nintendo-Sonic crossover sports games, and the rest were perennial Nintendo franchises. The top-selling first-party titles on Nintendo's home consoles, excluding those spent most of their lives bundled (e.g., SMB, SM World, Wii Sports), typically had an attach rate in the 30-40% range since the N64. MKWii and NSMBWii were at 36.7% and 29.8%, respectively. Speaking of attach rates, the Wii had an overall attach rate comparable to other Nintendo home consoles. Even excluding Wii Sports, its overall attach rate was over 8.2 games per system, just ahead of the NES, and second only to the GameCube.
While the Wii Series did bring in a significant periphery demographic, there's little to suggest that the vast majority of Wii sales came from non-gamers. Whether it's the Wii's overall sales trajectory, or what other kinds of games aside from Wii Sports and Wii Fit did on the system, or the general spending habits of the average Wii owner (~$650 spent per system, assuming $250 for the system and $50 for each game times eight gamers per system excluding Wii Sports), there's really not anything too far out of the ordinary with the Wii.
Finally, regarding the Aug. 2014 Nielsen survey, while the percentage of PS4 owners that owned a Gen 7 console going to 17% then to 9% in Jan. 2015 does seem odd, upon further inspection the disparity comes down to different questions being asked. While the Jan. 2015 survey asked "Which of the following video game systems have you ever owned?", whereas according to the original article reporting on the Aug. 2014 survey "it asked only what consumers currently owned at the time of the survey."
I would like to see the full data on that Aug. 2014 survey just to see a detailed breakdown. Unfortunately, none of the news outlets reporting on it link back to Nielsen, and Nielsen's website doesn't appear to have it, either, so my curiosity will remain unsatisfied.