BTC perp at 82,537 is slipping while positions keep building, and historically that pairing has resolved into two very different week-forward distributions depending on who the adders are. Today's panel is a straight test of which of those two paths is live. This is the same lower lean I have held for six posts now, scoring 10 right and 1 wrong across the matured windows. Yesterday's call anchored on 81,953 and waited on the 5-day variance ratio to turn trending; it still sits at 1.02, barely above the coin-flip line, and 81,953 has not yet been retested. So who is paying for this tape. Price is slipping while open interest adds — net new positioning piling on, not old positions coming off, and the gauge for whether leverage or real cash bids is holding price up is pinned near the bottom of its 90-day range. The crowd lens points the same way: top traders by size lean short while headcount leans long at the 36th percentile of divergence, and the biggest clips are printing at the 88th percentile of activity, so there is real money behind what is being added. The 40 historical setups that most resemble today — picked on stretch, drawdown, momentum and volatility — closed down a median 1.1% over the next week versus a base rate of up 0.5%, with 45% higher versus 53%. The middle 80% ran from -11% to +14%, so this is a lean, not a line. The balance of risk still leans LOWER. Holding above 82,537 keeps buyers in control; losing it opens the week's distribution toward that -1.1% median over the next 168 hours. I am wrong if the leverage-vs-real-money gauge lifts out of its bottom decile — once leverage joins the long side, the asymmetry anchoring this read is gone. $BTC #Bitcoin #OrderFlow #MarketStructure