Crank It to Eleven: Calls Where the Model Expects a Volume Step-Up
This study examines 47,036 earnings calls from a 165,182-call corpus (1990-2026) in which a language model answered YES to the battery item "Volume About to Step Up" - 28.5% of all calls (95% CI 28.3%-28.7%). Compared with other calls, these show higher confidence (7.51 vs 7.21), higher promotion (5.58 vs 5.05), and slightly lower stress (2.33 vs 2.43). Guidance was raised on 26.4% of these calls versus 21.1% of others. Language patterns such as "Scale-Dependent Advantage Claims" appear 1.61x more often than base rates. However, the one-day return profile is weaker: median -10.4% versus -7.2%, and a 37.6% beat rate versus 39.5% for the base sample.
- YES calls make up 28.5% of the 165,182-call corpus (n = 47,036; CI 28.3%-28.7%).
- Confidence is higher on YES calls (7.51 vs 7.21) and promotion is up 0.53 points, while stress is lower (2.33 vs 2.43).
- Guidance was raised on 26.4% of YES calls versus 21.1% of other calls, and lowered on 8.7% versus 11.6%.
- Despite the upbeat profile, the median next-day return is -10.4% versus -7.2% for the base sample, with a 37.6% beat rate versus 39.5%.
1Introduction
Earnings calls are where companies talk about demand, and few claims are more consequential than the suggestion that volume is about to accelerate. A model flagging "Volume About to Step Up" identifies a large slice of the record - 28.5% of 165,182 calls - and those calls carry a distinctive tone: more confidence, more promotion, less stress. Anyone who follows calls closely knows that confident volume talk can mark either genuine inflection or the peak of enthusiasm. This study describes the language, guidance behavior, and subsequent return distribution of these calls, without asserting that any of it predicts anything.
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 "Volume About to Step Up" (n = 47,036; 28.5% of the reference set, 95% Wilson interval 28.3%–28.7%). 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
The tone profile is unambiguously promotional: confidence runs 7.51 versus 7.21, promotion 5.58 versus 5.05, and stress 2.33 versus 2.43. Guidance actions skew positive - 26.4% raised versus 21.1% base - and lowered guidance is rarer (8.7% vs 11.6%). Linguistic lifts concentrate on scale stories: "Scale-Dependent Advantage Claims" at 1.61x, "Underused Fixed Costs" at 1.46x, and "A Tiny Fraction of the Market" at 1.42x. The trend peaks in 2021 at 37.74% of calls. The returns table complicates the story: median one-day return of -10.4% versus -7.2% base, and beat rate 37.6% versus 39.5%.
| Meter | Study group | Baseline | Δ |
|---|---|---|---|
| Candor | 6.78 | 6.86 | -0.08 |
| Evasion | 2.67 | 2.70 | -0.03 |
| Specificity | 7.58 | 7.56 | +0.02 |
| Stress | 2.33 | 2.43 | -0.10 |
| Promotion | 5.58 | 5.05 | +0.53 |
| Confidence | 7.51 | 7.21 | +0.30 |
| Action | Study group | Baseline |
|---|---|---|
| Raised | 26.4% | 21.1% |
| Maintained | 47.4% | 48.8% |
| Lowered | 8.7% | 11.6% |
| Withdrawn | 2.0% | 2.7% |
| Signal | Lift | In group | Baseline |
|---|---|---|---|
| Scale-Dependent Advantage Claims | 1.61× | 17.9% | 11.1% |
| Underused Fixed Costs | 1.46× | 60.6% | 41.6% |
| Early Products Growing Fast | 1.45× | 55.7% | 38.5% |
| A Tiny Fraction of the Market | 1.42× | 42.5% | 30.0% |
| Deferred Revenue Growing | 1.29× | 11.4% | 8.9% |
| When the CFO Dominates | 0.65× | 9.2% | 14.1% |
| Statistic | Study group | Returns sample |
|---|---|---|
| Median excess return | -10.4% | -7.2% |
| Interquartile range | -32.0% to +12.2% | — |
| Share beating SPY | 37.6% (95% CI 36%–39%) | 39.5% |
| Observations | 5,213 | 22,449 |
| Ticker | Quarter | Call date | Call grade |
|---|---|---|---|
| FLG | Q2 2025 | 2025-07-25 | B |
| FRST | Q2 2025 | 2025-07-25 | A |
| HMDPF | Q2 2025 | 2025-07-25 | B |
| FFBC | Q2 2025 | 2025-07-25 | B+ |
| DBOEY | Q2 2025 | 2025-07-25 | B+ |
| TBBK | Q2 2025 | 2025-07-25 | C |
| SSB | Q2 2025 | 2025-07-25 | B+ |
| STEL | Q2 2025 | 2025-07-25 | B |
4Discussion
A careful reader should conclude that YES calls sound more confident and are more likely to accompany raised guidance. They should not conclude that the model's YES identifies future volume growth, good investments, or mispriced expectations. The return table is descriptive: a lower median (-10.4% vs -7.2%) coexists with an interquartile range spanning -32.0% to +12.2%, so outcomes vary widely. The 2021 peak (37.74%) reflects the period's overall enthusiasm as much as anything. These are correlations in historical text, not signals, and no directional edge is claimed or implied.
5Limitations
All fields are AI-read and noisy; tone deltas of a few tenths of a point may not survive annotation error. The returns sample covers 5,213 YES calls against a base of 22,449 calls, skewed toward liquid names, so the -10.4% median is not representative of the full corpus. Our own forward tests falsified directional prediction from these features, and LLMs partially remember famous stocks' histories, contaminating any backtest. The 2025 trend row (25.0%, n = 6,012) is partial-year data. Nothing here should be read as causal or as a trading edge. See the full methodology, including the C1 pattern’s forward-test failure and the LLM-memorization finding.