A little over two years ago, I watched my first Counter-Strike tournament. It was DreamHack Winter 2014. StarCraft 2 had fewer tournaments than before, so I had time to watch other games, and it happened to be on. They were pretty good games, though the most I understood was Vincent “Happy” Schopenhauer and Fabien “kioShiMa” Fiey were good at shooting people. Afterward, I started to watch all of the CS:GO LANs as long as the teams were of high skill.
A year and a half later, I wrote my first CS:GO article because the Fnatic win over Virtus.Pro and EnVyUs at ESL Cologne was electrifying. I hadn’t planned to, but the story kept cropping up in my head. How would I frame it? With little to no theoretical knowledge, I wrote about the miracle Fnatic pauses that secured them the victory at that Major.
At Duncan “Thorin” Shields’ recommendation I started to write about CS:GO more regularly. It was a great game, great scene and a new challenge. I assumed I wouldn’t be paid (at that point I’d been writing for years without getting paid), but I enjoyed it for the sake of it. The game I had studied the most by then was Starcraft 2, which was a 1-v-1 game. CS:GO was the first team game I had studied in earnest. I had watched Dota 2 but relied on the thousands of hours I had played in my youth to understand the game. This was a completely different challenge as I’d be starting in the middle of Counter-Strike’s history, so had to catch up on what was, what is and what could be.
Today, all I’ve learned is how much I didn’t know about the game. I’m better than when I started, but there is still so much left to go. Because this is a team game, there is a lot more unknowable information than a 1-v-1 game. We don’t get to see a player play every kind of role and every position on every map. We don’t know how comms are or the internal issues of the team that affects play. We don’t know if a team called for a player to make a certain decision or if the player took the reins themselves.
CS:GO seems simplistic on the surface, but it gets more complicated the more you think about and research the game. I’ve constructed and deconstructed multiple models to help me study the game to give me a better grasp of what can happen in any single game. After two years of study, this is what I’ve learned so far. It’s been helpful in helping me analyze teams, but it is far from complete.
I analyze matches and series in eight broad categories:
The highest level of analysis that technically covers all other aspects is map analysis. There are seven maps in the pool and you play two sides every map. I consider T side and CT side wholly independent of each other, so in my mind a map pool of seven works out to 14 in my head. For instance, the former SK Gaming team was the greatest at Train because they could dominate both CT and T sides of the map. They did it by having their two star players, Gabriel “FalleN” Toledo and Marcelo “coldzera” David, play A, making it a nightmare to take. Epitácio “TACO” de Melo facilitated their stardom by being a great role player and holding inner by himself. If they went toward him, he could get a kill, get out and buy time for the rotation to happen and make a successful retake. This was filled out by Fernando “fer” Alvarenga’s aggression, which bought them information, and Lincoln “fnx” Lau supporting the others.
This level of analysis is even more important because of the limitations of practice in a team game. In a 1-v-1 game you can practice however much or little as you want, whereas in a team game practice will be limited by the one player who plays the least on a team. I asked Jason “Moses” O’Toole and Janko “YNk” Paunović how long it took to become good at a map without distractions. The answer was usually two to three weeks. This is without counting the fact that teams need to keep updated on their current map pools so others can’t surpass them, traveling to tournaments and playing in online leagues. It isn’t surprising we’ve entered a parity era as there is no way for teams to shore up their map pools without downtime.
So map analysis can go as broad or as deep as you want, and it’s a great starting point in trying to analyze a specific team.
The next category is the team identity. This is what the personnel on the team and how the five players fit into a larger picture. You need to understand what each of the five players brings to a particular team. To do that, you need to know what maps each player was good on, what role they played, how they could theoretically fit with each other, their top performance, average performance and worst performance. It’s always a bit harder to know because there is still a lot of unknown data, but you can usually tell who will be the star player of a specific team based on form and results. Sometimes the problem is obvious, such as FaZe earlier this year. It was obvious to everyone they needed a good leader and once they got it, the results dramatically improved.
The next category is roles, a topic tied closely to team identity because there are certain roles a team needs to have. This is divided into two sub-categories: the overall roles of the team and the specific positions they play. How each person analyzes the first is different. Some separate it into star players and role players. Others go with riflers and AWPer, support players and fraggers. In the end, it comes down to who is killing the most people consistently on the team and helping their teammates kill even more people.
The other way to look at it is what positions they play on any specific map. With 14 different sides to play and only five players, there are bound to be positional conflicts. Where do they place each player? Does the team have a fixed position for a player in every map or do they play more loosely and trade positions? What triggers a team to have players switch positions in any given round? Do the positions make sense for a particular player’s skillset? For instance, Emil “Magiskb0Y” Reif and Timothy “autimatic” Ta are both star players for their respective teams. For Dignitas to activate Magiskb0y, they have to place him in restrooms and give him space to be initially aggressive before switching to a more passive hold later in the round. For Cloud9 to activate autimatic, they have to put him in the rotator role and support his teammates with nades, play the reinforcer and be the key player on retakes.
The next category is teamwork. I’ve found that there are two ways for a team to function well. One is to do it instinctively like what the old Fnatic teams used to do. It’s hard to know what the comms were like, but the way they were always in sync across the map or in situations where they didn’t have time to coordinate seemed inhuman. They were the ultimate form of communication, the type that doesn’t need routine or practice or complicated instructions. They knew what the other was doing with minimal talking and acted accordingly.
The other is to have a strong routine. This can either come in the form of a strong leader like Mathias “MSL” Lauridsen, who tries to figure out the game plan before going in, and because every piece knows their place, they function well. The other way to build a strong routine is what SK did, which is to have every player understand the entire game situation. What positions they play, how that affects the map, what is the best thing they can do from a particular position. This is why their power play situations were so successful, as was their ability to trade roles based on circumstance or position.
The fifth category is tactics. This takes a combination of teamwork, roles and tactics and map analysis to comprise a unified whole. What players do I have? What players do they have? What is the best possible way for me to win a round? When EnVyUs was at its best, the players used their superior firepower with force buy style to completely catch other teams off guard. When Natus Vincere rose to be one of the best teams in the world, they worked map control, slowly suffocated the team and let Ladislav “GuardiaN” Kovács try to take a pick before executing on a site. The concept depends on the leader, the personnel he has and the team he is playing against.
Another way I’ve come to think about tactics is the maximization of your own team’s firepower and the minimization of the other team’s chances to shoot you in the head. It doesn’t matter if FalleN is on the other team on Overpass if you just run towards B instead. After a few rounds, FalleN will realize they’ve given up even challenging him and will then rotate to B to make them afraid again. Tactics are the interplay between two teams on a large macro level between multiple factors. Who is in form? Who isn’t? What parts of the map can I take? What information can I steal? What I will do, what they will do in response, and the mind games therein.
The sixth category is skill. Broadly speaking, I’ve put firepower, impact, psychology and game sense into this category. It is what makes up a single individual’s ability to affect the game. Peter “stanislaw” Jarguz is the worst individual player on OpTic, but relative to other in-game leaders, he has huge impact on individual rounds for his team and can often get his team out of bad spots. MSL, in contrast, won’t give you much in this category but will give you more in the tactics, the roles and teamplay. Egor “flamie” Vasilyev, when he’s on fire, can get you two kills easily, but because of who he is, he’ll go for the third kill too. Sometimes that means he’ll die, and what was once a certain round victory is now back in contention. This category measures the advantages and drawbacks of a single individual player.
The seventh category is minutiae. These are very small adjustments that can catch other teams off guard and help you steal or win a round. This can be something like studying a team’s tendencies and abusing them, creating one-way smokes or specific nade executes on a site. A new boost that no one has ever seen before.
Finally, there is what can only be classified as Voodoo Shit. This is what others deem momentum, confidence and form. There is no tangible way to analyze it, but everyone believes in it. Sometimes you’re on and sometimes you’re not, and there seems to be little rhyme or reason as to why. It’s strongly correlated with individual skill and individual’s emotions. I avoid using these terms because they’re huge blanket terms that don’t tell you much. Nevertheless, it is something to consider as well.
This is how I’ve come to analyze CS:GO after watching multiple tournaments in the last few years. Doubtless I will change what I’ve come to think again, but I thought it would be an interesting mental exercise in examining the game.