The Lab ran 49 experiments across 7 research threads this cycle (2026-07-21 → 2026-07-27) on the idle GPUs — every verdict below is published as the data said it, wins, losses and coin-flips alike. No cherry-picking.
Does crypto direction get more predictable at longer horizons?
- run of 2026-07-21 — '15m' leads on mean_auc: +0.526
- run of 2026-07-22 — '15m' leads on mean_auc: +0.526
- run of 2026-07-24 — '15m' leads on mean_auc: +0.526
- run of 2026-07-25 — '15m' leads on mean_auc: +0.526
- run of 2026-07-26 — '15m' leads on mean_auc: +0.526
Wide-feature XGBoost direction
- majors pack, 5m bars, 1-bar horizon (fresh-data re-run) — Marginal: AUC slightly above coin-flip; not bankable on its own
- alts pack, 5m bars, 1-bar horizon (fresh-data re-run) — Marginal: AUC slightly above coin-flip; not bankable on its own
- meme pack, 5m bars, 1-bar horizon (fresh-data re-run) — Marginal: AUC slightly above coin-flip; not bankable on its own
- broad pack, 5m bars, 1-bar horizon (fresh-data re-run) — Marginal: AUC slightly above coin-flip; not bankable on its own
- majors pack, 5m bars, 3-bar horizon (fresh-data re-run) — Marginal: AUC slightly above coin-flip; not bankable on its own
- alts pack, 5m bars, 3-bar horizon (fresh-data re-run) — Marginal: AUC slightly above coin-flip; not bankable on its own
- meme pack, 5m bars, 1-bar horizon (fresh-data re-run) — Marginal: AUC slightly above coin-flip; not bankable on its own
- meme pack, 5m bars, 3-bar horizon (fresh-data re-run) — Marginal: AUC slightly above coin-flip; not bankable on its own
- broad pack, 5m bars, 1-bar horizon (fresh-data re-run) — Marginal: AUC slightly above coin-flip; not bankable on its own
- majors pack, 5m bars, 3-bar horizon (fresh-data re-run) — Marginal: AUC slightly above coin-flip; not bankable on its own
- alts pack, 5m bars, 3-bar horizon (fresh-data re-run) — Marginal: AUC slightly above coin-flip; not bankable on its own
- meme pack, 5m bars, 3-bar horizon (fresh-data re-run) — Marginal: AUC slightly above coin-flip; not bankable on its own
- broad pack, 5m bars, 3-bar horizon (fresh-data re-run) — Marginal: AUC slightly above coin-flip; not bankable on its own
- majors pack, 5m bars, 6-bar horizon (fresh-data re-run) — Marginal: AUC slightly above coin-flip; not bankable on its own
- alts pack, 5m bars, 6-bar horizon (fresh-data re-run) — Marginal: AUC slightly above coin-flip; not bankable on its own
- broad pack, 5m bars, 3-bar horizon (fresh-data re-run) — Marginal: AUC slightly above coin-flip; not bankable on its own
- majors pack, 5m bars, 6-bar horizon (fresh-data re-run) — Marginal: AUC slightly above coin-flip; not bankable on its own
- alts pack, 5m bars, 6-bar horizon (fresh-data re-run) — Marginal: AUC slightly above coin-flip; not bankable on its own
- alts pack, 5m bars, 6-bar horizon (fresh-data re-run) — Marginal: AUC slightly above coin-flip; not bankable on its own
- meme pack, 5m bars, 6-bar horizon (fresh-data re-run) — Marginal: AUC slightly above coin-flip; not bankable on its own
- meme pack, 5m bars, 6-bar horizon (fresh-data re-run) — Marginal: AUC slightly above coin-flip; not bankable on its own
- broad pack, 5m bars, 6-bar horizon (fresh-data re-run) — Marginal: AUC slightly above coin-flip; not bankable on its own
- majors pack, 15m bars, 1-bar horizon (fresh-data re-run) — Marginal: AUC slightly above coin-flip; not bankable on its own
- meme pack, 5m bars, 6-bar horizon (fresh-data re-run) — Marginal: AUC slightly above coin-flip; not bankable on its own
Are calm or wild coins easier to predict?
- run of 2026-07-21 — 'low-vol' leads on mean_auc: +0.511
- run of 2026-07-23 — 'low-vol' leads on mean_auc: +0.511
- run of 2026-07-24 — 'low-vol' leads on mean_auc: +0.511
- run of 2026-07-26 — 'low-vol' leads on mean_auc: +0.511
Would pausing grids in a BTC downtrend stop the bleed?
- run of 2026-07-21 — 'BTC-downtrend' leads on total_pnl: +118
- run of 2026-07-23 — 'BTC-downtrend' leads on total_pnl: +118
- run of 2026-07-25 — 'BTC-downtrend' leads on total_pnl: +118
- run of 2026-07-27 — 'BTC-downtrend' leads on total_pnl: +118
Which coins are actually good for grids?
- run of 2026-07-22 — 'bottom (cull?)' leads on mean_grid_pnl: -2.83
- run of 2026-07-23 — 'bottom (cull?)' leads on mean_grid_pnl: -2.83
- run of 2026-07-25 — 'bottom (cull?)' leads on mean_grid_pnl: -2.83
Is a simple trend filter better than the grid?
- run of 2026-07-22 — 'trend_follow_50' leads on mean_return_pct: +64.00
- run of 2026-07-24 — 'trend_follow_50' leads on mean_return_pct: +64.00
- run of 2026-07-26 — 'trend_follow_50' leads on mean_return_pct: +64.00
Which crypto coins can a simple ML model actually predict?
- run of 2026-07-22 — 'ARB' leads on auc: +0.532
- run of 2026-07-23 — 'ARB' leads on auc: +0.532
- run of 2026-07-24 — 'ARB' leads on auc: +0.532
- run of 2026-07-25 — 'ARB' leads on auc: +0.532
- run of 2026-07-26 — 'ARB' leads on auc: +0.532
- run of 2026-07-27 — 'ARB' leads on auc: +0.532
Auto-published weekly by Gordon Lab. Each entry is a real backtest or model run on real market data; methods and full result tables live in the lab archive. This is research, not financial advice.