ENTRY 54
The SPEED tests are now complete. For the most part, all of my discoveries fit my predicted models. However, I have come to some…problematic conclusions.
SPEED is relatively unresponsive to environmental factors, but there are some cases where a subject's SPEED will shift slightly. For instance, H2 responds very well to challenges. If she is pushed to a limit that is just beyond her normal capabilities, she will rise to the occasion and eke out an extra point of SPEED. In contrast, some of the 1's do not respond well to that kind of stress. K1 and O1 are particularly notorious examples – when pressured, their SPEED noticeably drops.
These are just some of the more basic cases. Some subjects simply do not respond to pressure and expectation. Others respond in wildly different ways depending on the situation. As far as I can tell, there is no common link between any of these cases. Personality likely has something to do with it, but I can't study the nebulous, undefinable factor that is "personality". Even the humans seem to be having problems figuring that one out.
This isn't even the only variable in the SPEED equations. It is rare, but a subject will occasionally experience a day with a significantly higher or lower SPEED than usual. The next, they will be back to normal. It's baffling. It's frustrating.
Unfortunately, finding the cause of these SPEED variations could require looking into the finer details of the subjects' magic systems. I have already done this extensively in my biological survey of the subjects, but there is only so much one can learn from spectergraph scans and magic samples. Sometimes, one needs to open up the body, observe the lay lines directly, peel apart the different structures of the nodes…
Of course, I would never resort to such measures. They are extraordinarily risky and entirely unnecessary. The SPEED variances may frustrate me, but the overall changes are small and insignificant. I can still approximate the subjects' SPEED scores, and I can do so with confidence.
It worries me, however. The ATK tests are next, and ATK is a particularly difficult stat to pin down. What if my controlled environment is not "controlled" enough? What if I cannot draw any meaningful conclusions due to the inconsistent nature of the data? What measures might I have to resort to in order to find the answers I require?
…I suppose it's foolish to attempt crossing the bridge before one even finds the river. All I can do is perform the ATK tests to the best of my ability. If my best is not good enough, then…somehow, I'll find a way. Somehow.
