In an earlier post, I briefly touched on ways that guns in real life are different from guns in fiction. But it's important to note that not only guns are limited in this way: all weapons, and all damage from weapons, are compromised in terms of realism in almost every game they appear in. But before we get into the "how" issues, let's examine the "why".
War is, to put it lightly, fairly lethal. Soldiers can meet death at any minute, and while there is a huge ratio of shots fired to casualties inflicted, the risk of the situation is not really diminished because of that. However, most media operate on the concept of a small, central cast of characters. This means that one "major" death will drastically alter the dynamic of the show; if you have five characters, losing one is a big deal. Games end up going for the same thing: a few inexplicably powerful protagonists, and a bunch of expendable redshirts. The alternative is either a game like X-COM or a show like Band of Brothers: focused on a unit, with a lot of characters coming in and going out. This "Ship of Theseus" setup allows for the unit to remain central to the plot while still depicting death and replacement accurately. Of course, this requires a game to be about a group, and not an individual, to make sense, which means that single-player FPS games are right out.
So how would a game that "realistically" deals with the threat of death go about things? Well, in real life, military technology is based highly on a stimulus-response scenario: this technology leads to new advancements, so a new technology has to be developed to counter it. There are plenty of examples of this history, from weapons to armor to mobility. This is all a very loose explanation, but its relevance to the current scenario is this: Without the full range of available tactics, and without the ability to improvise based on logical reasoning, games are creating an imbalanced system that often relies on artificial interference to overcome. Oftentimes, these measures are less immersive; in a fantasy setting, the need to balance "soldier" with "mage" results in fighters being given techniques and statistics that don't actually reflect a real human.
To return to the concept of casualties: In most battlefield situations, there's some way to avoid death. This includes camouflage, cover, and concealment. However, in many games these things are largely underrepresented, or used in a different way due to gameplay changes. Shields, for example, are largely used as a simple defensive bonus, rather than the vital life-saving piece of equipment that they are. Think of this: you are carrying a shield - a tool that, in addition to being able to block blows in melee combat, can also be put between yourself and the arrows. It's not just a "defense bonus", it's literally the only thing in that situation that can be counted on to have any effect against the arrows. Armor might help a bit, but "try to soak up a bunch of arrows in your armor" isn't really the best strategy ever devised (at least not until plate armor came along, which not coincidentally resulted in the decline of the shield).
Demon's Souls and Mount & Blade are two games that use shields correctly primarily because of how they present physical movement. In both cases, the direct "action" nature of the game makes it important to maneuver both weapons and armor. Therefore, rather than "my character has a shield, apply passive bonuses", it's up to the player to utilize their shield defensively. This makes it far more visceral - the shield, rather than an accessory, is now a major part of combat. Both games also work because they're highly fatal, and there's very little chance to "soak up" damage. One good hit from a projectile can result in a character's death, which makes the shield even more important.
For gun-based games, the key concepts of combat are related primarily to setting up defensible positions and overcoming the defensible positions set up by the enemy.Neither side wants to be hit, nor do they want to be exposed to enemy fire; therefore, maneuvering and flanking is necessary to break a stalemate. There are a few tools in a commander's arsenal to accomplish this.
One obstacle to this is the third-person camera found in many games; in real life, if you can't stick your head out without getting shot, you can't see the enemy either (unless you have a mirror or something, as seen in a few war movies). In a game with a third person camera, all that's necessary is to rotate the camera so you can see your assailants while your "body" remains safely behind cover. This is one example of a "game feature" changing the tactical dynamic of a situation.
2) Use smokescreens to mask movement. This can be accomplished through the deployment of smoke grenades, a fairly common - but rarely accurate - device in games. The idea here is that creating a field where the enemy can't see means that you can move out in the open (which is otherwise an unpleasant necessity) without them shooting at you. It's not perfect, but it is basically a wall of concealment, and that can never be underestimated. Smoke is used well in a few different games, but one game that deserves commentary is Valkyria Chronicles.
In Valkyria Chronicles, all soldiers are essentially operating under "sector of fire" principles - if an enemy enters your sector, you shoot at them. There's no machine guns, which means no suppression per se, but if a hostile unit moves into another unit's range they're going to get shot at. Smoke breaks this by making it impossible for units to see through. However, units will not attempt to shoot through the smoke, which means that because of the gameplay system "smoke" functions as an impenetrable wall behind which you can run around with impunity. Running through it, of course, is a different matter.
3) Use grenades to flush out enemies from behind cover. The limited throwing range of grenades (much shorter, obviously, than the huge ranges covered by most rifles) makes the prior tactics necessary for its execution; in fact, one key job of a squad's riflemen is picking off enemies trying to close to grenade range. The key advantage of grenades is that they're indirect, and taking advantage of this is a major measure when it comes to breaking stalemates. A longer-ranged alternative to the grenade is the mortar, which fulfills a similar role: launching an explosive projectile over (and behind) cover. Mortars can also be used to launch smoke rounds, which increases their importance in the tactical dynamic.
Many gamers complain about grenade spam in games like Modern Warfare, but in reality it's a perfectly valid strategy. The confounding factors of the game are that everything's much closer together and everyone spawns with a few grenades. Metal Gear Online "dealt" with this by making it so that grenades do barely 1/3rd of a soldier's health, but they do knock people around (and potentially out of cover) which may make up for it.
Even with these measures though, it's important to note that no strategy is 100% able to avoid casualties. What it does do is change probability; suppressing fire makes it less likely that the enemy will be able to return fire, grenades create an opportunity to attack enemies behind cover, and so on. However, keeping the potential for casualties in mind is a major issue as well. The goal of maneuvers is to minimize casualties, rather than believe they can be eliminated.
Some games, like Full Spectrum Warrior, get around this by making one side "static" - the enemy exists to be acted upon, and will not use any tactics of their own. However, in a real fight, both sides will be using appropriate maneuvers and whatever support they can bring to bear. Expecting to win without losing any soldiers is as unfair as expecting to lose all your soldiers without inflicting any enemy casualties (and for the opposing commander, that's exactly what would happen).
In short, this issue can be summed up in two major points:
1: Character attachment means that developers and writers are loathe to let characters die realistically. This is a problem in games because games are an interactive medium, and this means that characters who are in gameplay are going to be (for the most part) exposed to some kind of danger. Therefore, those characters being at no risk makes the game feel artificial, because the logic of the situation is being suspended to protect those characters.
2: The solution is to allow for character death while promoting and including tactics that allow the player to minimize their casualties. One complaint of perma-death games is that it's often very hard to do anything about it, and this is a valid concern when it comes to player involvement. While this cannot always be avoided, as in real life, there are ways that the player can have a hand in reducing casualties. The reason that these tactics were discussed - ranging from medieval to modern - was to show ways in which actual logical systems allow for this to occur without the need for fake game systems. The goal is to create a scenario where everything's internally consistent - soldiers can die, and the player can have a hand in using actual in-universe logic to protect them. This creates a stronger narrative than either soldiers being totally invulnerable or players being forced to use unrealistic bonuses and techniques to "protect" their soldiers ("+5 to defense, now he's immune to bullets!")