We only use essential cookies and local storage required for authentication, preferences, and core functionality. We do not use analytics, advertising, or third-party tracking technologies. Privacy · Terms
Discover

Find the trade before you have a ticker.

Morning Picks builds the day's trade plan from 300 large caps before the bell. The Gap Analyzer rules on why the biggest movers moved. Sector and Industry Analysis rank where money is going, at two resolutions. The Stock Picker ranks a curated top-300 on 35 factors — and six sentiment and flow boards read the market around the idea. Every scanner names the population it read.

Morning Picks

The day's trade plan, before the bell.

Up to twenty setups — each with a verdict, entry, stop, targets, a 0.5–10% position size and the reasoning behind it, already worked out — plus the market-regime read that sizes them. It is what the app's Today section opens on.

300 large caps are scored by a 35-factor model across seven categories — the same model as the Stock Picker — then the pool is tilted toward the sectors the rotation model favors (×1.8) and away from the ones fading (×0.6), with no sector allowed more than 40% of it. The top seventy fan out to a validator pass — each collecting a validator verdict, a news and sentiment read, and momentum ranks measured within that day's pool — are narrowed at the downselect, and a synthesis stage publishes up to twenty setups. Only the twenty that clear the downselect get the vision model that reads the actual chart — it is the expensive step, so it never runs on all seventy. The regime call above them is rule code, not a prompt, which is why the bias label and exposure band are reproducible.

  • Five slots roll from pre-market to the evening, so the briefing is never stale
  • A step that fails is named rather than papered over: the reason is recorded and threaded into the synthesis prompt, which is told not to infer a read it never got
  • A failed chart read is disclosed rather than hidden — a “Some analyses failed” banner shows the count on screen, so a briefing built without chart input can no longer be mistaken for a healthy one
  • Equities only
35 factors · 7 categories
300 → 70 → 20
Entry · stop · target · size
Ready before the open
Morning Picks
Market Regime
Today's Trade Plan
TickerVerdictSetupEntryStopTargetR:RSize
NVDASTRONG BUYmomentum-breakout182.40176.20198.001:2.53.0%
LLYBUYpullback-to-support812.50792.00868.001:2.72.5%
COPBUYtrend-continuation118.20114.60127.401:2.62.0%
AMDBUYgap-fill174.80169.90186.501:2.42.0%
WMTBUYoversold-bounce96.4093.80102.601:2.41.5%
UNHHOLDrange-trade308.00298.50322.001:1.51.0%
CVXAVOIDnews-reaction—————
Market Regime
VIX16.8
SPY+0.34%
QQQ+0.52%
DIA+0.11%
SPY 5-day +1.28% · above its 20-day SMA
bearishcautiously-bullishbullish
Equity exposure55–80%
SPY +0.34% today and trading above its 20-day SMA — constructive but not euphoric. Derived by rule, not by an LLM.
Candidate Funnel
35 factors · 7 categories
Value5Quality7Growth5Health5Momentum5Earnings4Analyst4
Scored universe300
Top 300 by mkt cap + ≤10 catalysts
Rotation-tilted pool70
×1.8 favored · ×0.6 fading
AI downselect20
today's setups
Sectors the rotation model favors get ×1.8 weight, fading ones ×0.6 — and no single sector may hold more than 40% of the pool.
Evidence Fan-out
Per-symbol evidence gathered
Trade Validatorverdict + confidence×70
News + sentimentup to 5 headlines×70
Fear / Greedper-symbol index×70
Momentum ranks1M / 3M / 6M / 12M×70
Vol + ATR-1420-day realized×70
Chart patternsvision LLM, 260 bars×20
The vision chart read is expensive, so it runs only on the 20 names that survive the downselect — never on all 70.
07:00 ET weekdays| Append-only history
1/4
Gap Analyzer

Why did it move — and should you fade it?

The day's biggest movers, each ruled PANIC, JUSTIFIED, or UNCLEAR against the news that actually moved it. It sits beside Morning Picks in the app's Today section.

247 large caps are polled for absolute moves of 0.75% or more, above a $10 share price and a $2B market cap, and the twelve biggest by default are enriched with up to five material headlines each, the analyst price-target gap, sector context, and the day's VIX and SPY regime. One batched call then has to name a mechanism for all twelve: a real fundamental break you should not fight, a sentiment-driven overshoot worth fading, or genuinely mixed signals that say wait.

  • Every row carries a setup bias — buy the dip, short the rip, avoid, or wait (BUY_DIP / SHORT_RIP / AVOID / WAIT)
  • Every ruling cites the headlines behind it — up to five material ones — as its evidence
  • UNCLEAR is a first-class verdict rather than a failure
  • Each row keeps the thesis and catalyst it was drawn from, so you can disagree with the call without leaving the screen
  • Equities only
247 large caps polled
Headlines · targets · regime
PANIC / JUSTIFIED / UNCLEAR
Thesis kept on every row
Gap Analyzer
Universe Sweep
Universe Sweep · movers ≥ 0.75%
SymbolMoveMkt capSectorPT gapNews
MRNA-9.4%$12BHealth
+41.2
-0.339 art
SMCI+8.1%$24BTech
-12.6
+0.2511 art
PYPL-6.8%$71BFinanc
+28.4
-0.187 art
NKE-5.2%$104BCons Cyc
+19.7
-0.4212 art
DAL+4.6%$41BIndust
+6.3
+0.406 art
MU-3.9%$132BTech
+22.1
-0.0910 art
FDX+2.4%$63BIndust
+4.8
+0.175 art
Evidence Pull
MRNA-9.4%Healthcare
Analyst PT gap+41.2%
avg target vs pricepanic tell
News sentiment-0.33
(pos − neg) ÷ 9 articlesnegative
A target gap well past 20% against a down move reads as panic. Heavy negative news with the sector red reads as justified.
Panic vs Justified
panicfade5
justifiedfollow4
unclearwait3
One batch call rules on all 12 candidates — thesis and catalyst written first, the verdict after, confidence scored last.
Candidates
MRNApanicBUY DIP
84%
PYPLpanicBUY DIP
71%
SMCIunclearWAIT
58%
NKEjustifiedAVOID
66%
MUjustifiedSHORT RIP
62%
panicjustifiedunclear
Ranked panic first, then confidence, then move size — every row keeps its own thesis and catalyst.
07:06 ET weekdays| One shared result
1/4
AI Sector Analysis

Where the money is going, sector by sector.

All 11 SPDR sectors on one ladder, with the constituent breadth behind every ranking on the same row — the AI Analysis tab on the Sectors page.

Each sector ETF gets a composite flow score built from its 5- and 10-day trend, today's move, its volume against its own 50-day average, and the share of its seven-name large-cap basket trading above its own 50-day. It only reads INFLOW or OUTFLOW past ±15, so a flat tape stays neutral instead of inventing a rotation — and it is a composite read, not reported fund flows.

  • Every paired rotation call has to cite a number already on the ladder
  • Each stock idea comes from a named sector's own basket, with its technical claims re-read against real bars before it is published
All 11 sectors ranked
Constituent breadth per row
±15 before it calls a flow
Paired rotation calls
Sector Rotation
Sector Flow Ladder
Sector Flow Ladder · 11 GICS ETFs
ETFSectorToday5d20dVolBreadthFlow
XLVHealth Care+1.24%+3.8%+6.1%1.42×
86%
+58INFLOW
XLPConsumer Staples+0.72%+2.4%+4.0%1.28×
71%
+41INFLOW
XLUUtilities+0.58%+1.9%+3.2%1.19×
71%
+33INFLOW
XLEEnergy+0.94%+1.1%-0.4%1.34×
57%
+22INFLOW
XLFFinancials+0.21%+0.6%+1.8%1.05×
57%
+12NEUTRAL
XLIIndustrials+0.14%+0.3%+1.1%0.98×
43%
+6NEUTRAL
XLBMaterials-0.18%-0.4%+0.5%0.91×
43%
-4NEUTRAL
XLREReal Estate-0.32%-0.9%-1.6%0.88×
29%
-11NEUTRAL
XLCCommunication Services-0.61%-1.4%-0.8%1.02×
29%
-19OUTFLOW
XLYConsumer Discretionary-0.88%-2.6%-3.4%1.16×
29%
-34OUTFLOW
XLKTechnology-1.36%-3.9%-5.2%1.51×
14%
-52OUTFLOW
Flow score blends the 5d and 10d trend, today's move, the ETF's volume against its own 50-day average, and constituent breadth. It is a composite read, not reported fund flows.
Constituent Breadth
XLV
6/786%
LLY · JNJ · UNH · PFE · MRK · ABBV · TMO
XLP
5/771%
PG · KO · PEP · WMT · COST · PM · MDLZ
XLY
2/729%
AMZN · TSLA · HD · MCD · NKE · LOW · TJX
XLK
1/714%
AAPL · MSFT · NVDA · AVGO · ORCL · CRM · AMD
Breadth is the share of each sector's seven-name large-cap basket trading above its own 50-day average — so every reading lands on a seventh, and a name with no 50DMA is dropped rather than counted against the sector.
Rotation Calls
−XLK
+XLVpair 1
flow −52 → +58; XLV basket 6/7 above 50DMA against XLK 1/7
−XLY
+XLPpair 2
5d −2.6% vs +2.4%, staples printing 1.28× their 50d volume
−XLC
+XLUpair 3
XLU +0.58% on 1.19× volume while XLC fades −0.61% on 1.02×
Every pair has to cite a number that is already on the ladder — a flow score, a trend, a volume pulse or a breadth reading. Up to four pairs survive per run.
AI Rotation Read
Rotation themegrowth-to-defensive
Confidence74%
Tilt+XLV+XLP+XLU−XLK−XLY
LLYlongXLVsector flow +58, basket 6/7 checked
COSTlongXLPstaples pulse 1.28× avg vol checked
AMDshortXLKXLK 20d −5.2%, breadth 14% checked
Ideas may only come from a named sector's own basket, and each technical claim is re-read against that stock's bars before it reaches the page.
07:03 ET weekdays| One run, all users
1/4
AI Industry Analysis

The same read, one level down.

Industries ranked under their parent sectors, drawn from a curated 203-name industry map — because "Technology" is not a trade, and Semiconductors and IT Services rarely move together. Two resolutions turn "Technology is bid" into "Semiconductors are bid, IT Services are not."

Each industry is scored from its own performance snapshot plus a representative large-cap basket: basket move, volume pulse, and the share trading above the 50- and 200-day. Valuation is judged against that industry's own trailing P/E range rather than against another industry's, so "expensive" means expensive for it.

  • Click any row to drill into leadership and named picks
  • Every pick is re-read against live quotes, the 50- and 200-day, and RSI before it reaches the page — a bad claim comes back stamped corrected, with the real number beside it
  • Lives as the AI Analysis tab on the Industries page
203-name industry map
P/E vs its own history
50- and 200-day breadth
Ideas re-read against bars
Industry Flow
Industry Flow Ladder
Industry Flow Ladder · inflow → outflow
#IndustryFMP ΔBasketVol>50DMAFlow
1Semiconductors+1.84%+2.31%1.41×
71%
+62
2Biotechnology+1.12%+1.48%1.26×
60%
+37
3Aerospace & Defense+0.74%+0.91%1.11×
60%
+25
4Gold+0.52%+0.63%1.08×
67%
+23
5Restaurants+0.08%-0.12%0.94×
50%
+1
6REIT - Office-1.06%-1.34%1.19×
33%
-38
7Banks - Regional-1.62%-1.95%1.37×
20%
-61
P/E vs Own 2Y Range
6M1Y2Yown range
Semiconductors34.2×
lo 19.4× · avg 27.8× · hi 38.1×+23.0% premium
Biotechnology21.6×
lo 17.2× · avg 24.9× · hi 33.5×-13.3% discount
Banks - Regional9.8×
lo 7.4× · avg 11.6× · hi 14.9×-15.5% discount
Each band is that industry's own trailing P/E range, so “expensive” is measured against its history — not against another industry.
Constituent Breadth
Semiconductors7 constituents
NVDA
+4.12>50D
AVGO
+3.05>50D
AMD
+3.48>50D
MU
+3.26>50D
TSM
+1.90>50D
AMAT
+1.42<50D
INTC
-1.06<50D
basket+2.31%A/D6↑/1↓>50D71%>200D57%vol1.41×
AI Rotation Verdict
rotation themesemis-to-biotech
confidence74%
−Banks - Regional+Semiconductors
−REIT - Office+Aerospace & Defense
NVDAlongSemiconductorsverified
VRTXlongBiotechnologyverified
KEYshortBanks - Regionalcorrected
Every idea is re-read against live quotes and RSI before you see it — a bad claim comes back stamped corrected, with the real number next to it.
11 parent sectors| Newest read for today
1/4
AI Scanners

How every scanner here behaves.

Does Versatum scan the whole market, or just the ticker you typed? Neither: each of the four boards above answers the question you have before you have a ticker, and every one of them names the population it read — because none of them is the whole market.

The population is named

Morning Picks scores 300 large caps; the Gap Analyzer polls 247 for the day’s biggest moves and rules on the twelve largest; sector work covers all 11 SPDR sectors and industry work a curated 203-name industry map. Each board says which of those it was drawn from, so “nothing here” never reads as “nothing in the market”.

The model’s numbers get checked

Sector and industry stock ideas pass through a technical validator that re-reads price, the 50- and 200-day averages, and RSI(14) from real quotes. A claim that does not survive comes back stamped corrected with the true number beside it, rather than being deleted or quietly published.

One run serves everyone

A market-wide read is identical for every user, so it is computed once and cached under a system account instead of per person. That is what makes full-universe scans affordable at all — and why the board is usually already on screen when you open the page.

Stock Picker

How does Versatum's ML factor model score a stock?

Rank a curated top-300 by market cap across 35 factors in 7 categories, with per-factor attribution and a backtest — every pick backtestable against SPY.

Value, Quality, Growth, Financial Health, Momentum, Earnings Quality and Analyst Sentiment: each factor is percentile-ranked against the rest of the universe, then the ranks are weight-summed into one score. Not a neural net and not a black box — not a trained model at all — so a score is reproducible, explainable one factor at a time, and every factor can be inspected on its own. Each score carries a confidence equal to the share of factor weight that actually had data behind it. How many factors are active depends on your trading style: 10 for Day Trade, 25 for Swing Trade, 18 for Long-term (Position).

  • Sector-neutral rebalancing
  • The backtest window follows your trading mode — three months, six, or a year — with Sharpe, Sortino, Calmar, alpha, and beta visible while the model runs
  • A custom strategy builder lets you re-weight the factors to match your thesis and run backtests with returns, Sharpe, and max drawdown
  • Morning Picks is scored by this same model — its 25-factor figure is the Swing Trade default
  • Equities only
35 factors · 7 categories
10 day · 25 swing · 18 long-term active
Backtest vs SPY
Confidence = weight with data
Stock Picker
Factor Analysis
Factor Groups
Momentum12f
78
Value10f
65
Quality9f
82
Growth8f
71
Sentiment11f
88
Scoring Matrix
MOM
VAL
QUA
GRO
SEN
AVG
NVDA
92
58
85
94
91
88
AAPL
75
82
90
68
78
82
MSFT
70
78
88
72
75
79
AMZN
80
55
72
88
82
76
GOOGL
65
85
80
60
70
73
Rankings
1
NVDANVIDIA
884.2
2
AAPLApple
821.8
3
MSFTMicrosoft
790.5
4
AMZNAmazon
762.1
5
GOOGLAlphabet
731.2
Weights
Factor Weights
Momentum
25%
Value
15%
Quality
22%
Growth
20%
Sentiment
18%
Model Stats
Illustrative
Accuracy
84.2%
Sharpe
2.14
Win Rate
68%
35 Factors|ML-powered scoring
1/4
Sentiment & Flow

What are the crowd, the money and the calendar saying?

Six boards read the market around an idea: which names the cross-stock factors favor, how fearful the crowd is, where the institutional volume is going, what the news says, what the macro calendar holds, and what Congress is trading.

Scanner

Factor Alpha

Cross-stock alpha rankings across momentum, volume anomaly, volatility setup, mean reversion, and trend strength.

Coverage: Stocks + crypto — crypto’s twin is the Alpha Scanner

Sentiment

Fear & Greed

A per-stock, multi-factor sentiment composite built from breadth, news, volatility and volume, with an AI contrarian analysis on top.

Coverage: Stocks and crypto

Flow

Whale Tracker

Volume microstructure for equities — institutional accumulation and distribution read through an A/D score and OBV, an institutional-vs-retail split, and divergence alerts.

Coverage: Stocks; crypto has its own Whale Detector

NLP

News Sentiment

News NLP polled every 15 minutes, with multi-dimensional sentiment scoring, narrative arc tracking, and theme clustering. An untagged article stays untagged — it reads as unmeasured, never counted as neutral.

Coverage: Stocks, crypto and FX — equity news from FMP, crypto and forex news from Alpaca

Prediction

Economic Calendar

A macro calendar with AI impact predictions for Fed meetings (FOMC), CPI, NFP, earnings seasons, and geopolitical catalysts.

Congress Trades

Track congressional stock trades, with AI analysis of their timing and potential alpha.

Start with the idea.
The ticker comes after.