Score / Overall
0-39 low, 40-59 watch, 60-74 early, 75+ active, 90+ mania risk.
Direction score only. It ranks research priority; it is not a buy/sell signal.
Choose a focused dashboard. Meme and short-crowding research are kept separate from Congress disclosure filings.
Track meme-stock setups, short-crowding signals, score methodology, source gaps, and timing checks.
0-39 low, 40-59 watch, 60-74 early, 75+ active, 90+ mania risk.
Direction score only. It ranks research priority; it is not a buy/sell signal.
75+ high, 55-74 usable but incomplete, under 55 means major source gaps.
Missing borrow, options, Reddit/news firehose, or filings checks lower confidence instead of being guessed.
Under 10% normal, 10-20% elevated, 20-40% crowded, 40%+ extreme.
Higher means more of tradable float is sold short.
Under 5% normal, 5-10% notable, 10-25% high, 25%+ extreme.
Useful cross-check when float data looks noisy.
Under 2x easy cover, 2-5x watch, 5-10x tight, 10x+ crowded exit risk.
Days to cover estimates how many trading days shorts need at average volume.
Negative easing, 0-10% rising, 10%+ shorts pressing, 25%+ aggressive build.
Rising short interest can matter more when price and volume also rise.
Under 1x quiet, 1-1.5x active, 1.5-3x hot tape, 3x+ unusual activity.
Higher relative volume confirms crowd attention is current.
Public archive post matches. 1M above the 12M monthly baseline suggests fresh acceleration.
Sample only, not full Reddit firehose or comment volume.
Under $500M is microcap risk, $500M-$2B small cap, $2B-$10B tradable mid, $10B+ needs stronger flow.
Smaller caps can move faster but have more liquidity and dilution risk.
Research first → watchlist → early candidate → active setup → mania / short-risk.
No-trade risk means penalties or missing checks are too large.
Heuristic watch probability and likely attention window from social acceleration, author breadth, short pressure, liquidity fit, tape, catalyst, confidence, options, and penalties.
Missing options, weak tape, thin source coverage, and filing checks cap probability and widen the date range. This is not a forecast or trade signal.
Borrow fee under 5% normal, 5-20% tightening, 20%+ expensive, 50%+ stressed.
If borrow feed is missing, score confidence drops instead of assuming squeeze pressure.
Call volume ratio, IV rank, OTM call clusters, and 0-7 DTE flow should confirm attention.
Missing options feed reserves risk in confidence and next checks.
0-7 days strongest, 8-14 strong, 15-30 useful, 31-60 weak, 60+ stale.
Earnings, FDA, litigation, refinancing, product events, and filings can all matter.
Microcap, low price, SPAC/warrant/unit, dilution, bankruptcy, and going-concern risk reduce rank.
Penalty means "verify first," not "ignore forever."
Window 1
Window 2
Topic 1
Topic 2
Window 4
No public dashboard can judge meme-stock risk perfectly without live borrow, options, full social firehose, filings, and tape feeds. This model separates direction score from source confidence so missing feeds cannot look better than real evidence.
Used for the short-crowding rows. Each subscore is percentile-ranked against the current dashboard universe, then reweighted away from unavailable fields.
Used for broader discovery. It mixes public universe context with known short-crowding rows, then lowers confidence when required evidence is absent.