Mounted Enemies

That comes across as an incredibly dismissive reply, and isn't actually helpful. I already said the section on mounted combat in the PHB isn't useful to my issues; please don't assume that I just didn't read it. I'll more fully spell out the reasons why it doesn't help me here.

Being mounted as a player and running a mounted creature as a DM are fundamentally different things, largely because of what an "independent" mount entails for a player. If choosing a mount to act independently as a player, you are yielding your direct control to the DM in exchange for the ability of that creature to act as a true ally in combat, with the ability to attack and do damage to your enemies. If instead you choose to control it, its ability to attack is sacrificed in order for you to have direct control over its actions. That is a serious tactical choice, as I see it. For example, if I let a griffin I'm flying be independent, it becomes a serious contender in the combat, but I lose some degree of control over my own movement. I also am allowing the DM (when rationally consistent, assuming he isn't a jerk) to make my mount impede me rather than aid me, as per the examples given in the PHB. My griffin may decide to chow down on the corpse of a fallen foe rather than charge my next enemy, for example, potentially putting me in a tactically difficult situation.

As a DM, no such problems exist. I don't yield any control by choosing to make the mount "independent," because I'll still be in complete control of it as the DM. If I send my hobgoblins mounted on direwolves up against my party of PCs, then the direwolves can act exactly as the hobs would want them to (because, ultimately, I'm in control of both) thus gaining the ability to attack without in any way hampering me tactically. It's essentially cheating. The hobs gain additional movement speed and tactical options, while simultaneously adding allies to the combat. This, if a decent DM is running it, will likely make the encounter with the mounted hobs a significantly harder encounter than an encounter with an equal number of unmounted hobs who are merely *accompanied* by direwolves. And that throws off the encounter difficulty math.

One way to fix this problem is to try to treat the independent mounts the same way you would a mount used by a player. Most of the time they'll do what you want, but sometimes you make them do something stupid. That's hard to make convincing, though. And if you do it often, the players may feel like you're handicapping things in their favor. On the other hand, if your mounts always do exactly what their riders would want, players may start to recognize that your mounted creatures are essentially breaking the rules of the game. And that balance is hard. It's like being a teacher whose own kid is in his class. It's nearly impossible to be completely fair in a situation like that.

Another way to deal with it is to treat the mounts as controlled. They don't get to attack, they just add movement speed and some additional tactical options to the creatures riding them. But then why even add them? What's the point of even giving my hobgoblins direwolves for mounts if they give neither more nor less tactical advantage than a horse would? And now we're REALLY breaking the encounter math. If my direwolf is essentially just adding movement speed to the hobs, then it's basically just free XP waiting for my players to kill it. It isn't adding to the difficulty of the encounter nearly as much as a free direwolf that wasn't being ridden. And if the players are smart enough, they'll realize that and target the wolf first and then the hobgoblin only after the wolf is dead. So the encounter which is calculated as being a difficulty level of, say, 1200 XP (2 hobs, 2 direwolves, x2 modifier for 4 creatures) and thus considered to be a "Hard" encounter for a party of 6th level adventurers is actually a walk in the park for them.

So as far as I can see my options in the rules as written are a) cripple my mounted creatures by treating the mounts as controlled, b) treat the mounts as independent and thus effectively cheat the system, despite attempting to make it fairer to the players as well as I can. But neither of those is ideal. I was writing because I hoped other people had come up with better answers to this question than I already had myself.

Don't treat the encounter XP calculator as a definitive authority on building appropriate encounters. You're essentially asking how the calculator deals with monsters tactics, and it quite simply does not. The actual difficulty of an encounter can change with a number of factors the calculator does not account for, such as terrain, traps, equipment, morale, etc. Tactics, including whether monsters involved in the calculation use one another as mounts, are one such factor.

IMO, your option 1 is best - treat the mounts as independents who always do the tactically right thing. It gives a closer approximation to the intended encounter difficulty calculation than option 2. As a DM, I don't think of it as cheating anymore than I think it's cheating to give my NPC wizard a spell that makes fish fall from the sky. The rules that apply to PCs don't have to apply to NPCs.

No harm no foul. "Apology accepted, Captain Needa."

My reason for posting the RAW is two-fold.

1.) It helps to cite the RAW when discussing something like this, regardless of pre-reading it or not. Not only for the people posting, but for those reading that might have the same question in the future.

2.) If I had a dollar for every time someone said they read something on the subject, then RAW is cited and they say "Oh, I missed/didn't see that part.", I wouldn't need the Powerball. Especially in something like D&D, where various sections of the book cite things for different reasons. The index might have multiple entries for a given topic. Many times people read one section, and miss another. It happens all the time.

On your second point, I think to me it still makes perfect sense to have the player control the independent mount, even given those statements on examples of what the mount might do. A typical mount is trained, that's true, but while it might be trained to take on Orcs without hesitation, a 15' Balor demon might be too much for its animal psyche and it might flee "against your wishes". Typical mounts aren't things like "Summon Creature" spells, where the summoned thing must obey you.

Personally, if you let the mount act independently, you as a player need to then think like a DM when controlling it, and decide how it will act based on what the mount is, and the circumstances. What it might do in one case, might not be what it does in another. Especially if it's just a regular animal.

That's just my personal $0.02. What happens at your table is completely up to you and your DM. I don't think there's a right or wrong here.

My personal approach would be to allow both to be implemented, but to leverage animal handling to bridge the gap between them. If you allow a mount to be independent, but then want to do something that falls into the controlled section then call for an animal handling check. The same would go for a controlled mount that you want to attack this turn, call for an animal handling check to see if the mount is OK with raiding the ire of the demon in front of it or not. As a DM I wouldn't want to fully take control of a players mount if they were to allow it to act independently, but instead carry veto power on what the player says that mount does. Also, I would likely home brew that the mount always goes on the same initiative order as the player it is tied to, so that I keep a little extra sanity and so does the player. Granted this limits some of the independence because the mount can just bail on the initiative turn right before the player is going to swing down and smite the BBEG, haha.

The issue as I see it is that the "controlled"/"independent" dichotomy in the RAW actually doesn't make logical sense. Why buy and maintain a warhorse instead of a vastly cheaper riding horse if you can't actually purposefully direct it to use its attack in combat?

The reason that the designers made the choice they made, as far as I can tell, is because of the way in which a mount's attacks would increase the overall damage output of the PCs. The game is balanced pretty intentionally around a particular approximate damage output at any given level. And (to pick a random example) a tamed griffin ridden by your level 5 warrior is going to throw that damage output off severely. So the easiest way to prevent that problem was for the RAW to exclude the simultaneous use of the mount's attacks along with your own. (As far as I can see, that's the same reason they completely screwed over the Beastmaster subclass for Rangers.)

If you don't care about the RAW game balance being off and are willing to compensate for the difference, there's no reason not to just houserule that "controlled" mounts can still attack. But if you're the kind of DM who wants to the game system to be fairly numerically sound while still allowing the mount to feel like a real advantage in combat, then it isn't quite satisfying. I'll post my own ideas once I'm finished pulling them together.