ψ Psi Quant Research

Signal Report · US

TXN BUY

Composite decision report grounded in 1-XH strategy signals and a completed multi-agent debate (agents-3node-v1), generated at 2026-08-13 09:29.

Final Agent Decision

Action
BUY
Weight
35%
Confidence
55%
Variant
xiaohan_gogo_v2

Reasons

Bull Case (Analyst)

- The technical signal is a clean buy (score 0.60) with a rare full-stack bottoming pattern — 1h/2h/3h/4h all flagged as bottoms — suggesting a coordinated multi-timeframe reversal setup at $273.4, well below both fast EMA (277.5) and slow EMA (282.2), leaving room for mean reversion upside.
- The cross-sectional factor view is dominated by xh_gogo_score at +3.00 z — the single strongest positive factor in the entire catalog for TXN — which directly corroborates the technical buy signal and indicates the price-action bottoming is not idiosyncratic noise but a regime-consistent long trigger.
- Momentum_6m (+0.29 z) and beta_60d (+0.75 z) are both positive contributors, meaning TXN retains intermediate-term upward drift and offers leveraged upside if the broader BULL regime continues to lift high-beta semis.
- Composite z of +0.21 with rank 78/252 (top 31%) is constructive — not a top-decile basket name, but still above the median in a BULL regime with zero direction-flips, so the factor backdrop is not fighting the technical buy.

Bear Case (Analyst)

- The cross-sectional view is only mildly positive (composite z +0.21, rank 78/252) and TXN is NOT in the production basket — the quant model explicitly declined to put it in the top decile, meaning the technical buy is not backed by the strongest factor conviction this period.
- The biggest drags are severe: close_to_high_60d at -1.72 z and bollinger_b_20d at -1.57 z — TXN is far from its 60-day high and sitting near the lower Bollinger band, which in a BULL regime often signals persistent relative weakness rather than a cheap entry, as laggards tend to stay laggards.
- sharpe_60d at -0.86 z confirms the recent risk-adjusted return profile is poor — the stock has been bleeding efficiency over the past two months, and a single bottoming signal does not erase that deterioration.
- The only news item is about the Texas Stock Exchange opening — a headline event for the exchange venue, not for TXN fundamentals; there is zero company-specific catalyst in the news feed to justify stepping in front of a falling knife, and the technical signal may simply be catching a dead-cat bounce in a name the cross-section is underweighting.

Devil's Advocate

COUNTER TO BULL:
- A "full-stack bottoming pattern" is a lagging, curve-fit artifact: all four timeframes flagging bottoms simultaneously often means price has already fallen through every support level, and the pattern is simply describing where price is, not where it's going. The setup at $273.4 is only "cheap" relative to EMAs that are themselves declining — mean reversion to a falling average is not upside, it's just less downside.
- xh_gogo_score at +3.00 z is a single factor, and its dominance in the catalog is exactly why it's suspect: if it were a reliable regime-consistent long trigger, the composite z would be far higher than +0.21. The model is telling you this one factor is an outlier, not a confirmation — and outliers in factor space are often the first to revert.
- Momentum_6m (+0.29 z) and beta_60d (+0.75 z) are weak positives that cut both ways: high beta in a BULL regime means amplified downside if the regime wobbles, and a +0.29 z momentum is barely above zero — it does not establish "intermediate-term upward drift," it establishes that TXN is not a momentum leader.
- Rank 78/252 is not "constructive" — it is the definition of mediocrity. Zero direction-flips is a low bar; a stock can be consistently mediocre. The technical buy is a standalone signal with no independent fundamental or flow catalyst, and the factor backdrop is neutral at best, not supportive.

COUNTER TO BEAR:
- The "not in the production basket" argument is circular: the basket is a point-in-time construct, and the technical signal is explicitly flagging a reversal that the cross-section has not yet caught up to. Using a lagging factor rank to invalidate a leading price signal is backwards — the whole point of a bottoming pattern is that the cross-section is still negative.
- close_to_high_60d at -1.72 z and bollinger_b_20d at -1.57 z are exactly what a bottom looks like: price far from highs and pinned to the lower band. Calling this "persistent relative weakness" assumes the 60-day trend is the permanent state, but that is precisely the assumption a reversal pattern is designed to break. Laggards stay laggards only until they don't.
- sharpe_60d at -0.86 z is a trailing measure that will improve mechanically if price stabilizes — it is not a leading indicator. Using it to argue against a buy is like saying a stock can't rebound because its recent returns were bad; that is the definition of buying low, not a flaw.
- The "no catalyst" point is a strawman: the quant signal is the catalyst. The Texas Stock Exchange headline is irrelevant, but so is the absence of a headline — the technical and factor signals are the tradeable information. Demanding a news catalyst on top of a multi-timeframe bottom and a +3.00 z factor is moving the goalposts, not stress-testing the bear case.

Technical Signal (1-XH)

action
buy
score
0.60
variant
xiaohan_gogo_v2
timestamp
2026-08-13T15:30:00-04:00
close
273.4
ema_low_fast
277.5
ema_high_slow
282.2
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
3
Win rate
33%
Avg / trade
+17.7%
Total (compounded)
+51%
best / worst
+56.2% / -1.9%
avg holding
27.7 days
open position
yes — marked-to-market -1.3%
Entry Exit Entry Exit Return Exit reason
2026-08-12 open 277.10 273.43 -1.3% open_position_mark_to_market
2026-08-04 2026-08-12 283.40 278.04 -1.9% adaptive_breakdown_1h
2026-03-20 2026-06-02 192.52 300.70 +56.2% 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 (1 items)