Research on trading systems that hold up in live markets.https://marketmaker.cc/https://marketmaker.cc/blog/arbitrage-prereq-3-mean-stdev/https://marketmaker.cc/blog/arbitrage-prereq-3-mean-stdev/A short from-zero lesson for the arbitrage course. The mean, the standard deviation and the z-score on ten numbers, why one odd value shifts everything, and how a z-score tells you when something is unusual.Mon, 09 Nov 2026 00:00:00 GMTarbitragecourseprerequisitestatisticsz-scorepythonhttps://marketmaker.cc/blog/arbitrage-prereq-2-graphs/https://marketmaker.cc/blog/arbitrage-prereq-2-graphs/A short from-zero lesson for the arbitrage course. What a graph is, how to store one in Python, how to list every route between two points, and why trying every route stops working when the map grows.Mon, 09 Nov 2026 00:00:00 GMTarbitragecourseprerequisitegraphshortest pathpythonhttps://marketmaker.cc/blog/arbitrage-prereq-1-logarithms/https://marketmaker.cc/blog/arbitrage-prereq-1-logarithms/A short from-zero lesson for the arbitrage course. What a logarithm does to a chain of rates, why the sum of logs replaces the product, and how to read the minus sign. Tiny Python code with output you can check by hand.Mon, 09 Nov 2026 00:00:00 GMTarbitragecourseprerequisitelogarithmmathpythonhttps://marketmaker.cc/blog/arbitrage-course-16-release/https://marketmaker.cc/blog/arbitrage-course-16-release/Lesson 16 of the arbitrage course, the last one. Turn the paper-trading bot into something you can show: a trade log, a report page, a README, a clean public repository with no secrets, and an honest way to talk about the results.Mon, 09 Nov 2026 00:00:00 GMTarbitragecoursebeginnerreleaseREADMEgithubpaper tradingpythonhttps://marketmaker.cc/blog/arbitrage-course-15-pairs-trading/https://marketmaker.cc/blog/arbitrage-course-15-pairs-trading/Lesson 15 of the arbitrage course. Why comparing two raw prices means nothing, how the spread and its z-score show when two linked coins drifted apart, and a small backtest that shows how fees eat the edge. Python code with output you can reproduce.Mon, 09 Nov 2026 00:00:00 GMTarbitragecoursebeginnerpairs tradingstatistical arbitragez-scoremean reversionpythonhttps://marketmaker.cc/blog/arbitrage-course-14-futures-spot-funding/https://marketmaker.cc/blog/arbitrage-course-14-futures-spot-funding/Lesson 14 of the arbitrage course. What a perpetual future and its funding rate are, how to hold one coin on the spot market and the opposite position on the future, and what the fees, a negative rate and a liquidation do to the result. Short Python code with numbers you can check by hand.Mon, 09 Nov 2026 00:00:00 GMTarbitragecoursebeginnerfuturesfunding rateperpetualcash and carrypythonhttps://marketmaker.cc/blog/arbitrage-course-13-speed/https://marketmaker.cc/blog/arbitrage-course-13-speed/Lesson 13 of the arbitrage course. Check every three-currency circle with loops, then with a numpy matrix trick, then with a short Rust program, and measure all three. Runnable code and real timings.Mon, 09 Nov 2026 00:00:00 GMTarbitragecoursebeginnernumpyvectorizationrustperformancepythonhttps://marketmaker.cc/blog/arbitrage-course-12-paper-trading-risk/https://marketmaker.cc/blog/arbitrage-course-12-paper-trading-risk/Lesson 12 of the arbitrage course. Run your bot on live prices with fake money, move every number into a config file, and add risk limits that stop the bot before a bad hour does damage. Short Python code with checkable output.Mon, 09 Nov 2026 00:00:00 GMTarbitragecoursebeginnerpaper tradingrisk managementconfigpythonhttps://marketmaker.cc/blog/arbitrage-course-11-reality-check/https://marketmaker.cc/blog/arbitrage-course-11-reality-check/Lesson 11 of the arbitrage course. Latency, partial fills, a failed second leg and stale prices, with a small simulation that shows how a profitable plan turns into an average loss. Short Python code you can run and change.Mon, 09 Nov 2026 00:00:00 GMTarbitragecoursebeginnerlatencyexecution risksimulationpythonhttps://marketmaker.cc/blog/arbitrage-course-10-size-fees-slippage/https://marketmaker.cc/blog/arbitrage-course-10-size-fees-slippage/Lesson 10 of the arbitrage course. Put fees and order book walking together, compute profit as a function of trade size, and find the size where the profit is highest. Short Python code with numbers you can check by hand.Mon, 09 Nov 2026 00:00:00 GMTarbitragecoursebeginnerfeesslippageposition sizepythonhttps://marketmaker.cc/blog/arbitrage-course-09-order-book/https://marketmaker.cc/blog/arbitrage-course-09-order-book/Lesson 9 of the arbitrage course. What an order book is, how to read bids and asks with sizes, how to walk it to find the real price of a trade, and a short Python function you can check by hand.Mon, 09 Nov 2026 00:00:00 GMTarbitragecoursebeginnerorder bookslippagepythonhttps://marketmaker.cc/blog/arbitrage-course-08-rich/https://marketmaker.cc/blog/arbitrage-course-08-rich/Lesson 8 of the arbitrage course. Why a bot wants circles with a limited number of steps, the colour-coding trick behind the RICH algorithm (VLDB 2025), a small working version in Python, and an honest note on what is simplified.Mon, 09 Nov 2026 00:00:00 GMTarbitragecoursebeginnerRICHcolor-codingnegative cyclek-hoppythonhttps://marketmaker.cc/blog/arbitrage-course-07-spfa/https://marketmaker.cc/blog/arbitrage-course-07-spfa/Lesson 7 of the arbitrage course. How SPFA only rechecks the roads that can have changed, a step-by-step queue example, working Python code with a negative cycle alarm, and a speed comparison against Bellman-Ford.Mon, 09 Nov 2026 00:00:00 GMTarbitragecoursebeginnerSPFABellman-Fordqueuenegative cyclepythonhttps://marketmaker.cc/blog/arbitrage-course-06-bellman-ford/https://marketmaker.cc/blog/arbitrage-course-06-bellman-ford/Lesson 6 of the arbitrage course. The Bellman-Ford algorithm explained with a tiny worked example, why one extra round reveals a negative cycle, working Python code that returns the profitable circle, and how to check it by hand.Mon, 09 Nov 2026 00:00:00 GMTarbitragecoursebeginnerBellman-Fordnegative cyclegraph algorithmspythonhttps://marketmaker.cc/blog/arbitrage-course-05-rates-as-a-graph/https://marketmaker.cc/blog/arbitrage-course-05-rates-as-a-graph/Lesson 5 of the arbitrage course. Currencies as points, exchange rates as arrows, why multiplication turns into addition with logarithms, and why a profitable circle becomes a circle with a negative total.Mon, 09 Nov 2026 00:00:00 GMTarbitragecoursebeginnergraphlogarithmtriangular arbitragehttps://marketmaker.cc/blog/arbitrage-course-04-compare-best-prices/https://marketmaker.cc/blog/arbitrage-course-04-compare-best-prices/Lesson 4 of the arbitrage course. Spread, fees, quote currencies and units, and a short Python program that scans five exchanges and reports which gaps survive the fees.Mon, 09 Nov 2026 00:00:00 GMTarbitragecoursebeginnerspreadfeespythonhttps://marketmaker.cc/blog/arbitrage-course-03-stages-and-design/https://marketmaker.cc/blog/arbitrage-course-03-stages-and-design/Lesson 3 of the arbitrage course. Why a bot works in stages, what each stage does, and how to ask an AI agent for a design you can read and check.Mon, 09 Nov 2026 00:00:00 GMTarbitragecoursebeginnerarchitectureorder bookpaper tradinghttps://marketmaker.cc/blog/arbitrage-course-02-kinds-of-arbitrage/https://marketmaker.cc/blog/arbitrage-course-02-kinds-of-arbitrage/Lesson 2 of the arbitrage course. The four main kinds of arbitrage (cross-exchange, cyclic, statistical, futures-spot), one small example of each, and how to choose which one your bot will hunt.Mon, 09 Nov 2026 00:00:00 GMTarbitragecoursebeginnertriangularstatistical arbitragefunding ratehttps://marketmaker.cc/blog/arbitrage-course-01-what-is-arbitrage/https://marketmaker.cc/blog/arbitrage-course-01-what-is-arbitrage/Lesson 1 of the arbitrage course. What arbitrage means, why price gaps exist, and a ten-line Python scanner that compares the best bid and best ask of one coin on two exchanges.Mon, 09 Nov 2026 00:00:00 GMTarbitragecoursebeginnerbidaskpythonhttps://marketmaker.cc/blog/jev-bots-three-year-profitability/https://marketmaker.cc/blog/jev-bots-three-year-profitability/We ran a trend strategy across 15 crypto pairs, checked costs and access to future candles, and tested Jev separately. Where did profits appear, and why can't we promise them in live trading yet?Sat, 26 Sep 2026 00:00:00 GMTalgorithmic tradingJevbacktestingcryptocurrencytrading botshttps://marketmaker.cc/blog/xgboost-gradient-boosting-trading/https://marketmaker.cc/blog/xgboost-gradient-boosting-trading/Return-direction classifiers are imbalanced problems, and 0.5 is the wrong decision threshold. Comparing scale_pos_weight, focal loss, and precision-constrained threshold optimization on crypto data - plus the engineering differences between XGBoost, LightGBM, and CatBoost.Tue, 25 Aug 2026 00:00:00 GMTmachine-learningxgboostgradient-boostingquantpredictionhttps://marketmaker.cc/blog/transfer-entropy-crypto-information-flow/https://marketmaker.cc/blog/transfer-entropy-crypto-information-flow/DCC-GARCH tells you when crypto dependence tightens. Transfer entropy tells you which way it points. A directed information-flow measure, its null calibration, and an honest account of what it does and does not add over average pairwise correlation.Mon, 24 Aug 2026 00:00:00 GMTcausal-inferencetransfer-entropyinformation-theorycryptonetworkhttps://marketmaker.cc/blog/trade-classification-algorithms/https://marketmaker.cc/blog/trade-classification-algorithms/Reconstructing trade direction from OHLCV, bar history, or venues without an aggressor flag - the classic rules, the stale-quote problem, bulk volume classification, and how to measure them against crypto's free ground truth.Sun, 23 Aug 2026 00:00:00 GMTmicrostructuretrade-classificationLee-Readyorder-flowdeep-learninghttps://marketmaker.cc/blog/tick-data-forecasting-deep-learning/https://marketmaker.cc/blog/tick-data-forecasting-deep-learning/Sequence models fed tick data still assume regular spacing. Three ways to tell a Transformer when a tick actually happened - learnable-timescale continuous encoding, ODE-RNN latent state, and delta_t as a plain feature - and the ablation that decides between them.Sat, 22 Aug 2026 00:00:00 GMTHFTtick-datadeep-learninghigh-frequencypredictionhttps://marketmaker.cc/blog/synthetic-control-trading-evaluation/https://marketmaker.cc/blog/synthetic-control-trading-evaluation/This series has priced the selection route to a false edge - DSR prices the winner, PBO prices the search. Neither touches confounding: the strategy that made money because volatility doubled the week you deployed it. The Synthetic Control Method builds a weighted counterfactual from a donor pool of untouched instruments and gives you a falsification criterion and a placebo p-value.Fri, 21 Aug 2026 00:00:00 GMTcausal-inferencesynthetic-controlcounterfactualevaluationeconometricshttps://marketmaker.cc/blog/probabilistic-forecasting-trading/https://marketmaker.cc/blog/probabilistic-forecasting-trading/How to evaluate a predictive distribution honestly - CRPS as a proper scoring rule, the PIT histogram as a calibration diagnostic, and DeepAR sampling in GluonTS.Thu, 20 Aug 2026 00:00:00 GMTforecastingprobabilisticquantile-regressionDeepARuncertaintyhttps://marketmaker.cc/blog/pinn-options-pricing/https://marketmaker.cc/blog/pinn-options-pricing/Putting the Heston, free-boundary, and jump-diffusion pricing PDEs into a neural network loss - the log-price residual, the mixed-partial autograd trick, and what still needs measuring.Wed, 19 Aug 2026 00:00:00 GMTdeep-learningPINNoptionsBlack-ScholesPDEhttps://marketmaker.cc/blog/pcmci-causal-discovery-crypto/https://marketmaker.cc/blog/pcmci-causal-discovery-crypto/How PCMCI's two-stage MCI test recovers directed causal links between crypto assets where correlation and bivariate Granger cannot - the construction, the tigramite pipeline, and the real-data study it still needs.Tue, 18 Aug 2026 00:00:00 GMTcausal-inferencePCMCIcausal-discoverytime-seriescryptohttps://marketmaker.cc/blog/order-flow-imbalance-prediction/https://marketmaker.cc/blog/order-flow-imbalance-prediction/Turning raw book updates into a signed flow quantity: the CKS event decomposition, multi-level OFI with PCA reduction, and Lee-Ready trade classification - plus an honest accounting of what the headline R-squared actually measures.Mon, 17 Aug 2026 00:00:00 GMTmicrostructureorder-flowimbalancepredictionHFThttps://marketmaker.cc/blog/neural-ode-continuous-finance/https://marketmaker.cc/blog/neural-ode-continuous-finance/Neural ODEs, Neural SDEs and continuous normalizing flows for irregularly-sampled market data - and the one ablation that decides whether continuous dynamics are worth their solver cost.Sun, 16 Aug 2026 00:00:00 GMTdeep-learningneural-ODEcontinuous-timeSDEdynamicshttps://marketmaker.cc/blog/multi-task-learning-trading/https://marketmaker.cc/blog/multi-task-learning-trading/Does jointly predicting return, volume, and volatility actually help? Measuring loss-balancing schemes and diagnosing negative transfer through gradient cosine similarity - with a classical baseline and purged walk-forward folds.Sat, 15 Aug 2026 00:00:00 GMTdeep-learningmulti-taskshared-representationpredictionauxiliaryhttps://marketmaker.cc/blog/model-pruning-low-latency-trading/https://marketmaker.cc/blog/model-pruning-low-latency-trading/Magnitude and structured pruning, the Lottery Ticket Hypothesis, movement pruning, distillation and 2:4 sparsity - the methods behind shrinking a trading model, and what still has to be measured before any of it ships.Fri, 14 Aug 2026 00:00:00 GMTmodel-compressionpruninglottery-ticketlatencydeploymenthttps://marketmaker.cc/blog/liquidity-prediction-execution/https://marketmaker.cc/blog/liquidity-prediction-execution/Our VWAP article shipped a static, weekly-refit volume curve and called the forecaster the weakest link. This is the follow-up: a Bayesian intraday updater run against the same 500-parent BTCUSDT harness, with the IS delta conditioned on realized curve error.Thu, 13 Aug 2026 00:00:00 GMTmicrostructureliquidityexecutionpredictionorder-bookhttps://marketmaker.cc/blog/koopman-operator-market-dynamics/https://marketmaker.cc/blog/koopman-operator-market-dynamics/Dynamic Mode Decomposition fits a linear operator to nonlinear market dynamics. The only question that matters: do the fitted modes persist from one window to the next, and does the rolling spectral radius lead realised volatility? Here is the measurement protocol and the code to run it.Wed, 12 Aug 2026 00:00:00 GMTmathematicsKoopmandynamical-systemsspectralpredictionhttps://marketmaker.cc/blog/knowledge-distillation-trading-models/https://marketmaker.cc/blog/knowledge-distillation-trading-models/The blog's standing answer to the accuracy-vs-latency tension is a two-stage fast/slow split. Distillation is a different answer: train one small model to mimic the ensemble. The KD loss, temperature, born-again nets, early exits for a variable latency budget, and the distill-to-FPGA pipeline - plus the measurements that would decide whether it beats the two-stage split.Tue, 11 Aug 2026 00:00:00 GMTmodel-compressiondistillationlatencyHFTdeploymenthttps://marketmaker.cc/blog/instrumental-variables-quantitative-finance/https://marketmaker.cc/blog/instrumental-variables-quantitative-finance/Selection bias in the search is not the only way a regression fools you. When the regressor is correlated with the error, more data makes the estimate more confidently wrong. Instrumental variables applied to the one endogeneity problem this blog has already left open: permanent impact from net taker flow.Mon, 10 Aug 2026 00:00:00 GMTcausal-inferenceinstrumental-variables2SLSeconometricsendogeneityhttps://marketmaker.cc/blog/hawkes-process-order-arrival/https://marketmaker.cc/blog/hawkes-process-order-arrival/Fitting a self-exciting point process to real crypto trade tape: where the three numbers (mu, alpha, beta) come from, how to estimate the branching ratio n, whether the exponential kernel survives a goodness-of-fit test, and how much n moves when you change the estimation window.Sun, 09 Aug 2026 00:00:00 GMTmicrostructureHawkes-processpoint-processorder-arrivalcryptoHFThttps://marketmaker.cc/blog/hamiltonian-neural-networks-finance/https://marketmaker.cc/blog/hamiltonian-neural-networks-finance/Hamiltonian Neural Networks are provably stable - but stability is worthless if the conserved quantity does not exist. Learning a scalar H via autograd, symplectic integration, and the falsification test that decides whether a financial (q, p) pair is canonical at all.Sat, 08 Aug 2026 00:00:00 GMTdeep-learningHamiltonianphysics-informeddynamicsconservationhttps://marketmaker.cc/blog/granger-causality-lead-lag-crypto/https://marketmaker.cc/blog/granger-causality-lead-lag-crypto/Granger causality on crypto prices done two ways - differenced returns and Toda-Yamamoto on levels - with a correct Wald implementation, an effective-N corrected causality matrix, and a rolling-stability test of whether the lag is tradeable at all.Fri, 07 Aug 2026 00:00:00 GMTcausal-inferenceGranger-causalitylead-lagcryptoVARhttps://marketmaker.cc/blog/gaussian-processes-trading/https://marketmaker.cc/blog/gaussian-processes-trading/Kernel design for financial time series - Matern roughness, locally periodic composition, spectral mixtures - and the marginal likelihood as a regularizer that needs no validation set. Plus the honest list of what still has to be measured before any of it is tradeable.Thu, 06 Aug 2026 00:00:00 GMTbayesiangaussian-processkerneluncertaintynon-parametrichttps://marketmaker.cc/blog/fourier-neural-operator-finance/https://marketmaker.cc/blog/fourier-neural-operator-finance/Operator learning maps whole function spaces, not points - how the Fourier Neural Operator parameterizes a PDE solution operator in frequency space, what that buys for option pricing, and which of its claims still need measuring on real hardware.Wed, 05 Aug 2026 00:00:00 GMTdeep-learningFNOPDEoperator-learningoptionshttps://marketmaker.cc/blog/flash-attention-trading-models/https://marketmaker.cc/blog/flash-attention-trading-models/Flash Attention makes a 23,400-step trading context computationally free - tiling, online softmax, and an IO bound of N^2 d^2 / M. Whether that longer context makes the model better is a separate question, and it has to be measured, not asserted.Tue, 04 Aug 2026 00:00:00 GMTdeep-learningattentionFlash-AttentionGPUoptimizationhttps://marketmaker.cc/blog/ensemble-methods-trading/https://marketmaker.cc/blog/ensemble-methods-trading/Three things about trading ensembles this blog has not covered: why the covariance term - not model count - dominates ensemble error, how stacking works as a signal-combination layer with out-of-fold meta-features, and whether turnover netting across many signals actually reduces execution cost.Mon, 03 Aug 2026 00:00:00 GMTmachine-learningensemblebaggingstackingalphahttps://marketmaker.cc/blog/double-ml-causal-trading/https://marketmaker.cc/blog/double-ml-causal-trading/Every model on this blog so far answers 'what predicts what?'. Double ML answers 'what causes what?' - with a standard error you can defend. The partially linear model, Neyman orthogonality, purged cross-fitting on order book data, and an honest account of why a valid DML confidence interval survives exactly one pre-specified question.Sun, 02 Aug 2026 00:00:00 GMTcausal-inferencedouble-MLtreatment-effecteconometricsquanthttps://marketmaker.cc/blog/causal-forests-treatment-effects-trading/https://marketmaker.cc/blog/causal-forests-treatment-effects-trading/Every backtest on this blog estimates a conditional mean. Causal forests estimate a conditional treatment effect instead - tau(x) rather than mu(x) - with honest splitting, an adaptive-kernel weight representation, and a calibration test that tells you whether the heterogeneity you found is real.Sat, 01 Aug 2026 00:00:00 GMTcausal-inferencecausal-forestheterogeneous-effectseconometricstreatmenthttps://marketmaker.cc/blog/bayesian-neural-networks-trading/https://marketmaker.cc/blog/bayesian-neural-networks-trading/Every sizing rule on this blog treats uncertainty as one number. MC Dropout and deep ensembles split it into model ignorance and market noise - and those two deserve different position sizes.Fri, 31 Jul 2026 00:00:00 GMTbayesianuncertaintyneural-networkriskposition-sizinghttps://marketmaker.cc/blog/automl-systematic-trading/https://marketmaker.cc/blog/automl-systematic-trading/Automated feature generation (tsfresh, Featuretools), budget-aware model search (FLAML's cost-frugal optimizer), and the WorldQuant formulaic alpha factory - the parts of the research pipeline this blog's search-and-overfit arc never covered, and what the arc's own results say about them.Thu, 30 Jul 2026 00:00:00 GMTAutoMLNASfeature-engineeringautomationquanthttps://marketmaker.cc/blog/liquidation-cascades-trading-signal/https://marketmaker.cc/blog/liquidation-cascades-trading-signal/Every leveraged position advertises the price at which it must be sold. How to build the on-chain liquidation depth chart and the CEX liquidation heatmap, model cascade dynamics as a reproduction number, and trade forced flow as a signal instead of only fearing it as a risk.Wed, 29 Jul 2026 00:00:00 GMTliquidationscascadetrading signaldefiperpetualsorder bookmarket microstructurecryptohttps://marketmaker.cc/blog/uniswap-v3-lp-strategies-hedging/https://marketmaker.cc/blog/uniswap-v3-lp-strategies-hedging/An active v3 LP is running a market-making book with gas costs and no cancel button. Size ranges from a GARCH vol forecast, frame rebalancing as fee-gain vs realized-cost, derive the position delta from tick math and hedge it with perps, and understand JIT liquidity and the honest pitfalls of backtesting LP from on-chain data.Tue, 28 Jul 2026 00:00:00 GMTuniswapdefimarket makingdelta hedginglvrjit liquidityquantethereumhttps://marketmaker.cc/blog/mev-supply-chain-pbs-mevboost/https://marketmaker.cc/blog/mev-supply-chain-pbs-mevboost/MEV stopped being a lone bot and became an assembly line: searcher, builder, relay, proposer. A technical walk through proposer-builder separation, MEV-Boost's sealed-bid block auction, why the searcher's margin gets bid away, order-flow auctions as the new moat, and how Solana's Jito model differs.Mon, 27 Jul 2026 00:00:00 GMTmevpbsmev-boostflashbotsjitoorder flow auctiondefiethereum