I didn't come to systematic trading by vocation, but by elimination. I know the emotions markets can trigger — euphoria, fear, impatience, the urge to "win it back" — because I've felt every one of them first-hand, and paid for each. At some point I stopped asking how to control them and changed the question: how do I take them out of the equation?
The answer was to take myself out of the decisions. Since then every trading choice comes from a written rule, tested on data and applied without exception. If a rule can't be tested, as far as I'm concerned it doesn't exist.
I work almost exclusively in the US markets: for efficiency, costs and liquidity — especially in options — they have no rivals. I look for statistical edges and structural inefficiencies, when they truly exist; and when they stop existing, it's the data that says so, not my opinion.
The method is always the same, whatever the instrument. First, a hypothesis that can be measured. Then a backtest on long historical series, with realistic costs and slippage, looking at the drawdown before the return. Then a check that the logic holds outside the sample it was born on. Only at the end the live market, with my own capital: it's the only test that counts, and that's where every strategy has to prove it holds up. I'm someone who gets his hands dirty, for better or worse.
I work along three lines. Automated trading systems on stocks and futures, developed in EasyLanguage (TradeStation) and PowerLanguage (MultiCharts). Options strategies on SPX and XSP, always with defined risk, where the maximum loss is written in the strikes before entering: structures that collect from time, volatility and the market's most frequent moves, not bets on tomorrow's direction. And quantitative analysis in Python: in recent years — with the now-decisive help of artificial intelligence — I've built dozens of tools for working with large amounts of data: backtesting, seasonality, volatility, market breadth, dealer positioning in options. They are the same tools that power the systems of the model portfolio.
Kriterion Quant is where all of this becomes public and verifiable.
The signals service. Three quantitative systems on US large-cap stocks — a Donchian-style breakout, a Z-Score mean reversion and an intraweek bias — running in a model portfolio with real-time notifications. They are the same three systems I execute on my own account. Every trade, open or closed, is visible on the public dashboard, drawdowns included.
The research. In-depth articles on the blog, weekly market analysis reports, videos on the YouTube channel and the Quantcast podcast. No motivational content: numbers, method and code.
Training and consulting. In-person courses on operational finance — from ETF investing to advanced portfolios, from signal-based trading to options — and EasyLanguage strategies developed on request.
Promises of easy profits, doctored screenshots, untested theories, backtests without costs, a "secret method" that only works in the slides.
A track record you can verify at any time on the live dashboard; the methodology behind every signal, explained; drawdowns shown before returns; a person who answers questions personally.
I live and work in Sofia, Bulgaria, where the company behind Kriterion Quant is based. I trade the American markets, so my working day follows New York hours.
The model portfolio's track record is public and always up to date. For any question, write to me: I answer personally.