The explanation engine
Why did it move?
A tracked name drops 6%. You check the news and read "risk-off sentiment" or "profit-taking" — phrases that describe the move without establishing a cause. Farlens checks twelve measurements and identifies the factor most consistent with the observed move.
Why “sentiment” may be an incomplete explanation
A price move has a limited set of plausible contributing factors, and one or two of them tend to dominate the day's observations. It might be the whole market. It might be the sector rotating out. It might be money leaving the name specifically while the price holds up. It might be short-selling pressure that doesn't show on the chart at all. It might be a world event repricing the sector's inputs. In crypto it is frequently leverage — a funding rate that got expensive and a cascade of liquidations that followed.
Each of those is measurable, and each can leave a different pattern. The analysis identifies contribution and correlation; it does not prove cause.
What gets checked
All twelve measurements in the signal stack, on every name — equity or crypto — every day:
- Is it the tape? Trend, momentum, and where price sits against its moving averages.
- Is money actually leaving? Money flow index, on-balance volume, accumulation/distribution — price and money flow can disagree, and that disagreement is an analytical indicator rather than a trading signal.
- Is it the sector? Relative strength against the sector ETF separates "this name is in trouble" from "everything like it fell".
- Is someone shorting it? FINRA off-exchange short volume — a measurement many retail dashboards do not surface.
- Is it leverage? For crypto: open interest, funding rates and liquidations. A move amplified by forced selling is a different event from a move driven by spot selling.
- Is it the world? Chokepoint disruption, conflict and sanctions, mapped to the sectors each event touches.
- Is it the conversation? Engagement-weighted discussion volume, measured against that name's own baseline rather than an absolute.
What a read looks like
The output is the composite score, the components that contributed to it, and a plain-language reading of which one is most consistent with the observed move — followed by the case for and the case against case, together.
A worked example of the format. Not a real reading, not about any actual security, and not a recommendation.
| Measurement | Reading | Contribution |
|---|---|---|
| Sector rotation | Sector ETF flat; the name underperformed it sharply | dominant |
| Short volume | Off-exchange short share well above its own average | contributing |
| Money flow | Outflow, but milder than the price move implies | minor |
| Trend | Still above the 200-day | neutral |
| Geopolitical exposure | No active event mapped to this sector | none |
| Narrative | Discussion volume at baseline | none |
Reading: this looks name-specific rather than sector-wide, with shorting pressure rather than broad distribution. The world-events layer contributed nothing — evidence that the model did not attribute a contribution to that layer in this illustrative example.
Stocks and crypto, same question
Crypto can have positioning context as well as news context. Funding rates, open interest and liquidation data can help describe leverage and forced-selling conditions. The planned Farlens crypto layer will use those inputs only after the required data rights and technical evidence are verified. It is not available today. Funding rates explained · liquidations explained.
What this is not
- Not a prediction or proof of cause. Analysing measured contributions to a move is different from establishing why it happened or forecasting what happens next.
- Not a recommendation. No buy or sell calls, no price targets, no entry or exit levels — on any surface, ever.
- Not one-sided. Every read carries the bull case and the bear case together.
- Not infallible. Our current internal assessment grades the composite WEAK and not robust. The limitations are published on the methodology page.
Frequently asked
Why is my stock down today?
One or two measured factors can tend to dominate the day's observations — market, sector, money flow, shorting, an event, or discussion. Farlens reports which factor is most consistent with the move; it does not establish cause.
Why is bitcoin down today?
Leverage data can add context alongside news. Funding, open interest and liquidations may be consistent with amplification by forced selling.
Do I get today's reading for free?
No live reading is offered today. Public educational material and methodology are open — every measurement and weight is published. Live readings on names you follow are planned for eligible account holders at launch. Explore a synthetic sample.