"Just for reference, here's a short list of some of the mistakes to which I refer:
Maneuvers not overlapping
Squares not square -- for instance, rising a bit at the bottom (ditto triangles, and the bottoms of square eights)
Over-turning the inside loop of the eights, and under-turning the outside -- so there's a little "hook" in the flight path at the transition
Loop sizes not matched
Making the exit from the inside of the vertical eight too high, and making up for it by making the exit from the outside too low"
Here's a few comments based on your list of repeating mistakes which you might want to consider, Tim.
First, be sure you are facing the center of each maneuver and your handle is centered on your chest and "generally" at level flight height...this goes hand in hand with Brett's initial comments about placement, head tilt, etc. Your body/arm posture should be as close to "neutral" with respect to the maneuver as possible. This minimizes the potential for asymmetrical inputs due to...well...not doing so!
Ted
Insure your CG is optimally located and your handle angle at neutral is not biased down in the "natural" grip so that equal inputs inside and outside result in equal responses. Remember how carefully you aligned the on board components of your control system to be square and recognize that the handle is simply an additional part of the control system and shouldn't have built in bias.
You can make an aircraft fly "acceptable patterns with less than optional flight trim...particularly in pitch. You can take an airplane with a CG well ahead of the optimum location, open up the spacing on your handle and think "there, that's good enough." Nope! Your CG is best located as far aft as the airplane remains stable throughout the pattern and glides positively when the engine quits. How far aft that will be depends on the area and configuration of the lifting surfaces (wing and tail) but rocket science math isn't necessary to determine the optimum aftmost location.
After first flights with the CG at a sure safe location (~15% of the average chord for any "normal" design) gradually movement the CG aft a quarter inch or so on consecutive flights. If the airplane starts to respond to quickly but still glides positively (no tendency to balloon into a breeze) reduce the handle spacing as necessary to retain a response rate with which you're comfortable. Adjust gradually until the glide tends to get light on the lines and would make you uneasy trying to land in a breeze. Move the CG a quarter inch or so forward until the glide is again comfortable. Adjust handle spacing to once again optimize response rate.
(Quick aside. The optimum CG is important because it will result in the most consistent response rate from calm to very windy conditions. A forward of optimum CG will tend to demand more and more input as winds increase to maintain a safe rate of turn and allow patterns of similar size to best air conditions. An airplane forced to fly patterns with a far forward CG can easily reach a point where patterns are no longer possible in high winds as the forward CG and increasing airspeed as the ship accelerates exceeds the pitch force necessary to no run into the ground in consecutive loops!!)
Now, if flying an optimally CG trimmed ship still exhibits the type of faults you've listed look critically at the items such as posture, handle neutral, etc. If loops are consistently tighter one way than the other or corners overturn consistently one direction check handle neutral and front of chest and centered location of the handle in maneuvers. (p.s. don't be too concerned if your hand and handle rise with the airplane as it transits above level flight (and in concert with your eye/head...keeping roughly the same relative view of handle and airplane in your peripheral vision)
If you've satisfied these criteria and the airplane is still exhibiting the same sort of consistently diverse response rate/entrance/exit it might well be time to consider flap/elevator alignment relative to one another (your hinge lines are already, of course, sealed, right). This part is worth a discussion of its own but, in the most basic sense, if the other considerations are pretty much in place and the ship turns more aggressively inside than outside (the most likely historically) trim in a turn of down elevator at a time and see if the response becomes more equal...if so, another turn (or half turn if you've the capability) at a time until the response is equal both ways.
Sorry, this got way out of hand. Let me know if any of this might apply to your situation and the discussion could be refined a bit. Bottom line is that the airplane/pilot mechanism is more likely to be the major factor in repeated errors than a hitch in your shoulder's get-a-long!
Ted
p.s. Two other trim musts: very little or no rudder offset (don't get Brett started) and leadouts close together (half inch or so) and their centerline exiting the wing tip no more than an inch or a bit less aft of the CG at the wingtip with the airplane dry (if IC...if powered by overly macho AAAs I'm not the guy to ask).