This illustration shows how the line reacts at turn over when the energy in the line is not properly dissapated. The line continues to move in a direction 180 degrees from the top leg (line leg) of the loop. It pulls the leader and fly with it and ends up "dumping" as much as 10-12 feet short of the intended target on a long cast.
What causes this "kick" is that most leaders are not designed properly for these sinking lines. Even the store bought kind. They have insufficient mass in the thick "butt" end of the leader to match the mass in the tip of the fly line. Unless the mass in the 2 materials is similar, the enegy in the line cannot effectively transfer to the leader. Since the energy has to go somewhere it causes the line to keep moving in the direction it was last sent, down.
Why isn't this effect seen with floating lines you might ask. Actually, it can be imitated by overpowering a false cast. As the line turns over you will feel a bump as the fly turns over. That bump is the fly being kicked. You can actually see this happen if you look carefully.
So why is this not a problem using a floating line under normal casting power? Because the mass of the tip of the line is usually matched to the mass of the butt section. The rule of thumb for putting a butt section on a floating line (and many imporoperly think this applies also to sinking lines) is to use a mono line that is approximately 2/3 the diameter of the line it is being attached to. As it tuns out, the tips of floating lines are about 50% larger in diameter than a piece of mono line of the same mass. That is where the rule comes from and most shop personnel would know this rule.
With a sinking line, when you attempt to match the mass of both lines you start to run into prblems of having to use a very large diameter mono line in order to match the mass. With a fast sink rate line you might find that you need to use 80# test line or greater in order to get proper transfer of energy. But using mono that thick causes 2 obvious problems: First the mono is so thick it is hard to knot to the thin fly line and second you still have the same issue of matching masses. Only this time it is from the leader to the butt section. So all you have managed to do is move the problem a few inches down the line.
So how is the problem solved? First, use a braided loop at the end of the fly line. Instructions for braided loops can be found here. Second, use a leader that is properly designed to handle the difference in mass in the materials, while effectively transferring energy into continuously smaller diameter sections allowing good turn over without the kick.