Live实时 ·

Intelligence Hub情报中心

What still has room to run — surfaced before the crowd arrives.仍有上行空间的机会 —— 抢在人群之前浮出水面。

Regime格局 Stagflation Q3 Cycle周期 late Liquidity流动性 expanding Fed美联储 Hawkish
Prices from the latest settled build价格来自最近一次结算构建
News新闻流 accruing评分中 Alt-Data另类数据 graded已评分 Radar背离雷达 graded已评分 Buy-Board买入榜 ungraded未评分 Policy政策观察 accruing评分中 Special特殊事件 ungraded未评分

Emerging edge萌芽优势

1
A leading desk is positioned; the tape has not confirmed yet. Watch — don't chase.领先信号台已建仓，盘面尚未确认。观察 —— 不要追高。
TKO$189.64 early早期⚡ entry入场 T2 14
RS -8% vs SPY · buy-board bottoming (blocked)
room left剩余空间 81%

Discovery发掘

what the desks miss各台遗漏之处 14
Leads no desk has picked up yet. Watch — a lead is not a signal.尚无信号台采纳的线索。观察 —— 线索不等于信号。

Exhausted — already run透支 —— 已跑完 12

Desks lean bearish and price is rolling over. De-risk — not a long.各台偏空且价格见顶回落。降低风险 —— 不宜做多。

Dated catalysts有日期的催化 12

Events with a known date. Context for timing — not a reason to buy.有确定日期的事件。用于把握时点 —— 而非买入理由。

Command指挥榜

room left剩余空间 30
Most room left first. Nothing here is an entry unless it carries ⚡.剩余空间最大者居前。未带 ⚡ 者均非入场信号。

Track record历史战绩

is the edge real?优势是否真实？
5d天 measuring测量中
rank skill排名预测力 -0.012
scored已评分 134,675
emerging beat SPY?萌芽跑赢标普？ -0.5% no否
exhausted lagged SPY?透支跑输标普？ -0.3% yes是
10d天 measuring测量中
rank skill排名预测力 -0.009
scored已评分 127,592
emerging beat SPY?萌芽跑赢标普？ -1.3% no否
exhausted lagged SPY?透支跑输标普？ -0.7% yes是
21d天 measuring测量中
rank skill排名预测力 -0.013
scored已评分 109,997
emerging beat SPY?萌芽跑赢标普？ -2.7% no否
exhausted lagged SPY?透支跑输标普？ -0.7% yes是
63d天 measuring测量中
rank skill排名预测力 -0.100
scored已评分 42,818
emerging beat SPY?萌芽跑赢标普？ -0.6% no否
exhausted lagged SPY?透支跑输标普？ -3.8% yes是
Desk scorecard研究台记分卡measuring测量中Not enough evidence to call it yet证据尚不足以下结论 detail详情

Every name this desk surfaces is scored against what the stock actually did next, measured against SPY. Context only — never a trade instruction. 本台每个入选标的都以其之后的实际走势评分，并与标普对比。仅作参考 —— 绝非交易指令。 223,494 names recorded · 12,886 drops tracked 已记录 223,494 个标的 · 已追踪 12,886 次剔除

Horizon周期 graded已评分 rank skill排名预测力 trend skill趋势预测力 vs SPY对比标普 status状态
5d日 129420 -0.016 +0.005 +3.4% measuring测量中
10d日 118562 -0.012 +0.007 +5.5% measuring测量中
20d日 103384 -0.023 +0.007 +5.5% measuring测量中
30d日 82693 -0.058 -0.009 +0.3% measuring测量中
60d日 0 — — — measuring测量中
90d日 0 — — — measuring测量中

Before and after the engine overhaul引擎改版前后对比

5d日 hub-v3-trajectory -0.003 n=95080 · +4.8%
10d日 hub-v3-trajectory +0.011 n=84245 · +8.5%
20d日 hub-v3-trajectory +0.013 n=69261 · +4.4%
30d日 hub-v3-trajectory -0.018 n=48737 · +2.9%

What the numbers say数据说明了什么

  • ● 5d: score VELOCITY IC (+0.005) exceeds LEVEL IC (-0.016) — the change in the desk's score ranks winners better than its level.5d: score VELOCITY IC (+0.005) exceeds LEVEL IC (-0.016) — the change in the desk's score ranks winners better than its level.
  • ▲ 10d: 'exhausted' (-0.7%) outperforms 'emerging' (-2.4%) — possible residual inversion in the lifecycle staging.10d: 'exhausted' (-0.7%) outperforms 'emerging' (-2.4%) — possible residual inversion in the lifecycle staging.
  • ● 20d: score VELOCITY IC (+0.007) exceeds LEVEL IC (-0.023) — the change in the desk's score ranks winners better than its level.20d: score VELOCITY IC (+0.007) exceeds LEVEL IC (-0.023) — the change in the desk's score ranks winners better than its level.
  • ● 20d: departed names (+0.2% avg) underperform on-desk names (+5.5%) — DROPS ARE CORRECT; the desk sheds names before they lag.20d: departed names (+0.2% avg) underperform on-desk names (+5.5%) — DROPS ARE CORRECT; the desk sheds names before they lag.
  • ▲ 20d: 'exhausted' (-0.7%) outperforms 'emerging' (-1.9%) — possible residual inversion in the lifecycle staging.20d: 'exhausted' (-0.7%) outperforms 'emerging' (-1.9%) — possible residual inversion in the lifecycle staging.

Lessons logged经验记录

  • 2026-07-04 2026-07-04 inversion finding: the OLD hub engine was anti-predictive — pooled opportunity IC ~-0.09 @5d, and 'exhausted' names (+1.84% avg fwd) BEAT 'emerging' (-0.77%). The hub-v3-trajectory overhaul is the fix under test; this grader's engine_version A/B is the experiment that proves whether it worked.2026-07-04 inversion finding: the OLD hub engine was anti-predictive — pooled opportunity IC ~-0.09 @5d, and 'exhausted' names (+1.84% avg fwd) BEAT 'emerging' (-0.77%). The hub-v3-trajectory overhaul is the fix under test; this grader's engine_version A/B is the experiment that proves whether it worked.
  • 2026-07-04 PRICE LESSON: grade off data/yahoo/<T>.parquet adjusted closes ONLY. The S&P-1500 breadth close cache (data/breadth/_closes_cache.parquet) is SPLIT-CORRUPTED — using it silently poisons every RS / forward-return number. A name missing from yahoo is a coverage skip here, never a wrong return.PRICE LESSON: grade off data/yahoo/<T>.parquet adjusted closes ONLY. The S&P-1500 breadth close cache (data/breadth/_closes_cache.parquet) is SPLIT-CORRUPTED — using it silently poisons every RS / forward-return number. A name missing from yahoo is a coverage skip here, never a wrong return.
  • 2026-07-04 NOTE on the engine_version A/B: the historical hub ledger carries NO engine_version stamp (all ~63k rows pre-date the stamp), so the pre/post-overhaul split only begins accruing from 2026-07-04 forward. Until enough hub-v3-trajectory rows mature, the A/B reads one-sided; that is expected, not a bug.NOTE on the engine_version A/B: the historical hub ledger carries NO engine_version stamp (all ~63k rows pre-date the stamp), so the pre/post-overhaul split only begins accruing from 2026-07-04 forward. Until enough hub-v3-trajectory rows mature, the A/B reads one-sided; that is expected, not a bug.

~ marks a provisional reading from a single cross-section; the "proven" bar needs at least six separate grading dates. An accountable scorecard of this desk's own surfaced names — its ranking score (level AND velocity) graded against realized SPY-relative forward returns at 5/10/20/30/60/90 trading days, split on-desk vs after-leaving. A signal is 'proven' only past n>=40 with a significant Newey-West t of the right sign; a significant negative IC is an 'alarm'. Context-only — nothing here sizes a position.~ 表示来自单一横截面的暂定读数；"已验证"门槛需至少六个独立评分日。An accountable scorecard of this desk's own surfaced names — its ranking score (level AND velocity) graded against realized SPY-relative forward returns at 5/10/20/30/60/90 trading days, split on-desk vs after-leaving. A signal is 'proven' only past n>=40 with a significant Newey-West t of the right sign; a significant negative IC is an 'alarm'. Context-only — nothing here sizes a position.

Sector heat板块热度

where the desks cluster各台聚集之处
XLK Technology 31 names标的
XLV Health Care 19 names标的
XLE Energypolicy政策 ↑ 6 names标的
XLI Industrialspolicy政策 ↑ 18 names标的
XLF Financials 20 names标的
XLC Communication 9 names标的
XLU Utilities 8 names标的
XLP Staples 7 names标的
XLB Materials 2 names标的
XLRE Real Estate 2 names标的
XLY Discretionary 10 names标的