Technical analysis becomes AI-driven market forecasting
Our forecasting engine combines trend, momentum, volatility, volume, market structure and relative strength into daily probabilities for stocks, indices, commodities and cryptocurrencies. No single indicator decides alone — what matters is the combination of many signals across multiple time horizons.
Trend & moving averages
Identify direction, evaluate trend quality and reduce market noise
Trend indicators help the AI distinguish short-term noise from more durable price movement. Short-, mid- and long-term trend windows are combined to improve the interpretation of UP and DOWN signals and to separate trend continuation, reversals and overextended moves.
Price data & extremes
Open, high, low and close are the foundation of technical analysis. Rolling highs and lows show whether an asset is building strength, testing important areas or remaining in a weaker structure.
SMA, EMA & EWM
Moving averages smooth price action and make direction and trend stability easier to read. Faster averages react earlier, slower averages show the broader structure.
MACD, PPO & trend dynamics
MACD and PPO capture whether a trend is gaining or losing momentum. This helps the model distinguish trend-aligned acceleration from fading strength and possible reversal points.
Crossovers & trend changes
Crossovers between fast and slow trend lines indicate possible new market phases. The model evaluates not only the event itself but also how long the structure has persisted.
Momentum & oscillators
Detect acceleration, strength and overbought or oversold market phases
Momentum indicators show how strong a move is and whether a market is entering an extreme zone. They help the AI separate strong trends from short-term overextensions and detect possible turning points earlier.
RSI & stochastic
These oscillators show whether price action appears strong, extended or exhausted. Combined with trend data, the model can judge whether a high reading confirms strength or signals pullback risk.
ROC, CCI & Williams %R
These indicators measure speed and position within the current trading range. They are especially useful for evaluating fast impulses, capitulation moves or overheated trend phases.
ADX & directional dominance
ADX helps detect whether a market is truly trending or only moving sideways. This improves signal interpretation because momentum in a clear trend behaves differently from momentum in choppy ranges.
Volatility & price bands
Evaluate fluctuation, risk zones and realistic movement ranges
Volatility features help the model estimate typical move size and whether a signal emerges in a calm or risky environment. Price bands show when prices move unusually far or break out of compressed zones.
ATR & typical move size
ATR describes how strongly an asset typically fluctuates. This helps evaluate signals more realistically and interpret risk/upside ranges more effectively.
Bollinger, Keltner & Donchian
Bands and channels show whether price remains within normal fluctuation or moves into a breakout zone. This is especially important for breakout setups and early trend starts.
Range and event volatility
High/low-based volatility detects turbulence that close-only data may miss. This helps the model separate risky market phases from calmer trend conditions.
Volume & money flow
Identify whether a move is supported by real market activity
Volume and money-flow indicators show whether a move is confirmed by rising activity. They help the AI distinguish thin price moves from trends supported by broader buying interest.
Volume & volume spikes
Rising volume can indicate institutional activity or strong market interest. This often makes breakouts and trend changes more meaningful.
Money flow, OBV & Chaikin
Money-flow signals combine price action and volume. They help detect whether capital is flowing into or out of an asset.
VWAP & volume-weighted price zones
VWAP shows where an asset traded when volume is taken into account. These zones can provide useful reference points for institutionally relevant levels.
Market structure & price levels
Capture support, resistance, breakouts and structural turning points
Structure features show where the market reacted in the past and which price areas may be relevant. This helps the AI evaluate signals in the context of support, resistance, breakouts and trend breaks.
Swing structure & break of structure
Higher highs, higher lows and breaks of structure help detect whether a trend remains intact or whether a new market phase is starting.
Order blocks & fair value gaps
Price areas with notable reactions or incomplete price discovery can mark future decision zones. The model uses these zones as context for trend continuation and reversals.
Pivot points & Fibonacci levels
Mathematically derived price levels help identify support, resistance and retracement areas where signals may become especially relevant.
Regime, returns & relative strength
Combine return path, drawdown, benchmark context and market phase
This group complements classic technical indicators with broader market and return context. The AI detects whether an asset shows relative strength, whether a pullback appears healthy or risky and whether the market environment points to trend, range or risk-off conditions.
Return lags & multi-horizon performance
Historical returns across several windows show whether an asset is short-term overextended, mid-term strong or structurally weak over the long term.
Drawdown & recovery
Drawdown and recovery features help separate normal pullbacks from deeper weakness. This is central for risk, timing and position sizing.
Relative strength, beta & correlation
Benchmark comparison shows whether an asset develops its own strength or is merely carried by the broad market. This improves the quality of trend and reversal signals.
Important: the platform does not generate signals from individual standard indicators. The model evaluates many features jointly and learns from historical patterns which combinations were statistically relevant for different forecast horizons.
Five Pillars of Model Evaluation
To make forecast quality tangible, we focus on five key metrics: ACC, BALACC, AUC, BRIER and MCC – complemented by a weighted average across all relevant horizons. Each metric highlights a different dimension of model performance.
Accuracy
Accuracy describes the share of forecasts where the model got the direction right. It is the most intuitive metric: how often the forecast was correct overall – regardless of whether the market moved up or down.
Interpretation: A solid ACC is a good starting point, but it is not sufficient on its own – especially when markets trend in one direction for extended periods.
Balanced Accuracy
Balanced accuracy gives equal importance to up and down phases. Rather than just counting all correct predictions, it separately measures performance in rising and falling regimes and averages them.
Interpretation: BALACC is especially important in markets that are one-sided for long periods. It reveals whether the model genuinely handles both directions instead of just ‘liking’ one specific environment.
AUC – Discriminative Power
AUC shows how well the model separates cases that later move up from those that move down. Instead of focusing on a single threshold, AUC evaluates the full distribution of model scores.
Interpretation: High AUC values mean that high model scores are typically associated with better outcomes than low scores. It is ideal for judging the quality of ranking and selection strategies.
Brier Score – Probability Quality
The Brier score measures how well predicted probabilities match reality. It does not just look at whether an event occurred, but how far the predicted probability deviated from the actual outcome.
Interpretation: The lower the Brier score, the better the calibration: a model where 70% signals actually hit roughly 70% of the time has high probabilistic quality.
MCC – Matthews Correlation
The Matthews correlation coefficient condenses all entries of the confusion matrix into a single robust value. It takes correct and incorrect signals in both directions into account and is sensitive to class imbalance.
Interpretation: MCC is excellent for fair comparison of models across markets and time periods. Values near 0 indicate randomness, while positive values point to genuine information content in the signal.
Weighted Average Across Horizons
Instead of focusing on a single horizon, multiple horizons are combined into one overall score. Short-term and longer-term forecasts can be weighted differently, depending on strategy and usage profile.
Interpretation: Weighted average metrics help summarise a model with a single number without losing the depth of individual horizons. They are particularly useful for comparing models or markets.
Together, these metrics paint a clear picture: how often the model is right, how fairly it treats both directions, how well it separates strong from weak signals, and how reliable its probabilities really are.
FAQ
Answers to the most common questions about our technology, probabilities, metrics and how to interpret the stock forecasts.
What technology powers your stock predictions?+
Our forecasts are powered by an ensemble of modern machine-learning models specifically tuned for time-dependent financial data. The models process hundreds of technical indicators, market structure signals and volume information in parallel. Training is highly compute-intensive and runs on GPU-accelerated cloud infrastructure. Models are typically retrained or updated every night once new end-of-day data is available so the system can continuously adapt to current market conditions.
What historical data do you use for the forecasts?+
We work with high-quality historical market data, including closing prices, open, high and low prices, daily trading volume, and derived quantities such as returns, volatility and a wide range of technical indicators. Data is typically processed on an end-of-day basis after the market close, then cleaned and normalised. This ensures the models are built on consistent time series without distortions from outliers, splits or bad ticks.
What does the displayed probability mean and why are signals only shown above 70% or below 30%?+
The probability indicates how likely, according to the model, the price is to rise (UP) or fall (DOWN) over a defined horizon. For example, a value of 72% means that in historically similar situations the market went up in roughly 72% of cases. To avoid noise, we only show signals when probabilities are clearly elevated or clearly reduced, typically ≥ 70% for upward moves or ≤ 30% for downward moves. In the neutral zone between these thresholds the market is statistically hard to distinguish, so we deliberately avoid issuing signals there.
What is an UP signal and what is a DOWN signal?+
An UP signal means the model sees an elevated probability that the price will rise over the chosen horizon. A DOWN signal indicates an elevated probability of a price decrease. Both signals are purely statistical statements from technical analysis and are not direct buy or sell recommendations. They are intended as an additional building block within your own decision process, which should also consider risk profile, diversification and fundamentals.
What exactly does the forecast horizon mean?+
The forecast horizon specifies over how many trading days the model’s expectation extends. A 1D horizon refers to the next trading day, 5D to roughly one week, 20D to several weeks and 60D to a mid- to longer-term period. Horizons are based on trading days, not calendar days. Very short horizons are much noisier, while longer horizons tend to reflect more structural trends and regimes.
How should I interpret the probabilities correctly?+
Probabilities are not promises of a specific price outcome but statistical statements based on historically similar market situations. A probability of 70% does not mean the market will ‘certainly’ rise, but that it rose more often than it fell in comparable cases. Individual outcomes can always differ. Probabilities are best used as weighting or contextual input, not as guarantees.
What does it mean when different horizons show conflicting signals?+
Different horizons reflect different market dynamics. A positive 5D signal may indicate short-term momentum, while a negative 60D signal points to a broader downtrend. Such constellations are common during corrections or regime transitions. Many users use this information to adjust position size, stagger entries, or trade short-term setups more cautiously.
How often do signals occur?+
Signals are generated deliberately selectively. In calm or statistically ambiguous market phases, there may be extended periods without clear signals. During strong trends or high volatility, signals tend to appear more frequently. Exact frequency depends on the market, horizon and prevailing market dynamics.
Do the signals work in sideways markets or during crises?+
Sideways markets and extreme crisis phases are among the hardest environments for any form of forecasting. In such phases, the separation between UP and DOWN scenarios often deteriorates, which is reflected in fewer or more cautious signals. The system is designed to surface this uncertainty rather than forcing artificial clarity.
How accurate are AI-based stock predictions?+
Accuracy strongly depends on horizon, market regime and the specific stock. Financial markets always contain a large amount of randomness, especially in the short term. The goal of the models is therefore not perfect prediction, but a statistically measurable improvement over chance. What matters is whether probabilities are consistently better than neutral expectations across many cases — which is exactly what we measure with multiple metrics.
How do you prevent overfitting and unrealistic backtests?+
We strictly separate training, validation and test periods in time and rely consistently on out-of-sample evaluation. In addition, we use walk-forward testing, where models are repeatedly retrained and evaluated on subsequent, previously unseen data. This helps ensure that measured performance is not the result of hindsight optimisation but of genuine generalisation.
What do ACC, BALACC, AUC, Brier Score and MCC mean in your evaluations?+
ACC (accuracy) describes in what fraction of cases the model predicted the direction correctly – i.e. how often UP and DOWN signals turned out to be right. BALACC (balanced accuracy) gives equal weight to up and down regimes so models do not look ‘good’ simply because they favour one trend direction. AUC measures the discriminative power of the model scores: cases that later move up should, on average, receive higher scores than those that move down. The Brier score evaluates the quality of probability forecasts: the lower the score, the better the predicted probabilities match realised outcomes. MCC (Matthews correlation coefficient) condenses all entries of the confusion matrix into a single robust number and is particularly useful for fair comparison of models across markets and time periods.
What is a hit rate and how is it calculated?+
The hit rate measures how often a given signal turned out to be correct. For example, an UP signal with a high probability is counted as a ‘hit’ if the price rises over the chosen horizon, and as a ‘miss’ if it falls. Similarly, a DOWN signal must be confirmed by a subsequent price decline. The hit rate is the number of hits divided by the total number of signals. It can be computed separately for each horizon and provides an intuitive sense of how often signals have worked historically.
What does “walk-forward” (walk-forward validation) mean?+
Walk-forward validation is a continuous evaluation approach for time series. Instead of training once on the full history and evaluating ‘in hindsight’, the model is advanced through time step by step: it is trained only on data that would have been available at that point, and then tested on the subsequent period. This produces a more realistic estimate of forecast quality across changing market regimes.
What does “out-of-sample” mean for you — is it real market data?+
Out-of-sample means the evaluation is performed on periods the model had not ‘seen’ at training time. Our metrics are computed on real, continuously accruing market data as new trading days occur. This way, the numbers are not just an in-sample backtest, but an ongoing, timely measurement under real market conditions.
How do you prevent look-ahead bias and data leakage?+
With financial time series, it is critical to keep features and labels strictly separated in time. We generate all input features only from information available up to the respective market close, and evaluate via walk-forward on subsequent periods. This prevents future information from unintentionally leaking into training, normalisation, or feature computation.
How often are models updated and when are signals published?+
Forecasts are typically updated daily once new end-of-day data (after market close) is available. Signals are then ready before the next market open. This keeps analyses consistent and reproducible, relying on stable closing data rather than short-lived intraday noise.
Do you account for splits, dividends, and corporate actions?+
Yes. For robust time series, price data and derived indicators are processed so common breaks (e.g., from splits) do not create artificial jumps in features and evaluations. The goal is a consistent history so signals and metrics remain comparable over long periods.
Why do you use separate models per stock and forecast horizon?+
Stocks behave differently — liquidity, volatility, trend strength and reaction patterns can vary significantly. Also, a 1D forecast is statistically a different problem than a 60D forecast. That’s why models are trained per horizon and per asset, so forecasts better match the respective behaviour and time scale.
Why are some accuracies or metrics relatively low on certain horizons?+
Financial markets are non-stationary and influenced by many external factors such as macroeconomic releases, interest rate decisions, geopolitical events and company news. Especially on very short horizons random noise dominates, so metrics like ACC or BALACC naturally tend to sit closer to the chance level. Lower values therefore do not automatically mean the model is ‘bad’, but often that the corresponding market segment was statistically hard to forecast. It is important to interpret metrics in the context of horizon, market regime and your own strategy.
What role do lags play and why do you normalise the features?+
Lags capture how past values of an indicator influence future price movements. For example, if a specific momentum signal three days ago has historically been followed by rising prices, the model can learn this lagged relationship and adjust the current UP probability accordingly. To make such relationships comparable across stocks with different price levels, volatilities and liquidity, all features are normalised. This prevents very expensive, highly volatile or illiquid names from dominating the model and ensures that probability scales remain consistent across the universe.
Why do you use weighted averages of metrics across multiple horizons?+
Instead of focusing on a single horizon, we combine multiple horizons into an overall measure. Short-term horizons are typically noisier and more volatile, while longer horizons capture structural predictive power more reliably. In many cases, longer horizons therefore receive slightly higher weight to produce a more robust overall picture, while shorter horizons still contribute information. The result is a metric that reflects both trading and investment perspectives and is well suited for comparing models or markets.
Who are the forecasts most suitable for?+
The forecasts are designed for informed retail investors and traders seeking data-driven decision support. They are particularly suitable for users who already work with technical analysis, fundamentals or systematic approaches and want to integrate probabilities into their own strategy.
Are the forecasts investment advice or a buy/sell recommendation?+
No. Our forecasts do not constitute investment advice and do not replace personalised guidance from qualified professionals. They are automated, technically-oriented evaluations of historical price and indicator data. Your personal financial situation, risk profile and objectives are not taken into account. You should therefore treat the information as one building block within a broader decision and risk management framework.
How can I use the signals for short-term trades?+
For short-term setups, the 5D and 20D horizons are often more useful because they are less dominated by noise than pure 1D moves. In practice, signals work best as an additional filter — for example to confirm your own setup based on trend, support/resistance, volume, or a news context. It also helps to consider fundamentals and events such as earnings, and to manage risk via position sizing, stop levels, and diversification. Signals are statistical probabilities and should never be used as the sole decision basis.
How can I use the signals for mid- to long-term investing?+
For mid- to longer-term decisions, the 20D and 60D signals are often most relevant because they better reflect trend and regime phases. Many users apply these horizons to time entries and adds, to scale into positions, or to reduce risk during weaker phases. The most robust approach is combining signals with fundamentals (quality, valuation, growth) and a clear portfolio framework. Again, signals are not buy/sell recommendations — they are a data-driven complement to your own strategy.
How can I try the platform and what does it cost?+
We plan to offer new users a free one-month trial so you can explore the core forecasting features without any risk. After that, several options are conceivable, such as an inexpensive day pass for occasional access or a monthly / yearly subscription for regular users. Details of the final pricing model will be communicated transparently in the interface. In any case, the core goal is to provide high-quality, AI-powered technical analysis at a fair price.