Educational analysis only. Not financial advice. Signals are statistical patterns only.Full disclaimer

    Methodology Disclosure
    Public dataset on GitHub

    How Krentium's Signals Are Measured

    Methodology disclosure for Krentium's quantitative signal engines.

    18,808
    Total OOS Signals
    across four validation surfaces
    649
    Stock Universe
    237 current · 412 delisted
    219
    Crypto Tokens
    194 with signals
    12.55%
    Combined CAGR
    2006-2026 · gross of sub
    -32.0%
    Max Drawdown
    combined portfolio sim
    0.729
    Sharpe Ratio
    19.5-year period

    Overview

    Orientation to the three engines, the published hold horizons, and what "validated" means on this site.

    The three engines are: Wyckoff, a forward-return validation engine; Mean Reversion, a closed-loop strategy; and Crypto Dip Radar, a closed-loop strategy. Each engine carries one published hold horizon used for headline statistics; multi-horizon results for engines that support them are documented in Multi-Horizon Disclosure.

    "Validated" on this site means out-of-sample. Numbers shown are computed on data the engine did not see during construction. Significance is reported at the 95% level with Bonferroni correction across the validation battery; per-engine details live in Validation Framework.

    Validation Framework

    The nine statistical tests used across engines — walk-forward, block bootstrap, Carhart 4-factor regression, Monte Carlo, transaction-cost sensitivity.

    The validation battery spans five test families. Daily Wyckoff at the 20-day horizon carries the most extensive stack — nine tests applied. Other engines apply subsets appropriate to their claim type and data availability. Per-engine results are reported in Per-Engine Validation.

    Walk-forward temporal validation

    Train and test on disjoint time periods. The engine is built on data ending at one date; out-of-sample performance is measured on data after that date. Walk-forward win rate and signal counts per engine are reported in the Per-Engine Validation section.

    Block bootstrap confidence intervals

    Resample contiguous blocks of returns to construct confidence intervals around point estimates. Block resampling preserves the autocorrelation structure that simple resampling destroys. Reported as 95% CI bounds on win rate.

    Carhart 4-factor regression

    Per-signal alpha measured against a Carhart 4-factor model (market, size, value, momentum) using Fama-French total-return data. Alpha, t-statistic, p-value, and N reported per engine. Significance threshold: 95% with Bonferroni correction.

    Monte Carlo simulation

    Two distinct Monte Carlo tests: stock-selection (does the engine pick stocks better than chance?) and timing (does the engine time entries better than chance?). Both report percentile rank against a null distribution.

    Transaction-cost sensitivity

    Headline numbers are net of realistic transaction costs. The sensitivity test recomputes win rate across a cost grid to confirm no published metric depends on a cost assumption below realistic bounds.

    The full per-test pass/fail record at the 20-day horizon is documented in the 9-test extended validation stack in Wyckoff Daily. Multiple-testing correction is applied across the battery: Bonferroni-corrected, significant at α = 0.05.

    Combined Portfolio Simulation

    The combined outcome from all three engines running inside a single 10-slot account over 2006-2026.

    The combined-portfolio simulation runs Krentium's three engines — Wyckoff (daily and weekly), Mean Reversion, and Crypto — inside a single 10-slot portfolio over 2006-2026. Signals compete for the same capital pool under each engine's entry and exit rules. Wyckoff signals use the published 20-trading-day hold; Mean Reversion and Crypto exit conditions are defined per engine (see Per-Engine Validation).

    Round-trip transaction cost is 10 basis points. Idle cash between signals accrues at the Fama-French daily risk-free rate. The published value is the median of 20 random-seed runs. Values are denominated in EUR, matching the starting-capital currency.

    €100,598
    Final Value (gross)
    from €10,000 start
    12.55%
    CAGR
    compound annual
    -32.04%
    Max Drawdown
    peak-to-trough
    0.729
    Sharpe
    monthly returns
    1.041
    Sortino
    downside-only

    2006-2026 · 19.5-year period · €10,000 starting capital.

    Benchmark Comparison

    StrategyFinal ValueCAGRMax DD
    Krentium (3-engine combined)€100,59812.55%-32.04%
    100% SPY buy & hold€53,9459.02%-56.47%
    80/20 SPY/BTC (annual rebalance)€93,61212.14%-56.47%

    100% SPY price-only (no dividends). SPY with dividends would add ~1.5-2pp/yr CAGR. The 80/20 benchmark rebalances annually; its BTC sleeve starts 2018-01-01, before which it mirrors 100% SPY.

    How the simulation works

    Each trading day, up to ten concurrent signals occupy slots in a shared capital pool. All three engines draw from the same pool under their respective entry and exit rules.

    Round-trip transaction cost is 10 basis points. Idle cash earns the Fama-French daily risk-free rate.

    The same period was re-run under two alternative architectures: a sleeve-allocated variant (per-engine capital allocations of 40/35/25%) and a 40-day Wyckoff-hold variant. Final values across the three architectures span a €7,356 range.

    Seed variance band: €65,679 (5th percentile) to €161,739 (95th percentile) across 20 seeds — a 146.3% spread. The published value is the median canonical path.

    Per-Engine Validation

    Engine-by-engine validation. Each engine publishes a different claim type — the italic subtitle under each engine's header describes what its published number actually measures. Wyckoff and Mean Reversion share the same 649-stock survivorship-corrected universe; Crypto runs on a 219-token universe described in Universe Construction.

    Wyckoff Daily

    Forward-return validation · engine emits state, user picks hold

    20d published hold · live in production

    Wyckoff Daily is the daily-timeframe signal engine in the Wyckoff framework. Validation universe: 649-stock survivorship-corrected (see Universe Construction). Same-ticker signals deduplicated inside a 28-day window. All figures in this section are on the 20-trading-day forward-return horizon.

    Signal economics

    58.40%
    Win rate (20d)
    forward-return
    4,276
    Deduped signals
    28-day per-ticker window
    [55.9%, 60.7%]
    95% bootstrap CI
    block bootstrap
    +1.23%
    Median signal return
    per-signal 20d

    Carhart 4-factor regression

    +0.509%per signal
    t = 3.70
    p = 0.0002
    N = 4,014

    262 signals (of the 4,276 deduped total) are dropped from the regression: Fama-French factor values unavailable at signal date.

    Walk-forward + bootstrap

    Walk-forward (temporal out-of-sample, train 2006-2015 / test 2016-2025) win rate: 59.0%. Block-bootstrap 95% confidence interval on the 20d win rate (shown in the stat grid above) excludes 50% regardless of block size.

    9 of 9 statistical tests pass at the 20-day horizon

    Expand to see the list.

    Show the nine tests
    • Carhart 4-factor
    • Block bootstrap CI
    • Walk-forward temporal
    • MC stock-selection
    • MC timing
    • Survivorship-bias
    • Multiple testing (Bonferroni)
    • Transaction costs sensitivity
    • Effective N (Bartlett)

    28-day deduplication window

    A 28-day per-ticker deduplication window is applied across the Carhart regression, block bootstrap, and regime analyses. Bartlett autocorrelation analysis on per-ticker signal residuals returns the first insignificant lag at 28 days. Bartlett-adjusted effective sample size: 1,560 (22.16× inflation factor over the naive count).

    Bull vs bear regime

    Bull regime Carhart alpha statistically significant at 95% (N = 3,781). Bear regime estimate underpowered (N = 493); bull–bear alpha delta not significant at 95%. Earlier published values from a smaller historical universe showed a larger bear-regime alpha; current values include previously-delisted tickers.

    RegimeNCarhart αt-statp-valueSignificant at 95%
    Bull3,781+0.570%4.084.6e-5Yes
    Bear493+0.670%1.420.154No

    Bull–bear alpha delta +0.10 pp (pooled t = 0.20, p = 0.838). Regime definition: bull if Fama-French cumulative market return is above its 50-day SMA, bear otherwise.

    Crisis drawdowns

    CrisisWindowEngineSPY
    GFC 20082007-10-10 → 2009-06-2315.45%56.47%
    COVID 20202020-02-19 → 2020-03-2318.53%34.10%
    Bear 20222022-01-12 → 2022-03-079.38%25.36%

    Peak-to-trough drawdowns from the combined-portfolio simulation in Combined Portfolio Simulation. SPY values are price-only, without dividends.

    Risk metrics

    10.77%
    Half-Kelly
    capital fraction
    -4.71%
    Monthly VaR (95%)
    5th percentile
    -6.04%
    Monthly CVaR (95%)
    worst 5% mean

    Standalone simulation — Wyckoff in isolation

    12.98%
    CAGR
    compound annual
    0.762
    Sharpe
    monthly returns
    1.077
    Sortino
    downside-only
    $1,071
    Final value
    from $100 start

    Wyckoff Daily, standalone. $100 seed. See Combined Portfolio Simulation. Max drawdown and volatility deferred. Multi-horizon equity curves at 5d / 10d / 20d / 40d holds are visible on the Accumulation Daily tab of /performance.

    Wyckoff Weekly

    Forward-return validation · engine emits state, user picks hold

    20d forward-return horizon · weekly-bar signal emission

    Wyckoff Weekly runs the same framework on weekly bars. Validation universe: same 649-stock survivorship-corrected set applied across 2006-2026 with the 28-day per-ticker deduplication window. Per-signal statistical validation is complete; the standalone portfolio simulation (10-slot, 20d hold) is now populated and visible on the Accumulation Weekly tab of /performance. Crisis drawdowns and Weekly-specific risk metrics remain deferred.

    Signal economics

    61.15%
    Win rate (20d)
    forward-return
    2,533
    Deduped signals
    28-day per-ticker window
    [58.43%, 63.96%]
    95% bootstrap CI
    block bootstrap
    +1.57%
    Median signal return
    per-signal 20d

    Carhart 4-factor regression

    +0.936%per signal
    t = 3.84
    p = 0.00012
    N = 2,533

    Walk-forward validation

    Walk-forward win rate: 61.15% (5-fold temporal OOS). Win rate above 50% under both temporal slicing and serial-correlation-aware bootstrap (CI above).

    Bull vs bear regime breakdown

    Bull regime: Carhart alpha +0.984% per signal (N = 2,485, t = 4.07, p = 4.8e-5). Bear regime: N = 16 signals across 19.5 years; sample size prohibits meaningful alpha estimation.

    Weekly regime classification uses a different validation-time convention from the daily engine. The 16-signal bear cohort is too small to support either a positive or negative alpha estimate. 32 additional signals fell past the regime classification window and were excluded from regime classification.

    Blocks deferred to a follow-up compute pass

    The standalone equity curve at the 20-day hold (and 5d / 10d / 40d as exploratory disclosure) is now populated and visible on the Accumulation Weekly tab of /performance. The following Weekly-specific blocks remain deferred.

    Crisis drawdowns
    Per-Weekly-distribution crisis-window peak-to- trough metrics. The Weekly equity curve does include a 20d-hold drawdown chart on/performance; per-crisis attribution (2008 GFC / 2020 COVID / 2022 Bear) on the Weekly distribution is the deferred element.
    Weekly risk metrics
    Weekly-specific Half-Kelly and monthly VaR / CVaR. Daily-tab values are reasonable proxies given identical 20d-hold mechanics; per-Weekly-distribution refinement is deferred.

    Mean Reversion

    Closed-loop strategy · engine-defined entries and exits

    Out-of-sample period from 2016-01-01 · 649-stock universe (shared with Wyckoff)

    Mean Reversion is a closed-loop engine: entry and exit are both engine-defined. Validation universe: 649-stock survivorship-corrected (shared with Wyckoff, see Universe Construction). Out-of-sample period: 2016-01-01 onward. Median hold 3 trading days (mean 3.89).

    Signal economics

    66.67%
    Win rate (OOS)
    closed positions
    10,707
    Out-of-sample signals
    from 2016-01-01
    +0.94%
    Median signal return
    per closed position
    3d
    Median hold
    mean 3.89d

    Carhart 4-factor — annualized

    +16.3%annualized (full sample)
    Out-of-sample only: +12.6%

    Reported annualized to match signal frequency. Full-sample figure includes pre-OOS tuning period; OOS figure uses 2016-01-01 onward.

    Walk-forward validation

    Walk-forward win rate: 66.65% (5-fold temporal OOS, N = 10,614). Multiple-testing correction applied across the validation battery: Bonferroni-corrected, significant at α = 0.05. Bartlett effective sample size: 2,078 (1.81× autocorrelation inflation).

    Bull vs bear regime

    RegimeNWin rateMedian returnSignificant at 95%
    Bull9,56966.52%+0.90%No (delta CI crosses zero)
    Bear1,13367.96%+1.35%

    Broad-market trend classifier applied at signal entry date.

    Bull-bear delta not statistically significant at 95% (bootstrap CI crosses zero). Earlier published values from a smaller historical universe were approximately 2 percentage points higher in both regimes; current values include previously-delisted tickers.

    Crisis drawdowns

    CrisisWindowEngineSPY
    GFC 20082008-06-25 → 2009-06-2230.44%56.47%
    COVID 20202020-02-14 → 2020-03-1819.17%34.10%
    Bear 20222022-01-03 → 2022-09-2613.47%25.36%

    Peak-to-trough drawdowns, standalone simulation. SPY values price-only, without dividends.

    Kelly sizing by exit category

    Exits fall into three engine-defined categories. Half-Kelly reported per category.

    Exit categoryN% of exitsWin rateMean returnHalf-Kelly
    Momentum-recovery exit18,82871.9%76.15%+1.303%26.53%
    Price-recovery exit6,67125.3%56.47%0.404%0%
    Time-limit exit7962.8%0.0%9.61%0%

    Momentum-recovery exit: Engine-defined momentum-based early-exit. Positive expected value.

    Price-recovery exit: Engine-defined price-based early-exit. Mean expected return slightly negative; Kelly allocation rounds to zero.

    Time-limit exit: Engine-defined time-limit exit, fired when neither earlier exit condition resolves within the maximum hold window. Win rate 0% is structural (positions reaching the limit are forced exits).

    Risk metrics

    11.86%
    Half-Kelly
    capital fraction (aggregate)
    -6.14%
    Monthly VaR (95%)
    5th percentile
    -8.47%
    Monthly CVaR (95%)
    worst 5% mean

    Standalone simulation — Mean Reversion in isolation

    7.79%
    CAGR
    compound annual
    0.454
    Sharpe
    monthly returns
    0.647
    Sortino
    downside-only
    $415
    Final value
    from $100 start

    Mean Reversion, standalone. $100 seed. See Combined Portfolio Simulation. Max drawdown and volatility deferred.

    Crypto

    Closed-loop strategy · target + deadline at entry

    219-token universe · walk-forward OOS primary

    Closed-loop engine on a 219-token universe. Not survivorship-corrected (asset-class data constraint; see Universe Construction). Excluded categories: memecoins, NFTs, wrapped tokens, stablecoins. Median hold 2 trading days (mean 4.34).

    Signal economics

    73.0%
    Win rate (WF OOS)
    walk-forward
    1,292
    Walk-forward signals
    across all folds
    +7.24%
    Median signal return
    per closed position
    2d
    Median hold
    mean 4.34d

    Signal count breakdown

    Three population definitions: 948 in-sample signals using the production threshold over the full period; 1,292 walk-forward OOS signals across all folds, each fold's threshold re-derived from training data rather than held fixed; 1,901 raw signals before threshold filter. Per-fold re-derivation accounts for the walk-forward count exceeding the in-sample count.

    Factor-model regression

    Factor models (Fama-French) are equity-market constructs; not applicable to crypto asset class.

    Walk-forward validation

    Walk-forward win rate: 72.99% (N = 1,292). Bootstrap 95% CI on the win rate: [72.6%, 80.82%]. Multiple-testing correction applied across the validation battery: significant at α = 0.05. Bartlett effective sample size: 543 (1.75× autocorrelation inflation).

    Crisis drawdowns

    CrisisWindowEngineBTC
    GFC 2008data starts 2018-01-01; GFC predates crypto dataset
    COVID 20202020-03-06 → 2020-03-1212.09%52.54%
    Bear 20222022-01-14 → 2022-11-0913.24%66.99%

    Peak-to-trough drawdowns, standalone simulation. BTC benchmark is buy-and-hold.

    Kelly sizing by score band

    Win rate and median return reported across score bands. Half-Kelly shown for the canonical band; alternative bands reported for sensitivity.

    ThresholdNWin rateMedian return
    ≥ 5.094876.90%+7.24%
    ≥ 6.054978.70%+7.87%
    ≥ 7.027980.30%+7.91%
    ≥ 8.09187.90%+9.47%

    Kelly sizing by entry trigger

    Three engine-defined entry triggers. Half-Kelly reported per trigger.

    TriggerNWin rateMean returnAvg holdHalf-Kelly
    Trigger A13964.03%+5.62%3.4d19.61%
    Trigger B74376.99%+4.81%4.6d20.56%
    Trigger A+B6681.82%+8.30%3.9d28.93%

    Trigger A: Slow-developing entry. Lower win rate than Trigger B; positive expectancy. Shorter time-limit exit window.

    Trigger B: Sharper-developing entry. Largest signal count and high win rate. Longer time-limit exit window.

    Trigger A+B: Combined entry — both A and B triggered together. Highest win rate and magnitude. Fires rarely.

    Kelly sizing by exit category

    Five engine-defined exit categories. Counts and returns reported per category.

    Exit categoryN% of exitsWin rateMean return
    Profit-target exit61865.2%100.00%+11.61%
    Time-limit exit (long-window triggers)19120.1%4.19%16.59%
    Momentum-recovery exit909.5%74.44%+12.86%
    Price-recovery exit394.1%56.41%+5.55%
    Time-limit exit (short-window triggers)101.1%0.00%15.27%

    Profit-target exit: Resolves at the engine's preset upside threshold; win rate 100% is structural by construction. Most signals exit here.

    Time-limit exit (long-window triggers): Positions forced out when no earlier exit condition resolves. Category dominated by losses.

    Momentum-recovery exit: Momentum-based early-exit, used by the shorter-window trigger family. Second-highest expected value after profit-target exits.

    Price-recovery exit: Price-based early-exit. Small sample; modest expected value.

    Time-limit exit (short-window triggers): Same structural pattern as the long-window variant; category dominated by losses. Small sample.

    Risk metrics

    21.01%
    Half-Kelly
    capital fraction (aggregate)
    -8.19%
    Monthly VaR (95%)
    5th percentile
    -11.69%
    Monthly CVaR (95%)
    worst 5% mean

    Standalone simulation — Crypto in isolation

    15.78%
    CAGR
    compound annual
    0.554
    Sharpe
    monthly returns
    0.944
    Sortino
    downside-only
    $565
    Final value
    from $100 start

    Crypto, standalone. $100 seed. Benchmark: BTC HODL (CAGR 14.54%). See Combined Portfolio Simulation. Max drawdown and volatility deferred.

    Engine active 25.5% of trading days. Remaining time gated off by a market-condition meta-rule on the broader crypto trend. Cash on hold during gated-off periods accrues a risk-free yield.

    Asset-class distinctions

    Blocks rendered for Wyckoff and Mean Reversion that do not apply to Crypto as an asset class.

    Sector breakdown
    not applicable — crypto has no traditional sector structure. Token category breakdown (L1/L2/DeFi/etc.) could be added as a separate schema extension in future but is not currently computed.
    Regime classification
    No equity-market regime analog applies to crypto.
    Survivorship correction
    Not applied — asset-class data constraint. See Universe Construction and Limitations.

    Multi-Horizon Disclosure

    Forward-return validation at four holding horizons. The full grid, including horizons where the engine does not pass.

    The Wyckoff daily engine was tested at four forward-return horizons: 5 trading days, 10 days, 20 days, and 40 days. Each row below is an independent Carhart 4-factor regression on the 649-stock survivorship-corrected universe, with a block-bootstrap 95% confidence interval on the win rate at that horizon. All four rows are shown, including horizons where the engine does not produce a statistically significant edge. Verdicts follow a four-tier convention (fail, marginal, pass, strong pass) defined in the footnote below the table.

    HorizonNWin rate95% CICarhart αtpVerdict
    5d4,29753.29%[51.35%, 55.20%]+0.075%1.120.261FAIL
    10d4,29155.40%[53.25%, 57.53%]+0.192%1.950.051MARGINAL
    20dpublished hold4,01458.40%[55.90%, 60.70%]+0.509%3.700.0002PASS
    40d4,24660.88%[57.67%, 64.01%]+0.961%4.742.2e-6STRONG PASS

    Verdict tiers: FAIL — no significant edge at the 5% level. MARGINAL — edge near the threshold of statistical significance. PASS — clear statistical significance at 95% confidence. STRONG PASS — edge at p-values well below 0.001. The 20-day row additionally passes a 9-test extended validation stack (see Wyckoff Daily).

    Why 20 days is the published hold

    The published hold is 20 days. 40d shows a larger per-signal alpha (+0.961% vs +0.509%) but is not adopted because (a) the full 9-test validation stack is complete only at 20d, and (b) substituting 40d in the combined portfolio simulation adds +€1,864 (+0.11 pp CAGR) over 2006-2026 — within the seed-variance noise band.

    Tested horizon vs published hold

    Tested horizon: forward-return measurement interval. Every row in the table above is a tested horizon.

    Published hold: portfolio-simulation exit convention. At 40 days these diverge — 40d is tested but not adopted as the portfolio-sim hold per the reasoning above.

    Multi-horizon analysis for weekly Wyckoff is deferred to the weekly compute pass (weekly horizon: 20d only).

    Sector Breakdowns

    Per-sector performance for Wyckoff Daily, Wyckoff Weekly, and Mean Reversion (OOS-aligned).

    Yahoo 11-sector taxonomy. Threshold for reliable inference: N ≥ 30 per sector. Sectors below threshold flagged "N too small" and rendered with "—" for win rate, mean return, and median return. Aggregate metrics (coverage, overall WR) include all signals in the rollup; only per-sector display is gated. Crypto has no equity-sector analog.

    Wyckoff Daily

    Coverage 100.00%. Deduped signals N = 3,928. Overall WR 58.40%.

    Sector-breakdown source set (N = 3,928) is 348 signals smaller than the canonical Wyckoff Daily deduped count (N = 4,276). The gap reflects two-stage gating: 262 signals dropped for Fama-French factor unavailability (matching the Carhart regression N = 4,014), plus 86 additional signals dropped for SPY-calendar alignment in the sector breakdown only. Aggregate WR reproduces canonical 58.40%.

    SectorN signalsTickersWin rateMean returnMedian return
    Technology88510160.23%+2.32%+2.06%
    Healthcare6635856.26%+1.11%+1.01%
    Industrials6456158.60%+1.32%+1.43%
    Consumer Cyclical5375861.08%+1.94%+1.64%
    Financial Services5206353.27%+0.92%+0.67%
    Consumer Defensive2563358.59%+1.27%+0.80%
    Basic Materials2192761.64%+1.77%+1.86%
    Communication Services1932060.10%+2.00%+1.87%
    EnergyN too small102

    28-day per-ticker dedup applied before sector grouping.

    Wyckoff Weekly

    Coverage 96.88%. Deduped signals N = 2,533. Overall WR 61.15%.

    SectorN signalsTickersWin rateMean returnMedian return
    Technology5127663.67%+2.02%+2.04%
    Consumer Cyclical4596757.08%+1.01%+1.31%
    Healthcare3854262.60%+1.94%+1.61%
    Financial Services3305061.52%+1.66%+1.86%
    Industrials2694762.08%+1.63%+1.70%
    Communication Services1693256.80%+1.04%+1.22%
    Consumer Defensive1392558.27%+1.88%+1.23%
    Basic Materials1272063.78%+4.74%+1.37%
    Unknown792170.89%+3.71%+2.67%
    UtilitiesN too small276
    EnergyN too small2212
    Real EstateN too small154

    Tickers whose sector could not be resolved are shown in the Unknown bucket.

    The Unknown bucket comprises tickers whose sector could not be resolved via SEC EDGAR or yfinance lookups; the elevated WR likely reflects selection effects on the unresolvable subpopulation rather than a tradeable cross-sector signal.

    Mean Reversion

    OOS-aligned (from 2016-01-01). Coverage 98.09%. Deduped OOS signals N = 10,707. Overall WR 66.67%.

    SectorN signalsTickersWin rateMean returnMedian return
    Technology2,3808766.39%+0.55%+1.07%
    Financial Services1,6815568.47%+0.43%+0.99%
    Consumer Cyclical1,6226666.77%+0.32%+0.96%
    Industrials1,5525367.20%+0.34%+0.94%
    Healthcare1,2094364.52%+0.18%+0.76%
    Communication Services7213368.10%+0.53%+1.12%
    Basic Materials5372466.29%+0.10%+0.83%
    Consumer Defensive5352267.85%+0.34%+0.70%
    Unknown2041561.76%+0.25%+0.71%
    Energy1531660.78%-0.46%+1.09%
    Real Estate95861.05%+0.28%+0.84%
    UtilitiesN too small184

    OOS-only sector breakdown differs from the full-period signal set on several sectors, notably Energy, the Unknown bucket, and Healthcare. Drift driven by post-2016 delistings concentrated in Energy. Canonical table uses OOS-only; headline win rate computed on OOS-only.

    Crypto

    not applicable — crypto has no traditional sector structure. Token category breakdown (L1/L2/DeFi/etc.) could be added as a separate schema extension in future but is not currently computed.

    Universe Construction

    Stock universe construction (backtest and live scanner) and crypto universe construction.

    Backtest universe

    649 US stocks: 237 current S&P 500 constituents and 412 previously-delisted or demoted tickers. Survivorship correction is applied by including the 412 delisted constituents in the validation signal set. Same-ticker signals are deduplicated inside a 28-day window before any aggregate statistic. Date range covered: 2006-2026.

    Pre-backtest sector exclusions

    Energy, Utilities, and REITs are excluded from the backtest universe. Driven by commodity prices, interest rates, and regulatory factors rather than equity-specific dynamics.

    Live scanner universe

    The live scanner runs nightly on 225 symbols: 215 stocks and 10 ETFs (SPY, QQQ, DIA, IWM, VOO, VTI, XLK, XLF, XLE, XLV).

    The 10 ETFs are scanned for signals. None have surfaced at daily or weekly timeframes to date.

    ETF buying signals surfaced at daily or weekly to date: 0.

    Live vs backtest reconciliation

    Backtest universe (N = 649) and live scanner stocks (N = 215) are distinct constructs. Overlap: 200 stocks. Live scanner excludes 37 backtest survivors selected by a per-engine performance threshold. Live scanner includes 15 stocks not in the backtest universe (14 from a wider candle dataset; 1 ticker rename, MMC->MRSH).

    EnginePruned NPruned shareWR if pruned removedWR deltaα / mean-return delta
    Wyckoff Daily3035.1%58.03%+0.57 pp+0.066 pp (α)
    Wyckoff Weekly2709.3%61.05%+0.31 pp+0.067 pp (α)
    Mean Reversion (OOS)1,26711.8%66.82%+0.16 pp+0.0047 pp (mean ret)

    Removing the 37 pruned tickers from the backtest signal set shifts published win rate by sub-percentage-point amounts on each engine.

    Crypto universe

    219 tokens (universe version v4). 214 have sufficient historical data; 194 have at least one signal. Excluded categories: memecoins, NFTs, wrapped tokens, stablecoins. Asset-class data constraint: token historical data is constrained to currently-listed assets. Survivorship correction not applied.

    Limitations

    Documented limitations of the validation: CBE pre-merger gap, weekly cap-tier coverage, crypto survivorship, SPY price-only. Combined-portfolio seed-variance disclosure is in Combined Portfolio Simulation.

    • CBE pre-merger corporate-action gap. 12 pre-merger Wyckoff signals fired on CBE Nov 2012 before Eaton acquisition close.
    • Weekly cap_tier coverage gap (structural). 62.6% unknown — not resolvable via further SEC fetches. Weekly cap_tier breakdowns not surfaced.
    • Crypto universe not survivorship-corrected. Asset-class data constraint.
    • SPY benchmark is price-only (no dividends). With dividends, SPY total return is higher than the price-only series shown. Carhart regression uses Fama-French total-return data; reported alpha is dividend-adjusted.

    Regulatory Classification

    Krentium publishes investment-strategy recommendations within the meaning of § 86 Abs. 1 WpHG (Anlagestrategieempfehlungen) and complies with MAR Article 20 + Delegated Regulation (EU) 2016/958. Krentium is not an investment firm (Wertpapierdienstleistungsunternehmen) and does not provide individualised investment advice (Anlageberatung) within the meaning of MiFID II Art. 4(1)(4). Public legal disclosures are maintained on the following pages:

    Educational analysis only. Not financial advice. Signals are statistical patterns only.Full disclaimer

    Cookie Preferences

    We use cookies to enhance your experience and analyze site usage. Essential cookies are required for basic functionality. You can choose to accept all cookies or only essential ones. Learn more in our Privacy Policy

    Essential: Required for basic site functionality (always enabled)

    Functional: Remember your preferences and settings

    Analytics: Help us understand how you use our site