Well, during my long nights before sleep, I have on more than one occasion gone ahead and tested my reaction times. However, I’ve always been struck by how far below the bell-curve I am when it comes to the average– the average being around .2s.
I average at .3, .26 if I concentrate.
Now, it may be my ego convincing my consciousness that there is something amiss here, but I have 2 things that I think may be affecting the testing data I’ve seen-
1. Lag
2. Prediction
The most common form of test I’ve seen is one in which you are to wait for a light or other indicator change, red to green per se, and you are to click or press a button as soon as you can.
The most obvious skew here is that users with shoddy internet access could have either consistently or unevenly lower results.
However I would think if anything this effect would drop the average, even if it can account for a single person’s low score.
The second alteration I can see would be people attempting to preempt the clock, and in doing so get ridiculous times of .1 or less. Even with min-time limits, a lucky series of clicks, or a well-crafted system for timing probabilities can easily produce results that would destroy the bell-curve.
Testing the second point, in under ten minutes I was able to get 4 consecutive times in under .15s, one of which at .03s. But this is mostly because there is no feasible way to prevent false-start statistical skewing so long as they manage to get lucky– there is no system in place for many of these tests to prevent this, and even if there were, (deducting .5s for false-starts for example), one would simply have to restart whenever they mistime a click, and eventually a highly-above-average result is all but guaranteed.
now gimme my Ph.d.