We can’t hold on! An index tracking overseas banks in one week from September 18 dropped into negative territory: the number of analysts cutting profit forecasts for the first time in 23 weeks exceeded those raising them. This ended the longest streak of upward revisions since September 2021. This index counts the number of firms—not how much forecasts were changed. The most concentrated downgrades were in three sectors: consumer, materials, and financials.

The other part hasn’t budged. In the same week, a basket of U.S. large-cap stocks still had its profit-growth expectations written at about 32% this year. The forecast for Q3, calculated from June 30, continued to move up by 1.4%; historically, at the same point in time on average it declines by 2.2%. What’s been settled is the plus-and-minus adjustments along the side—the overall level didn’t fall.

The cost side lines up: on September 16, the U.S. monetary authority raised the policy rate by 25 basis points, lifting the range to 3.75% to 4%—the first time in three years. During the day, the yield on the 10-year Treasury note touched 5%. On Wednesday, an international organization raised its global inflation forecast for 2027. Once the base is pushed higher, the room to revise upward is naturally limited.

I lay out four names and read two places: how far each is from its historical record, and the direction of the adjustments in this round. Ethereum is still about 45.9% away from its record, and about 10.23% over the past seven days; Tron is about 20.3% away, about 2.42% over the past seven days, but has already turned negative over the past thirty days. Cardano is about 92.3% away, about 22.21% over the past seven days; Avalanche is about 92.9% away, about 36.22% over the past seven days. All four names—and also the platform token BNB—are on this list. The order over the past seven days matches the level sequence exactly: the closer one is to its record, the flatter the plus/minus adjustments.

This index’s plus-and-minus adjustments are only the side tweaks; the level will wait for the next rewrite. Whether it stays or not depends on two readings: if, over the next two weeks, the number of firms being cut falls back to the side that was increasing, while the full-year expectation still hovers near 32%, that would suggest it’s only a quarter-end calibration. If the full-year expectation is revised downward as well, and the downgrades spread from those three categories to more, then this way of reading is invalid. The same question can be checked on the Binance spot side yourself: prices change daily, and the payoff of balance-type products and the quarterly announcements for the platform token BNB won’t move until the next adjustment—these pages are publicly visible to the outside. This article is a record of viewpoints and does not constitute investment advice.$ALLO $SHELL #华尔街盈利预期23周来首现看空