They are probably leaving out a lot, which seems pretty much inevitable. One thing which doesn't seem to get mentioned is the on time it takes to install from a disc -- with the system quite possibly using more power during this time it could actually compensate for most of the power spent downloading. Then of course, there's background downloading, which they seem to at least mention and maybe account for in their study up to a point, but don't make it clear how that assumption is factored into their data.
One thing that really jumps out is that pretty much the entire environmental cost they calculate for a disc based game is based on the cost of the consumer traveling to buy the game.
For example, they set the overall cost for a disc game at 1.2 kg CO2 equivalents. But then they say that if the same consumer traveled specifically to buy a game, and not to also go shopping for ten other items, the cost would increase to 5.04! And if the consumer traveled by public transit, the total cost would decrease to .895.
So, if you think most people travel to a game store specifically to buy a game, the disc cost should actually be four or five times higher than what they calculate in the study.
If you think most people buy a game when they're already out because they're shopping for other things as well, the number becomes four or five times lower. Either way, it's enough to completely change the findings of the study.
As far as I can tell, they don't go into any detail on how they calculate what the average person does. They seem to have picked an arbitrary number -- that people are buying an average of ten things every time they go out to a game store and purchase a game. And then they divide the total environmental cost of driving to the game store by ten (I think?) because of the multitasking.
There's also this interesting paragraph relating to disc game distribution costs in the united states:
Weber and colleagues (2010) estimate that carbon equivalent emissions of compact disc (CD) production and distribution in the United States is almost 3 times that of PS3 BD production in Europe. The higher emissions for optical disc production in the United States are mainly the result of the differences in transport and also differences in the carbon intensity of energy production and fuel use. Assuming BD production would be similar to CDs, carbon equivalent emissions for BDs in 2010 would only fall categorically below downloading for files above 11 GB in the United States (based on lower bound intensity).
So, in parts of the world with different transport costs the data changes completely. They don't specify whether they're also thinking about the the added environmental cost for the consumers traveling to and from the game store in this paragraph, which as mentioned before makes up the majority of the overall disc cost in their model.
As mentioned by a couple people, rapidly increasing download speeds probably make data from 2010 counterproductive to think about at this point anyway. But I had fun reading it, thanks!