The mid-game is tricky; you've survived or even powered through the early stuff, you've got some cities, you've managed to defend yourself, and then... now what? The answer comes a bit later after you feel like you didn't really accomplish anything, and you notice that one or two AI is super-runaway. Not just because of a few techs, but they're in a seriously good spot and they're going to continue until they probably win. That's the problem with mid-game.
So how do you deal with it? Once you've gotten a few cities up and survived the initial stuff, say, around Medieval, you need to take stock.
Military
- Am I going to attack someone?
- Do I have an aggressive OR crazy neighbor (looking at you, Maria)?
- Does an AI that I might battle have units that are 1+ tier (or UU) higher than mine?
Yes? Build more units. Have plenty of ranged, and at least a few Melee units for soaking and a mounted unit or four to help pillage/capture/OHKO archers.
Economy
A common trap of the mid-game is the feeling that, now that things are smoothing out, you can finally build that big backlog of buildings you've been meaning to get. While you do want some, you still have to focus on priorities even if it means sacrificing things in the short term. (I'm not including things like a Monument/Well/Shrine which is probably one of the first things you build anyway)
Must Haves - All Cities
- Market - So much free money, so little build time. Perfect!
- Library - Fill that scientist slot ASAP, always.
- Writer's Guild x3 - Not just in your highest pop cities, but ones that have actual growth potential as well.
- Aquaduct - Only 1/gpt and the sooner you build them, the more powerful their long-term effect.
Situational Must-Haves
- Stone Works - If you can build it. Super cheap, more production.
- Stable - Expensive, so only if you have 2+ pastures to make it count, or if you're going to build a dozen mounted UUs.
- Forge - Always if you have Iron (and copper?) in that city. Otherwise less important, unless you need a workshop.
- University - ASAP in cities that have forest or jungle, less important otherwise (I know it has a GP slot but if it isn't a priority then you just have to wait).
- Workshop - If your city can work 3+ forest/jungle tiles (due to camps, lumbermills, or just blank tiles). Especially important to combo with the University in this kind of city. Otherwise, low priority.
- Harbor - Only if you can actually benefit from it. Will it connect two continents and give you 2+ city connections? Build it! You have an 8 pop island city with -2 isolated penalty? Boo-hoo, too bad, back of the line!
- Colosseum - Its most important feature is "Gives +2 Production to Barracks/Forge/Armory". If your city has any of those, get it (it's cheap), otherwise no.
That's it all the way to the start of the Industrial Era. No, really. Not counting UBs or Temples with max Piety, nothing else helps you setup for the oncoming storm. Yes, you'll build the other stuff but not until later. So now that you've got your shopping list, what do you actually do in order to not regret 100 turns of nothing?
#1 - What's the AI doing?
Your immediate concern are your neighbors. Yes, Russia is in first place and ahead by 8 techs, but she's got a buffer of 2 civs that you can't get through. That means ignore her for now but make a note of what that super AI is trying to steer towards. Meanwhile, figure out what you're going to do with the guy next door. Either buddy up or smash him.
#2 - I built my priorities, now what?
Start by getting more workers and just go crazy on all the things. Make sure you are maxed out on trade routes. Build some settlers or pioneers for those juicy island cities or to grab some more land, but don't build them in your Capitol.
#3 - I did all that NOW WHAT?
What's your plan? Think long term and look at your competition.
I want to believe! You've got a religion? Does Ethiopia have a mega-22 city religion while yours sits at 5? Give up on that plan. Otherwise now is a good time to grab some temples and setup missionaries/inquisitors. Again, only if you've got a religion and beliefs that have a big impact from lots of +faith or +followers.
I want to smash things! Get knights ASAP and borrow a few dozen horses if you have to. Let's see - available immediately at the start of Medieval? Yes! "Cheap" compared to later units? Yes! 4 move? Yes! 24 CS, hey where have I heard that, oh wait, Musketmen have 24 CS too, are an era later, and cost +40% more. Okay, sure melee have cover and defensive terrain blah blah, but your knights have flanking
and the movement to jump in for a flank and then back out, or even attack and back out. Ouch, my knight is down to 30 HP at the start of my turn! So what? Pillage your current tile, move and pillage the next one, and oh wow you still have 1 movement left. Sweet! Also some of your Knights want medic. Also - Cannons. Sweet, Sweet Cannons. Yes Please.
I want shinies! Focus on your Trade Route cities first to maximize gold (Caravanasaries etc), and then worry about things like banks. Money is only as good as the stuff you spend it on, so if you're making it rain then you should invest in a bigger army or helping that high-pop-low-production city to invest some buildings they need. Seriously, if you're sitting on 2k gold in the Renaissance, you're doing it wrong. Spend Spend Spend!
#4 - Help! My scientists are still trying to invent fire while the Dutch can into space!
Yeah, the Dutch are awesome. If you have your other priorities under control, now and only now should you worry about tech as a real priority. The real key to science is population. The biggest numbers don't come from the +3 science a building will give you, and only some of it comes from the specialist slot. The biggest boost is from population and all the "Get X Yield for every X Citizen" buildings. If you have to tech up, build up.
Keep in mind that each city
including puppets increases your tech costs, so if you capture make sure they're worth it (or burn baby burn!), and if you want new cities, make sure that you can jump-start them to help them earn their keep quickly (building investment, trade routes, and/or workers).
"But I literally want to win a space victory!" Okay fine, but my answer is still the same. You need a large population to probably convert into soylent green as fuel if you want to go to space. If you're really behind, snatch up a few cheap techs that your AI civs already know since they're cheaper. You
can build wide and win this way, but make sure you're building wide-and-tall at the same time. In most all cases, your victory isn't going to happen because you decided to pick one in the last 30 turns of the game. It's best decided in mid-game and adjusted as necessary, but this is the time to point your compass and go.
#5 - So important it gets its own number.
Before you research Scientific Theory, do whatever you can to be ready to build Public Schools. That means Universities in all cities regardless of forest/jungle, building up a reserve of faith (if you have Jesuit Reformation), and building up a large bank account (so you can invest for quick builds). Yes, I said don't sit on money, but this is the exception. Make flat gold trades from the AI if you have to, screw your GPT.
Why? You haven't had a Science-per-4 citizens building since the library. Back then, you probably had 3-6 pop in your cities. Now, Public Schools are going up in your cities with 10-20 population. The effect is staggering! As soon as you reach the end of the Industrial Era, the tech costs bloat like crazy. I've tested it before too, and getting late schools is a deathblow. Sure, early Industrial, they might make 8 turn techs cost 7. Big whoop. But by the time you get into the Modern Bloat Monster, those schools can make the difference between 14 turns and 10 turns. The bigger the tech cost, the more every bit of science matters. Yes, sometimes you get more science and you say "oh, same number of turns", but remember you get beaker overflow to the next tech, so it still adds up and saves turns somewhere down the line.
It doesn't have to be your first tech, but it darn well better not come
any later than your third tech, at
most. Admit it, Himjei Castle, the Leaning Tower, and Chicken Pizza were all built by the AI on turn 2 and you never had a shot. You already have Knights, so gunpowder cannons are about as low as you want to dip, barring tech breakthroughs or spies. Stick to the extremely boring top of the tech tree and go go go. The only time I ever delay SciTheory is if I have a shot at building Slater Mill in Steam Power.
#5.5 - So good that I had to edit this in!
Did I mention that Slater Mill is one of the most OP wonders in the game (at this stage at least, considering the investment-to-return ratio)? +5 production is almost a mini-Ironworks. +2 coal can actually matter! That means two more factories later, or two ungodly overpowered seaports. Unless you're playing on some hyped-up resource game, even with a large territory you'll often find yourself sitting on very little coal. It matters.
If you've got a lot of coastal cities with good sea resources, that coal for seaports is stupid OP. You also get a free factory in that city (4 GPT and 1 coal saved) which gives you another +3 production and +1Production-per-4-citizens! And you get that factory two techs early! Dear God, that thing is just insane. Oh, and if you're playing as England then this is the Number One Wonder in the whole game for you. It does everything I said plus gives you your UB factory which is 1Production-per-2-citizens (2x better than regular factory) making it a contender for the most powerful UB in the game. Getting it early can send your production (and thus units/buildings) skyrocketing while everyone else is playing with Lego©.
#6 - Wrapping it up
Mid-game is the foundation on which you build your end-game. If you have to stop a runaway AI, you have to setup your tools in the mid-game. If you have to catch-up, mid-game is the time. The reason mid-game screws people over is simply lack of a plan. Analyze what advantages you have, what the AI has, and what you can do to trump them. That's all. Analyze, plan, win.
