I would question this point. The captain voting is a subjective choice, not an objective one. And the one with the most votes isn't necessarily the highest-ranked. A simple example: me. I participated in two team tournaments, in one I was captain #1, and in the other, captain #3, although my ranking was clearly far from the top 3. The same hypothetical Gomunguls and LuckyF, based on the voting, could be ranked 20th and 1st, respectively. Therefore, the spread of captains can be considered partially random. This makes your calculations incorrect, or rather, the opposite, and the balance provides an even number of choices (not players). IMHO.

You can look at these things based on the results of past tournaments, but as a rule, everyone is too lazy. You can come up with your own way of verifying, but I suggest comparing the top 4 in the voting with the top 17-20 from the previous озв4 and see what the win rate is for each of these groups of players (not taking into account games within the group, since intra-group play always results in 1 win and 1 loss, which simply brings the indicators closer to 50%). If you want to recalculate based on other assumptions or the number of participants, feel free, but do it yourself:)
So, Top 1-4
maretti 3-0+0-1+2-0+0.5-2.5 = 5.5-3.5
stinger 2.5-2.5
hellight 3-5
amilu 6-2 +3-0 + 0-1 = 9-3
Total 20-14, win rate 59%

top 17-20
vova 3-5
nitro 1-7 +1-0 = 2-7
toga 4-4
aya 2.5-3.5 + 0-3 = 2.5-6.5
Total 11.5 - 22.5, win rate 34%
I may have made a mistake somewhere, but the error is not significant and will not affect the conclusion that captains who took the first 20% of places in the voting have almost twice the win rate compared to those who took the last 20%. And within the interpretation of the results, despite such a difference in the win rates of the captains, 2 teams out of 4 from these improvised groups made it to the playoffs, that is, the chances of the teams were more or less equal, despite the captain factor, and the system with 4 tiers is competitive.

And additionally, I'm not against 5 teams, although in theory, an even number of teams is more balanced. And there are 2 reasons for this:
1) Captains don't understand the strengths of players from the lower tiers, so at least for the 5th tier, you can expect a complete random selection, rather than a conscious choice of the strongest player during early picks.
2) I'm just too lazy to look for 20 people who are worthy of being captains; I barely scraped together 16. Especially considering the absence of such players as maretti and amilu.