I just want to apologize for my bad English, but I will hope you understand my proposal.
I admit at once that it seems to me that SM is somewhat stronger than other factions. And I see on the forum that I'm not the only one who thinks so. But since I understand that not one of my arguments or somebody else's argument will not be taken into account until they are confirmed by statistics. But how so it turns out that the percentage of wins on the statistics does not show the superiority of the SM. Perhaps the developers are right and just some of us did not fully understand how to confront the SM faction. But let us also consider the human factor that may violate the objectivity of statistics.
First, SM is often chosen by the first faction, so we can assume that there are many inexperienced players there.
Secondly, people want to enjoy the game by winning matches. But if a player often loses for a faction in most cases, he will stop playing for it. Thus, natural selection leaves only those who are able to maintain a competitive game for the faction for which he plays.
Is it possible to disregard these factors in the processing of statistical data? Of course you can. To do this, you just need to exclude people who play only for one faction.
How it should look like. First we take a lot of players who played for the last month for more than one faction. We exclude from them those who have a percentage of the game of one faction more than 70 (ideally it would be 55-60% but the base of the players I think this will not allow). To exclude those who specialize exclusively on one race. Next, consider the average for the pair:
1. Orc SM
2. Eldar SM.
For example, we get
1. 45% 55%
2. 47% 58%
We take the power of CM for a unit, and we calculate the ratio.
1. 0.82 1
2. 0.81 1
So we get the value by which we can already adjust the balance.
Orc: SM: Eldar
0.82: 1: 0.81.