I've been watching a number of Mao's past competitive and exhibition skates recently, she is an amazing skater! I'm going to ask a stupid question, but I've read that Mao has been reworking her jumps recently. Why? Her jumps are lovely and her 3A is just a delight to behold.
Her jumps became much better now, but she had severe issues with her jumping technique which sadly were ignored in her junior days, and they influenced her consistency when her body matured.
The most noticeable was "hammer toe" (borrowing te explanation from Wikipedia: hammer toe occurs when the free leg rises unusually high, typically near (in some cases above) hip height, before descending to strike the ice. This can make the jump easier to rotate but sacrifices height and some control).
As a young skater, Mao depended on the quick body rotation to fully rotate her jumps. When she started growing, it became more difficult for her to rotate her flip and Lutz jumps, and the "hammer toe" started becoming more and more pronounced:
http://jelly.ionic.pl/upload/3f-take-off.jpg
Because lifting the free leg to that position takes time and influences the skater's balance, the whole jump is affected. Instead of using the momentum gained during the mohawk turn, after executing the turn Mao used to slow down, trace a curve with the inside edge while lifting her free leg and getting arms in the position, and only then slam the free leg down and execute the take-off.
http://jelly.ionic.pl/upload/mao_2010oly_3f.gif
Note that executing the curve caused Mao to pick in such a way that her assisting leg was no longer aligned with the "edge" one; but instead placed inside the curve so that her entire body was tilted to the inside. This body tilt made it extremely difficult for her to maintain an outside edge while attempting 3Lz and, in 09-10 season, caused her flips to become unstable (as shown in the gif above).
Now here's a gif from her practice session:
http://jelly.ionic.pl/upload/mao_3f_nag_01_nor.gif
Note that she takes effort not to drop her upper body / lift the free leg too high and takes off immediately after the mohawk turn.
Another thing Mao's team is working on is to have more speed going into the jumps, to improve ice coverage and to avoid underrotations. This obviously will take more time, and so far is the most obvious while looking at her 2A:
http://jelly.ionic.pl/upload/mao4cc2a.gif
It used to be more "up-and-down", now it covers much more ice.
Then there's 3S, which she now includes in her Chopin's Waltz to improve consistency. According to the reports, she nailed every attempt this summer.
Mao has been "reduced" to just an "above average skater" on protocol because her jumps and consistency have gone awry. You can't expect her to be a viable competitor when she can't get herself together. Mao Asada lost to herself and her own strategies. Talent-wise, she and Yu-na are on equal plane, but Yu-na is the far superior student of CoP.
I've got to agree with dlgpffps. Even aside from the jumps, Mao's programs simply weren't CoP-smart. If you read the PCS guidelines, you'll notice that CoP awards programs with variations in mood and tempo, good ice coverage, a choreo that interprets the changes in the music, various linking elements and transitions. Mao's programs were set to gloomy, monotonous music with little change in tempo and mood, the transitions were scarce (aside from lovely reverse spreadeagle into 2A and nice steps into 3T in the LP) and all the interesting choreography was reserved for the long step sequences. To add insult to injury, the programs (and the costumes

)didn't even take advantage of Mao's greatest strenghts: her beautiful posture, elegant positions and arm movements. On the contrary, Mao mentioned that she found it extremely difficult to skate to Masquerade and Bells, because of the fast, monotonous tempo and no space to breathe. As TAT said, she was supposed to "Overcome" the music.
YuNa's programs, OTOH, were extremely CoP-smart and suited her style and strenghts. She didn't have to "overcome" the music; the music carried her.
I don't think the judges were bent on holding Mao down - they gave her equal or higher scores for spins, spirals and step sequences.
Mao put all her eggs in the 3A basket, let her other skills deteriorate
I don't think it's fair to accuse Mao of depending on 3A too much. Mao herself always wanted to have a full set of clean triples. A part of her difference of opinions with Arutunian was the 3Lz dispute - Mao wanted to fix it ASAP and even started to work on it on her own, Rafael believed it wasn't fair to introduce the flutz deduction in the first place.
TAT went to the other extreme - she believed that Mao is an incredible jumping talent and will be able to execute upper-tier jumps even with her old technique. To give TAT credit, she brought back Mao's 3S and improved her edges. But otherwise she concentrated on the upper tier jumps and combos (3A-3T, 3F-3T, 3L-3L, 3L-3L-2L 3F-3L-2L, 'Tano jumps...) without fixing the hammer toe first. Some of these combos didn't even make sense CoP-wise - why waste time practicing 3L-3L, if the planned repeated jumps were 3A and 3F? And 3Lz from that time was wonky, to say the least - while, technically, it took off the outside edge, the body position on the take-off was rather unconventional:
mao_lutz_2009_sp.gif
And then Mao hit a minor growth spurt. The hammer toe-ish jumps went AWOL - 3Lz was a goner and 3F became unstable. There was an interesting documentary aired after the Olympics which showed clips from Mao's practices with TAT and alone before the Olympics and it really showed how much she struggled with 3F jump. I can't say how impressed I am that she managed to get it almost consistent in time for Olys and Worlds (the only mistake being the turn-out in the Olympic LP).
Now because 3F was shaky at best, she could no longer do 3F-3T or 3F-3L. Because of that, she could no longer do 3S - it was more CoP-efficient to do solo 3L. And 3A remained in her programs not so much because Mao ignored other jumps in favour of it (judging from the practice clips she actually spent the most time practicing the elusive 3F), but because next to 3L and 2A it was actually the most consistent jump for her.
Take Miki Ando as another example. She is arguably the lesser skater of the two, but still beat Mao in every competition in 10-11 because she was CoP-smart.
And here I must disagree a bit. I liked Miki's programs and spins in the 10-11 season, and I think it was smart of her to let go of 3F and 3-3s for one season - she gets injured so often that she could end up burning out way before Sochi if she plans to continue until them.
But such a strategy wouldn't work in the long term. It's enough to look at the results from 4CC LP. Miki was flawless (not one -GoE) - Mao had multiple mistakes (on 3Lz, 3F-2L-2L, 2A-3T and 3S) and their scores were barely 1,87 apart. It would be enugh for Mao to land 3S cleanly to win the LP (and Mao already proved this off-season that 3S became much more consistent for her).
In that particular season Mao wasn't aiming for immediate win - she just started the process of making her programs and elements CoP-efficient. And, as Mao herself said, it might take a season, or two, or three, but she is willing to sacrifice temporary results to achieve her final goal.