ψ Psi Quant Research

Signal Report · US

BBAI HOLD

Composite decision report grounded in 1-XH strategy signals and a completed multi-agent debate (agents-3node-v1), generated at 2026-10-14 10:34.

Final Agent Decision

Action
HOLD
Weight
0%
Confidence
42%
Variant
xiaohan_gogo_v2

Reasons

Bull Case (Analyst)

- Technical signal is a fresh `buy` (score 0.60, xiaohan_gogo_v2) with `gogo_detail` showing synchronized bottoming across 1h/2h/3h/4h timeframes — a multi-timeframe reversal setup, not a single-bar blip.
- The strongest cross-sectional lifter is `xh_gogo_score` at +1.71 z, directly corroborating the technical buy signal — the same signal family that fired the action is also the top-ranked factor for this name.
- `beta_60d` (+1.50 z) is a positive lifter in this BULL catalog, meaning high-beta names are being rewarded in the current regime; BBAI's elevated beta is a tailwind rather than a drag here.
- Fundamental catalyst is real and quantified: >$225M in contract awards since July 2026 across homeland security, defense, intelligence and federal civilian customers, including 25+ new awards — and the market reacted with an 11.7% intraday surge and analyst upgrades.
- Sentiment flow is improving: pre-announcement consensus was "lukewarm" with most analysts sidelined, then multiple firms upgraded within hours — a positive revision cycle that can sustain momentum.

Bear Case (Analyst)

- Composite z = -0.37, rank 197/252 (only 78.2% percentile — i.e., bottom quartile), and BBAI is **not** in the production basket. The cross-section explicitly did not select it.
- Three heavy draggers dominate the bear case: `mean_revert_20d` (-1.74), `rs_1m_vs_spy` (-1.68), and `momentum_1m` (-1.68). The stock is a recent underperformer vs. SPY with negative 1-month momentum — the technical buy is fighting the cross-sectional trend.
- Price context contradicts the bull narrative: BBAI hit a **new 52-week low** at $2.15 on the same day (close $2.175), and the CFO had 34,050 shares withheld at $2.39 for taxes — insider-adjacent selling pressure near the lows.
- Valuation/fundamentals are flagged as stretched: one source notes the stock "looks above fair value" after a 97.2% 3-year gain, and another explicitly warns the company is unprofitable and fails to generate positive operating cash flow — the contract wins don't yet translate to earnings.
- The technical signal is a counter-trend bottom call against a bearish factor backdrop; with `mean_revert_20d` at -1.74, the cross-section is betting on continued mean reversion *down*, not up — the buy signal lacks basket or factor confirmation.

Devil's Advocate

COUNTER TO BULL:
- `xiaohan_gogo_v2` at 0.60 is a weak score, not a high-conviction one. "Fresh buy" says nothing about magnitude; a 0.60 is barely above threshold and could be noise dressed as signal.
- Multi-timeframe "synchronized bottoming" on 1h–4h is all short-horizon. These are the same timeframes that produce false bottoms in a downtrend. Synchronization across correlated intraday frames is not independent confirmation — it's one signal echoed four times.
- `xh_gogo_score` +1.71 z is not independent corroboration; it's the *same signal family* as the action. Citing it as confirmation is circular. You're double-counting one factor.
- `beta_60d` +1.50 z being a "positive lifter in this catalog" is a regime claim, not a stock-specific edge. If the regime flips, high beta becomes the drag. You're long a factor exposure you don't control.
- The contract catalyst is already priced: 11.7% intraday surge + upgrades within hours = the move happened. Buying after the pop on a stock at 52-week lows is chasing, not positioning. "Positive revision cycle" is asserted, not evidenced — upgrades ≠ sustained inflows.
- Fundamental catalyst is revenue-side only. No mention of margin, cash conversion, or backlog timing. Awards ≠ bookings ≠ revenue ≠ profit.

COUNTER TO BEAR:
- Composite z = -0.37 and rank 197/252 is a *mild* negative, not a strong short. Bottom-quartile ≠ broken; the score is near neutral, so "not in basket" is weak evidence of downside.
- The three draggers (`mean_revert_20d`, `rs_1m_vs_spy`, `momentum_1m`) are all *backward-looking* and highly collinear. They're essentially one bet: "it fell recently." That's exactly the setup where a genuine reversal signal has the most edge — mean reversion factors are weakest at turning points.
- 52-week low + technical buy is the classic capitulation setup. Dismissing it as "counter-trend" assumes the trend persists; that's the same assumption the draggers encode. Circular.
- The CFO share withholding at $2.39 is tax-related, not discretionary selling. Framing it as "insider-adjacent selling pressure" is misleading — withholding for taxes is mechanical, not a bearish signal.
- "Above fair value" and "unprofitable / no operating cash flow" are static fundamental flags. They were true before the contract awards and don't speak to the marginal change the bull is trading. Valuation flags don't time reversals.
- Bear case leans on the cross-section "betting on continued mean reversion down" — but the cross-section is a lagging composite. Using it to reject a fresh signal is assuming the factor model is right and the signal is wrong, with no independent basis.

Technical Signal (1-XH)

action
buy
score
0.60
variant
xiaohan_gogo_v2
timestamp
2026-10-14T15:30:00-04:00
close
2.175
ema_low_fast
2.291
ema_high_slow
2.522
gogo_score
0.6
gogo_detail
1h_bottom,2h_bottom,3h_bottom,4h_bottom

📈 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
-9.9%
Total (compounded)
-21%
best / worst
+3.7% / -23.4%
avg holding
28.1 days
open position
yes — marked-to-market -23.4%
Entry Exit Entry Exit Return Exit reason
2026-09-17 open 2.84 2.17 -23.4% open_position_mark_to_market
2026-07-22 2026-08-20 2.94 3.05 +3.7% adaptive_breakdown_1h

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 (7 items)