FelipeDiniz Polymarket trader just hit 80% win rate across 80 trades on Polymarket while somehow still down $4.2K — the weirdest contradiction in prediction markets right now. FelipeDiniz is a conservative Polymarket trader ranked 2,145,932 with an 80.6% win rate, the kind of stat that should print money. Instead: negative $4,237 total PnL, -91.78% ROI on $48,314 deposited. The math breaks because he's winning individual bets but losing on position sizing and entry prices that don't math. His edge hack is simple: low-risk bets across 81 different markets, averaging $2,550 per trade, 6.6 trades daily. He's not chasing moonshots. He's grinding. The problem? His average entry price sits at $1.4M on markets that don't exist at that level. That's not an edge — that's data noise or serious position accounting issues. His buy-to-sell ratio of 2.1 means he's accumulating more than exiting, which creates exit liquidity problems on smaller Polymarket markets. One red flag screams from the profile: his best trade pulled $6,188 on Heat vs Hawks while his worst single loss hit -$4,397 on an esports LoL bet. That 1.4x ratio between biggest win and loss on an 80% win rate trader tells you the market-picking is solid but the sizing discipline isn't locked. He's won most bets. He's just sized them all wrong. Across 67 closed positions, that compounds fast. Current state: 13 open positions, $683K total volume moved, $44K net deposits after $3.9K in withdrawals. He's still grinding the same strategy despite the drawdown — conservative, diversified, high-frequency. The realism check: this is what 80% accuracy looks like when you treat prediction markets like penny stocks and buy on bad entry prices. FelipeDiniz is the trader everyone should study to learn why win rate is not PnL. Not everyone survives watching their conviction bets lose.