ψ Psi Quant Research

Signal Report · US

FDX BUY

Composite decision report grounded in 1-XH strategy signals and multi-agent debate, generated by DeepSeek at 2026-07-16 09:00.

🎯 Final Decision

Action
BUY
Weight
35%
Confidence
55%
Variant
xiaohan_gogo_v2

Reasons

Bull Case (Analyst)

- Technical signal is a clear buy (score 0.70) with a rare multi-timeframe bottom setup: 1h, 2h, 3h, and 4h all showing bottoms simultaneously, plus daily "blue fully above yellow" — a pattern that historically precedes mean-reverting rallies from oversold levels.
- Cross-sectional factor view ranks FDX in the top 2.0% of the universe (rank 5/251) with a composite z-score of +0.58, indicating strong relative strength in the current BULL regime catalog.
- The top lifter is xh_gogo_score at +3.00 z — this directly corroborates the technical buy signal, creating a rare alignment between the quant timing model and the cross-sectional factor ranking.
- News sentiment is overwhelmingly positive: Jim Cramer has made multiple calls calling the recent pullback a buying opportunity, citing "a great set of results" and a "self-help story" after a 25% decline.
- Fundamental catalysts are present: the $1.4B supply chain sale to CMA CGM and the FedEx Freight spin-off with $2.4B in revenue and a new standalone LTL strategy provide tangible corporate actions that support the bull case.

Bear Case (Analyst)

- The technical signal is contradicted by the price structure: close at $313.4 is below the slow EMA of $317.5, meaning the stock remains in a downtrend despite the "buy" signal — the gogo model is catching a falling knife.
- Top draggers in the factor view are all momentum/mean-reversion indicators: bollinger_b_20d (-0.58 z), rsi_14 (-0.34 z), and sharpe_60d (-0.15 z) — these suggest the stock is still weak on a risk-adjusted basis and may not have found a durable bottom.
- FDX was explicitly NOT placed in the production basket despite its high composite rank — the cross-sectional model's portfolio construction algorithm chose to exclude it, implying the model sees better risk/reward elsewhere even within the top decile.
- The "buy" signal comes from the xiaohan_gogo_v2 variant, which is a high-frequency mean-reversion strategy — these signals have limited shelf life and can reverse quickly, especially when the daily trend (EMA structure) remains bearish.
- News is dominated by a single source (Jim Cramer on CNBC, 4 of 8 items) — this creates concentration risk in sentiment, and Cramer's calls are often contrarian indicators for short-term moves.

Devil's Advocate

COUNTER TO BULL:
- The "rare multi-timeframe bottom setup" is a pattern-mining artifact — with enough lookback periods (1h, 2h, 3h, 4h), spurious alignments are expected. Historical "precedence" is likely overfit to a handful of cherry-picked rallies.
- Composite z-score of +0.58 is barely above zero — calling this "top 2.0%" implies the universe is heavily compressed or the ranking is ordinal noise. A 0.58 z-score is not statistically significant for stock selection.
- xh_gogo_score at +3.00 z is the same model generating the technical buy — this is circular reasoning, not "rare alignment." You're double-counting the same signal.
- Cramer's calls are a known sentiment trap: his "buying opportunity" rhetoric on a 25% decline is a classic media narrative, not a catalyst. 4 of 8 news items from one source is noise concentration, not conviction.
- The $1.4B sale and spin-off are already priced in — these are backward-looking restructuring events, not forward earnings drivers. The spin-off creates execution risk and dilutes focus.

COUNTER TO BEAR:
- Price below slow EMA is a lagging indicator — the gogo model is explicitly designed to catch reversals *before* the EMA flips. Calling it a "falling knife" ignores the model's edge: it's mean-reversion, not trend-following.
- Momentum draggers (bollinger_b_20d, rsi_14, sharpe_60d) are exactly what you'd expect at a bottom — these are *features* of the setup, not bugs. The bear case is describing the entry condition as a flaw.
- Exclusion from the production basket is a portfolio construction artifact, not a signal on FDX — the model may have hit position limits, sector constraints, or liquidity filters. This doesn't invalidate the stock's standalone case.
- The "high-frequency" criticism is misapplied: xiaohan_gogo_v2 operates on 1h-4h bars, not tick data. Its shelf life is days to weeks, which is fine for a tactical trade. The daily EMA structure is irrelevant to the strategy's horizon.
- Cramer being a contrarian indicator is a meme, not a testable hypothesis — and even if true, 4 of 8 items from one source doesn't make the sentiment "concentrated" when the other 4 are likely from earnings calls or filings.

Technical Signal (1-XH)

action
buy
score
0.70
variant
xiaohan_gogo_v2
timestamp
2026-07-15 15:30:00
close
313.4
ema_low_fast
313.2
ema_high_slow
317.5
gogo_score
0.7
gogo_detail
1h_bottom,2h_bottom,3h_bottom,4h_bottom,daily_blue_fully_above_yellow

📈 Backtest Track Record xiaohan_gogo_v2 · signals simulated over available history (~3-month 30-min window, accumulating daily)

Trades
2
Win rate
50%
Avg / trade
+0.2%
Total (compounded)
+0%
best / worst
+4.0% / -3.6%
avg holding
17.6 days
open position
yes — marked-to-market -3.6%
Entry Exit Entry Exit Return Exit reason
2026-06-25 open 325.03 313.42 -3.6% open_position_mark_to_market
2026-03-31 2026-04-15 347.81 361.87 +4.0% adaptive_breakdown_30m

复盘 = each past xiaohan_gogo_v2 buy signal simulated: enter next bar, manage adds / adaptive sells / −25% hard stop. Demonstration only, not investment advice.

News & social (8 items)