The Empirical Rule is just a really basic rule of thumb for estimating the width of a bell curve based on standard deviation, or estimating standard deviation based on a bell curve (divide the width of the bell by 4).
This is basically just like the Empirical Rule except that it works for any distribution, not just bell-shaped normal-ish distributions. It even works for bimodal distributions, though it's so vague that it barely matters.
If you do not have an account, you should get one, because it is awesome! You can save a playlist for each test or each chapter, and save your "greatest hits" into a "watch right before the final" list (not that we recommend cramming, but when in Rome...)