Real, usable edge for systematic traders — strategies, backtesting and code, no fluff
98 articles
To backtest a trading strategy in Python is to build a faithful simulator of your own decision process under the information constraints you actually faced in real time. Most public tutorials get the arithmetic right and the epistemics wrong: they…
Read →A 99% oneday ValueatRisk forecast says that losses should exceed the threshold on about 1% of eligible days. It is a probability forecast, not a statement that losses cannot be larger. Backtesting asks two separate questions: did exceptions occur at…
Read →A realized Sharpe ratio is an estimate, not a property of a strategy. Finance returns are serially dependent, heteroskedastic, and asymmetric, so a normaltheory confidence interval can be misleading. Bootstrap inference resamples the observed return path to approximate the sampling…
Read →Paper trading (forward testing) runs a strategy in real time against live, unfolding data with simulated capital. Its unique value is epistemic: it is the only validation stage where the future genuinely does not exist yet, so it forcibly exposes…
Read →Execution algorithms are the scheduling layer that converts a parent order into a sequence of child orders timed to minimize the sum of market impact, timing risk, and signaling. For any strategy trading meaningful size, the execution algo is not…
Read →Order types are the protocollevel primitives your strategy uses to express intent to a matching engine, and the choice among them is a microstructure decision with direct PnL consequences. The same alpha, expressed through a marketable order versus a passive…
Read →Transaction costs are the wedge between a strategy's gross signal and its net PnL, and for most quantitative strategies that wedge — not the signal — decides profitability. Modeling them is not a finishing touch you bolt on after tuning;…
Read →Walkforward optimization repeatedly fits a strategy's parameters on a window of past data and grades them on the next, unseen window — then stitches those outofsample segments into a single equity curve. It is the closest a backtest gets to…
Read →Overfitting is the dominant failure mode of quantitative strategy research: the model fits the noise in your sample, and that noise does not recur. The problem is not merely "too many parameters" — it is the combination of a flexible…
Read →Backtesting biases are the systematic ways a historical simulation inflates expected performance relative to what you could actually have captured. They are not rare pathologies — they are the default state of a naive [backtest](/post.php?slug=backtesttradingstrategypython). Each bias below has a…
Read →Survivorship bias enters a strategy when its historical universe is defined using entities that exist today. The missing names are not random: bankrupt companies, acquired firms, delisted microcaps, failed funds, and discontinued products are disproportionately poor performers. A backtest that…
Read →Most failed trading research does not fail because the researcher cannot fit a model. It fails because the experiment quietly answers a weaker question than the one required for capital allocation. “Does this feature correlate with returns in my notebook?”…
Read →A profitable backtest is not evidence of an edge — it is a hypothesis that demands a statistical test. Random strategies produce winning equity curves all the time; the question is whether yours is distinguishable from luck. Significance testing gives…
Read →The way you structure a backtest determines which lies it lets you tell yourself. A vectorized backtest computes signals and returns across an entire price array at once — blazing fast, perfect for research sweeps, and dangerously easy to contaminate…
Read →Standard kfold crossvalidation is quietly broken for financial data, and it lies in your favor. In a typical [machine learning for trading](/post.php?slug=machinelearningtrading) pipeline you shuffle samples into folds, train on some, validate on the rest, and trust the average score.…
Read →The single most common reason a backtest lies to you is that it was the best of many. When you try a hundred parameter sets, indicators, or universes and report the winner, the headline [Sharpe ratio](/post.php?slug=sharperatioexplained) you found is not…
Read →Monte Carlo pricing is straightforward when a payoff is an average across simulated paths. Risk is harder: desks need sensitivities to hundreds or thousands of market inputs, and rerunning a bumped valuation for every input is slow and noisy. Pathwise…
Read →Monte Carlo estimates a derivative price as an average of discounted payoffs. A Greek is the derivative of that expectation with respect to spot, volatility, rates, or a model parameter. The numerical difficulty is that a small difference between two…
Read →Ordinary Monte Carlo draws independent pseudorandom points. Its error decreases at the familiar rate O(N^1/2) regardless of dimension, but the points can leave important regions unevenly covered. QuasiMonte Carlo (QMC) replaces random points with lowdiscrepancy sequences designed to fill the…
Read →Rareevent quantities are often the ones that matter most: a deep outofthemoney put, a portfolio expected shortfall contribution, barrier breach probability, or defaultcluster loss. Plain Monte Carlo is inefficient when only a handful of paths reach the event. Importance sampling…
Read →LongstaffSchwartz leastsquares Monte Carlo (LSMC) prices Bermudan and Americanstyle claims by learning an approximation to continuation value from simulated paths. It is a practical solution when a tree or PDE cannot accommodate the number of risk factors. The essential idea…
Read →Combinatorially symmetric crossvalidation (CSCV) asks a deliberately uncomfortable question: when a strategy wins on one half of history, does it still rank well on the other half? Unlike an ordinary train/test split, it repeats that experiment across every symmetric partition…
Read →The Probability of Backtest Overfitting (PBO) measures whether a research process repeatedly selects strategies that look exceptional in sample and ordinary out of sample. It is a property of the search procedure and candidate family, not a mystical probability that…
Read →White’s Reality Check asks whether the best strategy in a searched family is genuinely better than a benchmark once the search itself is acknowledged. Selecting the maximum observed Sharpe from dozens of rules changes its null distribution; ordinary onestrategy pvalues…
Read →When a research platform tests thousands of signals, asking whether each pvalue is below 5% is a machine for manufacturing false alphas. False discovery rate (FDR) control changes the target: rather than trying to prevent every false positive, it limits…
Read →Research Notebook Standards is a practical quant topic: specify it, estimate it, cost it, and only then allocate risk. What follows is a nofluff working note aimed at systematic researchers who need something they can implement and falsify. Why it…
Read →Capacity Dashboards is a practical quant topic: specify it, estimate it, cost it, and only then allocate risk. What follows is a nofluff working note aimed at systematic researchers who need something they can implement and falsify. Why it matters…
Read →Slippage Attribution Engines is a practical quant topic: specify it, estimate it, cost it, and only then allocate risk. What follows is a nofluff working note aimed at systematic researchers who need something they can implement and falsify. Why it…
Read →Factor Attribution Pipelines is a practical quant topic: specify it, estimate it, cost it, and only then allocate risk. What follows is a nofluff working note aimed at systematic researchers who need something they can implement and falsify. Why it…
Read →Brinson Attribution Code is a practical quant topic: specify it, estimate it, cost it, and only then allocate risk. What follows is a nofluff working note aimed at systematic researchers who need something they can implement and falsify. Why it…
Read →PnL Explain Systems is a practical quant topic: specify it, estimate it, cost it, and only then allocate risk. What follows is a nofluff working note aimed at systematic researchers who need something they can implement and falsify. Why it…
Read →Scenario Engines Intraday is a practical quant topic: specify it, estimate it, cost it, and only then allocate risk. What follows is a nofluff working note aimed at systematic researchers who need something they can implement and falsify. Why it…
Read →Greeks Aggregation Books is a practical quant topic: specify it, estimate it, cost it, and only then allocate risk. What follows is a nofluff working note aimed at systematic researchers who need something they can implement and falsify. Why it…
Read →Real Time Risk Servers is a practical quant topic: specify it, estimate it, cost it, and only then allocate risk. What follows is a nofluff working note aimed at systematic researchers who need something they can implement and falsify. Why…
Read →Position Limit Engines is a practical quant topic: specify it, estimate it, cost it, and only then allocate risk. What follows is a nofluff working note aimed at systematic researchers who need something they can implement and falsify. Why it…
Read →Kill Switch Design Patterns is a practical quant topic: specify it, estimate it, cost it, and only then allocate risk. What follows is a nofluff working note aimed at systematic researchers who need something they can implement and falsify. Why…
Read →Circuit Breaker Effects is a practical quant topic: specify it, estimate it, cost it, and only then allocate risk. What follows is a nofluff working note aimed at systematic researchers who need something they can implement and falsify. Why it…
Read →Limit Up Down Rules is a practical quant topic: specify it, estimate it, cost it, and only then allocate risk. What follows is a nofluff working note aimed at systematic researchers who need something they can implement and falsify. Why…
Read →Trading Halt Handling is a practical quant topic: specify it, estimate it, cost it, and only then allocate risk. What follows is a nofluff working note aimed at systematic researchers who need something they can implement and falsify. Why it…
Read →Corporate Event Calendars is a practical quant topic: specify it, estimate it, cost it, and only then allocate risk. What follows is a nofluff working note aimed at systematic researchers who need something they can implement and falsify. Why it…
Read →Volume Filters Capacity is a practical quant topic: specify it, estimate it, cost it, and only then allocate risk. What follows is a nofluff working note aimed at systematic researchers who need something they can implement and falsify. Why it…
Read →Advancement is a practical quant topic: specify it, estimate it, cost it, and only then allocate risk. What follows is a nofluff working note aimed at systematic researchers who need something they can implement and falsify. Why it matters Advancement…
Read →Liquidity Screens Research is a practical quant topic: specify it, estimate it, cost it, and only then allocate risk. What follows is a nofluff working note aimed at systematic researchers who need something they can implement and falsify. Why it…
Read →Penny Stock Filters is a practical quant topic: specify it, estimate it, cost it, and only then allocate risk. What follows is a nofluff working note aimed at systematic researchers who need something they can implement and falsify. Why it…
Read →Delisting Return Assumptions is a practical quant topic: specify it, estimate it, cost it, and only then allocate risk. What follows is a nofluff working note aimed at systematic researchers who need something they can implement and falsify. Why it…
Read →Survivorship Free Indexes is a practical quant topic: specify it, estimate it, cost it, and only then allocate risk. What follows is a nofluff working note aimed at systematic researchers who need something they can implement and falsify. Why it…
Read →Backfill Bias Detection is a practical quant topic: specify it, estimate it, cost it, and only then allocate risk. What follows is a nofluff working note aimed at systematic researchers who need something they can implement and falsify. Why it…
Read →Restatement Handling PIT is a practical quant topic: specify it, estimate it, cost it, and only then allocate risk. What follows is a nofluff working note aimed at systematic researchers who need something they can implement and falsify. Why it…
Read →Split Dividend Adjustment Bugs is a practical quant topic: specify it, estimate it, cost it, and only then allocate risk. What follows is a nofluff working note aimed at systematic researchers who need something they can implement and falsify. Why…
Read →Corporate Action Adjustments PIT is a practical quant topic: specify it, estimate it, cost it, and only then allocate risk. What follows is a nofluff working note aimed at systematic researchers who need something they can implement and falsify. Why…
Read →FIGI LEI Data Quality is a practical quant topic: specify it, estimate it, cost it, and only then allocate risk. What follows is a nofluff working note aimed at systematic researchers who need something they can implement and falsify. Why…
Read →Symbology Mapping Challenges is a practical quant topic: specify it, estimate it, cost it, and only then allocate risk. What follows is a nofluff working note aimed at systematic researchers who need something they can implement and falsify. Why it…
Read →Fuzzy Matching Security Masters is a practical quant topic: specify it, estimate it, cost it, and only then allocate risk. What follows is a nofluff working note aimed at systematic researchers who need something they can implement and falsify. Why…
Read →Record Linkage Corporate Actions is a practical quant topic: specify it, estimate it, cost it, and only then allocate risk. What follows is a nofluff working note aimed at systematic researchers who need something they can implement and falsify. Why…
Read →Entity Resolution Alt Data is a practical quant topic: specify it, estimate it, cost it, and only then allocate risk. What follows is a nofluff working note aimed at systematic researchers who need something they can implement and falsify. Why…
Read →Diffusion Models Synthetic Paths is a practical quant topic: specify it, estimate it, cost it, and only then allocate risk. What follows is a nofluff working note aimed at systematic researchers who need something they can implement and falsify. Why…
Read →Score Based Generative Models is a practical quant topic: specify it, estimate it, cost it, and only then allocate risk. What follows is a nofluff working note aimed at systematic researchers who need something they can implement and falsify. Why…
Read →Normalizing Flows Returns is a practical quant topic: specify it, estimate it, cost it, and only then allocate risk. What follows is a nofluff working note aimed at systematic researchers who need something they can implement and falsify. Why it…
Read →Isolation Forest Spoofing is a practical quant topic: specify it, estimate it, cost it, and only then allocate risk. What follows is a nofluff working note aimed at systematic researchers who need something they can implement and falsify. Why it…
Read →Anomaly Detection Market Data is a practical quant topic: specify it, estimate it, cost it, and only then allocate risk. What follows is a nofluff working note aimed at systematic researchers who need something they can implement and falsify. Why…
Read →Focal Loss Rare Events is a practical quant topic: specify it, estimate it, cost it, and only then allocate risk. What follows is a nofluff working note aimed at systematic researchers who need something they can implement and falsify. Why…
Read →SMOTE Finance Pitfalls is a practical quant topic: specify it, estimate it, cost it, and only then allocate risk. What follows is a nofluff working note aimed at systematic researchers who need something they can implement and falsify. Why it…
Read →Imbalanced Classification Rares is a practical quant topic: specify it, estimate it, cost it, and only then allocate risk. What follows is a nofluff working note aimed at systematic researchers who need something they can implement and falsify. Why it…
Read →Asymmetric Loss Functions is a practical quant topic: specify it, estimate it, cost it, and only then allocate risk. What follows is a nofluff working note aimed at systematic researchers who need something they can implement and falsify. Why it…
Read →Pinball Loss Trading is a practical quant topic: specify it, estimate it, cost it, and only then allocate risk. What follows is a nofluff working note aimed at systematic researchers who need something they can implement and falsify. Why it…
Read →Expectile Regression Risk is a practical quant topic: specify it, estimate it, cost it, and only then allocate risk. What follows is a nofluff working note aimed at systematic researchers who need something they can implement and falsify. Why it…
Read →Ordinal Regression Returns is a practical quant topic: specify it, estimate it, cost it, and only then allocate risk. What follows is a nofluff working note aimed at systematic researchers who need something they can implement and falsify. Why it…
Read →LambdaMART Trading is a practical quant topic: specify it, estimate it, cost it, and only then allocate risk. What follows is a nofluff working note aimed at systematic researchers who need something they can implement and falsify. Why it matters…
Read →NDCG for Cross Sectional Rank is a practical quant topic: specify it, estimate it, cost it, and only then allocate risk. What follows is a nofluff working note aimed at systematic researchers who need something they can implement and falsify.…
Read →Ranking Metrics IC Alternatives is a practical quant topic: specify it, estimate it, cost it, and only then allocate risk. What follows is a nofluff working note aimed at systematic researchers who need something they can implement and falsify. Why…
Read →Log Loss vs PnL Alignment is a practical quant topic: specify it, estimate it, cost it, and only then allocate risk. What follows is a nofluff working note aimed at systematic researchers who need something they can implement and falsify.…
Read →Brier Score Trading Models is a practical quant topic: specify it, estimate it, cost it, and only then allocate risk. What follows is a nofluff working note aimed at systematic researchers who need something they can implement and falsify. Why…
Read →Proper Scoring Rules is a practical quant topic: specify it, estimate it, cost it, and only then allocate risk. What follows is a nofluff working note aimed at systematic researchers who need something they can implement and falsify. Why it…
Read →Platt Scaling Trading Scores is a practical quant topic: specify it, estimate it, cost it, and only then allocate risk. What follows is a nofluff working note aimed at systematic researchers who need something they can implement and falsify. Why…
Read →Isotonic Calibration Probabilities is a practical quant topic: specify it, estimate it, cost it, and only then allocate risk. What follows is a nofluff working note aimed at systematic researchers who need something they can implement and falsify. Why it…
Read →Gradient Boosting Calibration is a practical quant topic: specify it, estimate it, cost it, and only then allocate risk. What follows is a nofluff working note aimed at systematic researchers who need something they can implement and falsify. Why it…
Read →Population Based Training is a practical quant topic: specify it, estimate it, cost it, and only then allocate risk. What follows is a nofluff working note aimed at systematic researchers who need something they can implement and falsify. Why it…
Read →Successive Halving Tuning is a practical quant topic: specify it, estimate it, cost it, and only then allocate risk. What follows is a nofluff working note aimed at systematic researchers who need something they can implement and falsify. Why it…
Read →Multiple Testing Hyperparameters is a practical quant topic: specify it, estimate it, cost it, and only then allocate risk. What follows is a nofluff working note aimed at systematic researchers who need something they can implement and falsify. Why it…
Read →Hyperparameter Overfitting Risk is a practical quant topic: specify it, estimate it, cost it, and only then allocate risk. What follows is a nofluff working note aimed at systematic researchers who need something they can implement and falsify. Why it…
Read →Nested Cross Validation Costs is a practical quant topic: specify it, estimate it, cost it, and only then allocate risk. What follows is a nofluff working note aimed at systematic researchers who need something they can implement and falsify. Why…
Read →Blocked Cross Validation Finance is a practical quant topic: specify it, estimate it, cost it, and only then allocate risk. What follows is a nofluff working note aimed at systematic researchers who need something they can implement and falsify. Why…
Read →Jackknife Plus Time Series is a practical quant topic: specify it, estimate it, cost it, and only then allocate risk. What follows is a nofluff working note aimed at systematic researchers who need something they can implement and falsify. Why…
Read →Distributionally Robust Optimization ML is a practical quant topic: specify it, estimate it, cost it, and only then allocate risk. What follows is a nofluff working note aimed at systematic researchers who need something they can implement and falsify. Why…
Read →Adversarial Robust Trading Models is a practical quant topic: specify it, estimate it, cost it, and only then allocate risk. What follows is a nofluff working note aimed at systematic researchers who need something they can implement and falsify. Why…
Read →Utility Based Deep Hedging is a practical quant topic: specify it, estimate it, cost it, and only then allocate risk. What follows is a nofluff working note aimed at systematic researchers who need something they can implement and falsify. Why…
Read →Deep Hedging Frameworks is a practical quant topic: specify it, estimate it, cost it, and only then allocate risk. What follows is a nofluff working note aimed at systematic researchers who need something they can implement and falsify. Why it…
Read →Continuous Time ML Models is a practical quant topic: specify it, estimate it, cost it, and only then allocate risk. What follows is a nofluff working note aimed at systematic researchers who need something they can implement and falsify. Why…
Read →Few Shot Regime Adaptation is a practical quant topic: specify it, estimate it, cost it, and only then allocate risk. What follows is a nofluff working note aimed at systematic researchers who need something they can implement and falsify. Why…
Read →Domain Adaptation Market Regimes is a practical quant topic: specify it, estimate it, cost it, and only then allocate risk. What follows is a nofluff working note aimed at systematic researchers who need something they can implement and falsify. Why…
Read →Label Shift in Finance is a practical quant topic: specify it, estimate it, cost it, and only then allocate risk. What follows is a nofluff working note aimed at systematic researchers who need something they can implement and falsify. Why…
Read →Covariate Shift Correction is a practical quant topic: specify it, estimate it, cost it, and only then allocate risk. What follows is a nofluff working note aimed at systematic researchers who need something they can implement and falsify. Why it…
Read →Concept Drift Adaptation is a practical quant topic: specify it, estimate it, cost it, and only then allocate risk. What follows is a nofluff working note aimed at systematic researchers who need something they can implement and falsify. Why it…
Read →Counterfactual Explanations Alpha is a practical quant topic: specify it, estimate it, cost it, and only then allocate risk. What follows is a nofluff working note aimed at systematic researchers who need something they can implement and falsify. Why it…
Read →Integrated Gradients Trading Models is a practical quant topic: specify it, estimate it, cost it, and only then allocate risk. What follows is a nofluff working note aimed at systematic researchers who need something they can implement and falsify. Why…
Read →Ablation Studies Quant Research is a practical quant topic: specify it, estimate it, cost it, and only then allocate risk. What follows is a nofluff working note aimed at systematic researchers who need something they can implement and falsify. Why…
Read →Spike and Slab Variable Selection is a practical quant topic: specify it, estimate it, cost it, and only then allocate risk. What follows is a nofluff working note aimed at systematic researchers who need something they can implement and falsify.…
Read →Elastic Net for Alpha Signals is a practical quant topic: specify it, estimate it, cost it, and only then allocate risk. What follows is a nofluff working note aimed at systematic researchers who need something they can implement and falsify.…
Read →