Alpha — the number
One score per name, 0 to 100, with the three dials behind it.
The scale
0–100, and 50 means nothing to see. Above 50 = relative leadership building; below = relative weakness. 80 is strong multi-horizon leadership; 20 is its mirror on the short side.
Direction · Strength next to the number (Bullish · Strong) is the same information in words. A name too weak to lean either way reads Neutral — no forced opinions.
What's inside it
Relative strength vs the market and vs the sector, over multiple lookbacks. Confirmation from peers. Trend health. Not every ingredient counts equally — the blend is ours.
Then the discounts: unusual volatility for that name pulls the score toward 50, a fresh shock is flagged before it's allowed to move the score, and a market mood that opposes the name trims the edge. Alpha is post-discount — the caution is already in the number.
Engine components — the three dials
Edge — how far this name's relative-strength picture sits from neutral. The raw material.
Conviction — how much the engine trusts the read after volatility and noise adjustments. High Edge + low Conviction = interesting but treacherous. The heatmap's Confidence view shows the raw material behind it: how many of the engine's independent checks currently agree with the call — relative strength across several horizons against both the market and the sector, peer breadth, trend and momentum. "9/10 signals" is a chorus; "3/10" is one loud voice and a lot of shrugs. The count is always out of ten, and the tenth check is stability — the name's short-term behavior. Steady (no abnormal one-to-two-day moves) counts as agreement: nothing violent is contradicting the read. A sustained shock makes stability vote directionally — with the call if the surge runs the read's way, against it if not (the ⚡ badge is reserved for that counter-trend case, because it's a warning). A fresh shock — sharp but not yet persistent — abstains until it proves itself, so a single wild day can't swing the count. And the chorus has been worth listening to — in our testing, bullish reads with near-unanimous agreement went on to beat the market by a meaningfully wider margin than split reads over the following weeks. A tested expectation with the usual caveats, and one that held for bullish reads: on the short side the evidence is mixed so far, so treat bearish agreement as description, not endorsement.
Sector rank — how much of the name's own sector it is currently beating, oriented to the side of the call (for a short, how much of the sector is beating it). High always means the evidence agrees with the direction.
It's a rank, not a health check: 97 in a hot sector is a monster; 97 in a struggling one is the best house on a rough street. The health check sits two cards over — Sector tone. Rank 97 + tone 19% isn't a contradiction; it's the full picture: leader of a weak group.
Sector rank 97 · Sector tone 19% → concentrated leadership — prime hedge-pair long leg, but crowded and lonely if it cracks.
The footnote, decoded
First part: what the score has done today. Second: the market's mood (Risk-On → Constructive → Mixed → Cautious → Risk-Off) and — when it appears — headwind/tailwind is disclosure, not advice: the engine has already trimmed (or credited) this name's edge for the tape. Don't discount it a second time yourself.
The score strip, card by card
The desk strip first — the sticky row under the header carries the five facts worth keeping in view on every tab: the session and last refresh, the tape, the market regime, Market α, and market tone. Market α is the whole market scored on the same 0–100 scale as every stock, and its chip carries a tiny range bar: the span is today's low-to-high of that score, the faint tick is where it opened, the bright dot is where it sits now. Dot near the top of the span = the market's strength has been building all session; dot at the bottom after opening high = a day that gave it back. It's an open-high-low-current of a score, not a price — which is why you won't find it anywhere else. Every stock and sector ticket carries the same bar beside its big Alpha number, so a name's day reads the same way: did its strength build or bleed since the open? The bar resets at each open and fills in as the session runs — early in the day (or right after a restart) it may show only the open and now, so give it an hour before reading much into it.
Selected Alpha — the focused name's score, direction, and where it opened. Identical to the big number in the detail panel, kept in view while you browse.
The price line follows broker convention: the day change is measured against the prior session's official close. After 16:00 ET the headline number freezes at the closing print, and any evening movement shows separately as a small AH figure against today's close — so a quiet after-hours drift never rewrites the day you actually traded.
Relative value state — the Extension read (Fair, Very Extended, Deep Discount…) vs the name's own sector norms. Position, not direction.
Tone — how bullish a scope is leaning: the share of names the engine calls Bullish. 50% is balanced; the descriptor says it in words (65% = Bullish, 19% = Bearish). With a stock selected the card shows its sector's tone — the denominator for that name's Breadth — against the market and universe tones below. At sector or market level it reads Market tone vs universe. Your opportunity-set gauge: 60% is a stock-picker's tape, 25% means longs are fighting the current.
Market regime — the tape's own health, read from our TM Breadth set: how many names are in uptrends, and whether fresh highs outnumber fresh lows. Same regime that feeds every name's headwind/tailwind.
How to trade it
Direction of Alpha beats level. Rising through 60 is a better entry than sitting at 80 — the move is being built, not already spent. And the same rising score is worth more from Fair or Discount than from Extended: strength that hasn't stretched yet has somewhere to go.
Level sets size. High Alpha with high Conviction earns a full position; high Alpha with Turbulent volatility badges earns a smaller one. Agreement sharpens this on the long side: in our testing, bullish reads where nearly all checks agreed went on to beat the market by a meaningfully wider margin than split reads — so a split-signal long deserves a smaller position, or patience until more checks line up. On the short side that pattern has not held so far, so size shorts by risk, never by our certainty.
Bleeding back toward 50 is the early exit — the score decays before the price chart confesses. Pair it with the chart's ribbon for the visual confirmation.
The directional exit that tested best: leave a bullish call when its Alpha closes back under 60. We made exits compete on identical fresh Strength-run entries. In the trending tape, the fixed 21-day hold earned the most in total, but the Alpha-under-60 exit kept most of that edge in less than half the time, with half the worst-case — the best return per day held. In the 2022 grinding bear, every fixed hold lost money, worse the longer you held, and this exit cut the loss by two-thirds. Robust in both weathers, so it's the rule the tracker flags: it's the directional twin of the pair book's three-Narrowing-days rule. Unlike pairs, there's no "payoff window" to wait for — the edge accrues while the run persists and stops when the score gives up 60.