Disclaimer: This is only my opinion-- you are welcome to disagree. I don't pretend to assume that my opinion is the only correct answer.
The calculation for WN8 is largely based on damage and kills in a particular game, and at this point ignores spotting damage and tracking assistance. This method of measurement encourages chai sniping, because it doesn't matter if you're spotting your own damage or not-- damage is damage as far as WN8 is concerned. In this way, in my opinion, WN8 is a sub-optimal way of measuring a person's effectiveness and proficiency in battle, and it creates an incentive to farm damage rather than putting oneself in danger (for instance, it's better for your WN8 to farm damage in a light tank from the back than it is to be spotting for others, even if this leads to the loss of the game).
For this reason, I think that base XP (which would ignore the effects of premium time and win/loss bonuses) should be the number that WN8 is based on for each tank (rather than damage an kills). A given tank would have a baseline base XP score that each game would be counted against and the WN8 number would be generated based on the difference between the end score and the baseline score. For instance, 1500 base XP is substantially better than, say, an average game that has something like 600 base XP, so that game would probably be somewhere around 4,000 WN8. In this way, I believe WN8 could have a more complete, holistic measure of a player's proficiency at certain tanks, and the measure would be less prone to padding. Best of all, legacy statistics have always included XP, so it could be possible to change things over to the new measuring methodology without losing data from the games that have been played since 2010.
What say you, wotlabs forum?