The front page actually trumpets the article as "The Myth About Exercise", however upon inspection you find that it's simply not true.
There is absolutely no doubt that losing weight requires that more calories be expended than are consumed. However, amid all the nonsensical and ill-explained studies this simple message gets lost among various anecdotes that relate to individuals working out and then consuming muffins.
Did it not occur to the investigators that bad habits do not make good correlations? It should be obvious to everyone that you cannot work out and then indulge in all the food that causes more calories be consumed. Although I'm afraid the message being conveyed is that all food is somehow equal so working out can't help.
I can't begin to fathom the reasoning that would lead someone to think that an hour's workout can be successful if it is followed-up with an ice cream sundae.
So let's eliminate the nonsense.
If you can't figure out that junk food is bad for you, then there's nothing that can ever help you. If you haven't figured out that more activity can help burn calories then you are also beyond help.
Fat and alcohol are higher in calories than carbohydrates and proteins, so do the math.
Exercise is NOT punishment and food is NOT a reward. Food is fuel and if you purchased gas for your car and each time you bought an extra 5 gallons to put in a gas container, you wouldn't be surprised to find a garage full of gas containers that you never use. Similarly with food. If you consume more than you can use, you end up with the equivalent of a garage full of fat.
Similarly you cannot diet and achieve any kind of results without corresponding activity to build muscle and redirect metabolism in support of muscle. Similarly consider the following ridiculous quote:
"...a pound of muscle burns approximately six calories a day in a resting body, compared with the two calories that a pound of fat burns."
It then goes on to say that if that 10 pounds of fat were converted to muscle, it would only result in allowing 40 more calories per day without weight gain.
What?!? That would be true, if the only thing you ever did was hibernate all year long. The operative phrase is "resting body". Since you aren't resting, your metabolism must increase to maintain the muscle and increased activities will allow for more strenous activities which will also burn calories.
Most importantly, the article makes the classic mistake of equating "weight" with the more nebulous result that most people actually envision; shape. In other words, "weight", in and of itself, is a completely meaningless metric. Generally when people want to "lose weight", they really want to lose inches and gain a more shapely body regardless of what the weight actually is.
This leads to the erroneous conclusion that weight and shape are related to each other. Muscle mass and the water it holds may be heavier than fat, so the scale is not a true indicator of how one looks. The majority of people wouldn't mind weighing more if they lost several inches from their waist or (pick your particular area).
What is probably the most amazing aspect of this, is that of all the species on this planet, it should be the "most intelligent" that has the greatest difficulty in figuring out how it should eat. It isn't that hard:
1. Increased activity to burn calories. Exercise is good to emphasize building muscle.
2. Reduce high calorie foods, especially those that metabolize too quickly (i.e. sugars).
3. Stay away from alcohol.
It is highly unlikely that your hunger levels will increase from exercise, unless you're already restricting calories from dieting. In such a situation you can readily consume nearly as much as you'd like in fruits which will be more than adequate in compensating for any energy needs and do not readily translate into fat. In addition, it's probably a major improvement over that muffin or pizza.
I remember that article as well, and even though I agree it was a bit too anecdotal, I was glad to see ideas that are all too often shielded from the public.
The rise of gyms is a public health dream come true. Over a matter of decades they have gone from specialized Olympic-level training facilities to a cornerstone in modern society. From a public health perspective, it parallels the steep drop in cigarette smoking since the 70's. The only difference is that while we have seen a dramatic decline in lung cancer, we have not seen a decline in obesity. This is cause for at least a grain of suspicion about exercise.
The article further brings intelligent challenges to the calories in/calories out paradigm for weight loss, which I've always found suspect. In a technical sense, expending more calories than you eat will produce weight loss; but in a technical sense as well, weight equals mass times gravity.
Of greater import are the moderating variables in caloric expenditure - metabolism, as you point out; but also quality of calories eaten, and quality of calories expended (eg, type of exercise). Technically you could reduce your calorie intake but just eat cotton candy; you can also get a cold and lose weight. Further interesting moderators are discussed in the article. It's common sense that you can't just pig out on an ice cream sundae after exercise; but the concern raised in the article is most subtle, it's that exercise just slightly lowers such appetite inhibitions towards eating unhealthfully. Over time, this could have an effect which could be more insidious and sly - say, a few extra potato chips after every workout. Perhaps of more import are to bring attention to decision making processes, rather than calorie-counting.
When things are simplified to just calories in/calories out, I'm suspicious that the paradigm loses more than it gains, producing a net-loss in meaning ;)