# The State of GameFi / Web3 Gaming, 2025 to Mid-2026: A Data-Driven Post-Mortem and Survivor Analysis

## TL;DR
- GameFi is in structural collapse, not a correction: the Caladan analysis (April 2026, via CoinDesk) found roughly 93% of about 3,200 Web3 gaming projects effectively dead, tokens down about 95% from 2022 peaks, and studio funding down 93% by 2025, after the sector burned through $12B to $15B because, in CoinDesk's words, it was "chasing a token-driven future that gamers never bought into."
- A small cohort survives by inverting the model: build the game first, treat the token as infrastructure. Survivors (Pixels, Gods Unchained, Off The Grid, Illuvium, Alien Worlds, Splinterlands, and Axie in evolved form) share gameplay-first design, hybrid monetization, and homes on gaming-optimized chains (Ronin, Immutable, opBNB, WAX).
- Headline traction numbers are heavily inflated. "World of Dypians 135M+ wallets" and "Lumiterra +9,451%" reflect smart-contract interactions and low-base incentive spikes, not unique humans. Genuine sector activity was about 4.66M daily unique active wallets in Q3 2025 (DappRadar), and the leading analytics provider, DappRadar, shut down in November 2025, removing the sector's reference dataset.

## Key Findings
1. The failure rate (~93%) exceeds the dot-com bust. Caladan, built on a ChainPlay review of 3,200+ titles, found the average GameFi title survives about four months before its token falls 90%+ and daily users drop below 100.
2. Capital rotated out wholesale. Gaming's share of Web3 venture investment fell from 62.5% in 2022 to single digits in 2025; per Caladan, "of the venture firms that invested in the sector, 58% have realized losses between 2.5% and 99%."
3. Annual gaming funding cratered from a $10B peak in 2022 and about $4B in 2021 to $293M in 2025 (BGA 2025 report), with the bulk going to infrastructure, not games.
4. Gaming remains the most-used Web3 category (25% of all dapp activity in Q3 2025) but on a shrinking base: 4.66M dUAW, down from 7.4M at end-2024.
5. The shutdown wave is broad and ongoing: 300+ gaming dapps went inactive in Q2 2025 alone (8% of listed gaming dapps), and closures continued into 2026 (Forgotten Runiverse, Pixiland, Wildcard).
6. Survivors are real but small: Illuvium at ~40,000 DAU, Pixels with Web2-comparable retention, Off The Grid on Steam. None has cracked mainstream scale.
7. The data layer itself broke. DappRadar shut down on November 17, 2025, so no standardized sector-wide unique-user data exists past Q3 2025. Every 2026 figure is directional.

## Section 1: Executive Summary
The blunt verdict as of mid-2026: GameFi as conceived in 2021 to 2022 has failed. The most authoritative recent post-mortem, the Caladan analysis published April 22 to 23, 2026 and covered by CoinDesk, concluded that roughly 93% of about 3,200 Web3 gaming projects are now effectively dead, average token prices are down about 95% from their 2022 peaks, and funding to studios collapsed 93% by 2025. Caladan estimates $12B to $15B in venture capital, token sales, and NFT proceeds flowed into the sector between 2020 and early 2026, and that nearly all of it was "wiped out." CoinDesk's framing was that "Web3 gaming burned through up to $15 billion chasing a token-driven future that gamers never bought into."

This is corroborated independently. DappRadar reported that 300+ gaming dapps went inactive in Q2 2025 alone, about 8% of all gaming dapps it listed. The Blockchain Game Alliance (BGA) 2025 State of the Industry Report documented annual funding falling to approximately $293M, "a massive drop from $4 billion in 2021 and the peak of $10 billion in 2022." ChainPlay's earlier sector study reached the same "93% of GameFi projects are dead" conclusion that Caladan echoes.

What remains is a thin layer of survivors that share one trait Caladan identifies: they built a game worth playing first and treated the token as supporting infrastructure rather than the product. The most telling structural signal of the sector's deterioration is that DappRadar, the de facto standard analytics provider for the entire dapp economy, shut down on November 17, 2025. Founders Skirmantas Januškas and Dragos Dunica announced via X that "running a platform of this scale became financially unsustainable in the current environment, and after exploring every option, we had to make the difficult decision to wind things down." There will be no DappRadar Q4 2025 or 2026 quarterly gaming reports; the Q3 2025 report (published October 16, 2025) is the last. For a sector that sold itself on transparent, verifiable on-chain metrics, losing its reference dataset is more than an inconvenience: it means the second half of 2026 will be analyzed on fragmented, self-reported, and bot-contaminated data.

The honest read: this is a reset, not a death, but the survivors are operating at a fraction of mainstream gaming scale, and the speculative cycle that defined the category is over.

## Section 2: Why Blockchain Gaming Has Struggled to Gain Traction

**UX and onboarding friction.** The single biggest practical barrier. Installing a wallet, funding it, signing transactions, and paying gas is brutal compared to a Web2 email signup. Industry funnel data shows only about 3% to 8% of visitors to a well-built Web3 site connect a wallet, and only 20% to 40% of those complete a first transaction; roughly 65% of Web3 users drop off immediately after connecting a wallet. The BGA's 2024 report had already flagged onboarding as a top challenge, though it noted improvement (from 79.5% of developers citing onboarding friction in 2023 to 53.9% in 2024).

**Unsustainable tokenomics and P2E death spirals.** The defining structural flaw. Axie Infinity is the canonical case study. Its SLP token had an uncapped supply with weak sinks; the economy had no real inflow except new players buying in. When player growth slowed, SLP emissions overwhelmed demand, rewards collapsed, and the model unwound. Per DappRadar data cited by Caladan, Axie peaked at 2.7 million daily active users in summer 2022, and by April 2026 had approximately 5,500, a loss of 99.8%. AXS fell from a November 2021 high near $155 to $165 down to roughly $2.40 to $2.90 by January 2026 (down 98%+); SLP fell 99%+. A CertiK analysis of 2022 tokenomics failures found that investigations into "Terra, Celsius Network, Beanstalk, Fortress, Axie Infinity, Solend, and Babylon Finance...have caused over $790 million in direct asset losses." The same mechanism killed the Telegram tap-to-earn wave: Hamster Kombat lost 96% of users within six months, its market cap falling from about $300M (August 2024) to about $12M (February 2025).

**Gameplay quality and misplaced priorities.** Studios built economies before games. Pixelmon raised $70M in a February 2022 NFT mint and shipped no public game four years later. Square Enix's Symbiogenesis offered a passive point-and-click treasure hunt with little gameplay and only 554 unique NFT holders across its 10,000-item collection. As Nicolas Pouard of Ubisoft argued in the BGA report, the "Web2.5" branding failed; games need to be genuinely fun with blockchain in a support role.

**Reputation and trust deficit.** In the BGA 2025 survey, scams, fraud, and rug pulls were cited by 36.0% of respondents as the greatest threat to industry credibility, followed by poor-quality games at 20.4% and Ponzi-style economic models at 12.6%. The Ronin bridge hack poisoned mainstream confidence: per CoinDesk, Sky Mavis's Ronin Network lost 173,600 ether and 25.5 million USDC in March 2022, worth in excess of $625 million, later attributed by the FBI to North Korea's Lazarus Group. Even at the peak, just 12% of gamers had ever tried a crypto game (Coda Labs survey cited by Caladan).

**Funding drought and high user-acquisition costs.** Funding collapsed (see Section 6). At the same time, acquiring real users is uniquely expensive in Web3: a wallet connection is not an engaged user, install attribution is broken on web, and Apple's ATT privacy changes plus the 30% App Store/Play Store cut on in-app purchases make mobile distribution hostile. The BGA found 32.6% of studios cited lack of funding as their single biggest company challenge.

**Regulatory uncertainty.** Over 60% of gaming studios cite unclear crypto regulations as a primary barrier (in-game token classification, tax treatment). Notably, the BGA found a shift in sentiment: 64.4% of respondents now expect regulation to positively influence the sector, viewing legal clarity as a route to institutional capital.

**Technical and ecosystem fragmentation.** High gas fees on Ethereum, fragmented Layer-2 adoption, and the proliferation of incompatible app-chains splintered liquidity and players. This drove migration to gaming-specific low-fee chains (Ronin, opBNB, Immutable zkEVM, WAX, Sei, Kaia) but at the cost of cross-ecosystem network effects.

Together these forces produced mass attrition: the average title burned its runway before achieving product-market fit, the token collapsed before the game shipped, and capital fled to AI and real-world assets.

## Section 3: Major GameFi Project Shutdowns (2025 to 2026)

The broader wave: DappRadar recorded 300+ gaming dapps going inactive in Q2 2025 (about 8% of all listed gaming dapps). BitPinas counted at least 22 crypto games that shut down or dropped Web3 features in 2025, with closures continuing into 2026.

| Project | Approx. date | Reason | Funding raised | Current status |
|---|---|---|---|---|
| Mystery Society (Deadrop) | Jan 2025 | Could not raise capital after 3 years of development | n/a | Shut down |
| Square Enix Symbiogenesis | July 31, 2025 | Limited engagement, weak gameplay; only 554 NFT holders | n/a (Square Enix) | Ended; later revived as final season on Sony's Soneium |
| Ember Sword | May 2025 | Burned $18M over 7 years, never shipped viable game | ~$18M | Shut down, no refunds |
| Nyan Heroes (9 Lives) | June 30, 2025 | Could not secure funding despite 1M+ playtest players | ~$13M | Shut down |
| Blast Royale | June 30, 2025 | Could not scale; open-sourced codebase | ~$5M | Shut down |
| Rumble Kong League | May 2025 | $FAME token troubles despite Steph Curry endorsement | part of ~$22M weekend cohort | Shut down |
| Tatsumeeko: Lumina Fates | 2025 | Funding/market conditions | n/a | Shut down |
| Pirate Nation (Proof of Play) | Aug 2025 | Insufficient player base, high on-chain costs | n/a | Fully on-chain game closed; pivoted to Arcade on Abstract |
| Champions Ascension | March 2025 | Paused to focus on REACH Labs | n/a | Paused |
| Age of Dino (Xterio/Binance Labs) | Sept 30, 2025 | Sunset; token/achievement snapshot for migration | n/a | Shut down |
| Captain Tsubasa Rivals | Nov 28, 2025 | Could not sustain ecosystem | n/a | Polygon/Oasys versions closed; Telegram/LINE mini-app continues |
| The Walking Dead: Empires (Gala) | July 2025 | Could not maintain momentum | n/a | Support ended |
| Wildcard (Playful Studios) | April 27, 2026 | Exhausted funding; launched one incomplete map | ~$55M ($46M Series A 2022 + $9M May 2025) | Multiplayer + Thousands Network shut down |
| Pixiland | Jan 15, 2026 | Market volatility, regulatory uncertainty; canceled TGE | bootstrapped | Pivoted to Web2 |
| Forgotten Runiverse | Jan 27, 2026 | Financially unsustainable to maintain live MMORPG | backed by Alexis Ohanian among others | Offline indefinitely (250K+ players, ~7K peak concurrent at launch) |
| Bloktopia | Jan 6, 2026 | Loss of traction | n/a | Shut down |
| Hamster Kombat | 2025 (collapse) | Tap-to-earn model unwound | n/a | Still live but lost 96% of users in 6 months |
| Pixelmon | n/a (zombie) | Raised $70M, never shipped public game | $70M | Effectively defunct |

Other 2025 closures named by tracker Tr3vor and BitPinas include Battlebound/Anterris, Blade of God X, Derby Race, Goombles, Junglexyz, Kryptomon, Loot Legends, and Valeria Studios (Land Before the War). Caladan also notes studios such as Mojo Melee/Planet Mojo and Treasure DAO publicly pivoting from gaming toward AI.

## Section 4: Successful Surviving GameFi Projects in 2026

A critical distinction must precede this section: "still online" is not "viable," and "wallet activity" is not "unique users." Several widely cited survivor metrics conflate smart-contract interactions or incentive-driven spikes with genuine human traction. With DappRadar gone after November 2025, the figures below are a mix of its final Q3 2025 data and post-shutdown self-reported or third-party estimates; weight them accordingly.

**Pixels (Ronin).** The clearest survivor among accessible Web3 games. The browser-based social farming MMO peaked above 1M daily active user wallets in 2024 and, crucially, posted Web2-comparable retention: month-2 retention above 40% and month-6 above 20% after migrating to Ronin. The PIXEL token reached a fully diluted valuation above $2B before falling about 95% from its all-time high. The team deliberately reduced daily active wallets to pursue higher-lifetime-value players, a VIP access model, and "net ecosystem spend" (in-game spending exceeding token distribution). Strength: genuine retention and economic discipline. Risk: token overhang and a smaller, monetization-focused user base. In 2026 it launched an AI economy-management platform, Stacked.

**Gods Unchained (Immutable zkEVM).** A mature trading-card game often cited as the quality benchmark. After fully migrating assets to Immutable zkEVM, its NFT trading volume rose 507% to $27.2M in Q3 2025 (DappRadar), a top gaming collection. But the underlying player base is modest: third-party tracker Cards Unchained showed around 10,000 concurrent players in recent months, and Immutable laid off about 20 staff (~6% of workforce) from the Gods Unchained team. Strength: skill-based gameplay, real card ownership, infrastructure backing. Risk: aging player base, no AAA catalyst yet.

**Off The Grid (GUNZ / Avalanche).** Caladan's flagship example of "game first." The Gunzilla Games cyberpunk battle royale, backed by $100M+ in funding and co-created with director Neill Blomkamp, became the first major Web3 title on Steam (July 17, 2025), accessing Steam's 130M+ users. It reportedly has around 450,000 daily players and 13M+ users; GUNZ has processed 740M+ transactions across 14M wallets. Crucially, blockchain features are optional and off-platform on Steam. Strength: AAA production, mainstream distribution, token-after-product sequencing. Risks: the GUN token is down 75%+ from its March 2025 launch, and in April 2026 multiple staff reported delayed salaries, a warning sign even for a flagship.

**Illuvium (Immutable).** The AAA-aspiring auto-battler/RPG. Caladan cites 40,000 daily active users with 35% to 45% monthly retention, close to the 40% to 50% benchmark of traditional mobile titles. Strength: production quality, deflationary asset-fusion economy, hybrid free-to-play. Risk: small absolute numbers, long development runway, token tied to broader altcoin sentiment.

**Alien Worlds (WAX).** A high-transaction survivor: about 420,000 wallets generating 599 transactions each per month in Q3 2025, leading WAX's 687M total transactions. But independent analysis (Naavik) estimates 30% to 40% of "players" may be bots, and the mining economy's sustainability depends on TLM demand outpacing inflation. Strength: longevity, low entry cost, DAO governance. Risk: bot-heavy activity, questionable economic durability.

**Axie Infinity, evolved state (Ronin).** The fallen flagship is restructuring rather than dead. In early 2026 it halted SLP emissions, introduced bAXS (an account-bound reward token to curb speculation and bot farming), and an Axie Score system linking rewards to engagement. AXS briefly rallied 60%+ on the reforms. An MMORPG, Atia's Legacy, is planned. Strength: brand, Ronin infrastructure, genuine attempt at sustainable redesign. Risk: DAU collapsed to roughly 99,000 by end-2025 (Caladan), high token concentration.

**Splinterlands (Hive).** The endurance case: 6+ years of continuous operation with a live-tested economy, historically around 590,000 wallets. Strength: proven sustainability, fast accessible matches. Risk: aging, niche.

**World of Dypians (opBNB / multichain) - metric scrutiny required.** WoD marketed itself as the "#1 most popular Web3 game" with "over 135M active wallets engaging during Q3 2025." This figure is a conflation. DappRadar's own wording is that "135 million wallets engage with its smart contracts," and the game averages only 12.8 transactions per wallet. CoinMarketCap's own analysis notes WoD's daily active users "can swing from over 100k to below 10k," and the project claims roughly 3.7M to 4M monthly on-chain players, not 135M unique users. The WOD token fell 77% over a 90-day window, with about 67% of supply still unreleased (structural inflation overhang). Verdict: real activity exists, but the 135M figure measures contract calls, not humans, and should not be read as unique-user traction.

**Lumiterra (Ronin) - metric scrutiny required.** DappRadar reported Lumiterra attracted 18.6M wallets in Q3 2025, "a huge increase of 9,451%." This is a low-base, incentive-driven spike off a near-zero starting point following its Ronin launch, with only 15.1 transactions per wallet. It is a promising title, but the percentage is a launch artifact, not evidence of durable mass adoption.

**Others.** Big Time reports 80,000+ active players with 2.3-hour average sessions (cosmetic-NFT, skill-based). Star Atlas remains a long-runway, high-production Solana bet still largely pre-full-release. The Sandbox restructured (about 50% staff laid off in Q3 2025, Animoca taking greater control, credit-card payments added) and never sustained more than about 4,500 daily on-chain users despite $93M from SoftBank. Seraph: In the Darkness sustains steady opBNB activity.

## Section 5: Common Success Factors of Survivors vs Failures

| Factor | Survivors | Failures |
|---|---|---|
| Design priority | Game-first; fun without token (Off The Grid, Illuvium, Gods Unchained, Pixels) | Token/economy-first; speculation as the product (Pixelmon, most P2E) |
| Economic model | Hybrid monetization, ownership, sinks; "net ecosystem spend" (Pixels); deflationary fusion (Illuvium); account-bound rewards (Axie bAXS) | Inflationary P2E with uncapped emissions and weak sinks (Axie SLP, Hamster Kombat) |
| Token sequencing | Token after a working product (Off The Grid: 6 months live + 480M+ testnet txs before GUN listed) | Token/NFT mint before any playable build (Pixelmon $70M, Ember Sword) |
| UX / friction | Optional/abstracted blockchain, mainstream distribution (Steam, Epic, browser), stablecoin payments | Mandatory wallet setup, gas friction, crypto-native-only |
| Retention mechanics | Web2-comparable retention (Pixels 40%+ month 2; Illuvium 35-45% monthly); live-ops, seasons | Retention collapses when token rewards fade (Matr1x FIRE <10%; Nyan Heroes) |
| Infrastructure / chain | Gaming-optimized chains: Ronin, Immutable zkEVM, opBNB, WAX, Avalanche L1; low fees, fast blocks | Expensive/fragmented chains, abandoned smart contracts, no home-chain alignment |
| Funding discipline | $100M+ deployed against shipped product (Gunzilla); fund-managed runway | Raised tens to hundreds of millions on hype, removed pressure to ship |
| Longevity proof | 6+ years live (Splinterlands); multi-year retention | Average title dead in ~4 months (Caladan) |

The lesson is consistent across both BGA survey sentiment and Caladan's quantitative post-mortem: survivors made the token optional and the game essential; failures did the reverse.

## Section 6: Current Market Metrics and Trends (2026)

**Read this before any number below.** The sector's primary measurement instrument no longer exists. DappRadar, the de facto standard for dapp and gaming analytics, shut down on November 17, 2025. Every sector-wide unique-active-wallet figure in this report is therefore from Q3 2025 (its final report, published October 16, 2025) or earlier. Post-Q3-2025 data is sparse, fragmented, and sourced from less standardized providers: project self-reporting, raw chain explorers, and smaller trackers, none of which apply DappRadar's deduplication or bot-filtering at scale. There will be no Q4 2025 or 2026 DappRadar quarterly. Treat every 2026 metric below as directional, not precise, and assume wallet counts overstate unique humans by an unknown but material margin. The disappearance of the category's reference dataset is itself a signal: analytics infrastructure follows the money, and the money left.

**User activity.** Gaming led all Web3 categories in Q3 2025 with 4.66M daily unique active wallets (DappRadar), down 4.4% QoQ, and held 25% of all dapp activity (up from 20.1% in Q2 2025) even as the broader dapp industry fell 22.4% to 18.7M dUAW. The trajectory through 2025 was downward: 7.4M dUAW at end-2024, 5.8M in Q1 2025, 4.8M in Q2 2025 (lowest since early 2023), 4.66M in Q3 2025. No DappRadar data exists past Q3 2025 because the company shut down on November 17, 2025.

**Top chains (Q3 2025, by dUAW).** opBNB led at 1.05M (World of Dypians, Seraph); Sei second at 802,000 (+86%); Nebula/SKALE about 647,000; Kaia 462,000 (+229%, driven by engagement platform KGeN); Ronin 419,000 (+55%; Axie, Lumiterra, Pixels, The Machines Arena). By raw transactions, WAX led with 687M (Alien Worlds), with GUNZ testnet second at 151M (Off The Grid). Immutable zkEVM led gaming NFT trading.

**NFT trading volume.** The top 15 gaming NFT collections generated $135M in Q3 2025, less than 10% of the $1.6B total Web3 NFT volume. Gods Unchained (+507% to $27.2M on Immutable) and Sorare (+347%) stood out. Metaverse NFT trading was just $17M, down 55% QoQ. BitPinas noted gaming token volume surpassed $10M on only three days in the 12 months to early 2026.

**Funding and capital rotation.** This is the starkest data. Annual gaming investment fell from a $10B peak in 2022 and about $4B in 2021 to $293M in 2025 (BGA/DappRadar). Quarterly 2025: Q1 $91M (down 71% YoY), Q2 $73M (down 93% YoY, 74.8% to infrastructure), Q3 $129M (best of a weak year, but only 53% to actual games). Caladan: gaming's share of all Web3 VC fell from 62.5% in 2022 to single digits in 2025, with displaced capital absorbed by AI (~$1.8B), real-world-asset tokenization (~$2B), and infrastructure (~$2.6B). Animoca Brands, the sector's most prolific backer with 380+ Web3 bets, cut pure gaming exposure to about 25% of its portfolio and pivoted toward stablecoins, tokenization, and treasury products.

**Structural shifts.** Five trends define 2026: (1) stablecoins replacing volatile native tokens as in-game currency (27.3% of BGA respondents cited stablecoin payments as a growth driver); (2) account abstraction and gasless/embedded wallets reducing onboarding friction; (3) migration to gaming-optimized chains and app-chains (SANDchain, GUNZ); (4) a decisive shift from user acquisition toward retention as the success metric; (5) hybrid Web2/Web3 monetization with optional, abstracted blockchain.

## Section 7: Outlook and Key Risks for H2 2026 and Beyond

**Best-positioned project types.** (1) Game-first titles with mainstream distribution and optional blockchain (Off The Grid model, traditional-publisher entries like MapleStory N, EVE Frontier). (2) Trading-card and collectible games with genuine ownership and skill (Gods Unchained, Sorare). (3) Accessible, retention-proven social/farming games with economic discipline (Pixels). (4) Infrastructure and chains rather than single-game tokens (Immutable, Ronin), which Caladan singles out as the survivors of the chain layer. A modest early-2026 token rally (AXS, SAND, GALA up roughly 300% off the floor) occurred, but data providers stress these are exceptions off a near-zero base, not a structural reversal; MEXC described the sector as "left for dead."

**Major risks of further failures.** More shutdowns are near-certain. Sky Mavis co-founder Jeffrey "Jihoz" Zirlin said in January 2026 to expect many more closures, including on Ronin, "because of how hard it is to make a successful crypto game." Runway exhaustion remains the proximate killer (32.6% of studios cite funding as their top challenge), compounded by continued capital flight to AI and RWAs and the new threat of AI-powered bots and cheating (38.9% of BGA respondents named this the most serious AI risk). Even flagship Gunzilla showed salary-delay warning signs in April 2026. A second-order risk is now epistemic: with DappRadar gone, weak projects can survive longer on unverifiable self-reported numbers, and investors lose the independent yardstick that previously exposed dead games quickly.

**What mainstream traction would require.** The consensus across BGA, Caladan, and Ubisoft's Pouard is consistent: (1) invisible blockchain, with wallets, gas, and signing abstracted away so Web2 players never see the plumbing; (2) AAA or genuinely fun gameplay that retains players with no economic reward attached; (3) sustainable economies where in-game spending exceeds token emissions and stablecoins replace volatile reward tokens; (4) trust rebuilding, given that 36% of industry professionals still name scams and rugs the top credibility threat; and (5) regulatory clarity, which 64.4% now expect to help rather than hinder.

**Bottom line.** The speculative GameFi thesis is dead. A narrower, healthier "play-and-own" thesis is alive but unproven at scale. The next 12 to 18 months will determine whether the survivors can convert disciplined retention into mainstream businesses, or whether Web3 gaming remains a permanent niche. The single most important signal to watch is whether any survivor demonstrates sustained, organic, unique-user growth (not contract-call counts, not incentive spikes) above prior peaks. As of mid-2026, none has, and the tooling to prove it credibly has itself thinned out.

## Recommendations

**For builders:**
1. Ship a genuinely fun game before issuing any token. The clearest dividing line between survivors and the dead is token-after-product (Off The Grid) versus token-before-product (Pixelmon, Ember Sword).
2. Make blockchain optional and invisible. Use embedded wallets, gasless transactions, stablecoin payments, and mainstream distribution (Steam, Epic, app stores) with off-platform Web3 for those who want it.
3. Engineer net-positive token sinks. Target "net ecosystem spend" (Pixels) and account-bound rewards (Axie bAXS) over uncapped emissions. If your economy needs constant new buyers to pay existing players, it will die; assume a four-month clock if you do not fix it.
4. Optimize for 30-day-plus retention, not wallet connections or airdrop-farmed sign-ups. Benchmark against the 40% to 50% month-2 retention of traditional mobile. Below ~10% (Matr1x FIRE) is a death sentence.
5. Manage runway conservatively. Funding will stay scarce; raise less, ship more, and assume no rescue round.

**For investors:**
1. Discount headline traction metrics aggressively, and more so now that DappRadar is gone. Demand unique-user data and per-wallet transaction depth; treat "wallets engaging with smart contracts" (World of Dypians 135M) and triple-digit percentage spikes off near-zero (Lumiterra +9,451%) as red flags, not green ones.
2. Underwrite the chain/infrastructure layer over single-game tokens. Immutable and Ronin survived as infrastructure; most single-game tokens did not.
3. Require a shipped, retained product before deploying. 58% of VC firms in this sector booked losses; the ones that lost least backed playable products with measurable retention.
4. Watch the rotation. Capital has moved to AI, RWA, and infrastructure; a gaming re-rating requires either a breakout AAA hit with mainstream numbers or a sustained user-growth signal that does not yet exist.

**Thresholds that would change these recommendations:** a survivor demonstrating sustained organic unique-user growth above prior peaks; gaming dUAW (if a credible replacement for DappRadar's tracking emerges) reversing its 2025 decline; quarterly gaming funding recovering past $500M with the majority going to games rather than infrastructure; or a major traditional publisher posting Web2-scale retention on a blockchain title.

## Caveats and Data Limitations
- **The primary data source for the sector no longer exists.** DappRadar shut down on November 17, 2025. All DappRadar figures here are from Q3 2025 (final report, October 16, 2025) or earlier. Post-Q3-2025 sector-wide unique-active-wallet data is sparse, non-standardized, and largely self-reported or estimated by smaller trackers. This limitation now leads Section 6 because it conditions how every 2026 number should be read.
- **Metric definitions are routinely conflated.** "Active wallets" can mean unique humans, contract interactions, or bot-controlled addresses. One person can run many wallets; 30% to 40% of some games' activity may be bots (Alien Worlds). Wallet counts systematically overstate human players.
- **Caladan's "93%" rests on ChainPlay's methodology** and a definition of "effectively dead" (token down 90%+, sub-100 DAU). It is a defensible but constructed threshold, and the $12B-to-$15B funding range is an estimate that varies by source.
- **Funding figures vary by provider.** Caladan/Delphi cite roughly $360M to $390M for 2025; DappRadar/BGA report $293M. The discrepancy reflects different scopes (infrastructure inclusion, deal definitions). Both show an 85% to 93% collapse from the 2022 peak.
- **Survivor metrics are self-reported or third-party-estimated** where primary on-chain data is unavailable, and several (Off The Grid player counts, World of Dypians MAU) originate partly from the projects themselves.
- **The early-2026 token rally is real but small and may not hold;** it is a recovery from the floor, not evidence of renewed fundamentals.
- **Market-research market-size forecasts** (e.g., $124B by 2032 from SNS Insider, and similar 19%+ CAGR projections from Fortune Business Insights and Coherent Market Insights) are projection-driven and conflict sharply with the on-chain collapse documented here; they should be treated as speculative and methodologically distinct from activity data.

---

*Sources cited inline: Caladan analysis (April 2026, via CoinDesk, published ~April 23, 2026); DappRadar State of Blockchain Gaming Q2 and Q3 2025 reports; Blockchain Game Alliance (BGA) 2025 State of the Industry Report; CoinDesk; ChainPlay; CertiK; Naavik; BitPinas Web3 game shutdown trackers (2025 and 2026); CoinMarketCap; Coda Labs survey data; Coherent Market Insights (March 2026). On-chain figures are pre-November 2025 unless otherwise noted.*