Articles on markets and machines
The I-Trade quant team — on how algorithmic trading, risk and returns work. Research, mechanics breakdowns and engineering post-mortems. No promises of "multiples".
How we train our models not to trade
The best trade is the one that never happened. We explain why our models reject four out of five signals and how the confidence threshold protects returns from market noise.
Drawdown — the price of returns
Any returns above a deposit come at the cost of periods when the portfolio is worth less than the day before. We break down how to read maximum drawdown and choose a strategy that matches your risk appetite.
I-Trade Chain: a year in production
184 million trades, 99.99% uptime, and a few unwelcome surprises. A review of the first year of the public network where algorithms record every trade.
Perpetual funding explained in plain terms
A perpetual futures contract has no expiry date — the funding rate keeps its price anchored to spot. We explain the mechanism in plain terms and show how market-neutral strategies are built on it.
What we write about
Research
How our models work under the hood: confidence thresholds, ensembles, reasons for rejecting trades. No formulas for their own sake — just actionable conclusions. Start with "How We Train Models Not to Trade".
Education
Market mechanics in plain language: drawdown, funding in perpetuals, fees, rebalancing. Content after which the page strategies reads like an open book.
Engineering
What's under the platform's hood: a trading core with a median latency of 0.4 ms, a public network I-Trade Chain and lessons from a year in production — 184 million trades on-chain.