The Hidden Segment: YES calls show 17.0% lowered guidance vs 11.6% overall
This study profiles 34,917 earnings calls (21.1% of a 165,182-call corpus spanning 1990-2026) where the model answered YES to the battery item 'The Hidden Segment'. Compared with the base corpus, these calls show slightly lower candor (6.84 vs 6.86), higher evasion (2.86 vs 2.70), higher stress (2.86 vs 2.43), and lower confidence (6.93 vs 7.21). Guidance actions differ too: 15.4% raised guidance versus 21.1% overall, while 17.0% lowered guidance versus 11.6%. The largest battery lifts appear on 'Results Worse Than Direction' (1.56x) and 'Early Products Growing Fast' (1.55x). Among 4,248 calls with measured post-call returns, the median return was -9.35% versus -7.16% for the 22,449-call base, and 38.2% beat versus 39.5%.
- YES calls make up 21.1% of the corpus (34,917 of 165,182 calls, 95% CI 20.9%-21.3%).
- Stress is higher on YES calls (2.86 vs 2.43) and confidence lower (6.93 vs 7.21).
- Lowered guidance is more common on YES calls: 17.0% vs 11.6% in the base corpus, while raised guidance is less common (15.4% vs 21.1%).
- Median post-call return among the 4,248 YES calls with returns was -9.35% versus -7.16% for the base sample of 22,449 calls.
1Introduction
Earnings calls often feature a management pivot: a business line, geography, or product said to be overlooked by the market. The 'The Hidden Segment' battery item captures calls where the model detected exactly that framing. If such framing co-occurs with measurable differences in tone, guidance behavior, or post-call outcomes, it would matter to anyone who parses calls for signal. Prior work suggests narrative flourishes cluster with defensive contexts. This study examines 34,917 YES calls out of 165,182 (21.1%) from 1990-2026, profiling their language, guidance actions, top battery lifts, annual frequency, and post-call returns.
2Data & methodology
The corpus comprises 165,182 earnings-call transcripts published between 1990 and 2026, each scored independently by a large language model on an identical 37-field battery: seven categorical business verdicts, eight 0–9 behavioral meters, and twenty yes/no judgments. The study group is defined as calls where the model answered YES to the battery item "The Hidden Segment" (n = 34,917; 21.1% of the reference set, 95% Wilson interval 20.9%–21.3%). Baseline figures use all scored calls. Market outcomes join a fixed sample of 22,449 calls with twelve-month total returns in excess of SPY, measured from the first close after each call; this sample skews toward liquid U.S. names and is reported as descriptive history only.
3Results
YES calls skew toward defensive contexts: evasion runs 2.86 vs 2.70 and stress 2.86 vs 2.43, while confidence sits at 6.93 vs 7.21. Guidance differs in both directions: raised guidance is rarer (15.4% vs 21.1%) and lowered guidance more common (17.0% vs 11.6%). The strongest battery lifts are 'Results Worse Than Direction' at 1.56x (51.1% vs 32.8% base rate) and 'Early Products Growing Fast' at 1.55x. Frequency peaked at 24.8% in 2015-2016, fell to 15.8% in 2021, and was 19.7% in 2025 (6,012 calls). Median post-call return was -9.35% vs -7.16% in the base, with 38.2% beating vs 39.5%.
| Meter | Study group | Baseline | Δ |
|---|---|---|---|
| Candor | 6.84 | 6.86 | -0.02 |
| Evasion | 2.86 | 2.70 | +0.16 |
| Specificity | 7.46 | 7.56 | -0.09 |
| Stress | 2.86 | 2.43 | +0.43 |
| Promotion | 5.10 | 5.05 | +0.05 |
| Confidence | 6.93 | 7.21 | -0.29 |
| Action | Study group | Baseline |
|---|---|---|
| Raised | 15.4% | 21.1% |
| Maintained | 49.2% | 48.8% |
| Lowered | 17.0% | 11.6% |
| Withdrawn | 3.0% | 2.7% |
| Signal | Lift | In group | Baseline |
|---|---|---|---|
| Results Worse Than Direction | 1.56× | 79.6% | 51.1% |
| Early Products Growing Fast | 1.55× | 59.6% | 38.5% |
| Scale-Dependent Advantage Claims | 1.48× | 16.4% | 11.1% |
| A Tiny Fraction of the Market | 1.33× | 39.9% | 30.0% |
| Underused Fixed Costs | 1.32× | 54.9% | 41.6% |
| Statistic | Study group | Returns sample |
|---|---|---|
| Median excess return | -9.3% | -7.2% |
| Interquartile range | -29.6% to +11.3% | — |
| Share beating SPY | 38.2% (95% CI 37%–40%) | 39.5% |
| Observations | 4,248 | 22,449 |
| Ticker | Quarter | Call date | Call grade |
|---|---|---|---|
| VRTS | Q2 2025 | 2025-07-25 | C+ |
| FLG | Q2 2025 | 2025-07-25 | B |
| FRST | Q2 2025 | 2025-07-25 | A |
| CHTR | Q2 2025 | 2025-07-25 | C+ |
| DBOEY | Q2 2025 | 2025-07-25 | B+ |
| KNSL | Q2 2025 | 2025-07-25 | C+ |
| BAH | Q1 2026 | 2025-07-25 | C+ |
| LBTSF | Q2 2025 | 2025-07-25 | C+ |
4Discussion
A careful reader should conclude only that YES calls are described differently: slightly more evasion and stress, less confidence, more lowered guidance, and modestly worse measured returns. These are co-occurrences in model-generated labels, not evidence that the 'hidden segment' framing causes anything. The 2025 figure covers only part of the year. Our own forward tests falsified directional prediction from these labels, so no trading conclusion is warranted. The returns gap (-9.35% vs -7.16% median) is descriptive and small in practical terms.
5Limitations
Battery items are AI-read fields and noisy; YES labels reflect model judgment, not ground truth. The returns sample covers 4,248 YES calls against a base of 22,449, skewed toward liquid names, so return comparisons may not generalize. Our own forward tests falsified directional prediction from these signals. Additionally, LLMs partially remember famous stocks' histories, which can contaminate any backtest of model-labeled earnings calls. All figures are descriptive of this labeled corpus and its measured outcomes. See the full methodology, including the C1 pattern’s forward-test failure and the LLM-memorization finding.