DURDENBTC — CANONICAL CONTENT (PLAIN TEXT) ============================================ Generated: 2026-06-06 Source: https://durdenbtc.com Index: https://durdenbtc.com/llms.txt Sitemap: https://durdenbtc.com/sitemap.xml This file concatenates the canonical content of durdenbtc.com — the thesis, the three core trading systems (MRE v06, Arsenal BTC, The 8th Rule), and the six pillar guides — as plain text. Suitable for LLM ingestion and citation. Performance figures cited reflect backtested results. The site is published by DurdenBTC (X: @DurdenBTC). Anonymous handle, but the engineering background is disclosed on https://durdenbtc.com/about (20+ yrs SWE, currently a Senior Engineering Manager at a Fortune-1000 SaaS firm). All systems run on public data (FRED, on-chain feeds, exchange APIs). Table of contents: 1. Survival Is Alpha — The Thesis — https://durdenbtc.com/thesis 2. Macro Regime Engine (MRE v06) — https://durdenbtc.com/macro_regime_engine 3. Arsenal Bitcoin Trend System — https://durdenbtc.com/arsenal_btc 4. The 8th Rule — https://durdenbtc.com/the_8th_rule 5. Guide — Macro Regime Investing for Bitcoin — https://durdenbtc.com/guide/macro-regime-bitcoin 6. Guide — Bitcoin Buy & Sell Signals — https://durdenbtc.com/guide/bitcoin-signals 7. Guide — Bitcoin Drawdown Management — https://durdenbtc.com/guide/bitcoin-drawdown-management 8. Guide — Global M2 & Bitcoin — https://durdenbtc.com/guide/global-m2-bitcoin 9. Guide — Bitcoin & the ISM PMI Cycle — https://durdenbtc.com/guide/bitcoin-ism 10. Guide — Trend Following on Bitcoin — https://durdenbtc.com/guide/trend-following-bitcoin ============================================ ============================================ 01. SURVIVAL IS ALPHA — THE THESIS URL: https://durdenbtc.com/thesis ============================================ DurdenBTC · The Thesis · Survival Is Alpha # Truncate the left tail. Compound on a higher base. The mathematical case for why avoiding drawdowns and compounding at higher portfolio values beats riding the full wave — even inside a secular uptrend. EngineMRE v06 AssetSPY · BTC SPY MaxDD−14% BTC MaxDD−34% Buy & Hold−56% / −83% Long form Read ≈ 12 min Editor's note: A −50% drawdown requires +100% to break even; a −83% drawdown requires +488%. Skip the deepest holes and compound from a higher base — that asymmetry is the entire game. Truncate the left tail → Compound on a higher base → More terminal wealth ## TL;DR - The thesis: truncate the left tail of the return distribution, compound on a higher base, generate more terminal wealth. Drawdown avoidance is the compounding strategy. - The math: a −50% drawdown requires +100% to break even. A −83% drawdown requires +488%. Two portfolios with identical average arithmetic returns can have wildly different terminal values once you account for volatility drag. - The mechanism: MRE v06 for SPY, Arsenal BTC for Bitcoin. Both are systematic, rule-based engines that step aside when the macro environment turns hostile — before the waterfall phase, not after. - Track record: MRE v06 max DD −14% vs SPY's −56%. Arsenal BTC max DD −34% vs BTC's −83%. Catching most of the upside while cutting drawdown by more than half over multiple cycles. - The edge is measurement, not prediction. No tops, no bottoms — just data, models, and stepping aside. 01 — The Problem ## Volatility Is Not Free There is a persistent myth in investing especially in Bitcoin and crypto, that volatility is simply the "price of admission" to an exponential asset. That you must stomach the drawdowns to earn the returns. That any attempt to manage risk is market-timing, and market-timing doesn't work. This is mathematically wrong. Volatility has a direct, quantifiable cost. It's called volatility drag, and it silently destroys wealth by forcing you to compound from lower values after every drawdown. The core arithmetic is simple: losses are asymmetric. A −50% drawdown requires a +100% gain just to break even. A −75% drawdown, which Bitcoin has experienced three times.. requires a +300% recovery. Every dollar that survives a drawdown is compounding from lower nominal values, not from a peak. Drawdown Recovery Required Asymmetry Ratio −10% +11.1% 1.1× −20% +25.0% 1.3× −30% +42.9% 1.4× −50% +100.0% 2.0× −75% +300.0% 4.0× −85% +566.7% 6.7× The relationship is non-linear. Below −50%, the recovery math becomes brutal. This isn't a matter of opinion or risk tolerance, it's compounding arithmetic. The Geometric Mean Problem Two portfolios can have the same average arithmetic return and wildly different terminal values. A portfolio that returns +50%, −40%, +50%, −40% has an average return of +5%/year.. but the geometric return (what you actually earn) is −5.1%/year. The volatility consumes the entire return and then some. This is volatility drag. 02 — The Thesis ## Compound at a Higher Base The thesis is not "time the market." The thesis is not "sell the top and buy the bottom." The thesis is structurally narrower and mathematically grounded: Core Thesis If you can remove the fat left tail from the return distribution, even at the cost of surrendering the tip-top of the right tail, the sequence of returns means you will compound at higher portfolio values, and higher portfolio values at every stage of compounding equals more terminal wealth. This is the key insight 95% of people miss. They see a risk-managed strategy underperform during a vertical move and conclude it "doesn't work." But they're measuring the wrong thing. They're measuring return per cycle when they should be measuring terminal portfolio value across all cycles. ### An Illustrative Example Risk-Managed $336K $100K → +40% → +50% → +60% Steady gains, no drawdowns. Compounding on a growing base. Unmanaged $187K $100K → +150% → −50% → +50% Higher peaks, deeper valleys. Compounding from a crater. Both portfolios are invested in the same asset class. The unmanaged portfolio actually had a higher peak ($250K vs. $210K). But the −50% drawdown forced it to compound from $125K instead of $210K. By Year 3, the risk-managed portfolio is ahead by $149K.. a 79% advantage.. not because it captured more upside, but because it avoided compounding from a devastated base. "The sequence of returns matters more than the average of returns." — The Volatility Drag Axiom This is the mathematical reality that the "just buy & hold" crowd ignores. They assume path-independence.. that only the destination matters. But compounding is path-dependent. The order in which returns arrive determines the terminal value. A −75% early in the sequence is catastrophically more damaging than the same −75% late, because it destroys the base upon which all future gains compound. 03 — The Rebuttal ## The "Secular Uptrend" Fallacy The most common objection: "Bitcoin is in a secular exponential uptrend. Why would you try to manage around drawdowns when the trend will eventually bail you out?" Let's examine this carefully. ### What Does "Secular" Actually Mean? A secular trend is typically defined as a long-duration market movement lasting 10 to 40 years, driven by structural forces (demographics, technology adoption, monetary regimes). Bitcoin has been in a secular uptrend for approximately 15 years (at time of this writing). That places us somewhere in the middle of the typical secular window.. not at the beginning, where you can assume infinite runway, and not at the end. Saying "we're in a secular uptrend" is true. But it's not an argument against drawdown management. It's a description of the environment in which you should be managing drawdowns. Secular uptrends don't eliminate volatility, they contain volatility. The question is: do you let that volatility compound against you, or do you manage it? ### The Drawdowns Inside Bitcoin's "Secular Uptrend" Year Drawdown Recovery Required Time to New ATH 2011 −93% +1,329% ~2 years 2014–15 −85% +567% ~3 years 2018 −84% +525% ~3 years 2022 −77% +335% ~2 years Every one of these occurred inside the secular uptrend. The trend didn't prevent them. And each one forced holders to compound from a devastated base for 2–3 years just to get back to where they started. The Real Cost A dollar that survived the 2022 drawdown needed to 4.35× itself just to break even. A dollar in a risk-managed portfolio that avoided the drawdown was compounding on its full value the entire time. After the recovery, the managed dollar is still ahead.. permanently.. because it never fell behind. ### Position Sizing Advice Debunked The typical advice: "If you can't handle the volatility, reduce your position size." This sounds sensible, but think about what it actually implies. If you size down your position enough to emotionally tolerate a −75% drawdown, you've also sized down your exposure to the upside by the exact same proportion. You've traded volatility drag for opportunity cost. The wealth destruction is just happening through a different mechanism. The alternative: keep the position meaningful, and manage the tail risk directly. This lets you participate in the secular trend with real capital while shaving off the catastrophic left-tail events that destroy compounding. 04 — The Implementation ## Two Systems, One Philosophy The philosophy is asset-agnostic. The implementation is asset-specific. Two purpose-built systems apply the "Survival Is Alpha" thesis across the two primary asset exposures: ### For Equities (S&P 500): MRE v06 — The Macro Regime Engine A multi-asset voting system that polls 26 global assets across equities, breakevens, duration, FX, the curve, and credit. Each asset calculates a volatility-adjusted trend signal and casts votes into four macro regime buckets: Goldilocks, Reflation, Inflation, and Deflation. The system makes a binary decision: is the global macroeconomic environment friendly to equities? If yes, hold SPY at 100%. If no, hold cash. No discretion. No "nibbling." No hope trades. Symmetric hysteresis prevents whipsaw; an Arsenal SPY veto layer catches the cases where macro looks fine but equities are imploding on their own. MRE v06 · SPY · Regime overlay with risk-on/off transitions SPY Backtest — 23 Years (2003–2026) MRE v06: +925.8% · CAGR: 10.52% · Sharpe: 0.96 · Max DD: −16.22% Buy & Hold SPY: +679.4% · CAGR: 9.22% · Max DD: −56.47% 5,867 trading days. 26 voters. Seven independent forward tests, zero failures. The strategy outperformed buy-and-hold while spending 30% of trading days in cash.. precisely the days that produced the buy-and-hold drawdowns. ### How It Dodged Every Major Crash Sep 2000 → Dec 2002 Dot-com crash. Engine exited September 2000. S&P fell ~49% over the next two years. Engine re-entered near the bottom in December 2002. Dec 2007 → Apr 2009 Global Financial Crisis. Engine exited December 2007. S&P fell −56%. Engine went flat through the waterfall, re-entered April 2009. Feb 2020 COVID crash. Engine exited before the crash.. February 20, 2020. S&P fell −34% over the next month. Engine re-entered May 2020. Nov 2021 → Jan 2023 2022 bear market. Engine exited November 2021. S&P fell ~25%. Engine re-entered January 2023, early in the recovery. The engine didn't predict these crashes. It measured that the macroeconomic environment had turned hostile: credit spreads widening, volatility spiking, growth currencies collapsing, safety assets rallying.. and stepped aside before the waterfall phase. The edge isn't prediction. It's measurement. ### For Bitcoin: Arsenal BTC — The Bitcoin Trend System A three-state trend system built on VAMS.. a composite of an SMA-crossover trend filter and a three-layer volatility regime filter (calm-bull, calm-bear, crisis). Output is a single vote: +2 BULLISH (100% long), 0 NEUTRAL (50% long), or −2 BEARISH (0%, cash). No leverage, no discretion. Bitcoin requires a different system than equities because its return distribution is fundamentally different: fatter tails, higher kurtosis, and vol spikes that reverse in a week. The vol-regime layer is a kill-switch that fires before the trend signal has a chance to give back gains.. which is exactly when a naive trend system gives back a year of upside in a flash event. Arsenal BTC · BTC-USD spot · VAMS trend + three-layer vol regime Bitcoin Backtest — 10.6 Years (2015–2026) Arsenal BTC: +78,930% · CAGR: 87.75% · Sharpe: 2.32 · Max DD: −34.46% Buy & Hold BTC: +32,563% · CAGR: 72.73% · Max DD: −83.40% Captured most of Bitcoin's upside while cutting drawdown by more than half. Three of four forward tests pass; the OOS walk-forward verdict is honestly disclosed as WEAK on the forward-testing page. One philosophy across both engines. MRE v06 reads breadth across 26 macro voters and outputs a binary risk-on/off SPY signal. Arsenal BTC reads price-and-vol on a single asset and outputs a three-state position (100% / 50% / 0%). Different problems, different tools, identical disciplines: daily-close execution, full backtest disclosure, and forward-test honesty.. including the WEAK results when they happen. The 8th Rule (our older, more aggressive Bitcoin system) is still available to subscribers who prefer it, but Arsenal BTC is the flagship. 05 — The Framework ## The Operating Commandments 01 #### Cash Is an Asset Class In periods of chop, regime conflict, or hostile macro, cash (via $SGOV or $USFR) is the active position. It preserves purchasing power and critically preserves the compounding base from which the next move compounds. 02 #### No Hope Trades If the system says Neutral or Bearish, the position is zero. No "nibbling," no "DCA into weakness," no "it looks like a bottom." Buy strength or hold cash. Drawdowns are not buying opportunities: they are the enemy of compounding. 03 #### Respect the Regime Never fight the macro tide. If the regime says Inflation or Deflation, do not hold a Goldilocks portfolio because you "like" the stocks. The engine's 20+ asset consensus overrides individual conviction. 04 #### Accept the Right-Tail Cost Any risk management framework will clip some upside during vertical rallies. This is the trade. You surrender the top 5% of the right tail to remove the bottom 20% of the left tail. The compounding math makes this trade overwhelmingly positive over time. 05 #### Sequence Over Average Never evaluate performance by average return. Evaluate by terminal wealth. Two portfolios with identical average returns can diverge by hundreds of percent based on the order of those returns. The sequence is everything. 06 — The Bottom Line ## Why This Wins The entire thesis reduces to one sentence: "You don't need to capture more upside per cycle. You need to avoid the catastrophic losses that destroy compounding." MRE v06's +925.8% vs. buy-and-hold SPY's +679.4% over 23 years is almost entirely explained by dodging the major drawdowns.. 2008, 2020, 2022, 2025. The system didn't capture more upside than the market per bull cycle. It just never compounded from a crater. A −16.22% max drawdown vs. buy-and-hold's −56.47% means the system was compounding from near its peak value after every single macro shock. Every dollar in the strategy had a dramatically higher starting base entering the next bull cycle. Over 23 years that differential.. that permanently higher base.. compounds into a +246 percentage-point outperformance gap, with a Sharpe nearly twice that of buy-and-hold. The Bitcoin side tells the same story with bigger numbers. Arsenal BTC's −34.46% max drawdown vs. buy-and-hold BTC's −83.40% means the system enters every new bull cycle compounding from less than half the buy-and-hold drawdown. That asymmetry is the entire reason its 78,930% total return blows past buy-and-hold's 32,563%.. despite spending material time in cash. This isn't market timing. This isn't mean reversion. This is drawdown management as a compounding strategy. Truncate the left tail, accept a modest right-tail cost, and let the math do the rest. In One Line Survival isn't a consolation prize. It's the highest-alpha strategy in existence because every dollar that survives compounds forever, and every dollar that drowns in a drawdown compounds from zero. Join DurdenBTC on Substack Free macro regime updates, strategy breakdowns & thesis posts. ============================================ 02. MACRO REGIME ENGINE (MRE V06) URL: https://durdenbtc.com/macro_regime_engine ============================================ MRE v06 — The Macro Regime Engine — DurdenBTC - - - - - - - - - - - DurdenBTC · Proprietary System · MRE v06 # 26 voters. 4 regimes. One binary decision. The Macro Regime Engine classifies the global macro environment into Goldilocks, Reflation, Inflation, or Deflation — and outputs one call: own SPY or hold cash. 23 years of backtests, seven independent forward tests, zero failures. CAGR+10.86% Sharpe1.01 Max DD−14.35% History23 yrs Forward Tests7 / 7 SPY v06 Editor's note: Built once, audited every night. The engine reads off fully closed daily data, never repaints, and matches its test framework byte-for-byte across 5,867 trading days. ## TL;DR - What this is: MRE v06 is a 26-voter macro regime engine for SPY. Classifies the global macro environment into Goldilocks, Reflation, Inflation, or Deflation, then outputs a binary risk-on / risk-off call. - Backtest (Jan 2003 – May 2026, 5,867 trading days): +10.86% CAGR, 1.007 Sharpe, –14.35% max drawdown vs SPY buy-and-hold +9.30% / 0.499 Sharpe / –56.47% drawdown. - Forward-tested: 7 / 7 independent tests passed. The OOS walk-forward test Sharpe (1.219) came in higher than the training Sharpe (0.754). - Does not repaint. Signal computed once after the daily close. Test 5 equivalence guard runs at startup and matches the test framework byte-for-byte. - Free to read; live signals + members dashboard at $7 / month. 01 — What It Does ## Two States. That's It. MRE v06 classifies today's global macroeconomic environment into one of four regimes.. Goldilocks, Reflation, Inflation, or Deflation.. and condenses that classification into a single decision: RISK-ON (own SPY) or RISK-OFF (hold cash). The classification draws on 26 independent voters covering equities, commodities, breakevens, FX, the yield curve, and credit. The decision is published once per day, before markets open. This page explains what's under the hood.. enough that you can decide whether to trust the signal.. without giving away how it's built. The engine is proprietary. I do publish the indicator output daily but I don't share the code. I'm also not going to ask you to trust a black box on faith. Risk-ON Goldilocks / Reflation Growth is positive. Macro is friendly. Own SPY at 100%.. the wind is at your back. Risk-OFF Inflation / Deflation Macro environment is hostile. Hold cash. Preserve capital until conditions improve. These four regimes are the same growth/inflation quadrant the largest institutional allocators in the world use to size risk. The difference is that MRE v06 measures it quantitatively, in real time, across 26 voters.. instead of reading press releases and making a discretionary call. 02 — How It Works ## The Thesis, Not the Code MRE v06 doesn't look at a single chart. It doesn't use a moving-average crossover on the S&P 500. It doesn't care about RSI, MACD, or any single-asset technical applied to the index it's deciding to own. Instead, it listens to the entire global financial system, then runs the consensus through a five-stage pipeline. When global equities rally, risk currencies strengthen, breakevens expand, credit spreads tighten, and the curve steepens all at the same time.. that's not a coincidence. That's 26 independent markets all saying "the macro environment is favorable to risk." Stage 1 reads the votes. Stage 2 classifies the regime. Stage 3 confirms the change isn't noise. Stage 4 catches the edge cases (the "bad news is good news" rallies). Stage 5 keeps a final risk-off veto in reserve. Below: each stage in plain English. 01 ### 26 Voters Across Four Asset Categories Roughly 10 growth voters (equity indices, growth-sensitive commodities), ~7 inflation voters (breakevens, oil, copper, silver), ~5 duration voters (2y / 10y yields, the curve), and ~4 FX voters (DXY plus the major risk crosses). Each voter computes a volatility-adjusted trend signal and casts a vote for one of the four regimes. Volatility-adjusted means a 1% move in DXY and a 5% move in copper carry roughly equal weight.. no single voter can dominate by virtue of being a more volatile asset. 02 ### 4 Regimes (Plus a 5th "Chop" State) Goldilocks = growth up, inflation down. Reflation = growth up, inflation up. Inflation = growth down, inflation up. Deflation = growth down, inflation down. When voters disagree enough that no regime clears the conviction floor, the engine reports a fifth "Chop" state and stays in the last confirmed regime. This is by design.. it's how the system avoids whipsaw on indecisive days. 03 ### Symmetric Hysteresis (7 Bars Each Way) A regime change requires seven consecutive bars of evidence before it's confirmed on tape.. both when going risk-on AND when going risk-off. Earlier versions used asymmetric hysteresis (faster to exit than enter); v06 is symmetric and cleaner. Symmetric hysteresis means the engine doesn't panic-exit on a single bad week and doesn't chase a one-week rally before the conditions confirm. 04 ### Growth Override + Min-Dwell If macro voters say "deflation" but growth-sensitive markets are overwhelmingly bullish.. the "bad news is good news" case that captured the late-2023 / early-2024 rate-cut rally.. the override flips to risk-on. Once any regime is established, a min-dwell rule keeps the engine inside it for at least N days to prevent flip-flop. The override and dwell rules together let the engine respect macro when it matters and ignore it when growth assets are calling the bluff. 05 ### Arsenal SPY Veto (Final Layer) Layered last and applied unconditionally: an independently-tuned Arsenal-style SPY engine acts as the final risk-off veto. If MRE's macro stack says risk-on but the Arsenal SPY engine reads 0% (deep equity bear), the veto flips the published signal to risk-off. The veto bypasses dwell lock. The two engines are independently parameterised and validated.. they're not the same signal applied twice. The Arsenal SPY veto is what catches the "macro looks fine but equities are imploding" case that broad-breadth voting can miss. A look under the hood.. a sample of the 26 voters MRE v06 polls across asset classes 03 — The Track Record ## 23 Years. 5,867 Trading Days. MRE v06 has been backtested on SPY from 2003-01-02 through 2026-05-05.. 5,867 trading days, 23.34 years. That window includes the 2003 reflation, the 2008 GFC, the 2010 / 2011 / 2015 / 2016 corrections, the 2020 COVID crash, the 2022 rate-shock drawdown, the 2023 banking stress, and the 2024-2026 expansion. The system was held frozen for the full backtest.. no parameters were tuned to in-sample data. +10.86% Annualised Return 1.01 Sharpe Ratio −14.35% Max Drawdown +925.8% Total Return Metric MRE v06 Buy & Hold SPY Annualised Return +10.86% +9.30% Sharpe Ratio 1.007 0.499 Sortino Ratio 1.093 — Max Drawdown −14.35% −56.47% Calmar Ratio 0.757 — Total Return +925.8% +679.4% Trades per Year 2.4 — Time Long 70.5% of days 100% Voter Assets 26 — The headline number isn't the +925%.. buy-and-hold gets +679% just by sitting in SPY. The headline number is the Sharpe gap (1.01 vs 0.50) and the drawdown gap (−14% vs −56%). MRE v06 captured most of the upside while spending material time in cash.. exactly where the avoided drawdowns come from. MRE v06 equity curve with regime overlay vs Buy & Hold SPY (2003.. 2026) 04 — Forward-Test Suite ## Seven Tests. Zero Failures. A backtest is necessary but not sufficient. The hard question isn't "does the system look good on the data it was tuned on".. it's "does the system hold up against everything I can throw at it that isn't the original training set." MRE v06 ships with a seven-test suite that probes seven different robustness questions independently. Every test passed. Test What It Asks Result Verdict 1. OOS Walk-Forward Does it hold up on data the engine has never seen? Train Sharpe 0.754 → Test Sharpe 1.219. The test period's Sharpe was 62% higher than the training period's. Pass 2. Monte Carlo (Block) Is the return sequence just lucky? Actual Sharpe lands at the 50.9th percentile of 10,000 block-bootstraps. Median, not lucky. Pass 3. Regime & Crisis Stress Does it behave sanely when markets break? Beat B&H in 6 of 10 named crises (2008 GFC +44.5% relative outperformance, 2020 COVID +26.7%). Pass 4. Parameter Sensitivity Does it collapse if a knob is one step off? ±1 step from default retains 96.1%+ of baseline Sharpe across every parameter probed. Pass 5. New Input Impact Are we missing any high-impact signals? No candidate input (DFII10, IG_OAS, T10Y3M) improves the composite score meaningfully. Pass 6. Synthetic Price (SPS) Would it work on a statistically-similar but different SPY history? MODE A (block-bootstrap) p5 Sharpe +0.088; MODE B (Gaussian GBM) p5 Sharpe +0.069. Pass 7. Rolling OOS (1y & 3y) Is performance concentrated in a few big years? 16 of 21 rolling years (76%) post a positive Sharpe; zero catastrophic years; 3-year rolling median Sharpe of 1.00. Pass The test that matters most for credibility is Test 1 (OOS walk-forward). I split the data in two: the engine was tuned on 2003-2015 and tested on 2016-2026. If the parameters were curve-fit to history, you'd expect the test period's Sharpe to collapse. It didn't.. it improved by 62%. The frozen v06 parameters handled the COVID crash, the 2022 rate shock, and the 2023 banking stress better than they handled the relatively-calm 2003-2015 training era. Full per-test detail and the underlying methodology lives on the forward-testing page. 05 — Crisis History ## Drawdown Avoidance Is the Edge MRE v06 doesn't need to outperform buy-and-hold on every up cycle. It needs to dodge the catastrophic drawdowns that destroy compounding. If you hold through a 56% drawdown, you need a 127% gain just to get back to even. If you hold through −14% (MRE v06's worst), you need 19% to recover. The engine enters each new bull cycle compounding from near its peak instead of digging out of a hole.. that asymmetry is most of the edge. 2003.. Reflation Captured the post-bust recovery. Engine went risk-on as growth and inflation voters confirmed the cycle had turned. Dec 2007 → Apr 2009 Global Financial Crisis. SPY fell −56%. MRE v06 was flat through the worst of the waterfall.. the breadth signal flipped before the price collapse showed up on equities. +44.5% relative outperformance. 2011.. Eurozone Crisis Mid-cycle correction. Engine reduced exposure as the breadth signal weakened, then re-engaged once macro stabilised. 2015 / 2016.. China & Oil Two consecutive corrections (China devaluation, then oil collapse). Engine sat out both and re-entered for the post-Feb-2016 rally. Q4 2018.. Vol Spike Equities corrected sharply on rate hikes plus US-China trade. Engine flipped to risk-off and avoided the December drawdown. Feb 2020.. COVID Engine exited the week of February 20, 2020.. before the crash began. SPY fell −34% over the following month. +26.7% relative outperformance. Re-entered late May as the recovery broadened across asset classes. Nov 2021 → Jan 2023 2022 rate-shock bear. Engine exited near the November 2021 top. SPY fell ~25% over the following year. Re-entered early 2023 as conditions improved. Mar 2023.. Bank Failures Brief flight to safety on regional bank failures. The growth-override caught the late-March rally as growth voters held bullish despite credit stress. 2024.. AI Cycle Engine stayed long throughout. The breadth signal correctly identified that the AI capex story was a real growth tailwind, not a narrative-driven rally. 2025 / 2026.. Tariff Wobble Brief risk-off in early 2025 on tariff escalation. Engine re-engaged in mid-2025 as breadth recovered. Currently risk-on (Reflation regime). Each major exit happened before the worst of the drawdown.. not during it, not after it. That's not luck. It's what happens when the signal is driven by cross-asset breadth: macro deterioration shows up across markets simultaneously, before the equity waterfall begins. The seven-bar hysteresis means the engine waits for confirmation, but breadth confirmation arrives well before single-asset technicals do. 06 — Why It Works ## The Honest Answer 01 ### Breadth Is Hard to Fake A single stock can be manipulated. A single index can be propped up. But the simultaneous behaviour of 26 voters across equities, breakevens, FX, the curve, and credit is real. It takes a genuine macroeconomic shift to flip 26 markets at once, and that's exactly the moment when an exit is most valuable. 02 ### Losses Are Shallow The largest single drawdown in 23 years is −14.35%. The reason is structural, not lucky: the engine exits on confirmed regime breaks, not on stop-outs. There's no "wait for the bottom to be in" logic. When breadth flips, the engine flips. Losses don't have time to compound. 03 ### Drawdown Avoidance Compounds The Sharpe gap (0.96 vs 0.49) is more important than the return gap. A higher Sharpe means you can size the same returns with less psychological tax. The engine spent 30% of trading days in cash.. that 30% is where the −56% buy-and-hold drawdowns happened. 04 ### Holdouts Are Long, Exits Are Decisive Trades per year: 2.9. The engine commits to a regime once it's confirmed and stays committed until breadth materially deteriorates. It doesn't whipsaw. The hysteresis isn't a bug.. it's the feature that lets winners run for months while still cutting losers in days. 07 — What Could Go Wrong ## Read This I'd rather you know the failure modes up front than discover them yourself. It Does Not Predict Crashes It identifies that macro has turned hostile and steps aside. There's a difference. The system didn't "call" COVID.. it measured that cross-asset breadth deteriorated in the third week of February 2020 and exited. The crash happened after the exit. The engine responds to the present, not the future. V-Shaped Recoveries Cost The seven-bar hysteresis that prevents whipsaw also delays re-entry. A fast V-shaped recovery (think: the late-March 2020 bottom) means the engine misses the first few weeks of upside while it waits for breadth to confirm. The crisis-stress test (Test 3) makes this visible: the engine wins on extended drawdowns and loses on one-week vol shocks. Trades Per Year Is Low 2.9 trades per year sounds modest because it is. If you're looking for daily / weekly action, this is the wrong tool. MRE v06 changes its mind a few times per year, sometimes only once, and the value is precisely in not changing more often than the macro changes. Backtests Always Look Better Than Reality The backtest assumes daily-close execution at the published price. Slippage, gap risk, and the messy reality of trading around holidays will introduce friction the backtest doesn't see. Treat the published numbers as the ceiling, not the floor. 08 — What You're Getting ## What Subscribers Receive ✓ ### The Live MRE v06 Signal Risk-on or risk-off, refreshed daily before market open. Plus the four-regime classification (Goldilocks / Reflation / Inflation / Deflation), the full 26-voter vote breakdown, and the Arsenal SPY veto state.. all on the members dashboard. ✓ ### Immediate Substack Alerts on Every Signal Change The dashboard is for monitoring. The Substack e-mail is the source of record.. every regime flip, every veto, every override is announced via e-mail with my analysis of what shifted and why. ✓ ### The Full Forward-Test Suite + Backtest History All seven forward tests, including methodology and results, plus the full 23-year equity curve and trade history. If you want to verify the engine yourself before trusting it, the dashboard gives you everything you need. ✗ ### What You're NOT Getting The voter list, the regime-classification thresholds, the hysteresis parameters, or the Arsenal SPY veto code. MRE v06 is proprietary. You get the signal and the analysis; the engine itself stays in-house. ### How to Use the Signal Risk-ON Own SPY at 100% SPY, $SPX, ES futures, broad-market ETFs.. the engine says the macro wind is at your back. Position size accordingly. Risk-OFF Cash or Hedge The macro environment is hostile. Preserve capital using $SGOV, $USFR or similar methods. Don't fight the regime. Signal changes are infrequent.. about three per year on average. Don't check every hour. When the signal flips, the e-mail will arrive. Combine MRE v06 with your own analysis.. the engine is one input, the strongest macro input I have, but it's not omniscient. For asset-level Bitcoin trend timing, see Arsenal BTC. 09 — Final Word ## I Don't Sell Hope I built MRE v06 to answer one question: is the world friendly to risk assets right now, or not? I didn't want to guess. I didn't want to rely on vibes, narratives, or X consensus on what the Fed is going to do. I wanted a system that measures the answer quantitatively, across enough independent data streams that the conclusion is robust.. and then tells me in plain terms whether to be in or out. Across 23 years of backtest, the answer was right enough that the Sharpe was >1.0, the max drawdown was −14%, and the engine outperformed buy-and-hold by 250+ percentage points cumulatively while spending almost a third of trading days in cash. Seven independent forward tests all passed. I trade the live signal with my own capital. I publish it because $7 a month is a fair price for a system I spent over a year building. If the edge ever disappears I'll tell you. If it needs fixing, I'll fix it and explain what changed. The Bottom Line A Decision-Support Tool, Not a Money Printer MRE v06 is a quantitative macro framework with a 23-year track record and seven forward tests behind it. It's not a money printer and I'll never sell it as one. But it's the best macro signal I've ever built, I trade it every day, and it's yours for less than the cost of a single bad trade. .. Durden out. Join DurdenBTC on Substack MRE v06 regime signals, Arsenal engines, full members dashboard — starting at $7/mo. Last updated: April 2026. Performance figures reflect backtested results on SPY from 2003-01-02 through 2026-04-28 (5,867 trading days, 23.32 years). Forward-test methodology and full results live on the forward-testing page. This content is for educational and informational purposes only. It does not constitute financial advice, investment advice, or a recommendation to buy or sell any asset. Trading equities involves substantial risk of loss. Past performance, whether backtested or live, does not guarantee future results. Backtested performance has inherent limitations: it is designed with the benefit of hindsight, does not reflect actual trading, and does not account for all factors that may affect real-world execution. The author is not a licensed financial advisor. Always do your own research and consult a qualified financial professional before making investment decisions. You are solely responsible for your own trading decisions.