TradeOS
BankrollP&LFree Margin Margin Used Notional RealizedWR
DRY
LIVE

Trades — spot

Open positions

PairP&LScoreStartEntryPriceStopTargetNotionalLevMarginFeesOpened

Signals

PairScoreTrendMomMRVolat PriceRSIATR%7d30dMCapVol/day

New listings — too new to score

Recently-listed Coinbase pairs without the ~60 days of history the confluence model needs. This is a lightweight momentum/volume read, not the validated score — treat it as a watchlist of fresh listings, not a buy signal.

PairAgePrice1d7d30dSince listVol/dayVol▲Bias

Scoring formula

Score = 0.30·Trend + 0.25·Momentum + 0.20·MeanReversion + 0.15·Volume + 0.10·Volatility
  • Trend 0.30 — EMA20 vs EMA50, price vs EMA200, EMA20 slope
  • Momentum 0.25 — RSI 50–68 zone, MACD vs signal
  • Mean reversion 0.20 — Bollinger %B (buy lower in band)
  • Volume 0.15 — latest vs 20-day average
  • Volatility 0.10 — ATR% (avg daily move); ~2–6% sweet spot

Each factor is scored 0–100 and shaded in the Signals table: strong ≥ 67 · neutral 34–66 · weak ≤ 33.

These factors feed a long score and a bear score (the bear score inverts mean-reversion — it shorts the overbought, not the oversold). They combine into one directional score = 50 + (Long − Bear)/2 — high = long, low = short, ~50 = neutral.
The market regime (BTC vs its 100-day MA) gates the side: Risk-on → LONG when score ≥ 65 · Risk-off → SHORT when score ≤ 35. Every position uses a 1.5×ATR stop and a 2:1 target.
The table is restricted to assets with a Coinbase perpetual-futures market — the only instruments these signals (especially the SHORTs) are executable on. Paper trades are costed with real Coinbase math: your account's live taker fee for the venue (spot vs US futures — they differ a lot) plus each perp's live funding rate, accrued over the hold — longs pay funding, shorts earn it. Entries whose funding bleeds >0.1%/day against the position are skipped (carry-expensive, crowded trades).