Fundamentals & analyst consensus come from Yahoo (cached 12h); news is live. Research tool — analyst targets are third-party opinions, not advice.
Fundamentals & analyst consensus come from Yahoo (cached 12h); news is live. Research tool — analyst targets are third-party opinions, not advice.
loading…
Install the free ntfy app from Google Play → Subscribe to the topic above. On Samsung: ntfy settings → Instant delivery, and Android Settings → Apps → ntfy → Battery → Unrestricted (One UI otherwise delays pushes). The watcher scans day + swing signals and pushes only NEW walk-forward-validated entries with entry/stop/target/leverage.
Every position here survived walk-forward out-of-sample validation with the parameters
chosen in the most recent window — in-sample-only winners never appear. Sizing: OOS-Sharpe ÷ volatility,
max 25% each; the remainder stays cash. Click a row's Chart button to see that strategy's entries on the candles.
Snapshots saved daily to reports/. Research tool — not financial advice; leverage suggestions
respect ESMA retail caps but margin calls are not simulated.
The account auto-executes the advised swing portfolio every watcher cycle (15 min default): fills at the latest price, logs slippage vs the advised entry, exits on stop / target / time. Every night at 23:15 the learning job re-fits all walk-forward parameters on data including today, recalibrates cost assumptions from measured slippage, and quarantines any strategy whose live expectancy turns negative (see the nightly push report). Overnight/CFD fees are not simulated yet. Virtual money — the point is an auditable track record and the live-vs-backtest gap.
Everything below assumes one goal: grow capital with validated edges while risking a drawdown you chose in advance. The app is built to be honest before it is optimistic.
The validated strategies here historically produce out-of-sample Sharpe ratios around 0.5–1.2 — that means realistic expectations of roughly 0.5–2% per month at moderate risk, with losing weeks and losing months along the way. Anything promising much more is either leverage (which multiplies losses identically) or overfitting. "Easily" in trading means: the process is easy to follow — the returns still require patience and discipline.
Pick a market in the top bar (categories: indices, stocks, ETFs, commodities, forex, crypto, eToro tradeable prices, your local CSVs). The chart auto-detects and draws:
Intervals: Daily for swing/long-term; 30m/15m for day-trading (≈1 month of history); Hourly in between; Weekly for the big picture.
Pick a strategy, tweak parameters (hover a field to see the walk-forward grid), choose leverage (1×/2×/5×) and costs, and run. You get equity vs buy-and-hold, metric tiles, and the trade list. Then press Validate (luck test):
Rule: a good-looking backtest that fails validation is a coincidence, not a strategy.
Optimizes on a training window, trades the NEXT window blind, slides, repeats — simulating how you would actually have traded. Read three numbers per row:
Click a row for the stitched OOS equity curve and per-window parameter choices. The weekly Saturday sweep (≈1.4 million backtests) does this across every strategy × symbol × 9 window schemes and judges each strategy by its median — never its best cell.
Search all ~15,000 eToro instruments, see leverage tiers and per-class fees (ESMA retail caps), and chart any of them on eToro's own price feed (ETORO:GOLD etc. also work in Backtest and Walk-Forward). Key rule: ×1 long stocks/ETFs/crypto = real asset, no overnight fee; any leverage or short = CFD with nightly fees (×3 on weekends) — that cost is real for multi-day holds. Crypto on eToro costs ~1% per side: set fee accordingly before believing a crypto backtest.
Every position passed walk-forward validation (OOS Sharpe ≥ 0.3, recent-window parameters),
is sized by edge ÷ volatility (max 25%), and shows entry/stop/target, suggested leverage
(auto-capped to 1× near FOMC), and stability. The unused remainder stays in cash — cash is a
position. Snapshots are saved to reports/ daily, so the advice is auditable.
A virtual $100k account auto-executes the advised swing portfolio every 15 minutes: real fills at current prices, logged slippage vs the advised entry, exits on stop/target/time. This produces the two numbers that matter more than any backtest: live win rate and live-vs-backtest gap. Give it 4–8 weeks before drawing conclusions — 10 trades mean nothing, 100 start to.
Install ntfy (Google Play, free) → Subscribe to the topic shown in the Portfolio tab. Samsung: enable Instant delivery in ntfy + set Battery to Unrestricted, or pushes arrive late. You'll receive: new validated entries (with all trade details), FOMC warnings, and the nightly learning report.
Sharpe: return per unit of risk; >1 is very good after costs. Drawdown: worst peak-to-valley loss — the number that makes people quit. OOS (out-of-sample): performance on data the optimizer never saw — the only kind that counts. WF efficiency: how much of the in-sample edge survives OOS. Expectancy: average %-gain per trade. CFD: a derivative position with financing costs — what leverage/shorts are on eToro. Quarantine: strategy suspended for losing real (paper) money despite good backtests.
The server listens on your home network. On your phone (same Wi-Fi), open Chrome → http://<your-Mac's-IP>:8642. Then Chrome menu → Add to Home screen — you get an app icon that opens full-screen. The whole UI is responsive: swipe the tab bar, all tables scroll. Away from home Wi-Fi the page won't load (the server is local) — but phone push alerts still arrive anywhere, because ntfy is cloud-delivered. For full remote access from anywhere, install Tailscale on Mac + phone (free, encrypted) and use the Mac's Tailscale IP.
Owner's part: go to login.tailscale.com/admin/machines → row "macbook-pro-di-matteo" → ⋯ → Share → copy the link. Send the friend that share link + the invite code (the owner sees the current code in the Portfolio tab → Account panel).
The friend's list:
Every account gets its own private paper-trading account, own alert topic and own settings — nobody can touch anybody else's. Heavy compute (the weekly million-backtest sweep) stays owner-only.
It does not place real orders (yet) — you execute manually on eToro. It does not model margin calls at 2×/5×, or CFD overnight fees in backtests. It cannot remove market risk: validated edges lose money on many individual days. Nothing here is financial advice — it is a research instrument that enforces discipline you would otherwise need willpower for.