About KruegerAlgorithms.

I work as a systematic trading researcher and algorithm engineer — guided by a principle that makes this field uncomfortable: every strategy claim must be falsifiable before it is worth anything. That is why my research is open: the studies on this site are fully readable, including the findings that did not work.

The backbone is a five-phase validation workflow: hypothesis → edge construction → validation gate (out-of-sample, walk-forward, Monte Carlo bootstrap, deflated Sharpe) → position sizing → live deployment. Costs are baked into every number from the first metric, and an edge has to hold cross-asset, not on one curve-fit market. Most candidates die in this funnel — that is its purpose, and that gets published too.

The current research agenda: intraday breakout structure across ten markets and eleven years of minute data, dealer positioning (gamma regimes) as a context layer, and exit management framed as a reinforcement learning problem on more than 120,000 historical trade episodes. The execution side — MQL5, Pine, Python, C# — is a tool, not an identity: it translates validated rules into code without loss.

The absence of return promises here has a technical reason, not a legal one: markets are non-stationary and edges measurably decay — my own research documents both. Hiding that is selling. Measuring and pricing it in is how you build systems worth trusting with capital.

My location in Konstanz is directly on the Swiss border at Lake Constance — in the heart of the three-country corner Germany · Switzerland · Austria. This central location enables close collaboration with customers from the entire DACH region.

The proximity to Swiss financial centers such as Zurich (60 minutes) and Zug creates ideal conditions for personal meetings and close project support. At the same time, I benefit from the central location in Europe with excellent accessibility.