It's often said that energy is lost when making a change of direction cast, like the single spey. I don't want to nit pick but I would say that the energy is not so much lost as dissipated, and that distinction leads to understanding and correction. It's all about keeping the available energy focused.
If we examine the pieces, the lift is the lift, but between that and the d-loop there is a corner around which the line must travel. In overhead casting and the forward Spey cast for that matter we know that straight lines are most efficient and carry focused energy. When we turn the corner on a large change of direction it's common to dissipate the energy a number of ways...
1) the line gliding from the lift wants to continue in it's original upriver path instead of making the turn, widening the d-loop with the anchor too far upriver
2) excessive dip of the rod causing the the energy in the line to dive-bomb the water, creating the dreaded bloody-L or perhaps a piled anchor, and/or an abbreviated d-loop, therefore a weak cast
3) under-or-over rotation thus the d-loop is out of 180 degree alignment to the target, resulting in loss of power even when the anchor was well-placed
4) power is not applied at the right time to form a fully energized d-loop
Problem #1 is worsened by an aggressive 'yank' verses a smooth lift.
The use of a slight dip provides a slow-down of the line, much like feathering the brakes to make a turn in your car. Too much and the anchor will pile up - a little goes a long way. The dip is less necessary when the power application is correct, or not necessary at all depending on line head length.
Problem #3 (over-rotation) is much more common than one would think. I found this to be true via video analysis of people doing single speys. You lose just as much power going too far around as you do by not going around enough, in fact maybe more since there is at least hope of tension in the latter. So if you do everything right and go too far, the cast will have no power due to a violation of Simon G's 180 degree rule.
A weak d-loop is usually caused by a lack of a secondary application of power that achieves the direction change. Keep in mind that the secondary A/P has to overcome the primary lift and sweep path so the key is s-m-o-o-t-h lift and sweep, or what I prefer to call "the carry".
It's important to note that the switch cast does not require a secondary application of power because the initial one does it all. Thus the two casts are not really as similar as they appear. Using the same application of power for these two casts is often the reason for difficulty in mastering the less frequently practiced one of the two. And guess which one that would be...
Things go a lot better when the body moves freely. I use the saying "let the body accomodate the arms, don't make the arms accomodate the body". This is because it's very difficult to maintain a stable load in the rod when the body is stiff and the arms are doing all the turning. If the arms are stiff and the body is turning, the rod will hold a more stable load. A stable load gives you more d-loop. Of course the best motion is a combination of both, well timed and well directed.
Shorter head lengths == easier direction change, simple physics.
Generally, I've found that when the d-loop is aligned and anchor is not forced upriver the cast will reach full line distances even at 90 degrees of change.
There is quite a lot that can be said on this topic, volumes in fact - but I hope these provide a good starting point for debugging the large angle change single spey.
IFFF Certified THCI @ 2005
Capeflyfisher Guide Service
Island Hopper, Guitarist, Incurable Dreamer
and Founder, Worldwide Flyfishing Forum