EPTLearning Chess from Starcraft
For a very long time, Chess was a game I had no interest in learning. It didn't
have cool graphics. It wasn't exciting like Goldeneye or MarioKart. It didn't
have Spider Mines. After losing my Broodwar friends to SC2 (with DoTA around
the corner...), after the Korean scene moved to SC2, after my last 1v1 on
Fighting Spirit, and after I stopped visiting TL.net on a daily basis,
Starcraft was all but dead in my mind.
Though I had played a handful of games in my childhood, I never thought about
Chess as more than random movement of goofy pieces around a checkered board. I
knew the names Bobby Fischer, Boris Spassky, Gary Kasparov, and Deep Blue.
Other than that, it wasn't a game that real people played. How wrong I was.
Chess is hardcore and serious business.
Let me start by dispelling the myth that Chess is only played by old Russian
men (and only men, for that matter). These myths are false. The game is
centuries old, and competitive all around the globe by men and women, boys and
girls. While the CSL (Collegiate Starleague) was successful in spreading across
North America in the span of some three years, most campuses probably already
had a Chess club (suffering from lack of membership). Some kids in elementary
school are encouraged to learn the game before they learn basic algebra. Unlike
Starcraft, Chess is a game you can learn for life.
I would like to introduce Chess to the Starcraft player, and hopefully
demonstrate the core similarities between the games. Moving from Starcraft to
Chess with these ideas in mind, I've found my experience at the keyboard to be
transferable to the chessboard.
As games, Chess and Starcraft fundamentally share:
- A balance between micromanagement and macromanagement.
- A time management component.
- Psychological components.
- Appreciation for player style.
In Starcraft, novice players are told first, "Macro macro macro," only to be
overrun by a 4-pool or cannon rush. The sneaky opponent forces the novice to
micro out of an uncomfortable situation. And as we know today, only a
combination of micro and macro will save the game. The importance of
multi-tasking is equally important in Chess. The Scholar's Mate Opening is the
Chess equivalent of the 5-pool (it can adapt to a normal macro game). White
plays the Queen and Bishop immediately, forcing Black to respond precisely. Of
course, we know that 4-pool is bad because it is simple to win if scouted.
Similarly, knowledge of the Scholar's Mate Opening prepares you against its
obvious pitfalls. But there are hundreds of known "trap openings" in Chess.
You cannot prepare for all of them, so there are always opportunities for
"cheese" in Chess. If you are a cheese player, Chess has many opportunities for
abusive play.
Chess strategy is divided into two major branches of thought: Tactical
Knowledge and Positional Knowledge. A tactical Broodwar player chooses 2-hatch
mutalisks or corsair-dt, relying on fast hands and fast decisions to handle
complications. In Chess, the tactical player creates complex puzzles involving
piece exchanges ("I take here, he takes there, but what if he takes there
first?"). Because Chess is usually played with a clock, the more experienced
tactical player will see the outcomes sooner, and walk away with more time (and
maybe some won pieces). In Blitz Chess (less than ten minutes per player for
the entire game), a very common and successful strategy is to play aggressive
openings that force your opponent to waste time solving puzzles, especially
ones that you have already thought through.
And then there are the Positional players. The value of position over tactics
was demonstrated in Broodwar's time by the evolution from a game of protracted
battles beginning early and ending late, to a game of subtle moves and delayed
engagement. Macro, technology, and information succeeded outright force. Chess
underwent the same evolution in the early 20th century, as players like
Nimzovich demonstrated that it is not obligatory to immediately occupy the
central squares with pawns. These classical openings are bloody slugfests, and
typically exhibit an early battle for the center with pawns. Modern Chess
principles amend the old "occupy the center" rule to the simpler, "control the
center." And so you see openings like the King's Indian, where a fiancetto-ed
bishop threatens the center on a diagonal, and advancement of the pawns is
reserved for careful timing. To play a positional game of Chess is to think for
the long term, placing your pieces on squares which may not be active at the
moment, but will become more valuable as the game develops.
I played Terran in Broodwar, so I learned the value of SimCity very early. The
placement of Bunkers, walls, Missile Turrets, Spider Mines: the Terran player
has many positional decisions to make throughout the game.
In Starcraft, unit upgrades play an important role in long-term strategy. The
timing attack is that window of time when your units are marginally stronger
than your opponent's, a sort of "power play." In Chess, the positional player
makes a similar investment; place your pieces on the squares that will become
important in the future. When your pieces are well-coordinated, impressive
combinations (that you could not have foreseen) become possible.
Chess is like real-time strategy games. How you budget your time is
tremendously important. In the opening game, when your moves are the easiest,
you want to move quickly, so a build-order is helpful. As the game grows more
complex in the middlegame, you need to decide when and how you are going to
manage your time. In Starcraft, we have multi-task routines (hotkey cycling)
and training programs to help us learn to manage multiple tasks simultaneously.
For experienced Starcraft players, the ability to execute many actions quickly
is predicated on the knowledge of what those actions are. Novices just aren't
aware that they need to be doing so many things at once, and so they end up
sitting on their hands watching battles unfold. In Chess, you see the same at
the beginner level; players move their pieces into completed undefended
positions, or fail to see upcoming threats, and are swept off the board because
they have no long-term plan for their pieces. Beginners are taught similar
"hotkey cycling"-like techniques: "Before moving, check all of your pieces for
capture." "Before moving, find your opponent's best forcing response to your
move." "Before moving, find a better move."
And for the longest time, my Broodwar game was preoccuppied with all of the
"wrong" principles. Stronger opponents (C+ to my C-) knew how to create
compelling attack timings; so compelling that I hid deeper inside my Terran
turtle shell, and wouldn't come out until I had found the precise weapon in my
arsenal to handle that specific defensive problem. My macro suffered terribly.
I once had the opportunity to play games against an A- Zerg. In every game I
scouted his build perfectly, knew what had to be done, and died immediately to
good mutalisk control. I might have survived if I had one more turret erected,
or if stim and U-238 were completed a few seconds earlier, or if I had a few
additional marines. Even then it would have taken most of my attention to micro
perfectly, but I might have stood a chance.
The problem with players of both games is that we spend too much time trying to
win battles. Chess, for beginners, is predominantly a game of tactics. A
stronger player will look at where your pieces are positioned, and (magically)
find ways to make your pieces disappear. There are simple ways to train this
knowledge outside of games (books, websites), so simple (and boring) that I
don't think it's the best way to improve your game. Fortunately, ninety percent
of sound positions will not have obvious tactical blunders, and only brutal
opponents (computers) will make you pay for these mistakes.
To paraphrase a Chess grandmaster, "Most Chess club players like to attack, but
have no understanding of Chess strategy. Create a position where there are no
attacking chances, and they will have no idea what to do." In my experience,
this is true; aggressive players are not prepared for a long game. Your best
shot is to defend your position before attempting making an attack. Of course
it is still a challenge to survive against superior tactics, just as it is
important not to die to good mutalisk control, but the winning strategy is the
deeper strategy.
This is why beginning Starcraft players are encouraged to think about macro
above all else, even if it means a-clicking all of your battles. The deeper
strategy is to play for the longer game. Tactics are something that you train
-- they can become rusty and out of shape if you don't practice them, but you
can improve them at any stage in your career. Strategy is learned. I may
overlook tactical vulnerabilities in my Chess games and lose a piece now and
then, but I have a hard time forgetting positional themes. Similarly, I might
over-stim my marines or misclick my attack, but I should never forget why I
need to upgrade stimpack! If my minerals should ever accumulate above 1000, it
should be because I forgot to build a supply depot or a barracks, not because I
don't value some 20 additional marines in my army. If this were so, then I
would be lacking deep understanding of the game itself. If my priority is on
the routine of building supply depots and barracks, and not the 20 marines,
then I would only be playing a game of build order memorization. For many
aspiring players (of all games), I believe this is a bad habit holding them
back.
Learning build orders must come with a warning. In Chess, you do not understand
an opening set of moves until you are also aware of all of dangerous variations
and pitfalls surrounding it. Similarly, when your opponent makes an incorrect
move, you are obligated to punish it if you want to maintain your advantage.
If you do not recognize the possible deviations away from your build order,
then you do not understand your build order. On the other hand, if you can
freestyle your build order, keep your money low, and win, then that is proof
that you understand Starcraft -- the metagame is not the game itself.
Avoiding Chess blunders is a similar routine to avoiding a surprise attack on
your base by scouting and watching the minimap. While you might think that the
two games are fundamentally different, in particular because of
incomplete/complete information, this is not the case. Chess is not a game
of complete information if you consider your opponent's secret intentions. In
Chess, a "quiet" move is one that appears to achieve a simple and natural
objective, but in fact accomplishes something unexpected. It is therefore
important to always be thinking about your opponent's secret plans and
anticipate their next move. If you can predict your opponent's next move, then
all that remains is calculation to checkmate. As it is, you are as much in the
dark on the Chessboard as you are in the fog of war. You can see where all of
the pieces are at any instant, but to see where they are going to be is the
crucial challenge. When your opponent activates a powerful piece (like a Bishop
or Rook), it is as though you have scanned the Protoss base, and in those ten
seconds glimpsed a Reaver loading onto a Shuttle, but you do not know when or
where it will strike.
As already mentioned, the key understanding of macro over micro is one way to
improve quickly when you move from Starcraft to Chess. There is also the
discipline of studying build orders, practicing your execution, and constantly
looking for weaknesses in your opponent's plans.
The psychological component is also there. Playing the Terran defensive taught
me the importance of discriminating desperate attacks (which require a change
in my plan) from bluffs (which require acceleration of my plan). In Chess,
these fundamental ideas are played out on a game board without the rigorous
demands of APM and dexterity, so it's easier to improve your game without rage
quitting.
There is also the important strategic component of controlling your impulse for
irrational play. Followers of the Day[9] Daily will recall that the best way to
convert a win is to consolidate your position, not to rush toward the finishing
blow. This is a major theme in Chess. The player with the advantage attempts to
constrict their opponent's options down to nothing in a safe and methodical
fashion. The player with the disadvantage will make risky attacks, sacrifice
pieces, anything to break out of a stranglehold and gain the initiative. Quiet
as it seems, Chess is actually a very dynamic game.
Chess is also deep. Deep in ways that, I believe, Starcraft can never be so
long as mechanics are the chief barrier to entry. It is impossible to play a
novel strategy in Starcraft without the mechanics to back it up; the physics of
the game world are too strict, and you will be punished quickly by stronger
players. The rules of Chess are more flexible. Powerful positions can be found
in unexpected places, and you can prove by analysis that they are strong. As
one grandmaster described another's exemplary play, "You could have seen any
one of these moves, but you could not have seen all of them." There are
probably more viable Chess openings backed up by decades of experience than can
be mastered in a lifetime, and certainly many yet-to-be-discovered or
popularized. All Chess players specialize by necessity, and one of the joys of
the game is meeting new players who can show you ideas you didn't expect.
Similarly, there is tremendous importance to the style of play. In Starcraft,
we look first to the race (Terran, Protoss, Zerg) as an indication of style.
The Terran are slow but versatile. The Zerg are numerous and aggressive. The
Protoss are beefy and deliberate. Chess only has two "races," White and Black,
and they share the same units. Fortunately, the game is still asymmetric, as
White's privilege of first move characterizes the game as a struggle by Black
to achieve an equal position. White plays for a win, while Black may be content
with a draw. Both players can play for a more tactical or more positional game,
so style determines every decision. While the Terran versus Terran matchup is
sometimes characterized as Chess-like due to tank rearrangement, this is
inaccurate. Terran versus Terran is an entirely symmetrical game since neither
player is given initiative; the closest analog is the Terran versus Terran
endgame, where the choice of air-superiority units (battlecruisers, wraiths,
valkyries, or goliaths) reflects the outcome of the middlegame. How would
Terran versus Terran look if one side had a 30 second head start, or if one
player was unallowed to build tanks?
This hopefully has sparked an appetite for the game of Chess. It is a game that
is analyzed in thick textbooks and takes a lifetime to master. There is an
international scene following the top players, and top games are as fascinating
to watch as masterful games by Boxer or Jaedong once you understand how the
game is played at the competitive level.
Give it a try!
