display | more...

Friends, noders, countrymen. I've come to bury Honour Roll, not to praise it.

In particular, I just want to say that it's not so damn complicated. I am not defending it. Not at all. I want to see it go away as much as almost everyone else does. It has its faults. But the way it works is not so complicated.

There are three things that have to be explained here first:

• Standard deviation is how much the data is spread out, in this case, how much the merit of all users above level 1 is spread out. You can find the current standard deviation taking into account all users above level 1 in the statistics nodelet.

• The unit used to compare your merit to the average merit is standard deviation. Thus, if the average or mean merit right now is 12, the standard deviation is 7, and your own merit is 19, then you're one (1) standard deviation above the average. Note that this unit fluctuates daily like the stock market depending on how the merit data for the entire noder population moves.

The level factor is then 1.00 if you are at the average merit or below. Your level factor then decreases if your merit rises above average, decreasing quickly towards 0.72 at first if you're still within 1 standard deviation of the average. As your merit gets much higher than the average, up until 2 standard deviations, then your level factor decreases more slowly, approaching 0.54. When your merit is 3 standard deviations above the average or more, then your level factor is close enough to 0.50 to effectively be there for all practical purposes (but never actually getting there). So:

```         # of standard deviations above average | level factor
---------------------------------------+--------------
0  or below average      |  1.00
1                        |  0.71
2                        |  0.54
3  or higher             |  0.50 (effectively)

```

And that's all I have to say about that.