The economy & markets — read in 60 seconds

Every AI stock tool is a black box. We tore ours open.

Type a ticker and get a senior-analyst read on the chart — conditioned on today's inflation-and-Fed regime, with every number that moved the score printed in plain sight, and a public scorecard that logs the losses too. Not a black box. Not a guarantee. A read you can audit.

Free to try · no card required · informational/educational only — not investment advice.

⌘ NVDA
68
/ 100
BULLISH · 90-DAY
confidence: medium · as of 2026-06-28
Show Your Work — every point in the score
+8
Above 200-day MATECHNICAL
+5
Above 50-day MATECHNICAL
+5
MACD positiveTECHNICAL
−4
Near resistanceTECHNICAL
+4
Macro regime base rateMACRO
= 68 / 100 · Cautiously Bullish
In rising-inflation + hawkish-Fed regimes, higher 90d later in 6 of 8 analogs (+4.2% median). Sample: 8 — medium confidence.
A historical/technical read — not financial advice.
Typical AI stock picker
★ STRONG BUY
🔒 reasoning hidden

No vanity metrics

What's on the table — in plain sight

Most money sites lead with “$37B+ traded” or “join 100k investors.” We're new, so we'd be making that up — and making things up is the exact thing we're against. Here's the honest version: four promises you can verify on this page before you trust us with a single decision.

🧾

Every score, itemized

The 0–100 number is broken into the exact signals that built it — each with its point value and a plain-English reason. No hidden weights; they live in config, not a vault.

📒

Wins and losses, public

Every call is timestamped and graded against the real outcome. We post the misses on the same page as the hits, with a calibration curve.

🧭

Conditioned on today

Every read is tagged to the current economy — inflation rising or falling, the Fed hiking, holding, or cutting — not a chart read frozen in 2019.

⚖️

A read, not advice

This is a historical and technical read for information and education only. You make the call. That sentence ships on every single card.

The transparency wedge

Every score is just arithmetic you can check.

Each signal is named, scored, and explained in one plain-English line — technicals and today's macro regime, side by side. A golden cross adds points and tells you so. An overbought RSI takes them away and tells you so. Start at 50. Add the green, subtract the red. The total is your score, and the weights live in config, not a vault. If you disagree with a number, you can see exactly which number to disagree with. Find a competitor who shows you this. We'll wait.

Show Your Work
+8
Above 200-day MATECHNICAL
Long-term trend is up — price holds above the 200-day.
+4
Golden cross (50/200)TECHNICAL
The 50-day crossed above the 200-day.
−6
RSI overbought (71)TECHNICAL
Above 70 — near-term pullback risk, docked out loud.
+4
Macro regime base rateMACRO
Favorable in regimes like today's, scaled for sample size.
start 50 → = 60 / 100

The macro edge

A great chart in the wrong economy is still a trap.

A chart doesn't mean the same thing when inflation is cooling and the Fed is on hold as it did when rates were at zero. So before we score anything, we read the room — then ask the only honest question: in past months that looked like today, how did this stock actually behave 90 days later?

01🧭

Read today's regime

We classify the economy from real macro series — inflation trend and Fed stance — and tag every read with it. Today: Rising inflation + Hawkish Fed.

02🔍

Find the analog months

We pull every past month whose regime matched today's, then measure this stock's forward 90-day return in each — relaxed to inflation-only if analogs are scarce.

03📊

Report the base rate — with its limits

Higher 90d later in 6 of 8 analogs · +4.2% median · sample 8 · confidence: medium. The caveat is part of the answer, never buried beneath it.

The trust moat

We keep score in public. Wins and losses.

Anyone can screenshot their wins. We log everything — every call timestamped, given a horizon and an entry price, then resolved against the real outcome 90 days later, right or wrong, and posted. No edits, no deletions. And we check our own honesty: when the read implies a 70% chance, does it land near 70%?

🎯

Hit rate

Shown live from the scorecard as calls resolve — never an invented round number. What matters isn't the brag; it's whether it's calibrated.

📈

Calibration curve

Does '70%' mean 70%? Per-bucket bars compare implied confidence to actual hit rate, each with its sample size (n=).

✓ ✗

Recent calls — incl. the misses

Red marks a miss, on equal footing with the wins. Backtested rows are flagged ·bf so you always know which is which.

Open the full live record →

A fair comparison

Why see your work when nobody else does?

To be fair: Danelfin, TrendSpider, AInvest and a ChatGPT prompt are polished, fast, and cover more tickers than we do today. Several are genuinely sharp tools. But on the two questions that decide whether you should believe an AI about your money — can you see the math, and can you see the misses — the honest answer for most is no. A number you can't audit is just a vibe with a decimal point.

CapabilityTallyreadTypical black box
Shows every point in the score Yes — every signal, with its pointsA score, no math
Conditions on today's Fed & inflation regime Yes — base rates + sample sizeRarely, never shown
Publishes losses, not just wins Public scorecard, misses includedWins only
Calibration you can check Yes — does 70% mean 70%?No
Tells you when the data is thin Flags 'low confidence' out loudSounds certain anyway
Promises returns NeverOften

The engine behind the channel

Every read is a post, ready in one tap.

The same card you read becomes a clean, branded image — score, regime, the ledger, the disclaimer — exported straight from the app, ready for the feed. It's how “read in 60 seconds” actually travels. Run your own ticker, share the receipt, and let the comments argue with the math instead of the vibes.

🎬

Same engine, two formats

The 60-second short is the headline; the app is the full receipt — the read we'd give on camera, on demand.

📤

One-tap shareable card

Every analysis exports as an auto-branded card with the CO logo, the gauge, the lean, and the disclaimer baked in — a post, not a screenshot.

TALLYREAD
68
/ 100
BULLISH · 90-DAY
tallyread.com · Not financial advice

Pricing

Audit a few reads free. Go Pro when you're tracking the board.

We don't hide the method behind a paywall. The free tier shows the whole glass box — you pay for volume and tracking, not transparency. No card required for Free. Cancel anytime. Not financial advice.

Free
$0
  • 3 analyses / day
  • Full Show-Your-Work breakdown
  • Macro regime base rates
  • Public scorecard access
  • Shareable cards
Pro · Most popular
$19 /mo
  • Unlimited analyses
  • Watchlist & alerts (“text me before PCE”)
  • Full personal call history
  • Priority data
Desk
$49 /mo
  • Everything in Pro
  • Sector-rotation & macro dashboard
  • Export-ready branded cards
  • Early access to new signals

No signup wall to see the math

Type a ticker. See the whole read — and the receipts.

Free scans, no card. Decide for yourself before you ever make an account.

Informational and educational only — not financial advice.

Questions, answered straight

Frequently asked questions

Is Tallyread financial advice?

No. Tallyread provides a historical and technical read for informational and educational purposes only. It is not investment advice, not a recommendation to buy or sell any security, and not a guarantee of any outcome. Tallyread is not a broker or a registered investment adviser, so you make every decision yourself.

How accurate is Tallyread?

Tallyread does not promise a fixed accuracy number or any returns. Instead, every call is timestamped, given a horizon and entry price, and resolved against the real 90-day outcome on a public scorecard that posts the misses next to the hits. The live hit rate and a calibration curve let you judge the track record for yourself.

What is a macro-regime base rate?

Tallyread classifies today's economy from macro data (whether inflation is rising or falling and whether the Fed is hiking, holding, or cutting), then finds every past month whose regime matched today's and measures how that specific stock actually moved over the next 90 days. The result is a hit rate, a median move, and the sample size of comparable periods.

Is Tallyread a black box?

No, and that is the whole point. The 0-100 score is shown as plain arithmetic: it starts at 50, then adds and subtracts named signals, each with its point value and a one-line reason. The weights live in public configuration rather than a hidden model, so you can see exactly which number to disagree with.

How much does Tallyread cost?

Tallyread has three tiers. Free is $0 and includes 3 analyses per day with the full Show-Your-Work breakdown and no card required. Pro is $19 per month for unlimited analyses, watchlist and alerts, and full call history. Desk is $49 per month and adds a sector-rotation and macro dashboard, export-ready cards, and early access to new signals.

How is the Tallyread score calculated?

The 0-100 score begins at 50, then adds points for bullish signals and subtracts points for bearish ones. Signals include standard technical indicators (50- and 200-day moving averages, RSI, MACD, Bollinger bands, ATR, and support/resistance) computed directly from price history, plus the macro-regime base rate. The full ledger of point values is displayed on every analysis.

What data does Tallyread use?

Prices are daily split- and dividend-adjusted history (about 12 years) from a market-data provider, Tiingo by default. Macro series such as CPI, core PCE, the federal funds rate, the 10-year yield, and unemployment come from FRED, the St. Louis Fed's database. When a live feed is unavailable, the app clearly labels any fallback data rather than presenting it as real.

What does the public Scorecard show?

The Scorecard logs every analysis with a timestamp, horizon, and entry price, then resolves it against the real outcome 90 days later, right or wrong, with no edits or deletions. It shows the live hit rate, a calibration curve that checks whether an implied 70% chance lands near 70%, and recent calls including the misses, with backtested rows flagged so you always know which is which.

How does Tallyread handle a small sample of comparable periods?

Tallyread labels the confidence of each macro base rate by how many analog months exist: high for 12 or more, medium for 6 or more, and low below that. It never dresses a thin sample up as certainty; when the data is limited, the read says so out loud and the caveat is shown alongside the number.