Dev Blog

Daily analyst notes on our AI strategies, the market, and sector trends.

• Strategy

  PanicFade v4: Buying the Overreaction, Patiently

  Sep 4, 2026 · PanicFade v4 is a live contrarian strategy that fades sentiment-driven selloffs in oversold large caps, backed by a clean four-fold validation but so far sitting mostly in cash with only one recent trade.

• Strategy

  Momentum SMA20 Rider: Riding Trends With a 37% Win Rate — and Why That's Fine

  Sep 4, 2026 · The live SMA20 Rider strategy has returned 81% in backtest with a defensible cross-validation profile, but its low win rate, high turnover, and one dead out-of-sample fold demand a clear-eyed read. Here's the balanced picture.

• Strategy

  channel-pullback: A Mean-Reversion Strategy That Backtests Positive but Fails Validation

  Sep 4, 2026 · The live channel-pullback strategy posts a 7.62% backtest return but flunks our walk-forward validation, with one brutal fold and a low deflated Sharpe undercutting the headline number. Recent paper trading has drifted the account below its $10k start.

• Strategy

  Donchian-Breakout: Textbook Trend-Following, Failing Its Own Validation Gate

  Sep 4, 2026 · The 20-day breakout strategy posts a modest 6.95% backtest return but flunks robust validation, and its live book has gone quiet — five straight scheduled runs rejected every order.

• Strategy

  bollinger-reversion: A Textbook Mean-Reversion Bet That Backtests Well and Validates Poorly

  Sep 4, 2026 · The strategy's full-sample backtest looks respectable, but walk-forward validation shows out-of-sample returns collapsing toward zero — and its live paper account has rejected every order for a week straight.

• Strategy

  Dual-Momentum Goes Quiet: Strong Trends, Failed Validation

  Sep 4, 2026 · The live dual-momentum strategy posted a 23.5% backtest return but failed its validation gate, and it has sat entirely in cash for the past week with zero executed trades.

• Strategy

  news-sentiment: A Sound Thesis Still Starved of Signal

  Sep 4, 2026 · The news-sentiment strategy runs live but has traded only twice in its backtest, and its cross-validation failed with just one of four folds positive. A promising idea whose edge remains unproven.

• Strategy

  Mean-Reversion: A 70% Win Rate That Validation Won't Sign Off On

  Sep 4, 2026 · The RSI-based mean-reversion strategy posts a strong 14.73% backtest return and a 70.59% win rate, but walk-forward validation fails on a negative out-of-sample fold. Live activity has been idle for months, with no signals firing since a lone WMT buy in May.

• Strategy

  momentum-code: Strong Backtest, Stalled Live — Why the Validation Gate Held

  Sep 4, 2026 · The live momentum strategy posted a 19.76% backtest return with four positive walk-forward folds, yet its validation gate failed and its live book has gone six sessions without executing a trade. We unpack the gap between the numbers and the reality.

• Strategy

  RSI Snap-Back: Solid Backtest, Quiet Live Tape

  Sep 4, 2026 · The RSI Snap-Back strategy pairs a respectable 20.95% backtested return with a two-thirds win rate, but a 23.7% drawdown, absent formal validation, and a week of zero-trade sessions temper the enthusiasm.

• Sector

  Reading the Room: What the Tracked Universe's Sector Mix Tells Us

  Sep 4, 2026 · A look at how the ~2,200 names in Headmars' tracked universe break down by sector, where the concentration sits, and why the headline market-cap figures deserve a skeptical eye.

• Market News

  Market Roundup — September 4, 2026: A Neutral Tape With a Bullish Lean

  Sep 4, 2026 · Today's flow skewed heavily neutral but with bullish headlines outnumbering bearish nearly two-to-one, as valuation calls dominated coverage of NVDA, GOOGL, AAPL and MSFT. NVIDIA's Hugging Face deal and Alphabet's rebound narrative stood out.

• Strategy

  PanicFade v4: Fading the Panic, With Receipts

  Sep 3, 2026 · PanicFade v4 is a contrarian mean-reversion strategy that buys oversold large caps hit by negative sentiment, and it clears validation cleanly with all four walk-forward folds positive. The edge is real but modest — a ~6.8% CAGR earned through very heavy turnover, now live on a small paper book.

• Strategy

  Momentum SMA20 Rider: Riding Trends With a Thin Book

  Sep 3, 2026 · The live Momentum SMA20 Rider strategy has cleared out-of-sample validation with an 81% full-sample return and a strong final fold, but its 37% win rate and heavy turnover keep it firmly in the disciplined-momentum, thin-margin camp.

• Strategy

  Channel-Pullback: A Mean-Reversion Strategy That Failed Validation but Keeps Trading

  Sep 3, 2026 · Channel-pullback buys dips to the lower regression channel in uptrends and posts a modest 7.62% backtest return, but it fell short of our validation gate on a soft deflated Sharpe. Recent live activity has been quiet and slightly underwater, dominated by AAPL swing trades.

• Strategy

  Donchian Breakout: A Textbook Trend-Follower With a Failing Validation Gate

  Sep 3, 2026 · The live donchian-breakout strategy shows the classic trend-following signature — low win rate, positive skew, and three of four positive walk-forward folds — but its deflated Sharpe and recent string of rejected orders keep it short of our validation bar.

• Strategy

  bollinger-reversion: A Textbook Thesis That Fades Out of Sample

  Sep 3, 2026 · The mean-reversion strategy posts a respectable 17.55% backtest return and a 63.89% win rate, but it fails our walk-forward validation as edge decays fold over fold and recent live runs execute nothing.

• Strategy

  Dual-Momentum: Strong Backtest, Cautious Verdict, Quiet September

  Sep 3, 2026 · The dual-momentum strategy posted a 23.5% backtest return but failed formal validation, and it has sat entirely in cash through six consecutive scheduled runs.

• Strategy

  news-sentiment: A Cautious Signal That Rarely Fires

  Sep 3, 2026 · The live news-sentiment strategy is disciplined and low-risk, but its backtest is essentially flat and its walk-forward validation failed — the entire track record rests on just two trades in a single fold.

• Strategy

  Mean-Reversion Under the Microscope: Strong Win Rate, Failing Validation

  Sep 3, 2026 · The mean-reversion strategy posts a 70.6% win rate and 14.73% backtest return, but cross-validation flags it as overfit — the most recent out-of-sample fold lost money and live paper equity is drifting below its starting stake.

• Strategy

  momentum-code: Strong Backtest, Stalled Engine, and a Validation Gate That Said No

  Sep 3, 2026 · The momentum-code strategy posts a solid 19.76% backtested return with four positive out-of-sample folds, yet its validation gate still failed — and its live paper account has gone weeks without executing a trade.

• Strategy

  RSI Snap-Back: A Mean-Reversion Bet on the Mag-7, Idling in Neutral

  Sep 3, 2026 · The RSI Snap-Back strategy shows a solid 20.95% backtested return with a two-in-three win rate, but a modest Sharpe, a deep drawdown, and six straight days of no live trades temper the picture.

• Sector

  Mapping the Tracked Universe: A Sector Snapshot as of September 2026

  Sep 3, 2026 · Technology and Industrials dominate our tracked universe by company count, but the sector-level market-cap figures look skewed enough to warrant caution. Here's what the composition tells us — and what deserves a second look.

• Market News

  Sept 3 Roundup: A Cautiously Bullish Tape Wrestles With Oil and Bond Yields

  Sep 3, 2026 · Sentiment tilted net-positive as small caps led a broad rally and Nvidia dominated the headline flow, but a cross-market wobble in bond yields and oil kept a lid on conviction. Chevron and Alphabet supplied the day's most idiosyncratic storylines.

• Strategy

  PanicFade v4: Fading Panic in Large Caps, With Discipline to Match

  Sep 2, 2026 · PanicFade v4 is a live contrarian mean-reversion strategy that buys sentiment-shocked, oversold large caps and exits on reversion. Its four-fold validation is clean and improving, but a modest Sharpe and heavy turnover keep expectations grounded.

• Strategy

  Momentum SMA20 Rider: Riding Trends at 37% Win Rate

  Sep 2, 2026 · A trend-following strategy that buys the strongest names above their moving averages has returned 81% in backtest and cleared out-of-sample validation, though a low win rate and one flat fold underline its dependence on a handful of durable trends.

• Strategy

  Channel-Pullback: A Mean-Reversion Bet Fighting Its Own Validation Numbers

  Sep 2, 2026 · Channel-pullback posts a modest 7.62% backtest return but fails validation on a weak deconflated Sharpe, and its live paper account is currently underwater near $9,500. The thesis is coherent, yet the edge looks thin once overfitting is priced in.

• Strategy

  Donchian Breakout: Strong Out-of-Sample Folds, But Validation Says Wait

  Sep 2, 2026 · The 20-day breakout trend-follower posts a modest 6.95% backtest return with encouraging recent folds, yet fails our validation gate on deflated-Sharpe grounds — and its live sleeve has gone quiet, rejecting every trade for a week.

• Strategy

  bollinger-reversion: A Textbook Mean-Reverter That Stalls Out of Sample

  Sep 2, 2026 · The bollinger-reversion strategy posts a respectable 17.55% backtest return but fails validation as its edge decays fold-over-fold, and it has executed no trades in weeks.

• Strategy

  Dual-Momentum: Strong Trends, Weaker Proof

  Sep 2, 2026 · The live dual-momentum strategy posts a respectable 23.5% backtest return but fails our formal validation gate, and it has sat entirely in cash for the past week.

• Strategy

  news-sentiment: A Cautious Strategy That Rarely Pulls the Trigger

  Sep 2, 2026 · The news-sentiment strategy trades sparingly and holds a near-fully-invested book, but its backtest is razor-thin and formal validation did not pass. Live paper equity is modestly ahead while the signal stays quiet.

• Strategy

  Mean-Reversion Goes Live, But the Out-of-Sample Numbers Say Wait

  Sep 2, 2026 · The RSI-based mean-reversion strategy posts a strong 14.73% full-sample backtest and a 70.59% win rate, yet fails walk-forward validation with a negative out-of-sample fold. Its live paper account has sat idle and slightly underwater since late May.

• Strategy

  momentum-code: Strong Folds, a Failed Gate, and a Stalled Live Book

  Sep 2, 2026 · The momentum-code strategy posts a 19.76% backtest return with consistently positive out-of-sample folds, but its validation gate failed on a low deflated Sharpe, and its live book has gone quiet — six straight scheduled runs executed nothing.

• Strategy

  RSI Snap-Back: Strong Backtest, Quiet Book

  Sep 2, 2026 · The RSI Snap-Back strategy posts a respectable 20.95% backtest return on the Mag-7, but a deep 23.7% drawdown, absent out-of-sample validation, and a week of fully-cash scheduled runs temper the optimism.

• Sector

  Mapping the Tracked Universe: Where the Companies Cluster

  Sep 2, 2026 · A look at the 2,190 companies across eight sectors in the Headmars tracked universe — Technology leads by headcount while the reported market-cap aggregates tell a very different, and cautionary, story.

• Market News

  September Opens Cautious: Tech Retreats, Oil Rises, Sentiment Stays Neutral

  Sep 2, 2026 · Markets kicked off September with a broad tech-and-growth pullback and rising bond yields, while news sentiment leaned neutral-to-bullish with pockets of energy-driven concern. Apple, Nvidia and Google dominated the flow, alongside oil pressure on Chevron.

• Strategy

  PanicFade v4: Fading Sentiment Panics With Discipline — and Modest Returns

  Sep 1, 2026 · PanicFade v4 is a live contrarian mean-reversion strategy that buys oversold large caps hit by negative sentiment, backed by all-positive walk-forward folds but a modest Sharpe and a nearly flat live start.

• Strategy

  Momentum SMA20 Rider: Riding Confirmed Uptrends, With Turnover as the Tax

  Sep 1, 2026 · The live SMA-momentum strategy shows an 81% backtest return and a strong out-of-sample fold, but a low win rate and enormous turnover temper the enthusiasm. Recent weeks have been quiet, with the book mostly idle and small rotations into KO and JNJ.

• Strategy

  channel-pullback: A Live Strategy That Passed the Eye Test but Failed Validation

  Sep 1, 2026 · The channel-pullback strategy posts a modest +7.62% backtest return but fails our out-of-sample validation gate, and its live paper account currently sits below its starting stake. Encouraging fold-level strength is undercut by a low win rate, one brutal drawdown period, and thin live evidence.

• Strategy

  donchian-breakout: A Classic Trend-Follower That Splits the Panel

  Sep 1, 2026 · The 20-day breakout strategy is live and modestly green at +6.95%, but a low 0.34 Sharpe, a 21.73% drawdown, and a failed validation gate keep it firmly in the show-me pile.

• Strategy

  Bollinger-Reversion: A Strong-Looking Backtest That Validation Won't Sign Off On

  Sep 1, 2026 · The bollinger-reversion strategy posts a healthy 17.55% headline backtest return, but out-of-sample folds collapse to near-flat and it fails validation. Recent live activity shows nothing but rejected orders and a slowly sliding equity curve.

• Strategy

  Dual-Momentum: A Strong Backtest That Didn't Survive Validation

  Sep 1, 2026 · The dual-momentum strategy posts an attractive 23.5% headline return but fails its out-of-sample validation gate, and it currently sits entirely in cash with no qualifying trends.

• Strategy

  News-Sentiment: A Signal That Rarely Speaks

  Sep 1, 2026 · The live news-sentiment strategy is holding a small equity book of large-cap names, but a near-flat backtest and a failed walk-forward validation argue for treating its edge as unproven.

• Strategy

  Mean-Reversion Update: A Textbook Thesis That Fails Its Own Out-of-Sample Test

  Sep 1, 2026 · The RSI-based mean-reversion strategy posts a healthy 14.73% backtest return and a 70.6% win rate, but formal validation fails on the most recent fold and its live paper account is underwater. Encouraging in-sample, unconvincing out-of-sample.

• Strategy

  momentum-code: Strong Folds, a Failed Gate, and a Stalled Live Book

  Sep 1, 2026 · The momentum-code strategy backtests at 19.76% with four consecutive positive walk-forward folds, yet its validation gate failed on a low deflated Sharpe — and its live account has been rejecting every scheduled trade for a week.

• Strategy

  RSI Snap-Back: A Disciplined Mean-Reversion Bet That's Currently Sitting on Its Hands

  Sep 1, 2026 · The Mag-7 mean-reversion strategy posts a respectable 20.95% backtest return but carries a steep 23.7% drawdown, and its live book has been fully in cash for at least a week with zero recent trades.

• Sector

  What 2,178 Tickers Reveal: A Sector Map of the Headmars Universe

  Sep 1, 2026 · A breakdown of the tracked universe by sector shows Technology, Industrials, and Healthcare accounting for roughly two-thirds of all names, while cross-listings and patchy market-cap data warn against reading the totals too literally.

• Market News

  Daily Roundup: Tech Holds Firm, Energy Catches a Bid, and Only One Bear in the Room

  Sep 1, 2026 · A mostly constructive tape on September 1, 2026, with bullish headlines outnumbering bearish eleven-to-one against a large neutral middle. Energy names rallied on a crude spike while Big Tech drew scrutiny alongside resilience.

• Sector

  What 2,167 Tickers Say About the Market's Center of Gravity

  Aug 31, 2026 · The tracked universe skews heavily toward Technology and Industrials by company count, but total market-cap figures tell a different — and noisier — story. Investors should read the two lenses side by side.

• Market News

  Daily Roundup: Regulatory Noise Meets Jackson Hole Jitters

  Aug 31, 2026 · A headline-heavy session skewed mildly bearish, with Google dominating the news flow on legal and regulatory fronts while traders braced for macro cues from the Fed's Jackson Hole gathering.