False. That's the conclusion that you've come to in order to support your own biased narrative (self-confirmation bias) due to you running a superficial analysis. You can't make any good (such) conclusions on merit distribution without running a quality analysis on all of groups (something which is not possible to do without some very complex machine learning).
Yes you are right, I forget to add up that all the data presented are for a chance of receiving merit (for probability). This post should not look on how the merit distribution looks like. I do only presented it way that merit is not that hard to acquired for it is not like a lottery that could be difficult to acquired. Sorry for the mess.