Review scores are broken. You know it, we know it, and the publishers who fly reviewers to preview events definitely know it. This page explains exactly how No Mercy Game reviews work — our six-point test, what we refuse to do, and why a 6/10 from us might be worth more than a 9/10 from somewhere shinier.

The rules we will not break

• We finish the game before scoring it. Credits or bust — no “played the first five hours” verdicts.
• We buy our own copies whenever review access comes with conditions attached.
• No sponsored scores. Ever. Sponsorships, if any, are labelled and never touch review content.
• Live-service games get re-reviewed after major patches — a launch score is a snapshot, not a life sentence.
• We test on the hardware our readers actually own, including budget phones and integrated graphics.

The six-point test

1. Combat honesty

When you die or fail, was it your fault, and did the game make that legible? Difficulty earns praise only when it is honest — the standard we set in our hardest games ranking applies to everything we score.

2. Respect for time

Not length — respect. A tight eight-hour campaign outranks a forty-hour checklist. Load times, unskippable cutscenes on retry, and daily-login psychology all get counted, and all get named in the review.

3. Performance floor

We benchmark on realistic hardware, not review rigs. A game that runs beautifully on machines our readers do not own gets scored on the machines they do. This is where many 2026 releases quietly fail, and it is half the reason our low-end PC list exists.

4. Systems depth vs. systems noise

Twelve interlocking mechanics that matter beat forty features that exist for the trailer. We ask one question of every system: if it were removed, would the game be worse? Count the noes, worry accordingly.

5. Monetisation dignity

Cosmetic-only with visible prices scores fine. Battle passes get scrutiny for time-pressure design. Anything that sells power in a competitive mode is an automatic cap on the score, stated in the first paragraph, because you deserve to know before anything else.

6. The week-later test

After finishing, we wait several days and ask: do we want to go back? Some games review well and evaporate; others lodge in your head. The score locks only after this test, and it has changed verdicts in both directions.

How the score maps to words

9–10 means drop what you are playing. 7–8 means excellent for its audience — read the review to learn if that is you. 5–6 means real flaws, real virtues; wait for a patch or a sale. 3–4 means broken promises. 1–2 means we suffered so you do not have to, as documented in our overhyped games post-mortem. Value verdicts for Indian wallets — is it worth it at launch price, regional price, or free — get their own line in every review, expanded in our free-to-play vs paid analysis.

Where the money comes from

Running a review site costs money, so here is the ledger: we may earn from clearly-marked affiliate links on hardware, never on the games being reviewed. Nobody on the team holds stock in publishers we cover. When a conflict is unavoidable, the review discloses it in the opening lines or we hand the game to someone unconflicted. Trust compounds slower than traffic — we are optimising for the former. Questions about any verdict are welcome via our contact page.

FAQ

Why do your scores sometimes differ wildly from the big sites?

Different tests. We weight performance on real hardware, monetisation dignity and the week-later test far more heavily than launch-week spectacle. A game optimised for review events and a game optimised for players are often different games.

Can readers request a review?

Yes — the most-requested titles each month jump the queue, particularly Indian releases and budget-hardware games the bigger outlets skip. Send requests through the contact form with the platform you play on.

Inside a review week: the process, hour by hour

Methodology pages usually stop at principles. Here is the actual production line — what happens between a game\u2019s arrival and its verdict, because transparency about process is the only real answer to “why should I trust you?”

Days one and two: the blind phase

The assigned reviewer plays before reading anything — no other reviews, no subreddit, no Discord discourse. First impressions get logged raw: onboarding quality, first-hour friction, the moment the game first delighted or insulted us. These notes are timestamped and quoted in the final review unedited, because first-contact experience is what a buying reader will actually feel, and it cannot be reconstructed after fifty hours of familiarity.

Days three to six: the grind phase

The middle of a review is deliberately unglamorous: main path to credits, side systems exercised properly, and the benchmark suite on our real-hardware bench — the budget phone, the mid-ranger, the integrated-graphics laptop and the mainstream desktop. Live-service titles get their economy audited with a spreadsheet: hours-to-unlock at zero spend, the real rupee price of the “recommended” path, and every timer measured against the question monetisation dignity asks — is this design, or is this pressure? Multiplayer games get queued at Indian peak and off-peak hours, because matchmaking health at 2 PM on a Tuesday is part of the product.

Day seven: the fight

Every score survives a team argument. The reviewer defends the draft against two colleagues whose job is disagreement: one attacks the score from the genre-veteran\u2019s seat, one from the value-for-money seat. Scores move in roughly a third of these sessions — both directions. Then the week-later test locks it. If a review misses embargo because the fight ran long, it misses embargo; the calendar works for the verdict, never the reverse.

Corrections, re-reviews and the paper trail

When we get something wrong — it happens — the correction is appended visibly, never silently patched. Factual fixes land within a day of verification. Verdict-level misses get a full re-review, marked as such, with the original preserved and linked; readers deserve the history, and we deserve the embarrassment. Live games are re-scored after any patch that changes the monetisation, the netcode or the core loop, and the score history stays public. A review site\u2019s errata page is its honesty measured in public; ours will never be empty, and that is the point.

• Blind first impressions, quoted unedited in every review
• Real-hardware benchmarks on four representative devices
• Economy audits with actual spreadsheets, not vibes
• Adversarial score defence before publication
• Public corrections and preserved re-review history

That is the machine. It is slower than the industry standard, more expensive than the industry standard, and it produces verdicts we can defend in any argument a reader brings. We consider that a fair trade for the only asset a review site actually owns.

Reader questions from the community

Why not just use aggregate scores from the big review sites?

Aggregates average away exactly the information a specific buyer needs. A game scoring 82 on the strength of visuals and narrative can be a terrible purchase for a competitive player on budget hardware, and a 68 with “repetitive but satisfying combat” complaints can be the best fifty hours of a grinder\u2019s year. Averages also inherit every structural bias of launch-week reviewing: preview-event goodwill, embargo pressure, and benchmarks run on machines no reader owns. We would rather be one clearly-reasoned opinion you can calibrate against than a decimal point summarising forty conflicting ones.

How do you handle review copies and early access?

Review access is accepted only when it arrives without conditions beyond a standard embargo date. Any request to preview scores, approve quotes or “align on messaging” converts the copy into a purchase — we buy it ourselves and note that in the review. Early-access titles get coverage clearly marked as provisional, never a locked score; scoring an unfinished game is reviewing a promise, and our entire methodology exists because promises are the industry\u2019s cheapest currency.

What qualifies someone to review for No Mercy Game?

Genre residency, not general expertise. The reviewer assigned to a fighting game has ranked seasons in fighting games; the soulslike reviewer has the scars of the back catalogue. A generalist can describe any game competently — only a resident can tell you whether it respects the genre\u2019s veterans, welcomes its newcomers, and where it sits against the ten titles its store page is imitating. Fluency is the qualification, and it shows in the details readers check us against.

Methodology is a promise made in public. This page is the contract, the reviews are the receipts, and every reader who checks one against the other is doing exactly what we built this site hoping you would. Hold us to it — via the contact page, loudly, whenever we drift.

One final note on independence: this methodology page is itself versioned. When the process changes — a new hardware tier on the bench, a new filter in the six-point test — the change is dated and explained here rather than applied silently. A rulebook that mutates quietly is not a rulebook; it is a mood. Bookmark this page, check it against any verdict that surprises you, and treat every mismatch you find as a bug report we owe you an answer for.