90% of coins go to zero. The 10% that 10x-100x share a few patterns. Here’s how to filter them:
1. The 4 filters every gem passes
Filter 1: Low market cap + real narrative fit
- Low cap: $10M-$500M. Below $100M is where 100x happens. Above $2B and you need billions in new money
- Narrative fit: Is it part of what’s hot right now? May 2026: AI agents, RWA, Bitcoin L2s, modular chains
- Red flag: $2B cap coin trying to be “the next ETH”. Too late
Filter 2: Tokenomics don’t screw you
Check TokenUnlocks before anything else:
- Supply: Low circulating supply vs FDV. If FDV is 10x circulating, VCs will dump on you
- Unlocks: No big unlocks in next 3 months. A 20% unlock = instant 15% drop usually
- Insider allocation: <20% to team/VCs. If it’s 40%, you’re exit liquidity
- Real example: BNB burns supply. DOGE inflates forever. One goes up, one doesn’t
Filter 3: Real usage or real catalysts
“Gem” doesn’t mean “good tech”. It means “people will buy it soon”.
- Usage: Daily active users growing, TVL rising, fees paid increasing. Check Dune, DefiLlama
- Catalysts coming: Token launch, CEX listing, mainnet, airdrop, partnership in next 30-60 days
- Red flag: “Revolutionary tech” with 50 DAU and no timeline
Filter 4: Liquidity + distribution
- Listed: On Binance, Bybit, OKX, or has $5M+ daily volume on DEX. If you can’t exit, it’s not a gem
- Holder distribution: Top 10 wallets don’t own 80%. Use Bubblemaps or Dextools to check
- No team wallets dumping: Check Etherscan. If team is selling into pumps, walk away
2. Where to find them before CEX listing
Gems are found early. By the time it’s on Binance, it’s already 5x.
Best hunting grounds:
1. Kaito/Yaps: Check what CT builders are shilling. If 10 smart accounts talk about it unprompted, look
2. Presales/Launchpads: BNB Launchpool, Base launchpads, Solana launchpads. Risky but 10-50x potential
3. Narrative Twitter: Search “AI agents”, “RWA”, “Bitcoin L2” and see what’s trending
4. Dexscreener/Dextools: Sort by volume spike + new pairs. Filter for >$500K liquidity and locked LP
5. Airdrop farmers: Projects with points systems often pump post-TGE. Check Dune for farm activity
3. Red flags that mean “scam” or “dead”
If you see 2+ of these, skip it:
- Team anonymous + no doxxed advisors
- Website made in 1 hour, whitepaper is 3 pages of buzzwords
- 90% of supply in 5 wallets
- LP not locked/burned on DEXs
- Only marketing, zero product or testnet
- Promises 100% APY “risk-free”
- Telegram full of bots, zero real discussion
4. The 15-minute due diligence checklist
Before putting $100 in, check:
1. CoinGecko/CoinMarketCap: Age, market cap, volume
2. TokenUnlocks: Any cliffs coming?
3. Dextools: Liquidity locked? Holder distribution?
4. Twitter: Real followers or bots? Are devs active?
5. Dune/DefiLlama: Real usage or fake volume?
6. Contract: Verified on Etherscan? No mint function?
If it fails 2+ checks, don’t touch it.
5. How to play it without getting wrecked
Even if you find a gem, 80% of people lose money because of execution.
Position sizing:
- Gem plays = 1-3% of portfolio max. If it goes to zero you don’t care
- Never go all-in on one. Have 5-10 shots
Entry/Exit rules:
- Entry: Buy dips, not green candles. If it’s up 80% today, wait
- Exit: At 3x, sell initial + 50%. Let rest ride to 10x or stop-loss
- Stop-loss: If thesis breaks, cut at -30%. Don’t “hold for the tech”
Rule: Gems make you rich on paper. Profit-taking makes you rich for real.
6. What “gem” looked like in past cycles
- 2020: UNI at $1 → $45. Uniswap had real usage
- 2021: SOL at $1 → $260. Fast chain + FTX backing + NFT narrative
- 2024: WIF at $0.01 → $4. Meme + timing + Binance listing
- 2026 so far: AI agent tokens, RWA coins, BTC L2s. If you caught them at $20M cap, they did 10-30x
Pattern: Real usage OR perfect timing + narrative + CEX listing.
Bottom line
A gem coin has 3 things:
1. Story: Fits current narrative people FOMO into
2. Math: Tokenomics don’t guarantee a dump
3. Timing: Catalyst coming before everyone notices
You find them by spending 30 min/day reading Twitter, checking Dune dashboards, and ignoring 99% of what you see.
Most important: Even the best gem hunters are wrong 80% of the time. Size small, take profits, and let the 1 winner pay for the 9 losers.
