Lots of thanks to famousringo for additional analysis and lapsed_gamer for graph math work.

Kenology recently posted a thread about a Nielsen graph being wrong.

The graph in question is just a number of total minutes.

The graph shows 360 higher than it which means that at around 4/5ths the install base, 360 gamers game about 20-30% more. It is important to note that wii had a big christmas which accounts for higher total minutes of gaming in January.

According to this graph: (analysis by lapsed_gamer)

Wii: 21% / 21M = 1.00% total playtime per million consoles
360: 18% / 16.2M = 1.11% total playtime per million consoles
PS3: 9% / 7.8M = 1.15% total playtime per million consoles

Conclusions:
According to this data the PS3 is played most per console with the 360 3.5% back and the Wii 13% back.

With wii hovering at 20% in January, it means that 1 out of every 5 minutes that people in america use their consoles are on the wii. You can see how install base is an important factor in this.

If you look back to the original Nielsen thread it showed these numbers roughly for how many minutes the average console owner  user games per month:

PS3 - 1067
360 - 1150
Wii - 540

A lot of you may be wondering if the math is wrong because it doesn't seem to coincide correctly with the graph. To answer that question you need to know how a Nielsen family works.

There is a box that registers what you are doing with the tv and who you are by making you enter a code for yourself when you start and end using the tv. Everyone in the family is given a code, as well as extra codes are available for guests.

Now, there does seem to be some slight difference between the numbers in the graph and the numbers that I've listed. It is more than likely that the graph is accounting for multiple users playing at the same time. Nielsen also works by totaling numbers rather than dividing the time by the amount of viewers. This means that 4 people playing wii for one hour, would equal 1 person playing for four hours by themselves. Naturally, since wii is meant to be more social compared to the FPS stations, there will be a higher total play time, but less actual play time per person.

Here is the gaming times

I'm not going to be posting in Ken's Nielsen correction page any longer because I don't want to keep bumping false information.

The numbers provided by Nielsen both originally and now, show that while wii's are selling, they are not played as much.

xbox 360 rules,.any way ?

not convinced with the asumptions, but good analysis

You should include lasped_gamer's analysis that the average 360 is played 10% more than the Wii, and the PS3 3% more than the 360. You should also replace the term "average console owner" with "average console user," since the Nielsen report describes them as user numbers rather than owner numbers.

User averages can be extremely misleading. Having more than one person in a household playing games ends up looking like a weakness if that person only spends a small amount of time gaming. For example, a Wii household with a main user playing 10 hours a week and a second user playing one hour a week looks worse than a single PS3 gamer playing 11 hours a week, even though the PS3 user is only gaming 10% more than the main Wii user and both consoles are being used just as much. The result looks the same as your average user analysis: The average Wii user plays half as much.