Summary of "MR"

Mean reversion — high-level concept

• Trade when price deviates substantially from a historical average with the expectation it will revert to the mean.
• Typical actions:
  • Go long on large negative deviations.
  • Go short on large positive deviations (shorts generally require more conservative parameters).
• Works across timeframes (1-minute to monthly). Lower timeframes produce faster signals; higher timeframes require more patience.

Assets, instruments and market data

• Crypto focus: Bitcoin (BTC), Ethereum (ETH), Solana (SOL — transcript typo “Salana”), Bitcoin Cash (BCH), ARC / Ark, UXL (UXLink?), Taiko, various altcoins (memecoins noted as avoidable).
• Instruments / market data used:
  • Spot order book, futures open interest, funding rates, liquidations, options expiries.
  • Volume-weighted averages, ATR, Bollinger Bands.
  • Volume profile, cluster data, big-trade detection.
• Macro/market indicators referenced: VIX, DXA (index), bond yields, oil prices.

Selection and screening criteria (filters)

• Stationarity: required when possible. Use unit-root / stationarity tests:
  • Augmented Dickey-Fuller (ADF), KPSS, Phillips-Perron.
  • Mean reversion assumptions only valid on stationary series.
• Liquidity & spreads:
  • 24h trading volume > ~100 million.

  • 100,000 transactions/day desirable.

  • Bid-ask spread < 0.1% of price.
  • Order book depth: at least $100k–$200k liquidity within ±2% of price (200k preferred).
  • Asset listed on ≥3 major exchanges for arbitrage efficiency.
• Age and volatility:
  • Asset >2 years preferred; ≥6 months on major exchanges as a minimum.
  • Stable volatility preferred; optimal daily volatility ~2%–5%.
• Correlation / beta:
  • Prefer correlation with Bitcoin < 0.7.
  • Optimal beta range loosely referenced (~0.8–1.5 implied).
• Avoid:
  • Trending assets, bull-market momentum plays, low-volatility assets, exotic/new listings, news-driven assets (memecoins).
• Additional monitors:
  • Funding rate (avoid extreme values e.g., >0.1%).
  • Order book imbalance (avoid entries when imbalance strongly against your side, e.g., >3:0).
  • Open interest / liquidation maps, manipulation detectors.

Indicators and parameter choices (methodology)

• Moving averages:
  • SMA, EMA, linear-weighted MA, VWMA, adaptive MAs (Kaufman, fractal/adaptive), triple/double exponential/smoothed MAs.
  • Presenter preference: VWMA with a long period (example: 240) as core mean.
• Bands and deviation measures:
  • Bollinger Bands applied to VWMA (example VWMA period 240); multi-deviation bands (2nd/3rd/4th deviations used).
  • Z-score / deviation indicator applied to chosen mean.
• Volatility and stops:
  • ATR used for stop sizing; example multiplier ≈ 3–3.4.
• Other signals:
  • Order-book imbalance, big-trade detectors, volume profile, cluster search, open interest changes.
• Stationarity screener (ADF/KPSS) built into the terminal for candidate selection.

Concrete trading rules / framework (step-by-step)

1. Asset selection
   • Run stationarity screener OR manually run ADF/KPSS.
   • Filter by volume, liquidity, spread, age, correlation.
2. Timing & timeframe
   • Choose timeframe consistent with risk tolerance:
     • Scalp: 1–5m
     • Intraday: 15–30m
     • Positional: hourly to daily
3. Entry rules
   • Enter long when price deviates significantly below the chosen mean (use Z-score / threshold).
   • Enter short when price deviates significantly above the mean (use more conservative parameters for shorts).
   • Confirm with order-book (no large opposing limit orders), liquidation map, and open interest imbalance in favor of the trade.
4. Position sizing
   • Base size: fixed 0.5%–1% of portfolio per trade (presenter example: 1%).
   • Increase size at extreme deviations (incrementally), but never average beyond 3 standard deviations.
   • Max concurrent positions: 3–4.
5. Averaging
   • Averaging allowed within rules to reduce average entry, but do not average if price moves beyond 3 standard deviations against you.
6. Stops & exits
   • Time stop: exit after a set time if no mean reversion (examples: 4–6 hours for shorter timeframes; several days for longer).
   • Statistical stop: exit if deviation exceeds 4 standard deviations.
   • ATR stop: ATR * ~3–3.4.
   • Trailing stop to lock gains when price moves toward the mean.
   • Exit on structural signals: change in correlation with market, large shifts in order-book/liquidity, major macro events.
   • Take-profit: typically on return toward the mean (often targets of a few percent per trade).
7. Risk limits
   • Retail maximum drawdown target: keep strategy drawdown <15%–25%.
   • Institutional standard: <10%.

Testing, validation and performance checks

• Backtesting:
  • Backtest over at least 1 year; include out-of-sample testing and stress tests on crisis periods.
  • Analyze maximum drawdown and worst-case scenarios; test for structural shifts.
• Performance metrics (presenter’s self-reported results):
  • Sample: ~20 trades/week with ~85% win rate in recent tests.
  • Claimed performance: ~25% return over two weeks (~10%/week) during a short testing period (40–50 transactions analyzed).
  • Presenter admitted a few large mistakes (averaging errors) and emphasized iterative learning.

Operational / monitoring checklist

• Morning prep:
  • Check overnight news, macro calendar, Asian session anomalies.
  • Update correlation matrix; compute daily support/resistance and extreme levels.
• During trading:
  • Monitor execution quality, order book anomalies, P&L per position, risk limits, open interest and liquidation clustering.
• Post-session:
  • Keep a trade journal (automatic or manual); compare plan vs outcome; prepare scenarios for next day.
• Event handling:
  • Reduce position size 50%–75% around macro releases.
  • Avoid new entries 30 minutes before/after major releases.
  • Macro shocks can disrupt mean reversion for 1–3 days.

Macro factors and calendar notes

• Macro drivers affecting mean-reversion behavior:
  • Inflation, unemployment, GDP, PMI, central bank decisions, QE, central bank speeches, FOMC.
• Calendar specifics:
  • CPI typically on the first Friday (mentioned).
  • FOMC on a regular schedule (presenter referenced second Wednesday; note transcript inconsistency).
  • Options expiries: weekly and monthly expiries, last Friday of month noted for abnormal moves.
• VIX and bond yields:
  • VIX up → stronger mean reversion in risk assets (general observation).
  • Bond yields affect risk assets and mean-reversion dynamics.

Execution / practical toolset

• Terminals / screeners (transcript contains many name variants):
  • “Spreadfighter” / “Spotfighter” / “Sportfighter” / “Spratfighter” / “Mediumfighter” — these names appear as modules; likely transcript typos for one or more product names.
    • Module features described: stationarity (ADF) screener, liquidity / 2% range liquidity indicator, net AI indicator (order imbalance / limit-player placing), liquidation map, big trades, power/open interest indicators.
• Other tools:
  • TradingView for charting and indicators.
  • Cluster search, volume profile, imbalance detectors.
• Automation:
  • Planned mean-reversion robots with customizable parameters and built-in stops/take-profits.

Key numeric thresholds / rules (explicit)

• Volume (24h): >100 million.
• Transactions/day: >100k.
• Bid-ask spread: <0.1% of price.
• Order book depth: $100k–$200k within ±2% of price.
• Daily volatility preferred: 2%–5%.
• Funding rate: avoid extremes >0.1%.
• Max averaging: do not average beyond 3 standard deviations.
• Statistical stop threshold example: 4 standard deviations.
• ATR stop multiplier: ≈ 3–3.4.
• VWMA main period example: 240.
• MA periods referenced: 10–20 (fast), 50–100 (standard), 200+ (long).
• Position sizing: base 0.5%–1% per trade; max 3–4 concurrent positions.
• Drawdown tolerance: retail 15%–25%; institutional <10%.

Main risks and cautions

• Stationarity risk: mean reversion only valid on stationary series; non-stationary series produce spurious signals.
• Structural shifts / regime changes: largest source of loss when an asset stops reverting to the mean.
• Liquidity & slippage: large trades can suffer slippage; order-book depth may be insufficient.
• Trend continuation: a deviation may be the start of a trend, not a reversion.
• News / manipulation: assets sensitive to news or manipulation are unreliable for mean reversion.
• Shorts: riskier due to stronger impulses and asymmetric losses.
• Execution risk: order-book imbalances and large opposing limit orders can severely affect trades.

Operational advice / recommended process to start

• Backtest for a substantial period: at least ~2 weeks in-sample/out-of-sample; a 1-year period recommended for stress testing.
• Start small: 0.3%–0.5% position sizing initially to learn mechanics, then scale up as you collect reliable stats.
• Keep a disciplined checklist and trade journal; automate checks where possible.
• Limit concurrent exposure to a handful of assets (3–4) and diversify across assets to smooth returns.

Disclaimer: The presenter framed the content as personal testing and results (claims are self-reported and tied to a short testing period with admitted mistakes). This summary reflects the transcript material and the presenter’s stated rules and thresholds; it is not financial advice.

Presenters / sources

• Presenter: unnamed individual (signed “M.” in the transcript).
• Platforms / tools mentioned: variants of a “fighter” terminal (see Execution section), TradingView, built-in stationarity screeners.

Category

Finance

---

Share this summary

Share on X/Twitter Share on Facebook Share on LinkedIn Share on Reddit Share on WhatsApp Share on Telegram

---

Is the summary off?

If you think the summary is inaccurate, you can reprocess it with the latest model.

Reprocess summary

Video