Experimental · we publish every call

An AI committee debates every trade.

NEW · AI Committee Bullshit Detector Market Backdrop Free

Five AI agents — technical, sentiment, fundamentals, risk, portfolio — argue over BTC, ETH and SOL, and a set of hard-coded (non-AI) rules can veto any of them. You read the reasoning, not just a verdict. Not a buy call. Not a prediction.

It's new and unproven — so we grade every call in public. As of 20 Oct 2026 the committee has 14 graded calls: 1 win, 10 losses, 3 mixed (plus 6 that never reached entry and 9 still open). A small, mostly-loss early record, not evidence of an edge. We already retired our old indicator engine after measuring it at ~41.5% and publishing the audit. Read it →

Get it on Google Play Download on App Store See how it works →
BTC · ETH · SOL — committee coverage Free — no account, no card Every call graded in public
Actual Pulse screen · iOS · v1.4.1
What's new — v1.4

Built for crypto skeptics.

For the people who roll their eyes at every "$X by EOY" tweet. Know when a dip is crypto's fault — or the whole market's. Call bullshit on predictions. Close the loop on yesterday's calls. Make every alert about coins you actually own. And lock in a price that never renews.

Market Backdrop

Is today a crypto move — or a whole-market move? The app now watches the macro tape — dollar index, Nasdaq, S&P 500, gold, the US 10-year yield, VIX — plus a live 30-day BTC↔Nasdaq correlation, refreshed hourly. One plain-language line at the top of Pulse tells you the backdrop: risk-on, risk-off, or mixed — and whether crypto has been moving with stocks lately or doing its own thing. Strictly a description of right now. Never a forecast.

Regime Risk-off
BTC↔Nasdaq · 30d 0.74 · moving together
Top driver VIX +9% · risk appetite down
Takeaway Whole-market move

Context, not prediction — held to the same honesty bar as everything else in the app. Free for everyone, on the Pulse tab.

Bullshit Detector

Pick a coin, direction, target price, and timeframe. The app returns engine read + crowd read + a verdict: Unlikely / Skeptical / Plausible / Likely. Paste any influencer prediction. Get an honest answer in 3 seconds.

Claim BTC → $200k by Dec
Engine Disagrees
Crowd Cooling (42% bullish)
Verdict ⚠ Unlikely

Illustrative layout. Same descriptive divergence math as the Pulse tab, which we measured as non-predictive.

Daily Pulse Brief

One card at the top of your morning. Today's market mood vs. yesterday. Today's biggest engine-vs-crowd gap. Yesterday's strong divergences resolved — who was right, the engine or the crowd? The ritual that closes the loop.

Mood Δ vs yday +8 pts
Biggest gap today SOL · 52-pt diverge
Yesterday's calls Crowd 2 · Engine 1
Brief for today Once / day

Portfolio · Connect Binance

Read-only API key. We pull your holdings, then prioritize alerts and divergences for coins you actually own. Skip the noise on coins you don't hold. Make every alert personal.

Connected Binance · read-only
Tracking BTC · ETH · SOL · LINK
Alert priority Your bag first
Permissions ✓ Read-only

Lifetime — $299, never renews

One-time payment. All Max features. Lifetime updates. Direct line to the founder. Founder's price, first 100 buyers only. The honest alternative to a subscription you'll forget to cancel.

Price $299 · one-time
Renewal Never
Includes All Max + founder access
Scarcity First 100 only
How it works

Two signals. One conversation.

We compute what 12 indicators say AND what the crowd is feeling — and surface the moments they disagree. The disagreement is the signal. Every input is auditable. No black box, no profit promise. And since v1.4, a third voice sits above the conversation: the Market Backdrop — so you know whether today's move is crypto-specific or the whole market.

01 · What the chart says

12 technical indicators, hourly, per coin.

RSI, MACD, EMA cross, Stochastic, Bollinger Bands, ATR, ADX, OBV, Ichimoku, CCI, Williams %R, Volume Profile — all configured to widely-published defaults. Each casts a bull / bear / neutral vote. The consensus direction is published, every vote is auditable.

RSIMACDEMA×Stoch BBATRADXOBV IchimokuCCIW %RVolume
02 · What the crowd thinks

Hourly sentiment from CoinGecko + Wikipedia.

Community engagement scores from CoinGecko (votes, watchlist activity, exchange listings) and interest spikes from Wikipedia pageviews. Aggregated into one 0–100 bullish % per coin. No scraped Twitter. No astrology. Refreshes every hour.

Bullish %
28% bullish 72% bearish
03 · Where they disagree

The gap is the signal.

When the engine says BULLISH but the crowd is 28% bullish (a 50+ point gap), we flag it as a strong divergence. Push notifications fire on these moments — not on every directional reading. We don't claim divergence predicts price. We claim it's worth a second look.

Pulse — the legacy indicator layer

The chart, the crowd. And the gap.

Most apps show you indicators on one screen and chat hype on another. Cryptochartics puts them side by side — and flags the moments they pull apart. Strong divergences fire a push alert; aligned readings stay quiet. This is the app's descriptive context layer: we measured it as non-predictive and retired it as a signal source. The AI committee covers BTC, ETH and SOL on the 1D and 1W timeframes.

Why we built this

Most crypto apps show you indicators in one tab and chat hype in another. The gap between those two screens is where the interesting questions live — and where most users miss the story.

Cryptochartics puts them side by side. When the technical consensus and the crowd point the same way, it's a clean read. When they strongly disagree — a 35+ point gap on our bullish-% scale — we flag it as a divergence and fire a push.

We're not claiming divergences predict price. We're claiming they're information worth pausing on, for any decision you're going to make anyway.

Sentiment refreshes hourly, sourced from CoinGecko community signals + Wikipedia interest spikes. No scraped Twitter, no astrology.

The technical layer

What the chart says — in 12 readings.

The technical half of the crowd-vs-chart comparison. No proprietary "AI score". Every formula is published, every threshold is the textbook default, every vote is auditable. The audit shows some of these are noisy or even reverse-predictive — we publish that too, on the audit.

RSI
Relative Strength (14)
Overbought / oversold momentum oscillator. Threshold 70 / 30.
Bullish
MACD
Moving Avg Convergence (12,26,9)
Trend-following momentum from the difference of two EMAs.
Bullish
EMA×
EMA Cross (50/200)
Death cross / golden cross. The macro trend signal everyone watches.
Bullish
Stoch
Stochastic (14,3,3)
Closing price relative to its range over the last N periods.
Neutral
BB
Bollinger Bands (20, 2σ)
Standard-deviation bands around a 20-period moving average.
Bullish
ATR
Average True Range (14)
Volatility — direction-agnostic. Used to size confidence.
Neutral
ADX
Directional Index (14)
Strength of the prevailing trend, regardless of direction.
Bullish
OBV
On-Balance Volume
Cumulative volume flow — confirms or contradicts a price move.
Bullish
Ichimoku
Cloud (9, 26, 52)
Multi-line trend system. Cloud above price = bullish bias, below = bearish.
Bullish
CCI
Commodity Channel Index (20)
Deviation from a typical price across the lookback window.
Bullish
W %R
Williams %R (14)
Faster Stochastic variant. The audit found this one reverse-predictive — when it disagrees with the rest, the signal wins 60% of the time.
Disagrees
Volume
Volume Profile
Where volume actually transacted — support and resistance from real trades.
Bullish
vs. the alternatives

The only crypto app honest about its own accuracy.

We respect what TradingView, LunarCrush, and the various signal services do well. But for "how does the chart's read stack up against what the crowd is feeling, and how often is the engine actually right?" — Cryptochartics is the only tool that puts both signals together AND publishes its measured win rate openly.

Feature Cryptochartics TradingView LunarCrush Signal services Telegram groups
Crowd sentiment + 12-indicator TA side-by-side✓✗~✗✗
Push alerts on engine-vs-crowd divergence✓✗✗✗✗
Bullshit Detector — check any influencer prediction✓✗✗✗✗
Market Backdrop — crypto move vs whole-market call✓~✗✗✗
Daily Pulse Brief — yesterday's resolution recap✓✗✗✗✗
Portfolio integration (Binance read-only)✓✗✗✗✗
Lifetime tier — one payment, no renewal✓✗✗✗✗
Engine win rate published openly✓✗✗✗✗
Indicator-by-indicator transparency✓✓✗✗✗
Plain-English explanations per vote✓✗✗✗✗
Mobile-first design✓~✓~✓
No "guaranteed profit" claims✓✓✓✗✗
Free tier✓✓✓✗✓
Annual price (top tier)$99$599+$240+$500–2k+varies
Pricing

It's free. All of it.

No account. No card. No tiers. The committee is experimental and unproven — charging for an edge we haven't demonstrated would be exactly the kind of thing this app exists to call out. So while we're proving it, everything is open to everyone.

Everything, everyone

Free

$0 / forever, for now
no account, no card
  • The full AI committee on BTC, ETH & SOL
  • Every agent's reasoning + the risk governor's vetoes
  • "Trades we walked away from" — the rejections feed
  • Market Backdrop — macro context, hourly
  • Bullshit Detector — check any influencer's call
  • Crowd-sentiment + cross-asset context
  • Accuracy tab — every committee call graded in public
  • Push alerts when the committee acts
Get the app — free
Questions

The fine print, up front.

What's the Bullshit Detector?
A 3-second sanity check on any crypto prediction. You enter the coin, the direction (UP / DOWN), the target price, and the timeframe — the app runs that claim against the live 12-indicator engine read AND the live crowd sentiment, and returns a verdict: Unlikely / Skeptical / Plausible / Likely. Paste any influencer's "$X by EOY" tweet and find out whether the data agrees. It compares the claim to current indicator and crowd readings (descriptive context, not a forecast), so it can say a claim looks plausible as well as unlikely. Free tier: 5 checks/day. Pro / Max / Lifetime: unlimited.
Is there a free trial?
Yes — 7 days of Pro or Max, free, no credit card required. Start the trial inside the app from the Paywall screen. At the end of 7 days you decide: pay annually in USDT, monthly via your app store, or drop back to the free tier (which is real, useful, and stays free forever). No auto-charge, no "we'll bill you unless you cancel" trap.
Is this financial advice?
No. Cryptochartics is a crypto intelligence layer — we show you what twelve well-known technical indicators say about a coin, what the crowd is feeling, and when those two stories disagree. That's it. We don't tell you to buy, sell, or hold. We don't manage money. We don't promise results. Decisions are yours. Crypto is volatile — trade money you can afford to lose.
How accurate are your signals?
Two separate records. The AI Investment Committee (BTC, ETH and SOL on 1D and 1W) is experimental and has a small, mostly-loss early record: as of 20 Oct 2026, 14 graded calls — 1 win, 10 losses, 3 mixed — plus 6 that never reached entry and 9 still open. That's too few calls to conclude anything, and so far it hasn't shown an edge; we publish it live on the Accuracy tab and will pull the committee if net-of-fees results stay negative after 100 graded trades. Our retired 12-indicator engine won about 41.5% of 12,362 evaluated trades, below the ~50% needed to break even after fees, and we stopped using it as a signal source. Full audit: cryptochartics.com/blog/we-measured-ta-failure. Nothing here predicts future results.
What does "divergence" mean here?
In the Pulse tab, a divergence is when the 12-indicator technical consensus and the crowd sentiment composite point in opposite directions, with a ≥35-point gap on a 0-100 bullish scale. It's descriptive context from our retired indicator engine, not a signal: when we measured divergence calls they did not beat a coin flip. The AI committee is a separate system and does not use divergence as its trigger.
If the old engine was below break-even, why show it?
As context, not instructions. Pulse shows what indicators and crowd sentiment say right now so you can see where the two stories differ. We measured neither layer as reliably predictive, so we don't present it as a signal. Use it as research.
What's the Market Backdrop?
A one-line macro context strip at the top of the Pulse tab, new in v1.4. Every hour we poll six cross-asset series — the dollar index (DXY), Nasdaq, S&P 500, gold, the US 10-year yield, and VIX — plus a rolling 30-day BTC↔Nasdaq correlation, and compose them into a risk regime: risk-on, risk-off, or mixed. The takeaway is written in plain language: is today's crypto move actually a crypto story, or is the whole market moving and crypto just along for the ride? Like everything else in the app, it's held to the same honesty bar — it describes the backdrop right now, it never predicts where prices go next. Free for everyone, no tier required.
Is there an iOS version?
Yes — Cryptochartics is live on both the App Store and Google Play. The free tier is identical across platforms. Paid tiers (Pro / Max) are currently Android-only while we finalize the business banking required for Apple In-App Purchases; iOS users get the full free experience today, and we'll enable Pro / Max on iOS in a later update. Download on the App Store or Get it on Google Play.
Why annual prepaid in crypto, not credit cards?
Two reasons. One: no auto-renewal means no surprise charges six months from now — you opt in actively when the year ends. Two: stablecoin payments cut card-processing fees, which is why we can charge $30/year instead of $50. Monthly card subscriptions through the Play Store are also available if you prefer.
What data sources do you use?
Binance Spot for OHLCV candles. The 12 indicators (RSI, MACD, EMA cross, Stochastic, Bollinger, ATR, ADX, OBV, Ichimoku, CCI, Williams %R, Volume Profile) computed from those candles using pandas-ta. Crowd sentiment is pulled hourly from CoinGecko (community engagement, votes, exchange listings) and Wikipedia (pageview velocity / interest spikes), then composed into a single 0–100 bullish % per coin. The Market Backdrop pulls DXY, Nasdaq, S&P 500, gold, US 10-year yield, and VIX quotes hourly from Yahoo Finance market data (via OpenBB), plus a 30-day BTC↔Nasdaq correlation. Everything is stored on Supabase Postgres and published over Supabase Realtime — the app sees new readings within a second of computation.
Do you sell my data?
No. We don't sell, share, or barter user data. The app is paid for by Pro/Max subscriptions and stablecoin payments. Free accounts cost us a few cents a month to run, which is fine. The full code is open at github.com/muratchary/cryptochartics if you want to verify.
Where can I read the full audit?
The full audit lives at cryptochartics.com/blog/we-measured-ta-failure. ~15 min read, includes the methodology, every chart, the confidence-paradox finding, the Williams %R reverse-predictive finding, and the academic literature citations. You can also grab the PDF version via the form below. We've made the methodology reproducible — if you find an error, please tell us at hello@cryptochartics.com.
Free audit · 8 pages

What 3,767 crypto signals taught us about technical analysis.

The original audit (since updated by a larger re-measurement at ~41.5%): 35.5% win rate on 3,767 trades, the confidence-paradox table, the Williams %R reverse-predictive finding, the academic literature backing it up — and why we retired the indicator engine as a signal source.

Written by the founder. Methodology included. Reproducible.

Read the audit online → · Download PDF directly ↓

We email you the PDF link and an occasional update — including new divergence findings as we publish them. Unsubscribe with one click. No spam, no resale.

Got it — the audit is on its way.
In the meantime, download the PDF now ↓ or read it online.

Read the minutes.
Decide for yourself.

An AI committee on BTC, ETH & SOL — the full reasoning, the calls it walks away from, and every result graded in public. Experimental, unproven, and free. No account, no card.

Get it on Google Play Download on App Store