Gamesense

treactis

What I mean by gamesense

I believe gamesense is the most important skill. That’s true in VALORANT, and in any other game too.

By gamesense I mean general macro knowledge of the game. Macro is the big-picture side of play: where to be, when, and why. In VALORANT, it means being able to:

• predict where the enemies are
• predict where you need to be yourself
• keep track of the economy (how many credits each team has)
• keep track of the ability economy (which abilities each side still has left)

That definition is short, though. In reality, macro is made up of many more parts than this.

How gamesense develops

Gamesense usually develops by playing a lot of games. Over time, patterns start to show up. You start making the right decisions.

But playing isn’t enough on its own. You have to stay aware of which decisions you’re making, and when. That awareness is what turns games played into real progress.

What took me to Radiant

Learning macro and building gamesense is what took me from Immortal 3 to Radiant. It wasn’t aim at all.

What helped me improve

There are many ways to improve gamesense. Below are the things that helped me most.

These aren’t separate methods where you have to pick one. They’re just things you can do, and you can do all of them at the same time.

Watching pros and experienced players

The most important one for me was watching stream VODs of pros and more experienced players. I picked stream VODs on purpose. I wanted to build my macro for ranked play. So I watched pros playing in that same ranked environment.

When I wanted to learn a specific role in a 5v5 team setting, I watched VODs from well-known tournaments instead.

In my opinion, this alone is pretty much enough to start improving. It was enough for me to reach Radiant.

Watching pros didn’t only build my gamesense. It also changed how I approach the game itself. Pros usually play in a much calmer, more calculated way. There’s no room for frustration in it. Tilting means letting frustration or other negative emotions affect your play. It has only drawbacks, and big ones. So watching pros also helps you build a mental image of the player you want to be.

These three players shaped that image for me the most:

• Eggster. A really strong Yoru player. He has a very good understanding of the game and of positioning, plus exceptional mechanics. What I love most is how he reacts to ranked games and to really unlucky moments. He usually laughs them off or turns them into a joke. He doesn’t spiral into negative thoughts, and his mental stays stable.
• Inspire. The true picture of calm and collected. He’s especially interesting to me because he plays Sentinel, the same role I play. You can learn a lot of line-ups, setups and protocols from him. Line-ups are fixed spots to throw an ability from, so it lands in an exact place.
• ALEKSANDAR. I just like his calm way of aiming. Nothing more to add.

In short, Eggster was my model for mentality and macro. Inspire was my main source for actual gameplay. ALEKSANDAR was my aiming inspiration. Aim has its own note, aim. Mentality will get its own note later.

Reviewing my own scrims

Another thing that improved me a lot was VOD-reviewing my scrims. A scrim is a practice match between two organized teams. Having a coach for the review is a big bonus.

Scrims aren’t the same environment as ranked. But they still build your general understanding of the game. What you learn in those reviews can realistically carry over into ranked.

Example: learning to lurk as Viper. I used to have no idea how to lurk as Viper. Lurking means playing away from your team, often alone, to catch enemies from an unexpected side. Viper usually isn’t picked in ranked, so I never had a reason to learn this.

Then I practiced her a lot on Lotus and Split. I also went through a very large number of VOD reviews, where we discussed my mistakes.

After that, I understood how to take space properly as Viper. Taking space means claiming parts of the map for your team. I could also play her in ranked. And I knew how to counter her when I faced her, in both scrims and ranked.

The bottom line

Just playing isn’t enough. Watching yourself and others play is the key to actively improving your macro and gamesense. And gamesense is the key to winning.

Scientific Evidence

In short, the research backs the core of the experience above. Gamesense looks like pattern recognition built from many games. It improves when you judge your decisions, not just the results. Decision skill can also be trained with video, though it carries over to real play less than it shows up in tests. But no study here tests whether gamesense matters more than aim. None tests learning by watching pros, either.

“Game sense” is not a mystical trait. It’s also not general intelligence. Sport science calls the same thing perceptual-cognitive expertise. It breaks down into measurable parts. Experts search the scene differently. They recognize situation patterns instead of reasoning through every option. They anticipate from partial information.

Esports-specific evidence agrees that experts differ from non-experts. But the advantage is narrower than popular claims suggest. It’s concentrated in spatial cognition and one kind of attention. It’s absent in most other cognitive areas. The evidence is also cross-sectional. That means it compares groups at one point in time. So it can’t tell us whether playing built the advantage, or whether people with the advantage were simply more likely to keep playing. And the main esports meta-analysis didn’t test decision-making at all.

The tactical-FPS layer on top of this is basically undocumented in peer-reviewed work. That layer covers timings, utility, information economy and map control. This note doesn’t cover it.

What expertise actually looks like, measured

Mann, Williams, Ward and Janelle (2007) combined 42 studies into one meta-analysis. That gave them 388 effect sizes from 1,288 participants, comparing expert and non-expert athletes on perceptual-cognitive measures. Results below are point-biserial correlations (r_pb), a way of measuring how strongly being an expert goes with a better score. All were statistically significant:

Measure	Effect (rpb)	Direction
Quiet eye duration	0.62	Experts longer
Response time	0.35	Experts faster
Response accuracy	0.31	Experts more accurate
Number of fixations	0.26	Experts fewer
Fixation duration	0.23	Experts longer

In traditional sports, experts make fewer and longer fixations. A fixation is a moment when your eyes stay still on one spot. Experts aren’t scanning harder. They’re looking at less, because they know what actually carries information.

The expert advantage also grew in more realistic test settings. For response accuracy, it was r_pb = 0.42 in field settings, 0.31 with video, and 0.25 with still images. The closer the test was to the real sport, the clearer the gap.

(These are correlations, not percentages. r_pb = 0.31 does not mean experts were 31% more accurate.)

In esports, the eye pattern looks different

Luo, Chen, Cho, Yan and Seo (2025) ran the first meta-analysis of eye movements in esports experts. It pooled 7 studies with 165 players.

• Experts had shorter fixations, not longer ones (SMD = −0.66, statistically significant, consistent across studies). SMD is a standardized mean difference, a common way to compare two groups.
• No clear difference in the number of fixations. The studies disagreed with each other a lot on this.
• No shooters were included. The games were Dota 2, League of Legends, StarCraft, FIFA, Gran Turismo and Assetto Corsa.

This complicates the sport pattern above. In fast screen-based games, experts may take in each spot more quickly, not linger longer. So “look for longer” is not a safe target for VALORANT. The useful question still holds, though: where is my attention going, and is that spot actually informative?

The esports cognitive advantage is real, narrow, and hard to explain causally

Miao, He, Hou, Wang and Chi (2024) ran a three-level meta-analysis in PeerJ. It pooled 15 studies, 142 effect sizes, and 1,085 participants, comparing esports experts with non-experts.

• Overall effect: Hedges’ g = 0.373 (small-to-medium, statistically significant). Hedges’ g is a standard measure of how far apart two groups are.
• Spatial cognition: g = 0.822. This is the standout result. It covers things like visual working memory and mental rotation.
• Bottom-up attention: g = 0.416. Top-down attention was not significant.
• No significant difference in perception, inhibition, problem-solving, task-switching, verbal cognition, or motor control
• Accuracy vs. speed: accuracy measures showed a clear effect (g = 0.560). Reaction-time measures showed a small effect (g = 0.215) that was not statistically significant.
• No difference by game genre. Shooters, strategy games, and mixed genres all showed similar results.
• It left out decision-making. The authors say their analysis included no studies on anticipation or decision-making. They note that esports research mostly tests general abilities and skips these game-specific skills.
• Cause and effect stays open. The authors discuss both directions. Gaming experience clearly shapes cognition. But training studies show smaller gains (g = 0.26–0.45) than the gap between experts and amateurs. So part of the advantage may have existed before those players became experts.

A 2026 scoping review of nine studies (Lu, Lee & Yoon) reached the same overall picture. Esports players may have advantages in some areas. But the evidence is still too weak to say whether esports causes them.

Three things worth writing down from this:

1. Generic “brain training” is a poor bet. Most cognitive areas showed no expert advantage at all. So training them has no clear target. Research in traditional sports points the same way. Kalén and colleagues (2021) found that tests of game-specific decision-making separated skill levels far more (g = 0.77) than tests of basic (g = 0.39) or higher general cognitive functions (g = 0.44).
2. Reaction time is the wrong thing to chase. Speed on these tasks didn’t reliably separate experts. Accuracy did. This is accuracy on general lab tasks, though, not accuracy of in-game decisions. The analysis didn’t test decisions.
3. No genre difference means the FPS-vs-strategy distinction didn’t matter for these cognitive measures. That’s a caution against assuming VALORANT builds some unique cognitive profile.

Decision-making can be trained, but it carries over less than it looks

Zhu, Zheng, Liu, Guo and Cao (2024) pooled 22 studies on training anticipation and decision-making in team sports. The training showed players game situations, usually on video, and had them decide what happens next or what to do.

• In lab tests, the gains were large. Accuracy improved with g = 1.51. Response time improved with g = −0.91 (negative means faster).
• On the real field, the gains were smaller, but still real. Accuracy improved with g = 0.65. Response time improved with g = −0.44.
• More realistic training carried over better. Virtual reality training transferred best (0.96). Projector-based training came next (0.46). Computer-screen training transferred least (0.19).
• Longer programs worked better. Training that lasted more than 4 weeks had larger effects.

Two limits matter here. These were football, rugby, basketball and similar team sports, with no esports. And most studies never tested whether the gains showed up in real matches at all.

For VALORANT, the useful part is this. Decision skill is trainable, not fixed. But getting better at spotting the right answer on a screen is not the same as getting better in a live round. The training that works is active. You have to decide, not just watch.

Experts don’t compare options — they recognize situations

Klein’s recognition-primed decision model comes from studying firefighters, military commanders, and other professionals under time pressure. It says experts don’t generate and compare a list of alternatives. Instead, they recognize the situation as an example of a familiar type. That recognition comes with a workable action already attached. They then mentally test that one action to see if it works. If it doesn’t hold up, they adjust it, or move on to the next option that fits the situation. They still don’t line up all the options side by side.

This is a much better model of “game sense” than careful deliberation is. A player who “just knows” the enemy is rotating isn’t calculating the odds. They’re pattern-matching against a large library of past situations. This means game sense gets built by exposure to many varied situations with a clear resolution, not by thinking harder in the moment.

Intuition is only trustworthy under specific conditions

Kahneman and Klein (2009) deliberately tried to reconcile their two opposing research programs. They concluded that expert intuition is only reliable when two conditions are both true. First, the environment must be regular enough to be predictable. Second, the person must have had the chance to learn those regularities through a lot of practice with fast, clear feedback.

Tactical FPS partly qualifies. The environment is highly regular. Maps, timings, economy, and utility all follow rules. Feedback is fast, too. Rounds resolve in about a hundred seconds. But the feedback is outcome feedback, not process feedback, and the two constantly disagree. A correct read can still lose the round to a lucky spray. A reckless peek can still win it. Learning only from outcomes, in a game with this much randomness, actively trains superstition instead of skill.

This is the single most useful idea in this note. It leads to one concrete rule for VOD review: judge the decision against the information available at the moment it was made, not against how the round ended. Without that rule, review just reinforces noise.

Cognitive load and communication

Working memory is limited. Adults can hold only about four chunks of information in mind at once (Cowan, 2001). Tactical FPS loads it constantly. You track positions, timings, utility states, economy, plus your own mechanical execution. Two conclusions follow from general cognitive-load theory (Sweller, 1988), not from esports research specifically:

• Automate whatever can be automated. Anything that becomes second nature frees up capacity for things that can’t be automated. Examples are default setups, standard timings, and pre-agreed rotations.
• Communication competes for the same mental capacity. Long, unstructured comms cost the listener working memory at exactly the moment they need it most. Short, standardized, information-dense callouts are cheap. Narration is expensive.

How this compares with the experience above

Claim from treactis	Verdict	Why
Gamesense is the most important skill, more than aim	Not covered	No study here weighs gamesense against aim, in VALORANT or anywhere else. Kalén et al. show decision-making skill separates skill levels strongly in traditional sports. That doesn’t say it matters more than mechanics.
Macro, not aim, took me from Immortal 3 to Radiant	Not covered	This is a personal account, and nothing here can test it. One possible reason it fits: at Immortal 3, aim may already be strong, so the remaining gap is elsewhere. That’s a guess, not a finding.
Gamesense is built by playing many games, as patterns show up	Agrees	This is exactly Klein’s recognition-primed model. Experts build a library of familiar situations and match new ones against it.
You must stay aware of which decisions you make, and when	Agrees	Kahneman & Klein say intuition only becomes reliable with clear feedback. In VALORANT, round results are noisy. Checking the decision itself, not the outcome, is the fix.
Watching pro stream VODs builds gamesense	Partly supported	Zhu et al. show video-based decision training works. But in those studies, players had to actively decide what happens next. Plain watching wasn’t tested. Motor-learning research also found video review works best with specific things to look for. Pausing a VOD and predicting the pro’s next move would be closer to what was tested.
Watch VODs from the same environment you want to improve in (ranked streams for ranked, tournaments for team roles)	Agrees in direction	Mann et al. found the expert gap was clearest in the most realistic settings. Zhu et al. found more realistic training carried over better. Both point the same way. Neither tested ranked vs. tournament VODs.
Watching calm pros builds a better mentality and less tilt	Not covered	No study here looks at emotions or tilt. This belongs in the future mentality note.
Scrim VOD review, with discussion of mistakes and a coach	Agrees in direction	Motor-learning research found video review works when it’s paired with specific, spoken feedback on errors. Discussing mistakes with a coach is that kind of review. That evidence comes from motor learning, not from decisions in esports.
What you learn in scrims carries over into ranked (the Viper example)	Not tested	No study here measures transfer from scrims to ranked. Zhu et al. show transfer between settings is real but smaller than it looks. The Viper example is one clear personal case, which is useful, but it’s n=1.

How strong is this evidence

Solid, from sport science and applied cognition: the expert visual-search profile in traditional sports (Mann et al., a meta-analysis). Game-specific decision skill separating skill levels more than general cognition (Kalén et al., a meta-analysis). Recognition-primed decision making (a large body of qualitative and applied work). The conditions-for-intuition framework (Kahneman & Klein). That last one was jointly written by two researchers who used to disagree, which is unusually strong ground for a contested topic.

Moderate, from traditional team sports: decision training with video or VR (Zhu et al.). The lab gains are clear. The real-match gains are smaller and were tested in fewer studies. None of it is from esports.

Real, but cross-sectional and narrow: the esports cognitive-expertise meta-analysis. It cannot support “playing VALORANT improves your cognition,” nor “improving your cognition improves your VALORANT.” It also didn’t include decision-making studies, so it says nothing direct about gamesense itself.

Early and small: the esports eye-movement meta-analysis. Seven studies, 165 players, no shooters. It’s the best esports data on visual search so far. But it’s enough to question the sport pattern, not to replace it.

Theory-driven reasoning, not tested here: the cognitive-load arguments about communication and automatic behavior. Working-memory limits are well established. Applying them to comms is reasoning, not a measured result.

Not covered by any evidence here: the tactical layer (timings, utility, economy decisions). The claim that gamesense matters more than aim. Learning by passively watching other players. Tilt and mentality. Transfer from scrims to ranked. These rest on personal experience only, which is honest n=1 evidence and nothing more.

Related Notes

• aim — the boundary case is crosshair placement. It’s a decision (where will they be?) that gets executed as a motor action.

Sources

• Mann, D. T. Y., Williams, A. M., Ward, P., & Janelle, C. M. (2007). Perceptual-cognitive expertise in sport: A meta-analysis. Journal of Sport & Exercise Psychology, 29(4), 457–478. https://pubmed.ncbi.nlm.nih.gov/17968048/
• Luo, Y., Chen, Y., Cho, J., Yan, C., & Seo, J. (2025). Differences in eye movement characteristics between expert and non-expert eSports players: a systematic review and meta-analysis. Scientific Reports, 15, 30185. https://www.nature.com/articles/s41598-025-12101-8
• Miao, H., He, H., Hou, X., Wang, J., & Chi, L. (2024). Cognitive expertise in esport experts: a three-level model meta-analysis. PeerJ, 12, e17857. doi:10.7717/peerj.17857. https://peerj.com/articles/17857/
• Lu, Lee, & Yoon (2026). The relationship between esports and cognitive function: A scoping review. PLOS ONE. https://doi.org/10.1371/journal.pone.0352875
• Kalén, A., Bisagno, E., Musculus, L., Raab, M., Pérez-Ferreirós, A., Williams, A. M., Araújo, D., Lindwall, M., & Ivarsson, A. (2021). The role of domain-specific and domain-general cognitive functions and skills in sports performance: A meta-analysis. Psychological Bulletin, 147(12), 1290–1308. https://doi.org/10.1037/bul0000355
• Zhu, R., Zheng, M., Liu, S., Guo, J., & Cao, C. (2024). Effects of perceptual-cognitive training on anticipation and decision-making skills in team sports: A systematic review and meta-analysis. Behavioral Sciences, 14(10), 919. https://doi.org/10.3390/bs14100919
• Klein, G. A. (1993). A recognition-primed decision (RPD) model of rapid decision making. In G. A. Klein, J. Orasanu, R. Calderwood, & C. E. Zsambok (Eds.), Decision Making in Action: Models and Methods. Ablex.
• Klein, G. (1998). Sources of Power: How People Make Decisions. MIT Press.
• Kahneman, D., & Klein, G. (2009). Conditions for intuitive expertise: A failure to disagree. American Psychologist, 64(6), 515–526. https://pubmed.ncbi.nlm.nih.gov/19739881/
• Cowan, N. (2001). The magical number 4 in short-term memory: A reconsideration of mental storage capacity. Behavioral and Brain Sciences, 24(1), 87–114. https://pubmed.ncbi.nlm.nih.gov/11515286/
• Sweller, J. (1988). Cognitive load during problem solving: Effects on learning. Cognitive Science, 12(2), 257–285. https://doi.org/10.1207/s15516709cog1202_4