Paul Barnett, creative director of Warhammer Online, warned in 2009 that the industry was "obsessively self-referencing" hit games—a take that aged into conventional wisdom while his own MMO shut down. Not all bad predictions die; not all good ones win. Here's how to sort signal from nostalgia when evaluating gaming's loudest forecasts.
How This Tier List Works
Three axes separate memorable predictions from useful ones: accuracy (did reality match?), consequence (who lost money or time believing it?), and meta-value (does the reasoning still teach us something even when wrong?). A prediction can be wrong yet S-tier if it reveals a structural blind spot the industry keeps repeating.
Scope limits: 2007–2025, industry figures with named roles, predictions about design trends or business models with verifiable outcomes. No anonymous forum prophets, no "X will fail" without documented source.
S-Tier: Wrong in the Specifics, Right in the Architecture
Paul Barnett — "Obsessively Self-Referencing" (2009)
The take: Barnett, then at EA Mythic steering Warhammer Online, argued that studios were copying World of Warcraft's surface rather than its structural innovations. The warning, originally reported in 2009, targeted an industry habit of cloning successful games without understanding why they worked.
What happened: Warhammer Online closed in 2013. Barnett now serves as chief creative officer at Wargaming. The MMO boom he criticized collapsed into a handful of survivors. Yet major live-service releases from 2016 onward solved the problem he identified by building systems rather than cloning content—seasonal event architecture, cross-progression structures, and social systems that transcend individual content drops.
Why S-tier despite personal failure: Barnett diagnosed the right disease, prescribed the wrong immunity. Warhammer Online itself borrowed heavily from established MMO conventions. The prediction's value lies in its failure mode—showing that recognizing a trap doesn't guarantee escape. This is decision archaeology: Barnett saw the cliff, built his own bridge toward it, and the bridge collapsed. The cliff was real.
Best for: Developers choosing between "inspired by" and "mechanically derivative." Skip if: You're seeking business strategy; Barnett's take is design diagnosis, not execution manual.
Hideki Kamiya — "Western Games Will Dominate, Japanese Studios Must Adapt" (circa 2010, various interviews)
Kamiya at PlatinumGames argued Japanese development had grown insular, producing "games for Japanese players that happen to sell elsewhere." The prediction seemed acute during the PS3/Xbox 360 era's western shooter ascendancy. Then Dark Souls (2011), Monster Hunter World (2018), and Nintendo's sustained hardware success complicated the narrative.
Meta-value: Kamiya was wrong about dominance but right about adaptation. FromSoftware's global success came precisely through Kamiya's prescribed mechanism—mechanical transparency, online integration, reduced narrative localization barriers—without sacrificing Japanese design identity. The prediction's survival required partial falsification.
A-Tier: Correct Enough, Costly Enough
Cliff Bleszinski — "Single-Player Games Are Dying" (various, 2012–2017)
Bleszinski's trajectory—Gears of War to LawBreakers collapse—makes this prediction tempting to dismiss as sour projection. Yet between 2012 and 2017, major publishers genuinely deprioritized linear campaigns: EA shuttered Visceral during Dead Space 3's pivot to co-op, Activision fused Call of Duty into multiplayer-first annual cycles, Sony's first-party output was the exception proving the rule.
The correction: Major single-player revivals in 2018–2023 didn't disprove Bleszinski—they arrived after the business model he described had already extracted its toll. Visceral never revived. The "death" was of a specific budget tier, not the form itself. Bleszinski conflated market contraction with extinction, a common error in trend prediction.
Trade-off: Bleszinski's urgency pushed some developers toward live-service prematurely; his accuracy made the warning self-fulfilling for mid-budget studios.
Industry Consensus — "VR Will Be Mainstream by 2016" (2014–2016)
Facebook's $2 billion Oculus acquisition, Valve's hardware investment, and major studio bundling deals created genuine momentum. The prediction failed on "mainstream"—VR remains niche in 2025—but succeeded in establishing persistent infrastructure: standalone headsets, hand tracking, medical and training applications.
Why not S: The reasoning was shallow (price drops solve everything) and the consequence was investor capital burned on false timelines. Meta's Reality Labs losses have been substantial. The prediction lacked Barnett's structural analysis; it was extrapolation dressed as insight.
B-Tier: Right for Wrong Reasons, or Wrong Without Residue
"Mobile Will Kill Dedicated Handhelds" (2010–2015)
Nintendo 3DS sales disappointed. PlayStation Vita died. The prediction seemed validated—except Nintendo Switch (2017) proved hybrid portability had value mobile couldn't replicate: physical controls, first-party software density, consistent performance. Mobile "killed" a form factor, not the use case.
Hidden variable: The prediction missed that "handheld" and "mobile" aren't competing on hardware but on attention economics. Free-to-play monetization and notification psychology won where raw portability alone couldn't explain outcomes.
"Cloud Gaming Will Replace Local Hardware" (2018–2022, Google Stadia era)
Google's investment lent credibility. Stadia's shutdown in 2023 and Xbox Cloud Gaming's supplementary—not replacement—role falsified the timeline. Latency economics and infrastructure unevenness proved more durable than assumed.
Why B, not lower: The underlying technology improved substantially; the prediction erred on adoption speed, not possibility. Some residue remains for future reassessment.
C-Tier: Correct but Obvious, or Obvious and Wrong
"Battle Royale Is a Fad" (2018–2019)
After PUBG and Fortnite exploded, contrarians predicted collapse. Apex Legends (2019) and continued Fortnite evolution proved endurance. The prediction's failure mode is instructive: it treated genre as fixed rather than template. Correctly identifying that specific implementations fade would have been useful; declaring the format dead was not.
"NFTs Will Revolutionize Game Ownership" (2021–2022)
Major publishers invested; player revolt and regulatory ambiguity followed. The prediction was testable quickly and collapsed fast. Low meta-value: Unlike Barnett's analysis, there's little structural insight to salvage. The mechanism (artificial scarcity in digital goods) contradicted gaming's actual economies (abundance as engagement driver).
D-Tier: Harmful When Believed, Forgettable Otherwise
"You Must Be on Every Platform" (2015–2019 Indie Consensus)
Porting to Switch, Xbox, PlayStation, and PC simultaneously burned small studios with divergent certification requirements and support obligations. The prediction came from platform holders and success-story outliers, not median economics. Failure state: A few breakout hits succeeded multiplatform; dozens of indies drowned in porting debt without equivalent marketing.
Elimination logic: Platform choice is audience-match, not presence-maximization. A focused release with community building often outperforms scattered availability.
F-Tier: Actively Misleading, Often Repeated
"Graphics Are Mature; Innovation Must Move Elsewhere" (Perennial, 2007–Present)
Repeated every hardware generation by commentators seeking contrarian positioning. Each cycle demonstrated that visual and systemic innovation are coupled, not sequential. Modern games with loading-free traversal or dimension-hopping mechanics require hardware advances; the graphical and mechanical are inseparable.
Why persistent: The take flatters writers' sophistication while requiring no technical engagement. It survives because it's unfalsifiable in the moment—"eventually" graphics must plateau—yet always wrong in the present.
Meta-Caveats: How to Use This List
Patch sensitivity: Predictions about business models age faster than design analysis. Barnett's 2009 warning required the 2016–2022 live-service explosion to fully validate. A "wrong" prediction may need 5–10 years to resolve.
Role-specific notes: Developers should weight Barnett and Kamiya heavily; investors should distrust Bleszinski's urgency and the VR consensus's timeline specificity; players should ignore platform-presence advice entirely.
Self-correction: An earlier draft placed Bleszinski in C-tier for personal LawBreakers failure. Reconsideration: the prediction's industry influence—actual studio closures, actual portfolio shifts—warrants A-tier despite his individual outcome. Predictions are social objects; their consequences exceed their authors.
Quick Reference: Who Each Tier Serves
| Tier | Best For | Skip If |
|---|---|---|
| S | Designers avoiding derivative traps; historians seeking structural patterns | You need actionable business timelines |
| A | Investors understanding market transitions; developers timing genre re-entries | You want comfort about single-player's health |
| B | Technologists tracking long infrastructure shifts | You need consumer purchase guidance |
| C | Case studies in format vs. implementation confusion | You want reliable forecasting models |
| D | Indies learning resource allocation; platform strategy students | You're a player choosing where to buy |
| F | Rhetoric analysis; identifying lazy contrarianism | You want any practical application |
Paul Barnett is chief creative officer at Wargaming; his 2009 comments were originally reported in 2009 and resurfaced in PC Gamer's April 2026 retrospective by Wes Fenlon with contributions from Mathew Kumar.
