Timing Upgrades

Upgrades are among the more abstract aspects of Starcraft II. They are certainly useful and often necessary to win matches, but beginners and even intermediate level Starcraft players have difficulty with implementing them into their game plan. When should I upgrade infantry weapons? Should I put down two Evolution Chambers or just one?

There’s no easy answers to these questions because the answer is universally “it depends”. Which race are you playing against? What composition are you going for? What composition is your opponent going for? Are you playing a macro game or do you want to end the game soon? The correct answer depends on these questions, and the answer might not matter at all if you don’t also micro and correctly position units on top of upgrades.

If you are lucky, the upgrade question is answered by your strategy. If you are going for Cloaked Banshee harassment, then you probably know to get cloak the moment the Starport hits that Tech Lab. Blink Stalkers, Burrowed Roaches, and Blue-Flame Hellion strategies are similar in this way.

 

Hmm. I wonder what upgrade I should get first?

 

Questions concerning upgrades arise more commonly in “standard” gameplay, when your entire strategy doesn’t hinge on one unit / upgrade combination. You can watch the pros and look at their upgrade timings, but that presents its own challenges. Pro players might have ideal timings for specific upgrades, but they tend to play reactively. Pros wait until they understand their opponent’s strategy, then act accordingly. Some early upgrades are standard, but implementing mid-game and late-game upgrades hinges on how the last stage of the game played out.

When you’re out there on the ladder and disruptions force your timings off, it’s good to have some general rules of when and what to upgrade.

Note: The following rules assume standard macro play (no all-ins).

Rule 1: The Relevancy Rule

Get upgrades that benefit your army. Before you scoff and roll your eyes at such an obvious statement, allow me to elaborate. Yes, even the lowliest beginner would never research Concussive Shells if he isn’t building Marauders, but it’s more common to see the opposite problem: players failing to research beneficial tech because they didn’t think through their army composition.

In a recent game of mine, I wanted to try Baneling drops against a Protoss opponent. I researched some Baneling damage upgrades and got Ventral Sacs, but I didn’t think about how that army would look when I actually tried to use it in a fight. Thus, I forgot Overlord speed. My Banelings could fly over force fields, but not quickly enough to be useful.

This rule only deals with upgrades tangentially. It’s actually about having a plan. Think about the army you’re building as you play. Think about how it will engage the enemy. Also think about what that army will look like as you enter the next stage of the game. Do this and it becomes obvious what upgrades you want.

Rule 2: The Investment Rule

Let’s say you’re playing a Terran vs. Terran match. Your opponent goes for cloaked Banshees. You, being ever so smart, build some Missile Turrets around your mineral line so you don’t have to waste your Orbital energy on scans. Good on you, guy. But to make those Turrets, you need an Engineering Bay. Now your turrets are up, and you’re thinking about upgrades.

“But I usually don’t get upgrades until later in the game,” you think. “What do I do?”

You adapt. Start those upgrades ASAP.

 

It's not like that Engineering Bay has anything better to do.

 

The investment rule is simple enough: if you have already spent money on a building that can get you upgrades, you might as well spend money to actually make the upgrades. The Engineering Bay, for instance, costs you 125 minerals to put down, so not using it for the purpose of upgrades means that those 125 minerals aren’t helping you as much as they can. Start your upgrades now and keep ‘em coming.

Think about this as it applies to some standard strategies. A Forge Fast Expand is commonly followed up with an early +1 weapons upgrade. The Protoss attack that hits when the weapons upgrade finishes is quite strong.

The gods of Starcraft didn’t magically make it so a +1 weapons attack was good. It just makes sense. The Protoss player has to put down a Forge in order to safely expand. Skipping the upgrade means the Forge isn't utilized fully. A One-Gate Expand strategy, on the other hand, doesn’t have an early weapons upgrade. That’s not because the upgrade is less useful, but because the Forge is unnecessary.

 

Early Zergling speed operates on the same basic principle.

 

Rule 3: The Army Rule

Sometimes, you’re not forced to put down an early Engineering Bay, Forge, or Evolution Chamber. You need them, of course, but it’s difficult to judge when they’re actually worth the money. A quick look at a unit tester will reveal the sad truth: until your army is a reasonable size, upgrades give you less bang for your buck than extra units.

Judging the exact army size when your upgrades become cost efficient is a tricky business. It depends on the unit mix out on the field, both for you and your opponent. In addition, level one weapon upgrades take 160 seconds to finish, so your analysis needs to incorporate what your army will look like nearly three minutes into the future.

That’s not an impossible prediction to make, but it requires substantial game knowledge. More substantial than I possess, in any case. For that reason, I abide by the “Army Rule”.

In short, you build upgrade structures when you build your army. An army isn’t six Zerglings used for scouting. It isn’t even five Helions for roasting workers. Those are units to keep your opponent distracted.

An army is that mess of units I told you to envision when we were back at Rule 1. It is strong enough to do serious damage, or even win you the game. If you’re still not sure how to define an army, then keep in mind that when you start an army you typically throw down extra production structures (or in the case of Zerg, spend larvae on not-drones). This is the time when you are trying to add power as quickly as possible. Upgrades come out more slowly than units, so by the time your upgrades are done, your army will be large enough to make full use of them.

 

When the upgrades start up. . .

. . . the 'racks go down.

 

Note: Pro players will sometimes start their upgrade structures before they start their army proper. Timing this sort of thing is more difficult than it looks. Early upgrades give you a jump on end-game upgrades, but mean your units will be further delayed. Early upgrades are fine, but you need to scout to make sure you’re safe.

The Rule of Two:

This isn’t actually a rule so much as a guideline. Once I’ve established my economy and have moved into the midgame, I make it a personal goal to have at least two upgrades going at all times. It might be overkill for some situations, but I find that sub-pro level players are far more likely to under-upgrade than over-upgrade.

 

Double Forge Power!

 

The rule isn’t perfect. It doesn’t take into account that certain strategies are more research-heavy than others. Still, I found when I started following this rule my army grew stronger, faster. It also helped me determine how many upgrade structures I needed.

Just as a bonus, this method also forced me to pay more attention to where I was on the tech tree. If you’re the kind of player who tends to sit at mid-tier tech for the entire game, this rule might help break you of the habit.

And Remember:

Research as many beneficial upgrades as your economy allows. When one upgrade ends, start the next as soon as soon as possible.

When your opponent has killed you, only then do you have my permission to stop upgrading.