But it’s a quite nice topic ? For once that we get a constructed thread, aiming for neutrality and comprehensiveness, it’d be a pity to let it go. Now, exchanging constructed thoughts on forums is timing consuming, and it’s understandable for one to wish to trim that down.
And it’s been a pleasure to read it as well. However, there were a significant amount of statements there and there which prevented me from wholeheartedly agreeing with it. So I will comment some of them a bit :
This, the amount of smurfs accounts is indeed the main reason why GM race’s proportions should be interpreted with caution. There are only 200 players in GM for each region, and sometimes (specially in the first weeks of a season). So 50 players among those each bring one or two smurfs with them, it can considerably alter the races’ proportions.
Statistically speaking, this parameter is described as the robustness of some data. If changing a bit one of the source variables greatly affects the final result, then that means the amount of uncertainty regarding this statistic is high, and it should be considered as “fragile” and interpreted with caution. That, IMO, is the case for the GM races’ proportion.
Interesting. This should be put in parallel with the amount of protoss tournament players though, but it’s interesting.
Indeed one unit could be unfair, but that’s compensated by the other unfair units of the other races. Which is why imbalanced units could lead to a balanced result, as long as the imbalances are spread in a balanced manner. 
Debatable. All the players should be taken in account when making a choice, but there is a certain level of execution required in order for your own ingame experience to pretend at starting to be relevant. A terran who can’t split shouldn’t whine about banelings for example. At the other hand, balance shouldn’t be based on only Maru’s level of split ; and the effort of splitting execution should be taken in account by the balance team.
Those would be massive nerfs to bio and also mech harass, and reflect in my opinion an exclusively protoss QoL point of view.
The harass mechanics are a way for the developers to balance the economics mechanics :
- Z has the best economy, followed by P and then T
- Terran has the best harass tools, follow by P and then Z
So if one were to considerably nerf the terrans’ harassing tools, then that would induce that mechanic not compensating the economical gaps anymore. Z or Protoss could get their 70+ workers economy unhindered, and then roll over the opponent. I understand it feels frustrating for a protoss player to lose 10 probes in 5s of inattention, but as you said it yourself, losing 10 probes isn’t game-ending for a protoss, it could even be considered like evening things out against a terran that will land his third base much later than the protoss could. If you lose more than that you screwed up, if you lost less, you’re still in a good position.
That also applies on reverse in PvZ, where the P should try to kill some workers against a macroing Zerg, and will be in a bad position if he doesn’t slow him enough. And so the harass is an additional mechanic to balance the spontaneous economy of the different races.
More discussion about the 3 races’ economy here (look at the workers amounts at 6’00, MULES included in particular).
Dealing with mine drops openers is most of the time a L2P issue, in particular regarding map control and minimap awareness skills. I provided advice to a master league protoss player complaining about this in my community, by analysing one of his replays (in particular regarding pylons and obs placement/sieging), and some weeks later he told me he did much better against those. Never saw him whining about those again ; and that without the least balance changes in this area.
That’s part of why the previous point was debatable. When arguing about imbalances in diamond and even masters, the players should first question their skills not being up to the task rather than the game being imbalanced. There is an awesome series about Harstem doing this kind of analysis, which I can recommend to everyone complaining about balance.
But what winrates ? As far as I know, upon selling the game the original balance team said that ±5% was acceptable for full ladder daily winrates, not for top of the ladder/progamer monthly average. Having PvZ (for example) at 45% one day with bronze to masters players isn’t the same than having progamers not managing better than 45% in a full month, IMO.
Humility is the first step in being able to see one’s own mistakes. And so in being able to get closer from the truth in the future. Hence those are wise words IMO, kudos for your appeased and open attitude, Dallarian ! 
PS : I saw the smurf you referred to in the other thread. if you give me a bit of time to anonymize the data, I think we can answer to your questions by ourselves. 