Bitcoin entered the Sept. 21–27 week above $80,000 after rising 4.82% over seven days and 3.82% over 30 days, with CryptoSlate recording $80,323 on Sept. 20. The rebound faces tests from US inflation expectations, Japanese funding costs and cryptocurrency investment flows. The Federal Reserve raised its target range by a quarter percentage point to 3.75%–4% on Sept. 16, while the University of Michigan’s final September consumer survey is due Sept. 25. Its preliminary reading showed year-ahead inflation expectations of 4.6% and consumer sentiment of 47.8%; a softer final reading could support risk appetite, while a firmer result could weigh on the recovery. The Bank of Japan’s overnight rate target of around 1.25%, announced Sept. 18, takes effect Sept. 24 and could increase the cost of yen-funded positions, although implementation alone does not establish that investors will sell Bitcoin. US Bitcoin ETFs recorded net inflows of $433 million on Sept. 18 after $450.4 million of outflows on Sept. 15, illustrating shifting demand rather than a confirmed weekly trend. Kraken’s Sept. 16 integration with X Cashtags makes it easier for users to move from supported tickers on X to Kraken and complete trades, but the announcement did not quantify additional Bitcoin purchases. CryptoSlate’s 90-day model, using an $81,233 reference close, projects a median Dec. 18 terminal price of $95,157, with 20th- and 80th-percentile estimates of $71,826 and $127,070. Its performance edge against the strongest simple benchmark was negative 0.4%, and Polymarket carries zero weight in the model. Separately, September prediction-market thresholds showed odds of 65.5% for $82,500, 34.5% for $85,000 and 52.5% for $77,500 at 10:42 UTC on Sept. 20. Those figures are not additive forecasts for Sept. 27 because contracts use Binance BTC/USDT one-minute highs or lows, have windows extending to September’s end and may use different starting points. The key question is whether new buying persists as inflation expectations and monetary-policy concerns evolve.