Hello all and welcome to my auto-battle guide. This guide exists for one simple reason; to help newer players or players looking to build an auto-battle team get up and going. Within you'll find a simple primer reviewing what autobattling is, why to do it or why not, how to approach it, build a decent team, and hopefully getting underway. If you like this guide and feel it's useful, please give it a sticky request.
Autobattling is a very interesting art within FEH. By pressing that little button control will be taken away from the player and let the game control their characters. Normally this is a pretty terrible idea. It is still a bad idea. However, it is a bad idea with MERIT when it's done in the tempest trials. By opting to autobattle the player can be freed from having to play through a series of five or more maps and, instead, watch TV, play with a friend, play a game, read a book, cook a meal, or many other things, all while earning points for the tempest trial. However this isn't always the easiest thing and this guide is meant to help you out.
Part 1: The A.I. and you.
The A.I. is a special little beast, in that... less than positive phrasing. There's no beating around the bush, it's stupid. It will make frequently dumb decisions and sucks at tactics, but you need to work with it if you want to make a viable team. The A.I. is predictable, however, and that means it can be worked with. Our goal is to minimize it's 'stupid' tendencies and to try and get it to... suck as little as possible. The A.I. will...
Always attack if it can get a kill.
Always attack if there is no other option, even if it means unit death.
Always attempt to use movement skills if possible.
Focus on ending a map at the first chance.
NOT plan ahead.
NOT bait, set-up, or anything else strategic.
WILL enter into infinite loops.
WILL leave units exposed.
These are all terrible things. A player can realize it's best to set up outside the enemy's range for a strike next turn, use a defensive unit to bait out opponents, heal, opt for safety over suicide, and so-forth. However you must deal with this in order to autobattle. As such...
Remove abilities like smite, drawback, and the like. If you catch the computer having entered into an infinite loop, remove the skill that caused it.
Remove attacking weapons from healers (or set them up with skills like Breath of Life)
Not pick units which require strategy to use properly over more... basic... units.
A frail, but powerful and ranged unit like Brave Lyn might be exceptional in the players hands but the computer will likely have them rush out into the open for a quick kill at which point they'll be exposed and quickly killed on map 1 leaving you at a disadvantage for the remaining maps. As such units that are frail (and not healers) are NOT ADVISED if possible. The A.I. simply can't use them properly. So, in simple terms...
The A.I. is a dunce.
Don't use units that require thought and planning on your auto-battle team.
Remove any ability that could result in an infinite loop.
Try to focus on simpler units over more complex ones.
Part 2: Building your team.
When it comes to actually building your team, it's important to remember that you are not building a team for the arena or even for PvE challenges. Your team is being built to last over a series of multiple maps without player input. Every player will have different units to top it off, so saying something like 'always put Hector on your team' is meaningless since many players don't have Hector. So, instead, I'm going to break down the various aspects and explain what you want.
A) Movement: There are four kinds of movement. Infantry are the baseline, moving at two squares per round and needing extra movement for forests but are unaffected by things like trenches. Armored units move at 1 space a turn but tend to be fairly strong and unaffected by forests and trenches. Mounted units move at three per turn but can't enter forests and struggle with trenches. Flying units move two spaces and are unaffected, movement-wise, by terrain BUT HAVE AN INNATE WEAKNESS TO BOWS! If you are making a team, DO NOT USE FLIERS! The player can deal with their weaknesses but the computer can not. All it takes is one screw-up and, suddenly, your flying unit is dead. The only reasons to ever use a flier are if they are the bonus unit and you have no others or if you don't have enough **** or ***** units to fill out a team. Mounted units can be a bit spotty as well since a simple forest can neuter them. By no means should you avoid them, just be careful. Infantry units are nothing special on the whole. Armored units, however, are where it's at IF you can get a solid team of them. Most people are not pumping Hectors, Winter Tharja's, and distant-counter Effie's. If you can get a solid team that reinforces each other, armored is the way to go. Otherwise, avoid fliers if at all possible.
B) Weapon type. The maps leading up to the final one are random. Only the final map will have anything resembling a weapon typing so it's probably not the best idea to try and design a team just for the final map. However, building a mono team is also a bad idea. In general, pick two colors (colorless doesn't count) and try to make sure you have at least one unit on the team. This should help at least a little bit. But, if at all possible, try to make sure that there is one staff user on the team. A healer will go a long way in the trials, even if they're just a *** with no weapon. Restoring that HP really adds up in the long run. There are ways around this, but it's never a bad idea to have one healer on your team.
C) Skills: Avoid skills with drawbacks like the plague! Things like Ruby Swords and Triangle Adepts are fine when the player is controlling them, but the computer will gladly suicide your ruby sword triangle adept into the only enemy lance user and ignore the rest of the team with axes. Likewise, skills which do things like damage the user are to be avoided at all costs. Mae's tome might be powerful, but she has to make it last over five maps, possibly without healing. That's not going to end well. Draw-backs in general are bad ideas for the trials no matter how minor or easy the player can work with them. However, some skills are very valuable. Weapons and skills that allow countering at both ranges are obvious godsends along with any weapon or skill that provides healing. This extends to renew and Breath of Life. Normally these skills aren't that good but being able to get back 10 HP every other turn when you're probably going to have 20-40 turns and can't afford to die even once? Awesome. Additionally skills that can reduce incoming damage are VERY valuable! Don't go out and load everyone down with Pavaise, but if you can get a few C-skills to reduce enemy damage or buff allies DEF/RES it will be very helpful in the long run. Also, abilities that deal automatic damage are... mixed. They aren't useless since you want to kill stuff before it can even launch a counter if possible and lower HP helps with that, but damage isn't one of the primary goals of an auto-team.
D) Rarity: Always try to have at least **** units if possible. The stats and skills really help out. Anything lower should only be used if it's the bonus unit. ***** units are the best to use always. Try and get the team full of them if possible, though it's not a game-ender if you have to use **** units.
E) Level: We're looking for at least level 30 if possible. If you don't have level 30 units hit up the training tower and/or do manual runs of the lower ranks of the tempest to get up there.
Put in simple terms, get gold or silver units, make sure none are fliers or have pesky drawbacks, put on your bonus unit and a healer of possible, and profit.
You will also want to build a SECOND team if possible. This teams goal is not to clear trials but, rather, clean-up after the first team. Even with a grade A team you'll lose on occasion. Because the A.I. is dumb and had your mighty sword user attack a Sapphire Lance Triangle Adept while at 2 HP. So you will occasionally fail, and that's when you bust out this team. It's goal is simple. Clean up the mess. Doesn't have to be good; just has to exist and do its job.
Part 3: Playing.
There are three primary difficulties for auto-battles. The hardest difficulty is the first and this should be avoided! A few players may have teams exceptional enough to handle this, but in general you'll want to avoid it as it's begging for the A.I. to fail and waste your time. The second hardest is far more likely. This is where players willing to gamble should go. You'll get more points per run, but also fail more often. If you find your team can't handle this, drop down one more. You'll still earn plenty of points and most certainly clear the 30 difficulty though this is the least efficient way to handle this trial. While lower difficulties can be done going for the highest difficulty your team can handle at least 60% of the time is what you should be doing.
As for set-up, turn off combat animations and support animations, turn on continuous auto-battle, and so-forth. You want to remove these things to speed it up (and it's not like you're watching anyways). Every 5 or 10 minutes look back and check on your team. If you notice them in an infinite loop figure out what you need to do to end it. If your team in struggling maybe watch a run or two to figure out what's going wrong (don't PLAY, just watch. The A.I. needs to handle it). Make sure to have plenty of energy bottles on-hand as well if you want to make this super-fast.
I hope this will help you build a solid auto-battle team for ease of trials. Going through by hand will always be the *most* efficient way, but for those who want to do other things, like work, this should help out a bit.
A new addition to the mix, Grand Conquests allow the player to pick up a large army of units to fling into battle with much more than 4 on the field at one time. This changes how to approach the situation immensely since new elements are on the field.
First off, camps and forts. A new addition to the game, camps and fortresses are unique unit-spawners that can also allow units to teleport between unoccupied squares. When under player control it's easy to simply block off a space by placing a unit on it, but the AI does not do this making the situation harder than it needs to be. Also, their focus tends to be on flipping the camp/fort, which is good... except that they can't assess risk and if their hard-fought turning will simply be undone by the AI next turn, possibly at the cost of a valuable unit and the like.
Secondly, the new system means that where a unit is placed matters even more-so than before. In tempest trials, while it mattered, every unit you sought to place would be fielded and was set. Now certain units may or may not be fielded and the balance is entirely different.
When making your army, do your best to ensure that you do NOT just go through from top to bottom to add units to your roster! The worst thing that can happen is starting a battle with, say, almost entirely red sword users against a rounded army and your mages and green units aren't coming up for another five or six losses. Additionally, the balance is drastically different. Armored units, while perfectly fine in the trials, will struggle to keep up with the main army while fliers don't have to worry about surviving multiple maps and how one bow-user can outright end an otherwise-successful run. As such here is what you should desire.
Movement: Mounted is the best, especially for the AI. The high movement will allow them to cross the map quickly to put pressure on enemy bases. Likewise, flying units, while still unfavorable to utilize in excess, are desirable to have for their ability to cross terrain and hit camps easy. Armored is the WORST due to the low movement and less need for durability.
Weapons: Unlike in the Tempest, where a lack of a healer could be a painful thing, in the GC there is no such worry or restriction. Maybe try to space it out so that you'll have one healer on the field most of the time, but offensive power, being able to ensure that the enemy is off of their camps so you can claim them, is essential. It is foolish to shirk defense for offensive prowess (a newly spawned unit has to walk to the front lines after all), but more defensively-inclined options can be more-or-less safely ignored. If anything movement and range are the two 'key' things. Several mounted mages can quickly close the gap on a fort, clear it, and claim it for other units to warp in for example. As such units who can reliably net kills will be essential.
Difficulty: Due to the number of units deployed the difference in the gap between the player and opponent will be greatly magnified. As such the AI should be able to survive on 30 and 35 with ease assuming you're deploying level 40 units. However, this does NOT hold true for actually CLEARING the map via getting all camps/forts. This is largely because it's poor AI struggles to muscle in on the final fort, especially without losing units which really hurts since there is a 10 turn time limit and they're likely on turn 5 or 6 by the time they even start to move towards the fort. Nevermind clogging passages with mages in front leaving them unable to actually attack the enemy fort (and exposed to fire). Even if the opponent was level 1 the AI would likely not have a 100% clear rate. However...
Map selection: Unlike the tempest trials (which only had the final map set and semi-randomized maps otherwise) the player can choose, at least somewhat, where to fight. If possible it is best to pick maps with minimal bottlenecks and barriers to the end. If you can pick the decent maps and avoid the ones that the AI struggles with you'll boost your clear rate immensely.
In short: Unit movement probably matters the most. Make sure your weapon types are mixed up throughout your battalion, an offensive tilt is nice, and make sure to pick the maps that the AI can handle. Do that and you'll be getting a solid full clear almost all the time.