I’ve been writing about games for over ten years, and I keep seeing the same two traps swallow smart people whole. On one side, the shill—the writer who treats every new release like a gift from the heavens, hand-waving broken mechanics because the art direction is pretty. On the other, the hater—the critic who can’t separate personal taste from structural failure, declaring a game garbage because it doesn’t cater to their specific muscle memory. Neither one does anybody any good. The real work of game analysis lives in a narrow, uncomfortable band between those poles. You have to be methodical. You have to back every claim with a specific example. And you have to admit when a game simply isn’t built for you without dismissing its value for someone else. Here’s how to do that.
Start with the Game’s Own Promises
Before you type a single word of judgment, figure out what the game claims to be. Not what you wanted it to be. Not what the marketing department exaggerated. Look at the tutorial, the first hour of gameplay, the mechanics the game actually teaches you. Dark Souls doesn’t promise a power fantasy; it promises methodical, punishing combat where patience gets rewarded. Stardew Valley doesn’t promise high-octane action; it promises a slow, systems-driven farming loop. If you criticize Dark Souls for lacking a relaxing fishing minigame, you’ve already failed the analysis. You’re not a critic; you’re just a person with a preference.
I use a simple framework: list the three core mechanics the game introduces in its first 30 minutes. For Hades, that’s dash-strike combat, boon-based build crafting, and narrative progression through failure. Every subsequent judgment I make about Hades has to reference at least one of those pillars. If the dash-strike feels unresponsive on a technical level, that’s a valid critique because it undermines a core promise. If I simply don’t enjoy roguelike repetition, that’s a personal note, not an objective flaw. The distinction matters.
Separate Technical Performance from Design Intent
This is where most game discussions collapse into shouting. A game can run at a locked 60 frames per second with zero bugs and still be poorly designed. Conversely, a game can be a technical disaster and still deliver on its design goals brilliantly. Conflating the two is lazy.
Take Cyberpunk 2077 at launch. The technical performance on base last-gen consoles was abysmal—sub-20 FPS, constant pop-in, crashes every hour. That’s a technical failure, period. But separate from that, the game’s narrative structure, side quest design, and environmental storytelling were genuinely ambitious and often successful. You can say both things in the same review without contradiction. I wrote at the time: “The game underneath the bugs is a sprawling, reactive noir RPG with some of the best side quests I’ve played in years. The game on my screen right now is a broken mess that shouldn’t have been sold.” Both statements are true, and both are backed by specific examples—the Delamain questline for the former, a crash log for the latter.
Use the “Three Examples” Rule
I never let myself make a broad claim about a game without three concrete examples. “The level design is repetitive” is a worthless sentence on its own. “The level design falls into a repetitive pattern: the three fortress dungeons in Elden Ring’s Mountaintops of the Giants—Castle Sol, the Guardians’ Garrison, and the Forge—all use the same enemy placement template of two banished knights at the entrance, a narrow corridor with fire traps, and a commander-type boss in a circular arena” is an argument. It names the locations, the pattern, and the specific elements that repeat. The reader can agree or disagree, but they can’t dismiss it as vague whining.
This rule also forces you to check your own biases. If you can’t find three specific examples to support a claim, the claim might be emotional rather than analytical. I’ve scrapped entire paragraphs after realizing I was reacting to a single frustrating death rather than a systemic issue. That’s the discipline.
Distinguish Between “I Don’t Like This” and “This Is Bad”
This is the hardest skill to develop, and I still catch myself slipping. The test is simple: can you imagine a player who would genuinely enjoy this mechanic, and can you articulate why? If the answer is yes, you’re dealing with a preference, not a flaw. Your job is to describe the mechanic accurately enough that the hypothetical player can recognize their own taste in your description.
I despise weapon degradation systems. Breath of the Wild’s fragile weapons drove me up a wall. But I can’t call it bad design, because I can articulate exactly what it achieves: it forces constant improvisation, prevents players from settling into a single dominant strategy, and turns every enemy encounter into a resource calculation. For a player who thrives on adaptability and moment-to-moment decision-making, that system is a feature, not a bug. My review said: “The weapon fragility will alienate players who prefer mastery and consistency, but it’s a deliberate tool to keep combat fluid and exploratory.” That’s analysis. “The weapons break too much” is a complaint.
Contextualize Within the Genre and the Developer’s History
No game exists in a vacuum. A competent but unremarkable open-world game from a AAA studio with a $200 million budget deserves harsher scrutiny than a competent but unremarkable open-world game from a five-person indie team. Resources, experience, and precedent all matter. When I reviewed Forspoken, I didn’t just compare it to some platonic ideal of an action RPG. I compared it to Final Fantasy XV, the previous major title from Luminous Productions, noting where it improved (traversal fluidity, spell variety) and where it regressed (side quest depth, world interactivity). I also placed it against contemporaries like Horizon Forbidden West, not to punish it for falling short, but to establish the standard that players in 2023 could reasonably expect from a $70 open-world action game.
This also means acknowledging when a developer is deliberately working against genre conventions. Disco Elysium removed combat entirely from the CRPG formula. Judging it for “lack of action” would be missing the point. Instead, the analysis should focus on whether the replacement systems—the skill-check dialogue, the Thought Cabinet—carry the weight that combat normally bears. They do, and I can prove it: the measure of a CRPG’s encounter design is whether it creates tension, meaningful choice, and character expression. The tribunal sequence in Disco Elysium achieves all three without a single health bar.
Address the Audience, Not the Developer
Too many critiques read like an open letter to the studio: “You should have done X,” “Why didn’t you fix Y?” That’s not analysis; that’s backseat game development. Your reader isn’t the developer. Your reader is someone trying to decide whether to spend $70 and 40 hours on this game. Write for them. Frame your observations as information the reader can use to make a decision, not as demands for a patch.
Instead of “The developers should have added a quest marker here,” write “The absence of quest markers in this section means you’ll need to rely on environmental clues and NPC dialogue. If you enjoy piecing together directions from a torn journal entry, this will feel immersive. If you find that tedious, you’ll be frustrated.” The first version assumes the developer made a mistake. The second version describes the experience and lets the reader decide if it matches their taste. The second version is also more accurate, because you don’t actually know whether the omission was a mistake or a deliberate choice.
When a Game Is Genuinely Bad, Show Your Work
Sometimes a game is just broken, shallow, or incompetently made. When that’s the case, you still need to prove it. “This game is terrible” is not a review; it’s a tweet. A proper negative analysis walks the reader through the failure step by step, with receipts.
I reviewed The Day Before when it briefly existed, and I didn’t just call it a scam. I documented: the Steam store page promised an open-world MMO survival game; the actual executable contained a small extraction-shooter map with no persistent world, no MMO features, and assets that matched a previously cancelled project from the same developer. I listed the missing features by name, cross-referenced them with the store description, and included timestamps from gameplay footage. That’s how you dismantle a bad game without sounding like a hater—you let the evidence do the shouting.
Credit What Works, Even in a Disaster
Conversely, almost no game is 100% irredeemable. Finding and acknowledging the one thing a game does right isn’t being soft; it’s being thorough. It also makes your criticism more credible, because it shows you’re paying attention rather than just piling on.
Anthem was a structural failure in nearly every respect—loot system, mission variety, endgame, narrative coherence. But the flight mechanics were genuinely excellent. The sense of weight, the transition from running to hovering to full jet-powered flight, the heat management that forced you to skim waterfalls and rivers to cool your thrusters—that was a well-designed system. I said so in my review, and I backed it with a description of a specific moment: diving into a canyon on a legendary contract, weaving through rock pillars while managing heat, popping up to unleash an ultimate ability, then diving again. That paragraph didn’t soften my overall verdict, which was harsh. It just made the verdict fair.
Build a Consistent Scoring or Evaluation Framework
You don’t need to publish a numerical score, but you should have an internal framework that you apply consistently across every game or piece of content you analyze. Mine has five categories: mechanical integrity (do the systems work as intended?), design coherence (do the systems support each other?), artistic execution (does the aesthetic serve the design?), technical performance (does it run properly?), and emotional resonance (does it leave a lasting impression?). I weigh these differently depending on the genre. For a competitive shooter, mechanical integrity and technical performance get more weight. For a narrative adventure, artistic execution and emotional resonance dominate.
This framework prevents me from reviewing a walking simulator and a battle royale by the same criteria, which would be absurd. It also forces me to articulate why a game succeeds or fails in terms that are relevant to its genre. When I reviewed What Remains of Edith Finch, I barely mentioned “mechanical integrity” because the mechanics are intentionally minimal—walking and interacting. Instead, I focused on how each vignette’s interaction design served its narrative theme, like the cannery sequence where you split your attention between a monotonous fish-chopping task and a fantasy world unfolding in your peripheral vision. That’s design coherence and artistic execution working in tandem.
FAQ: Common Questions About Game Critique
How do I review a game in a genre I don’t normally play?
You have two ethical options. Option one: decline the review. If you’ve never played a fighting game and can’t tell a good frame-data implementation from a bad one, your review will be useless to fighting game players. Option two: disclose your inexperience upfront and frame the review as a newcomer’s perspective. “I’ve spent five hours with Street Fighter 6 as someone whose last fighting game was SoulCalibur II in 2003. Here’s what the modern controls and tutorial systems did for me, and here’s where I still felt lost.” That’s honest and potentially valuable to other newcomers. Just don’t pretend to evaluate high-level balance when you can’t.
What if I genuinely can’t find anything positive to say?
First, triple-check your “Three Examples” rule. If you’ve documented specific failures across every relevant category and still have nothing positive, you’re probably dealing with a genuinely broken product. In that case, your review serves as a consumer warning, and that’s a legitimate function. But also ask yourself: did the game’s concept have potential? Can you articulate what a successful version of this game would look like? That’s not praise for the existing product, but it’s useful context that elevates the critique above pure negativity. “Babylon’s Fall had the kernel of an interesting idea—combining character-action combat with live-service progression—but the execution failed at every level. A game that actually delivered fluid PlatinumGames combat in a persistent multiplayer framework could work, but this isn’t it.”
How do I avoid being influenced by a game’s community or hype cycle?
I have a strict rule: I don’t read other reviews or engage with community discourse until after I’ve published my own analysis. Hype and backlash are both contagious. If you absorb a week of “this game is a masterpiece” discourse before playing, you’ll either conform to the consensus or overcorrect into contrarianism. Neither is honest. Play the game cold, form your own conclusions, write them down, and only then see what others are saying. If the community has identified a technical issue you missed—like a memory leak that only manifests after 20 hours—you can update your review with that information, clearly labeled as a post-publication addition. But your core analysis should be yours alone.
Final Thoughts: The Critic’s Responsibility
Game criticism isn’t about being right. It’s about being useful. A useful review helps a reader predict their own experience with the game, regardless of whether you personally enjoyed it. That requires precision, honesty about your own biases, and a willingness to do the unglamorous work of documenting specifics. The next time you sit down to write about a game, ask yourself: if someone who loves everything I hate reads this, will they still find it helpful? If the answer is yes, you’ve done your job.