As a result of past dialogues with you I've tried different things on the slopes. For instance, out of a discussion with you and Shepherd Wong about getting off the lift I've ridden a snowboard completely flat in a straight line to see if I could do, ridden a tail and nose manual the same way, and even ridden switch judo with a flat board (it works but with a torsionally stiff board I suggest a gentle slope and room to roam). In that spirit, let me suggest that you experiment with different turn shapes and reflect on where the load peaks for you in the turn. For instance, you should be able with judicious use of terrain to even generate your max gs while travelling uphill. I don't care what release or transition you use; simply reflect on where your max gs are in each turn. Also use your ears to note what your edges are telling you on this point. If you don't have people around you and feel comfortable doing so, even look to see at which point in the turn your skis throw the most snow. To be real clear, note where on AVERAGE your skis throw the most snow: if you go through a pile of snowgun sugar on transtion and then hit the loaded part of the turn on hardpack, obviously the snowthrow variance don't count.
Then, watch the Zahrobska video again. 1) Note where the snow shows up off her edges. 2) Reflect on the fact that she is racing, and trying to let the skis run as much as possible. She ain't trying to do C shaped turns. 3) Think where the max gs are in her turns. We could have an angels on a pin theoretical discussion on the meaning of "second half" of a turn, but as someone who's ridden switch judo with learning in mind I'd ask that you not abuse my time nor your own that way. If you agree, on reflection, that when the outside leg starts to bend again, edge angle backs off, and the skis stop throwing snow, that forces are diminishing, and that Zahrobska, or Ligety or Kostelic in the links in this post, is/are then entering a transition to the next turn, then arguing about whether the transition defines the beginning of the next turn or the end of the last become pointless anyway.
Then preferably find some gates or set some pine cone gates and run them with a focus on not being late and trying to let the skis run. Think again where YOUR peak loads are in that case. They won't be when you're going across the hill.
It's not a theory, the forces are real, outside leg is relatively straight during the loaded part of a turn for a reason. Actually the reason that Reilly stated, that biomechanically the straighter leg is much stronger when under load. Pretty basic.
Now think about your ski instructor biases that caused you initially not to see this.
As for Byggmark he's also demonstrating the same thing, but for me to break out where and how would take another paragraph.
A note regarding Reilly's point as to versatility: I am not saying that everyone should go out and try to ski like any of these three skiers. There are drills that involve exagerrated movements that could be similar to what they're displaying, but very few poeple are going to have the ability to do what they do exactly the way they're doing in those videos. If you put any of those 4 in bumps, they would not be sking eactly the same way themselves. The Mill/McNichols video series Reilly linked has Miller in bumps in one segment and that's a good example of his basics staying the same but the specific expression of them varying for circumstances.
The reality is that no matter what you individually may learn from this about what's going on with racer's turns, including the shocking truth that they ski in a pretty biomechanically efficient way, the same old stuff will get repeated on here anyway. Trust Lib Tech but don't trust the internet.