ψ Psi Quant Research

Signal Report · US

CHTR HOLD

Composite decision report grounded in 1-XH strategy signals and multi-agent debate, generated by DeepSeek at 2026-08-05 09:13.

🎯 Final Decision

Action
HOLD
Weight
0%
Confidence
40%
Variant
xiaohan_gogo_v2

Reasons

Bull Case (Analyst)

- Technical signal is a clean buy (score 0.70) with a rare full-stack bottom formation: 1h/2h/3h/4h all bottomed simultaneously, plus the 1h blue upper crossing above yellow upper — a momentum-inflection pattern that historically precedes sharp mean-reversion squeezes.
- Cross-sectional rank is elite: composite z = +0.80, rank 3/251 (top 1.2% of the entire universe). In a BULL regime catalog with zero direction-flips, this is a high-conviction long setup from the factor model's perspective.
- The dominant lifters are mean-reversion factors: xh_gogo_score (+2.37) and mean_revert_20d (+1.65) and bollinger_b_20d (+1.42) all point to a deeply oversold stock that is statistically due for a bounce. The +6.2% rally on Aug 4 (to $153.07) confirms the mean-reversion trigger is firing.
- News flow corroborates the technicals: the stock jumped ~5% on Monday as buyers leaned into a rebound attempt in Communication Services, and GuruFocus flags the stock as undervalued on GF Value after the 6.2% rally — suggesting the move has fundamental support, not just technical noise.
- Price action is constructive: close at $153.10 sits well above both EMA_low_fast ($145) and EMA_high_slow ($139.90), confirming the trend has flipped from bearish to bullish at the short-term horizon.

Bear Case (Analyst)

- The cross-sectional factor view is not a clean endorsement: CHTR is NOT in the production basket — the model itself did not put it in the top decile this period, which is a red flag that the composite z is driven by a narrow set of mean-reversion factors rather than broad-based strength.
- The top draggers are fundamental quality and momentum factors: momentum_6m (-0.60), sharpe_60d (-0.46), and beta_60d (-0.26) all negative. This means the stock's 6-month price trend is broken, its risk-adjusted returns are poor, and it has unfavorable beta characteristics — the mean-reversion bounce is fighting against a deteriorating fundamental trend.
- News is overwhelmingly bearish on the core business: Q2 2026 results showed loss of video and internet customers, with competitors like Starlink offering comparable service. Analysts explicitly warn that "broadband pressure will persist" and mobile growth is "losing steam" — the fundamental deterioration is ongoing, not cyclical.
- Insider selling is a direct negative signal: Director Thomas Rutledge sold 69,633 shares at $144.45, and Amundi sold 52,344 shares — both transactions occurring near current levels, suggesting those with the best information see limited upside.
- The stock is down 50% over the past year and 33% YTD, and Jim Cramer reports "troubling whispers" about the company. A single-day +6.2% bounce in a stock with this broken a trend is more consistent with a dead-cat bounce than a durable reversal, especially with the 6-month momentum factor still deeply negative.

Devil's Advocate

COUNTER TO BULL:
- A "full-stack bottom" on 1h–4h timeframes is a micro-structure artifact, not a regime shift. These align after any sharp selloff; the historical edge decays fast and is often just the first leg of a bear-market rally.
- Composite z of +0.80 with rank 3/251 is meaningless if the production basket excluded it. The factor model is telling you it lacks breadth — you are cherry-picking a z-score built on three mean-reversion inputs that all measure the same oversold condition (multicollinearity).
- The +6.2% bounce on Aug 4 is the *definition* of a mean-reversion trigger firing — but mean-reversion factors have near-zero expectancy when the 6-month trend is broken. You are treating a one-day snap as confirmation of a multi-week reversal.
- "Undervalued on GF Value" after a 6.2% rally is a lagging, price-relative label. It does not address the structural subscriber losses; value traps are full of stocks trading below GF Value.
- Price above EMA_low_fast/EMA_high_slow is a lagging confirmation — those EMAs were also above price in early July before the next leg down. Short-term trend flips in a -50% YTD name are noise until the 6m momentum turns.

COUNTER TO BEAR:
- Excluding CHTR from the production basket is a model governance artifact, not a signal. The composite z of +0.80 and rank 3/251 is computed on the same factor set — if the basket threshold is arbitrary (e.g., top decile), a rank of 3 is a rank of 3. You are overweighting a cutoff rule over the actual cross-sectional signal.
- Negative momentum_6m and sharpe_60d are *expected* inputs for a mean-reversion setup — they are the precondition, not the contradiction. The bull case does not claim trend-following strength; it claims oversold snap. Attacking it for lacking momentum is attacking the thesis for being what it is.
- The bearish news (subscriber losses, Starlink) is already priced into a -50% YTD decline. Using stale Q2 fundamentals to argue against a technical bounce ignores that the stock has already repriced the deterioration; the question is whether the *rate* of decline is decelerating.
- Insider selling at $144.45 is a weak signal: 69k shares is a rounding error for a director, and Amundi is an index/ETF manager selling for rebalancing, not informed conviction. Treating these as "those with best information" is a misread of the transaction types.
- "Dead-cat bounce" is a narrative, not a falsifiable claim. The 6.2% rally on above-average volume with a full-stack bottom across four timeframes is exactly what a durable reversal looks like at inception. Calling it a dead-cat before the next 5 sessions is unfalsifiable hindsight bias.

Technical Signal (1-XH)

action
buy
score
0.70
variant
xiaohan_gogo_v2
timestamp
2026-08-04 15:30:00
close
153.1
ema_low_fast
145
ema_high_slow
139.9
gogo_score
0.7
gogo_detail
1h_bottom,2h_bottom,3h_bottom,4h_bottom,1h_blue_upper_above_yellow_upper

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

Trades
3
Win rate
67%
Avg / trade
+3.6%
Total (compounded)
+10%
best / worst
+16.6% / -8.0%
avg holding
18.7 days
open position
yes — marked-to-market +16.6%
Entry Exit Entry Exit Return Exit reason
2026-06-23 open 131.27 153.07 +16.6% open_position_mark_to_market
2026-06-15 2026-06-17 145.70 134.05 -8.0% adaptive_breakdown_1h
2026-03-19 2026-03-31 211.27 215.68 +2.1% 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 (8 items)