Transcript Uncertainty Score
In plain terms
When management uses lots of hedging words ('may', 'could', 'approximately') on the earnings call, the stock tends to underperform over the next 1-3 months.
How it works
LM uncertainty-lexicon firing rate (modal verbs, approximations, hedges) on the full transcript predicts forward volatility + negative-tail returns. Bochkay-Brown-Leone-Tucker 2024 update reports a 30-90d negative drift on uncertainty-z high tail.
Live results
1 times picked on its own · 2 times inside a blend (2 beat the stock) · updated 2026-06-06Data dependencies
- Daily prices
Adjusted-close OHLCV for every US-listed ticker; primary price feed.
- Earnings call transcripts
Full earnings-call transcripts (prepared + Q&A), tokenised.
Expected edge
- Reported return
- ~2-4% over 60-90d on top-decile uncertainty
- Tested over
- T+1 to T+90d
Loughran-McDonald 2014 + Bochkay et al 2024: uncertainty z >= +1 -> ~2-4% drift over 60-90d.
Related families
Filters out analyst speech and just measures how often the CEO/CFO use uncertainty words ('may', 'could'). High uncertainty = bearish.
When managers' call language gets harder to read (rising Fog index) Q/Q, they're hiding bad news. Predicts 12-month underperformance.
Counts uncertainty/risk words ('may', 'could', 'challenging') in 10-K Item 1 (Business) YoY. A spike means management is privately more worried — bearish.
Explore Transcript Uncertainty Score on alphactor.ai
See which tickers this family is currently firing on, with live signals and rankings.