Live · mission control
Watch the platform run.
A read-only view of the system’s own health: container fleet, data spine, validation, self-healing, machine regime, honestly measured performance, and the live ML research matrix (walk-forward AUC across models and targets). No positions or account data are exposed.
loading snapshot…
11 sleeves · 2 live · 10 in shadow · deliberately different animals
SOL MAX-EV +13.0bp/wk · liveXRP MAX-EV +7.3bp/wk · live
single-owner guard · broker == executor books · 3 perps + tasty (0 vs book 0) · 17:15:07Z
Each sleeve is named for the Greek letter of the risk it monetizes. α Alpha · beta-neutral statistical edge in event contracts: small, repeatable, independent of market direction. β Beta · directional exposure to the underlying, with daily size set by a statistical regime classifier built on published regime-detection academics. χ Charm · Δ-decay (dΔ/dTime): deltas drift as expiry approaches; 0DTE pin dynamics. δ Delta · dPrice/dSpot: direction; index-futures trend exposure from range-break models. γ Gamma · dΔ/dSpot: convexity; long premium when realized vol is forecast above implied. ρ Rho · dPrice/dRate: rates; event contracts on rate decisions and macro prints. θ Theta · dPrice/dTime: decay; defined-risk short premium in forecast calm. ν Vega · dPrice/dVol: volatility level; term-structure calendars. Vanna · dΔ/dVol (equivalently dVega/dSpot): deltas shift as volatility jumps; skew structures. Veta · dVega/dTime (vega decay): vol exposure bleeds as expiry approaches. Vomma · dVega/dVol (volga): vega convexity; exposure that grows as vol itself jumps. All are 100% data-driven; every strategy earns capital through purged walk-forward validation, deflated Sharpe ratios, and null-control hypothesis tests, and is demoted the day the data says stop. They share no signal, no sizing, no failure mode.
| Sleeve | Strategy | Venue | Stage |
|---|---|---|---|
| α alpha | Cross-venue lead-lag xven_leadlag_shadow · live/shadow 2996 closed · +$279.95 · backtest N/A · blocked: dsr, lcb, beats_1N | Kalshi | shadow |
| α alpha | Hourly strike hourly_strike_paper · live/shadow 6791 closed · −$3.79 · backtest N/A · blocked: stage0, dsr, dsr_daily, walk_forward, lcb, beats_1N | Kalshi | shadow |
| α alpha | Perp mean-reversion (MAX-EV) perp mean-rev family (SOL/XRP/BTC/ETH/DOGE) · live/shadow 4864 forward closed · — · backtest N/A · real 1-contract maker canary live: SOL MAX-EV +13.0bp/wk; XRP MAX-EV +7.3bp/wk net d7 tier-adj · demoted: BTC SMOOTH, DOGE MAX-EV, DOGE SMOOTH, ETH MAX-EV, ETH SMOOTH SOL MAX-EV +13.0bp/wk liveXRP MAX-EV +7.3bp/wk liveBTC SMOOTH demotedDOGE MAX-EV demotedDOGE SMOOTH demotedETH MAX-EV demotedETH SMOOTH demoted | Kalshi | canary |
| α alpha | Range binary (breach vs stay 1σ) range_binary · live/shadow 589 closed · +$388.51 · backtest N/A · blocked: dsr, dsr_daily, walk_forward, lcb | Kalshi / opts | shadow |
| β beta | Winners Book (stacked) __winners_book__ · live/shadow 7 strategies · +$1,860.58 · backtest N/A · blocked: aggregate of member strategies | multi-venue | shadow |
| χ charm | 0DTE defined-risk condor volexp_condor_0dte · live/shadow 12 closed · +$152.69 · backtest N/A · blocked: stage0, dsr, dsr_daily, walk_forward, lcb, beats_1N · slice volexp_0dte_c25 (shadow) 10 closed · +$249.32 | Tastytrade | shadow |
| δ delta | 15m direction (ML) uf_direction_shadow · live/shadow 322 closed · +$2.03 · backtest N/A · blocked: stage0, dsr, dsr_daily, walk_forward, lcb, beats_1N | Kalshi | shadow |
| γ gamma | Long premium / tail convexity heatmap_long_vol · live/shadow 6 closed · −$15.50 · backtest −$149.17 (long strangle upper bound (mid), DoltHub) · blocked: stage0, dsr, dsr_daily, walk_forward, lcb, beats_1N | Tastytrade | shadow |
| ρ rho | Rates & macro event contracts no strategy has cleared into this sleeve yet · live/shadow N/A · backtest N/A | Kalshi | no candidate |
| θ theta | Short premium (naked strangle) heatmap_short_premium · live/shadow 25 closed · −$864.00 · backtest −$150.01 (exp/trade crossed, DoltHub 2020-26) · blocked: stage0, dsr, dsr_daily, walk_forward, lcb, beats_1N | Tastytrade | shadow |
| θ theta | Iron condor, monthly tenor (wings on) volexp_condor_monthly · live/shadow 18 closed · +$431.68 · backtest N/A · blocked: stage0, dsr, dsr_daily, walk_forward, lcb, beats_1N | Tastytrade | shadow |
| ν vega | Vol term-structure calendars calendar · live/shadow 20 closed · −$5,384.28 · backtest N/A · blocked: stage0, dsr, dsr_daily, walk_forward, lcb, beats_1N | Tastytrade | shadow |
| vanna | Skew & vol-acceleration structures no strategy has cleared into this sleeve yet · live/shadow N/A · backtest N/A | Tastytrade | no candidate |
| veta | Vega-decay structures no strategy has cleared into this sleeve yet · live/shadow N/A · backtest N/A | Tastytrade | no candidate |
| vomma | Vol-of-vol convexity no strategy has cleared into this sleeve yet · live/shadow N/A · backtest N/A | Tastytrade | no candidate |
What each sleeve trades · instrument, thesis, and the models that feed it
α alpha: Beta-neutral statistical edge in event contracts. Small, repeatable structures whose payoff does not depend on market direction; the flagship is cross-venue lead-lag on Kalshi, with maker and mean-reversion variants admitted only where net-of-fee expectancy survives real-fill accounting.
β beta: The Winners Book: a stacked portfolio of validated directional strategies whose daily gross exposure is set by the statistical regime classifier. Each constituent earned its place through purged walk-forward validation and is demoted the day the data says stop.
χ charm: 0DTE pin and delta-decay structures. Trades the pull of price toward high-open-interest strikes into expiration, gated by the fleet’s inside-move and gamma-state models.
δ delta: Directional index futures on CME micros. Range-break and trend continuation; the ML daily lane gates entries, and the sleeve makes no edge claim until that gate proves itself in shadow.
γ gamma: Vol-expansion long premium. Buys straddles and strangles when the vol-spike and IV-expansion models say realized vol will outrun implied. Long premium only pays when the forecast is right, so every structure earns its way through shadow first.
ρ rho: Rates and macro event contracts on Kalshi. A venue and a hypothesis; candidates are being built.
θ theta: Short premium. Sells rich implied volatility, gated on a vol-spike forecast rather than on high IVR alone. The proven track is a NAKED strangle book (undefined risk); defined-risk expressions of the same signal are still accruing evidence and are not yet proven.
ν vega: IV term-structure calendars. Trades the shape of the volatility curve across expiries, driven by the fleet’s IV-native targets (crush, expansion, premium survival).
vanna: Skew and vol-acceleration structures. Vanna-flow tags from the dealer-exposure engine time the entries; candidates in validation.
veta: Vega-decay structures. Studies how vega decays across the term structure; features live in the Greeks brain, candidates being spec’d for walk-forward.
vomma: Vol-of-vol convexity. Trades the price of vol-of-vol when it sits rich or cheap versus realized; tags context while candidates validate.
No per-day numbers here by design: they change nightly, the theses do not. Current evidence lives in the validation record.
Promotion is per-strategy through gate → shadow → canary → live, and demotion lands the day re-validation fails. No P&L decomposition, positions, or account data is shown.
model zoo
The board, ranked — every figure computed, refreshed as models retrain
Fleet scoreboard · 13 architectures
1. TabPFN v2.5500 titles · pod 43% · cov 86%
2. CatBoost410 titles · pod 48% · cov 93%
3. GRU362 titles · pod 35% · cov 97%
4. TabPFN v3227 titles · pod 28% · cov 80%
5. XGB222 titles · pod 36% · cov 93%
6. TFT-lite196 titles · pod 20% · cov 92%
7. PatchTST170 titles · pod 16% · cov 97%
8. GNN162 titles · pod 11% · cov 80%
9. N-BEATS128 titles · pod 13% · cov 85%
10. LGBM110 titles · pod 19% · cov 93%
11. Mamba97 titles · pod 10% · cov 80%
12. TiDE92 titles · pod 10% · cov 91%
13. JEPA40 titles · pod 4% · cov 80%
Model guide · workhorses and sequence nets
Gradient boosting (XGB · LGBM · CatBoost): tabular workhorses — the first bar every challenger must clear.
XGB: 222 championships (8.2%) · podium 36.1% · coverage 93% · best HYG Vol Spike 0.882
LGBM: 110 championships (4.1%) · podium 19.0% · coverage 93% · best HYG Vol Spike 0.905
CatBoost: 410 championships (15.1%) · podium 47.7% · coverage 93% · best HYG Vol Spike 0.899
GRU: the lightest neural net in the fleet.
GRU: 362 championships (13.3%) · podium 35.4% · coverage 97% · best HYG Vol Spike 0.911
Mamba (state-space): state-space sequence model; runs crypto 5m and the daily universe.
Mamba: 97 championships (3.6%) · podium 10.2% · coverage 80% · best opt_SNOW IV Crush 0.876
JEPA (pretrained encoder): pretrained joint-embedding encoder; strong on crash structure.
JEPA: 40 championships (1.5%) · podium 4.3% · coverage 80% · best XLY Vol Spike 0.843
Model guide · challengers and foundation models
Challengers (TFT-lite · N-BEATS · TiDE · PatchTST): neural challengers; each must beat the boosting bar to keep its lane.
TFT-lite: 196 championships (7.2%) · podium 19.7% · coverage 92% · best opt_SNOW IV Crush 0.885
N-BEATS: 128 championships (4.7%) · podium 13.4% · coverage 85% · best HYG Vol Spike 0.911
TiDE: 92 championships (3.4%) · podium 10.3% · coverage 91% · best HYG Vol Spike 0.884
PatchTST: 170 championships (6.3%) · podium 16.1% · coverage 97% · best HYG Vol Spike 0.874
TabPFN (tabular foundation model, no per-task training): the fleet's centre of gravity — two generations scored head-to-head.
TabPFN v2.5: 500 championships (18.4%) · podium 43.5% · coverage 86% · best XLF Crash 0.906
TabPFN v3: 227 championships (8.4%) · podium 28.4% · coverage 80% · best XLF Crash 0.908
family: 727 championships (26.8%) · v2-vs-v3 1,498–671 over 2,172 shared cells
GNN (cross-sectional graph): graph net over the cross-section; its 0DTE variant honestly failed net-of-cost and that negative is published.
GNN: 162 championships (6.0%) · podium 11.1% · coverage 80% · best ^VIX Crash 0.821
Target leaders
iv crushTFT-lite 0.885 · opt_SNOW
iv expansionCatBoost 0.850 · opt_SNOW
premium surviveXGB 0.807 · opt_CMCSA
inside movePatchTST 0.742 · 0dte_DBA
spread safeTabPFN v2.5 0.834 · 0dte_AMC
gamma stateTabPFN v2.5 0.886 · 0dte_AMC
vol spikeN-BEATS 0.911 · HYG
vrp signTFT-lite 0.736 · VIXY
Head-to-head · shared cells
CatBoost vs LGBM1,941–554 · 2,495 shared
XGB vs CatBoost965–1,521 · 2,486 shared
N-BEATS vs PatchTST1,396–907 · 2,303 shared
TabPFN v2.5 vs TabPFN v31,498–674 · 2,172 shared
Best cells on the board
HYG Vol SpikeN-BEATS 0.911
HYG Vol SpikeGRU 0.911
XLF Vol SpikeGRU 0.910
XLF CrashTabPFN v3 0.908
Hardest targets · nobody has cracked
trend chopbest 0.571 · 1,378 cells
directionbest 0.683 · 1,379 cells
buybest 0.709 · 1,379 cells
vrp signbest 0.736 · 1,378 cells
revbest 0.741 · 1,378 cells
This week on the board
scored cells2,716
coverage495 underlyings × 21 targets
leaderTabPFN v2.5 · 500 titles (18.4%)
title raceTabPFN v2.5 500 vs CatBoost 410
board refreshed2026-09-05 16:26 UTC
every figure above is computed from the live board (2,716 scored cells, updated 2026-09-05 16:26 UTC) on regeneration — descriptions carry no numbers, so nothing here can drift stale.
Every number above comes from the same purged walk-forward protocol as the matrix; published failures count as results too.