Table of Contents
In this interview, algorithmic trader Pavel Kycek sits down in Prague to unpack 18 years of market experience—transitioning from discretionary starts to fully systematic stock and crypto trading for his own account and for institutional/retail clients. Pavel explains why he ditched chart patterns and gut feel for research-led, rules-based execution, and how a lean stack of simple ideas (two–three conditions per strategy) scales when you think in portfolios, not single trades.
You’ll learn how Pavel builds robust, idea-first strategies that work across assets with adjusted exits, why he pairs momentum longs with mean-reversion shorts to balance exposure, and how he manages ~15 live systems plus an “incubation” bench using benchmarks and environment checks. He breaks down time-based exits, smarter stop-loss logic, tiny per-position sizing in crypto, and the mindset needed to survive drawdowns—all in a beginner-friendly way you can apply today.
Pavel Kycek Playbook & Strategy: How He Actually Trades
Core Philosophy: Keep Each Strategy Simple, Let the Portfolio Do the Heavy Lifting
Pavel Kycek builds strategies that are easy to understand on their own and sophisticated in combination. He uses an “idea-first” approach, defining what the edge should be and when it should and shouldn’t work, then keeps each system to just a couple of moving parts.
- Limit each strategy to 2–3 clear conditions; avoid piling on filters.
- Define the market regime where the edge should win and where it should struggle—expect some strategies to fail and plan around it.
- Focus complexity at the portfolio level, not the single-system level, to reduce over-optimization risk.
Portfolio Construction: 15 Live Systems, Benchmarks, and an Incubation Bench
He runs a diversified stable of strategies and manages risk from the top down. Live systems are compared to benchmarks and to current market environments; new ideas sit in “incubation” before going live.
- Target ~15 concurrently traded strategies; keep any single system to ~10–20% of portfolio impact.
- Maintain an incubation list running on your own capital before client or full-size deployment.
- Build regime-specific benchmarks (breakout, momentum, trend-following, mean-reversion) and compare each live system to its benchmark.
- Audit fit-to-environment continuously; if a strategy isn’t making money when its regime is favorable, dig deeper.
Markets & Timeframes: Crypto Automation, Stocks at the Close
Pavel trades crypto and stocks systematically, but execution differs. Stocks are scanned and placed around the daily close, while crypto requires automation to handle 24/7 bursts.
- For stocks: run daily scans near the close, place orders mechanically; keep the workflow to minutes, not hours.
- For crypto: automate entries/exits to capture fast moves; manual execution won’t scale.
- Aim to support dozens of small positions simultaneously (50–70 when needed) with portfolio tooling.
Entries: Momentum Longs, Mean-Reversion Shorts, and Simple Breakouts
Entry logic stays simple and testable. He blends momentum and mean-reversion concepts to balance exposures across regimes.
- Momentum systems: enter on strength; avoid profit targets (let trends run) and plan exits separately.
- Short-term breakouts: use daily timeframe triggers; design to catch the initial impulse only.
- Mean-reversion in crypto: cut size and hold time aggressively; think 1–2 days max if the bounce doesn’t show.
Exits: Time Stops, Close-Based Logic, and Context-Aware Conditions
He favors exits that reflect how markets actually move. Instead of tight hard stops or fixed targets, he uses time-based exits and close-based rules that statistically improve outcomes.
- Prefer closing-bar exits over intrabar hard stops to reduce fake-outs.
- Use time-stops (e.g., 1–3 days for short breakouts; ~6 days for momentum holds) depending on entry logic.
- Add simple “give-back” conditions (e.g., close back below/above prior day’s close after a breakout) to exit stretched moves.
- For mean-reversion systems, profit targets are acceptable; for momentum, skip targets and manage with time/condition exits.
Risk & Sizing: Tiny Per-Position Exposure, Hedge via Strategy Mix
Risk is constrained at the position level and diversified via the portfolio. In crypto, he intentionally sizes very small and pairs opposing systems to dampen swings.
- Cap each crypto position at ~1–2% of capital; rely on breadth, not size, for returns.
- Hedge momentum exposure with mean-reversion systems running simultaneously.
- Expect pump-and-dump dynamics in small coins; reduce weights and minimize time-in-trade.
Filters & Overfitting: Resist the Urge to Over-Engineer
He uses zero or one filter for most systems and accepts that some will go sideways for long stretches. The point is robustness across a portfolio, not perfect backtests.
- Use at most one filter in a minority of strategies; let the core idea carry the edge.
- Avoid optimizing to bus-data “perfection”; tolerate 1–2 years of chop if the system fills a portfolio role.
- Judge strategies by how they complement the whole, not by isolated metrics.
Regime Awareness: Know When a Strategy Should Get Paid
He constantly cross-checks live performance against the regime each system targets. If the market is ideal and the system still lags, something’s off.
- Define “green-light” regimes for each system (e.g., up-trend for long trend-followers) and track them daily.
- If returns diverge from expectations in a favorable regime, investigate and be ready to adjust or replace.
- Maintain readiness to pause or rotate systems if the environment changes materially.
Practical Daily Workflow: What to Do Each Day
His routine is light but deliberate: clarify regime, run scans/automation, and manage exits by rules. The emphasis is on consistency and portfolio-level decisions.
- Before the session: mark regime state and benchmark conditions relevant to your systems.
- During the session: let automation execute; avoid discretionary overrides unless a rule explicitly allows it.
- End of day: apply close-based exits, refresh time-stops, and rebalance exposures across strategies.
Origin Story & Mindset: From Discretionary Stress to Systematic Clarity
Pavel moved from discretionary day trading to fully systematic methods to reduce psychological strain and support family life. The mindset shift is the foundation of his process.
- Prefer rules over reactions; design processes that survive distractions and life events.
- Start with longer-term, rotation-style systems if needed, then evolve into fully systematic trading.
- Ditch pattern-chasing if it isn’t working; move to research-backed, testable ideas.
Research Prompts: What to Test Next
He often frames ideas with simple statistical questions about movement versus averages and MA context. Use these prompts to spark robust, testable edges.
- “What happens after a move that’s ~2× the long-term average?”
- “How does behavior change when price is above/below a key moving average?”
- “If a breakout closes back beyond yesterday’s close, does the edge decay?”
Build Simple Rules, Diversify Systems, Let the Portfolio Do Heavy Lifting
Pavel Kycek keeps each strategy brutally simple—two or three conditions, tops—then stacks multiple systems so the portfolio, not any single idea, carries the load. He cares less about finding a “perfect” setup and more about combining uncorrelated edges that earn in different weather. When one system stalls, another is designed to step up, so the equity curve depends on the mix. This approach trims overfitting and makes rules easier to follow on busy days.
In practice, Pavel Kycek defines what a strategy is supposed to capture—momentum burst, mean-reversion pop, or breakout impulse—then refuses to bolt on extra filters just to juice a backtest. He runs several small, clean systems at once and sizes them modestly, accepting that some will be cold while others are hot. Over time, the diversified engine smooths drawdowns and keeps him executing consistently. That consistency—not prediction—is the edge.
Size Tiny Per Trade, Scale Breadth, Survive Crypto-Level Volatility
Pavel Kycek treats position size like a blast shield: tiny per-trade risk so a single candle can’t wreck the day. Instead of swinging big on “A+” setups, he spreads small bets across many independent names and systems, letting breadth—not bravado—do the compounding. This is especially critical in crypto, where 5–10% intraday swings are just Tuesday, and slippage punishes oversized orders.
To apply it, Pavel Kycek caps risk per position at a modest fraction of equity, then limits correlated exposure so several coins or stocks in the same theme don’t move him as one. He scales the number of concurrent positions only as far as his execution and monitoring can stay flawless, preferring many tiny positions to a few large ones. When volatility spikes, he reduces unit size, shortens holding periods, and favors close-based exits over intrabar stops that can get whipped. The result is a portfolio that absorbs the wild days and keeps him trading tomorrow.
Pair Momentum Longs With Mean-Reversion Shorts To Balance Regimes
Pavel Kycek runs momentum longs to harvest strength when markets trend, then offsets that exposure with mean-reversion shorts that pick at overextensions when things snap back. The point isn’t to predict tomorrow’s weather—it’s to be paid in whichever regime actually shows up. When trends persist, the long side pulls; when the tape chops or fades, short mean-reversion cushions the hits. This pairing reduces reliance on any one edge and smooths the path of returns.
Operationally, Pavel Kycek keeps the rules clean: momentum longs enter on strength and avoid profit targets, using time-based or close-based exits to let winners breathe. Mean-reversion shorts enter after stretched moves into resistance or outside recent ranges, then aim for quick give-backs with tight time stops. He sizes both sides small and caps correlation so a single theme can’t dominate. With that structure, the portfolio doesn’t care whether the day trends or whipsaws—it has a rule-set ready for either.
Use Time-Based Exits And Close Logic, Not Fragile Hard Stops
Pavel Kycek favors exits that happen on the close or after a fixed number of bars because they reflect how trades statistically resolve, not how our nerves feel mid-candle. He’s seen tight intrabar stops get nicked by noise, only for the price to rip in the intended direction minutes later. By anchoring decisions to the close, he reduces fake-outs and keeps behavior consistent across markets and timeframes. Time-based exits also make testing honest: either the trade proved itself within the window, or it didn’t.
In practice, Pavel Kycek sets simple clocks—e.g., exit momentum positions after a set number of sessions unless a clear continuation condition is met, and cut mean-reversion trades quickly if the bounce doesn’t show. He uses give-back rules at the close (like “if price closes back through yesterday’s level, exit”) instead of arbitrary ticks that get whipped. Hard stops are still a last line of defense, but they’re placed wide and rarely touched; the real control comes from the clock and the close. This framework strips out drama, keeps execution mechanical, and protects the equity curve from death by a thousand cuts.
Test Ideas Against Benchmarks; Reward Process Discipline Over Prediction
Pavel Kycek judges every strategy against a simple, purpose-built benchmark so he knows if the idea is actually delivering edge or just noise. If a breakout system can’t beat a dumb breakout proxy during a friendly regime, he parks it and fixes the logic instead of forcing trades. He cares less about forecasting the next move and more about whether the rules produce repeatable, net-positive behavior over many samples. That mindset keeps him from chasing narratives and keeps capital aligned with what’s statistically working.
Day to day, Pavel Kycek maps each live system to its “should-get-paid” environment and checks performance relative to that backdrop. When results diverge, he investigates quickly: execution issues, parameter drift, or regime shifts that the rules don’t recognize. He rewards consistency by continuing to allocate to systems that follow rules cleanly, even through short-term chop, and he cuts allocation to rule-breaking ideas even if they just printed a lucky win. In his world, process discipline compounds; prediction just gets graded by the benchmark anyway.
In the end, Pavel Kycek’s edge isn’t a single magic setup—it’s a way of building and running a machine that keeps paying across changing markets. He keeps every strategy brutally simple, then lets diversification at the portfolio level do the heavy lifting. Momentum longs are allowed to breathe; mean-reversion shorts clip overextensions; and each system has a clear reason to exist. Position sizes stay tiny, so a bad candle can’t derail the week, and he scales with breadth, not hero trades.
Pavel Kycek’s exits are designed for reality, not nerves: close-based decisions and time stops reduce fake-outs, while give-back rules pull him out when the edge decays. He constantly checks whether a system is getting paid in the regime it was built for, comparing live results to straightforward benchmarks. If the numbers don’t line up, he fixes or shelves the idea—no narrative rescue missions. Day to day, workflow stays light: automate where speed matters (crypto), make close-time decisions for stocks, and maintain an incubation bench so new ideas prove themselves before earning capital. Taken together, the lesson is simple but powerful: build rules you can actually follow, measure them honestly, and let disciplined process—not prediction—compound.
