Thirteen Bps Selective Overheat Guard

Back to leaderboard · 2015-01-01 to 2025-06-01 · v3_top100_monthly

Overall top-100 hits
25.26
average across horizons
Predicted top-100 avg actual rank
234.61
lower is better
Rank IC
0.0489
picks the better of two stocks 51.6% of the time
IC reliability
6.48
t-stat; IC positive 62.9% of months
not adjusted for overlapping windows

Horizon Breakdown

top 100 only
Horizon Average Top-100 Hits Predicted Top-100 Avg Actual Rank Rank IC
3m 25.34 236.25 0.0451
6m 25.45 233.38 0.0572
12m 24.98 234.20 0.0444

Run Note

test summary
Thirteen Bps Selective Overheat Guard: 25.26 avg hits@100 (3m 25.34, 6m 25.45, 12m 24.98); avg actual rank 234.61; IC +0.0489 (t 6.48).

Run Details

strategy_test_results #68
id
68
strategy_id
strategy_000058
strategy_title
Thirteen Bps Selective Overheat Guard
strategy_hash
3f32ca1ddbf4ea650bb182792420a6b01d61feb976549b3d4ff2838aba432b90
evaluator_version
v3_top100_monthly
start_date
2015-01-01
end_date
2025-06-01
created_at
2026-07-25 09:09:42
updated_at
2026-07-25 13:16:27

Sample Top-100 Lists

2020-03-31 / 6m / 23 hits

Predicted top 100

A, AAL, AAP, AAPL, ABBV, ABC, ABMD, ABT, ACN, ADBE, ADI, ADM, ADP, ADSK, AEE, AEP, AES, AFL, AIG, AIZ, AJG, AKAM, ALB, ALGN, ALK, ALL, ALLE, ALXN, AMAT, AMCR, AMD, AME, AMGN, AMP, AMT, AMZN, ANET, ANSS, AON, AOS, APD, APH, APTV, ARE, ATO, ATVI, AVB, AVGO, AVY, AWK, AXP, AZO, BA, BAC, BIIB, BIO, BMY, CAG, CARR, CCI, CLX, CNC, CPB, CTLT, DLR, DPZ, DVA, DXCM, EA, EQIX, FTNT, GILD, GIS, HRL, HUM, INTC, JKHY, K, KR, LDOS, LLY, MSCI, MSFT, NEE, NEM, NFLX, NOW, NVDA, ODFL, OTIS, REGN, RMD, ROL, SBAC, SJM, TMUS, TSLA, TYL, VRTX, WST

Actual top 100

AAP, AAPL, ABMD, ADBE, ADSK, ALB, ALGN, AMD, AMP, AMZN, APD, APH, APTV, ATVI, AVGO, BBY, BWA, CARR, CDNS, CE, CHRW, CMG, CMI, CPRT, CRM, CTAS, CTLT, CTSH, DD, DE, DHI, DHR, DRI, DXCM, EBAY, EMN, ETSY, EXPE, FCX, FDX, GLW, GPS, HAL, HD, IDXX, IPGP, JCI, KMX, LEG, LEN, LH, LOW, MAS, MCHP, MGM, MOS, NCLH, NKE, NOW, NSC, NVDA, NWS, NWSA, ORLY, PAYC, PH, PHM, PNR, POOL, PVH, PWR, PYPL, QCOM, QRVO, RCL, ROK, ROL, SHW, SIVB, SNPS, SWK, SWKS, SYF, TDG, TEL, TER, TGT, TMO, TSCO, TSLA, TT, UPS, URI, VAR, VIAC, VTR, WHR, WMB, WST, WY

Matched tickers

AAP, AAPL, ABMD, ADBE, ADSK, ALB, ALGN, AMD, AMP, AMZN, APD, APH, APTV, ATVI, AVGO, CARR, CTLT, DXCM, NOW, NVDA, ROL, TSLA, WST

Strategy Script

3f32ca1ddb
"""Generated V3 stock-ranking strategy."""
from __future__ import annotations

import numpy as np


STRATEGY_ID = "strategy_000058"
STRATEGY_TITLE = "Thirteen Bps Selective Overheat Guard"
DESCRIPTION = """
Ranks candidates with date-local normalization, using the best steady-growth
weight while making the overheat guard trigger only on very hot price action.
"""
FACTORS_USED = [
    "return_1m_pct",
    "return_3m_pct",
    "return_6m_pct",
    "return_12m_pct",
    "momentum_12_1_pct",
    "eps_growth_pct",
    "revenue_growth_pct",
    "operating_income_growth_pct",
    "free_cash_flow_growth_pct",
    "operating_margin_pct",
    "free_cash_flow_margin_pct",
    "forward_pe",
    "pe",
    "vol_63d",
    "from_52w_high_pct",
    "from_200d_ma_pct",
    "market_cap",
]
PARAMETERS = {
    "r1": 0.102,
    "r3": 0.10,
    "r6": -0.12,
    "r12": -0.05,
    "m121": -0.21,
    "growth": 0.77,
    "growth_breadth": 0.27,
    "margin": -0.37,
    "value": -0.48,
    "vol": 0.81,
    "size": 0.58,
    "pull": 0.53,
    "sma": 0.72,
    "steady_growth": 0.13,
    "overheat_penalty": 0.005,
    "min_sma": -0.18,
}
COMPLEXITY = 8
PARENT_STRATEGY_ID = "strategy_000054"
GENERATION = 23
RANDOM_SEED = 2026072558
CREATED_AT = "2026-07-25T00:00:00+00:00"


def _date_zscore(df, column):
    clean = df[column].replace([np.inf, -np.inf], np.nan)
    if "trade_date" not in df.columns:
        std = clean.std()
        if not np.isfinite(std) or std == 0:
            return clean * 0.0
        return ((clean - clean.mean()) / std).fillna(0)

    grouped = clean.groupby(df["trade_date"])
    mean = grouped.transform("mean")
    std = grouped.transform("std").replace(0, np.nan)
    return ((clean - mean) / std).fillna(0)


def _date_zscore_series(df, series):
    clean = series.replace([np.inf, -np.inf], np.nan)
    if "trade_date" not in df.columns:
        std = clean.std()
        if not np.isfinite(std) or std == 0:
            return clean * 0.0
        return ((clean - clean.mean()) / std).fillna(0)

    grouped = clean.groupby(df["trade_date"])
    mean = grouped.transform("mean")
    std = grouped.transform("std").replace(0, np.nan)
    return ((clean - mean) / std).fillna(0)


def eligibility_filter(df):
    return df["from_200d_ma_pct"].fillna(0) >= PARAMETERS["min_sma"]


def calculate_score(df):
    growth_cols = [
        "eps_growth_pct",
        "revenue_growth_pct",
        "operating_income_growth_pct",
        "free_cash_flow_growth_pct",
    ]
    growth_frame = df[growth_cols].replace([np.inf, -np.inf], np.nan).clip(lower=-100, upper=300)
    growth_mean = growth_frame.mean(axis=1).fillna(0)
    growth_std = growth_frame.std(axis=1).fillna(0)
    positive_breadth = (growth_frame > 0).sum(axis=1) / len(growth_cols)
    steadiness = positive_breadth - (growth_std / (growth_mean.abs() + 50)).clip(lower=0, upper=3)

    max_growth = growth_frame.max(axis=1).fillna(0)
    quality_growth = growth_frame[["eps_growth_pct", "operating_income_growth_pct", "free_cash_flow_growth_pct"]].max(axis=1).fillna(0)
    revenue_growth = growth_frame["revenue_growth_pct"].fillna(0)
    unsupported_jump = (quality_growth - revenue_growth - 100).clip(lower=0, upper=300) / 100
    explosive_jump = (max_growth - 100).clip(lower=0, upper=300) / 100
    hot_price = (
        (df["return_1m_pct"].fillna(0) > 25).astype(float)
        + (df["return_3m_pct"].fillna(0) > 60).astype(float)
        + (df["from_52w_high_pct"].fillna(-1) > -2).astype(float)
    ) / 3
    overheat = explosive_jump * (1 + hot_price) + 0.7 * unsupported_jump

    growth = 0.55 * _date_zscore(df, "eps_growth_pct") + 0.45 * _date_zscore(df, "revenue_growth_pct")
    growth_breadth = (
        0.55 * _date_zscore(df, "operating_income_growth_pct")
        + 0.45 * _date_zscore(df, "free_cash_flow_growth_pct")
    )
    margin = 0.55 * _date_zscore(df, "operating_margin_pct") + 0.45 * _date_zscore(df, "free_cash_flow_margin_pct")
    value = -0.65 * _date_zscore(df, "forward_pe") - 0.35 * _date_zscore(df, "pe")
    pullback = (-df["from_52w_high_pct"].fillna(0)).clip(lower=0, upper=0.45)
    constructive_pullback = pullback.where(df["return_6m_pct"].fillna(0) > 0, 0)
    size = _date_zscore(df.assign(_log_market_cap=np.log1p(df["market_cap"].clip(lower=0).fillna(0))), "_log_market_cap")

    score = (
        PARAMETERS["r1"] * _date_zscore(df, "return_1m_pct")
        + PARAMETERS["r3"] * _date_zscore(df, "return_3m_pct")
        + PARAMETERS["r6"] * _date_zscore(df, "return_6m_pct")
        + PARAMETERS["r12"] * _date_zscore(df, "return_12m_pct")
        + PARAMETERS["m121"] * _date_zscore(df, "momentum_12_1_pct")
        + PARAMETERS["growth"] * growth
        + PARAMETERS["growth_breadth"] * growth_breadth
        + PARAMETERS["margin"] * margin
        + PARAMETERS["value"] * value
        + PARAMETERS["vol"] * _date_zscore(df, "vol_63d")
        + PARAMETERS["size"] * size
        + PARAMETERS["pull"] * constructive_pullback
        + PARAMETERS["sma"] * _date_zscore(df, "from_200d_ma_pct")
        + PARAMETERS["steady_growth"] * _date_zscore_series(df, steadiness)
        - PARAMETERS["overheat_penalty"] * _date_zscore_series(df, overheat)
    )
    score = score.where(eligibility_filter(df), -1e9)
    return score.replace([np.inf, -np.inf], np.nan).fillna(score.median()).fillna(0.0)