TF2 can sometimes be a little like Real Life™. There can actually be too much of a good thing. Not to mention too much of a bad thing. And since it’s “too much” even the good thing becomes basically a bad thing, savvy?
IMHO that applies very much to the numbers of specific classes used in a TF2 game, because - as indicated above - some classes are either too effective in higher numbers or not effective enough.
A very good example for the “too effective"-category are naturally our two powerhouse classes, soldier and demo. Imagine having four or five of those on a team and you can imagine the havoc they can wreck. It’s not without reason that leagues restrict the demo to one per team; and I could imagine if the competitive scene would play formats with larger teams they would probably put a limit on the soldier as well (though not necessarily to 1) [1].
But higher numbers can also be very effective with other classes.
I once played a game on Badlands where the enemy team went all scout. They would move with such an insane speed (not only for travelling, but with doubled capping time they capped CPs within seconds) that it was almost impossible to react appropriately. Of course the first round they took us by surprise, but even on the following two rounds they applied that tactic successfully.
I’m not suggesting such an attack could not be countered, but that would require a higher level of coordination than you usually have on a pub.
Similarly a high concentration of engies can be a problem. For many situations a higher number of engies is rather a liability than an asset, but if the team is playing defense, the engies are smart and cooperating, and the rest of the team is defending accordingly (e.g. covering blind spots of sentries, extensively spy checking and countering ubers) these engies can be highly effective.
But of course there’s the flipside to all this.
There are classes whose effectiveness does not relate linearly to their count (meaning that 4 are not twice as effective than 2).
The prime examples here are snipers and engies (when not primarily defensive), and to a slightly lesser degree spies.
All these are very specific classes, sniper and engy more of a supporting nature and the spy as a behind-the-frontline force. All of them are not too well suited to carry a fight on CP-maps or as attack on an A/D map. They can help, and their help can be invaluable, but they can’t do it as primary force [2]. Higher number of those classes also mean that there are lower numbers of other classes, and usually you can bet that on a team with 4 snipers you are missing some classes which you would desperately need to cap a point.
With spies I find that more of those only increase the enemy’s paranoia, usually resulting in more careful behavior and increased spy-checking which only makes it harder for the spies to operate effectively.
Sometimes I even think there can also be too many medics. Not because many of those would in itself be a bad thing, but a medic’s effectiveness depends heavily on the class he teams up with, and if there are too few useful classes to support the value of the medic decreases, especially if you take into account the line up of the rest of the team. And even while this is very situational I doubt a reasonable limit for medics would provide any problems.
Okay, so in summary I’m saying that too many of a class can either be too effective or not effective enough. So what’s the problem, you may ask. In the first case, the team will likely win ... good for them. In the second, bummer. Their fault of being stupid.
I have two problems with that. For one, as so often I’m concerned about in my articles, I think it kills the fun. If one team is clearly able to dominate the other team there is little fun for both sides. Even those on the winning side will only have some superficial fun, since there’s no challenge, no real struggle to prove themselves.
But also I claim that this disrupts the even playing field a game like TF2 should have, actually must have. Let’s take my example of the all-scout team on badlands: Unless the other team decides to imitate this tactic and go all-scout itself (resulting in a brute and stupid race with some deathmatch thrown in) this team needs a higher level of coordination and planning to counter that attack than the all-scout team (whose coordination started and ended with “Let’s all go scout”). Even by equal skill it demands more from the other team than the all-scout team. The playing field is no longer even.
Similarly each concentration of classes need some more coordinated choices to counter. I am strongly convinced that a good mix of classes is always stronger than a concentration of a few classes, but the demands to make that work is higher, probably much higher on the first than on the latter (disregarding skill as a factor).
But even if you disagree with that assessment I may still remind you that even the competitive scene recognizes that as a potential problem. And I’m sure the main reason is to maintain a reasonably well defined even playing field for both sides.
But before you get too worked up about that, allow me to define what IMO such class limits could look like:
In a very simple form I’d suggest to have a class limit of two for servers up to 10 players per team, and a limit of three for larger servers. I really don’t see at which point on any of the maps and modes you would need more than 2 resp. 3 of a single class to make it work. Or conversely I don’t see how this restriction could break a game in any way. Problems usually rather arise from a different level of skill on a team than by not having 4 or 5 soldiers (and I submit that in that case the additional soldiers won’t really help 
). You can fine-tune those limits more to get a better mix. Perhaps only 2 snipers, but allow for 4 soldiers or medics; or only 2 demos to keep the spam at bay. But that’s something for the more motivated and dedicated server admin
Based on all this I actually see little reason which would argue against class limits. One of the downsides one would occasionally encounter is that mediocre or even bad players would take the slots for a certain class and a more skilled player can’t make the difference he would otherwise.
But while this is true I’d argue that for one such a player’s skill is usually not limited to a single class and so he surely can throw around his weight on another class; but also on a normal pub server it’s only a question of time before a slot for a class would become available, so it’s all just a matter of a little patience.
IMO it’s a small sacrifice in turn for a greater advantage for the gameplay in general.
Of course there is some resistance against class limits which is stirred by some other motivation.
Some people feel they are entitled to play whatever class they want, whenever they feel like it, regardless if that choice makes any sense and how that choice will affect the game and their team. These people feel domineered over and act as if a god-given right was taken from them. [3]. But in most cases those people are self-centered assholes who will only think of themselves and how to satisfy their urges (sometimes at any cost).
But there is another golden rule, and that is that server rules are absolute and beyond contestation. And if somebody doesn’t like how the game is played on a particular server he can always move on to one of the hundreds of other TF2 servers.
Server admins shouldn’t shy away to cause some protest and disaffirmation by establishing class limits. I guarantee gameplay in general will improve and consequently attract more like-minded players. And naturally this will not subdue the skill of the players, actually quite in the contrary. Since the playing field is potentially a bit more balanced it should make for more challenging (and hopefully rewarding) games.
Server admins should give it a try. Especially with the engy update hitting us soon. This is the kind of update which will probably seriously disrupt gameplay, and class limits can certainly help to keep the game from degrading to an engineer construction site.
RSS 2.0