We combine historical returns, real-time data and machine learning into recommendations you can act on — without having to learn the models yourself. Sterk Sparingsel makes the complex understandable, one decision at a time.
Start your analysisOur platform is based on three principles that together will make trading decisions more robust: ongoing analysis, testing against historical data, and conscious risk management.
Our models retrieve and structure large amounts of market data continuously, so that patterns and deviations are captured while they are still relevant. The advantage of real-time is not speed for speed's sake — it's that your decision is based on the state the market is actually in now, not yesterday.
Before a strategy is recommended, it is tested against long time series of historical data from different market conditions. This gives a realistic expectation of how the strategy could have worked in the past, and contributes to the recommendations being rooted in actual patterns rather than assumptions.
Each recommendation is followed by an assessment of downside risk and position size. The aim is not to eliminate risk — that is impossible in trading — but to make it visible and manageable, so that decisions are made with open eyes rather than on gut feeling.
We believe a model is more trustworthy when the process behind it is explained, not hidden. Below are the four steps from collecting data to a finished strategy.
Market data, volumes and relevant macro factors are collected from established sources and updated continuously throughout the trading day.
The models identify statistical relationships and deviations in the data, regardless of market direction or sentiment.
Findings are tested against historical periods to assess how stable the patterns are over time, not just in a single section.
The result is presented as a concrete recommendation with associated risk assessment, ready for you to assess.
Transparency about data sources is part of our promise. We use publicly available market data and established financial data sources, and always state the time period on which an analysis is based.
The predictive models adapt to your time horizon. Below you can see how three types of investors typically use the platform in practice.
Day traders use the real-time models to identify short-term deviations and volume changes throughout the trading day, with risk parameters updated hourly.
For swing traders, emphasis is placed on patterns that extend over several days, combined with backtested history to assess the likely duration of a movement.
Long-term investors use the platform to assess risk exposure over time and test how a portfolio has historically responded to various market conditions.
The dashboard is put together to show what is relevant then and there, without unnecessary alerts or distractions. We have emphasized a calm graphic expression, so that your attention is on the data and not on flat effects.
Navigation and visualizations are simplified so that both new and experienced traders find relevant information quickly.
If necessary, you can move on from a recommendation and see which data points and time periods are the basis for it.
We answer directly what we are often asked about, without promising more than the platform actually delivers.
Market data is processed continuously throughout the trading day, but there will always be a slight delay from the data source to display in the interface. This delay varies with data source and market type, and we always state which period an analysis is based on.
We work with API integrations against established trading platforms. The scope of available integrations varies, and we recommend getting in touch to clarify which systems are relevant to you before you start.
No model can guarantee future returns. What we can say is that each recommendation is tested against historical data from multiple market conditions, and that the results of this testing are available for review before you make a decision.
We use established and publicly available financial data sources, and have routines for checking deviations and missing data before they enter the models. Data quality is an ongoing task, not a one-off check.
We are happy to discuss how the models can be set up for your time horizon and risk profile.