What is the Predictability Score?

The Predictability Score is a number from 0% to 100% that measures the **consistency** of a process, not just its average or spread.

• A **high score (e.g., 95%)** means the data points are stable and behave as expected. The process is "in control."
• A **low score (e.g., 40%)** means the data is chaotic and unpredictable. The process is "drifting" or "out of control."

Key Idea: Standard deviation tells you how wide the data is. We tell you how well-behaved it is. A stock that goes up exactly $1 every day is perfectly predictable (100%), even though its price is changing.

Getting Your First Score

1. Enter Data: Paste a list of numbers into the main text box. They can be separated by commas, spaces, or new lines. Example: 10, 12, 15, 11, 13
2. Calculate: Click the "Calculate Score" button.
3. View Results: You'll instantly see the score, a chart of your data, and a dashed line representing the average.

Advanced Analysis Tools

The K-Factor: Tuning Sensitivity

The "Sensitivity Mode (k)" is your most powerful tuning knob. It tells the algorithm how much to punish inconsistency.

• Sports (k=0.5): Forgiving. Good for player stats where a single bad game shouldn't tank the score.
• Standard (k=1.0): Balanced. The default for most use cases.
• Finance (k=2.0): Strict. Penalizes stock or crypto volatility more heavily.
• Pharma (k=15.0): Extremely strict. For high-precision industrial processes where tiny deviations matter.

Target Calibration: Stability vs. Accuracy

If your process has a specific goal (e.g., a machine must produce parts at 500g), enter it in the "Target" box. The tool will calculate two things:

1. Predictability Score: Is your process stable?
2. Target Deviation: Is your stable process accurate, or is it consistently missing the target?

Sliding Window Mode: Finding *When* Things Broke

This is our most advanced feature. Instead of one score for the whole dataset, it calculates a score for a moving "window" of data. The resulting chart shows you the **evolution of predictability over time.**

• Purple Line: High predictability. The process is stable and "in control."
• Yellow Line: Moderate drift. The process is becoming less stable.
• Red Line: Critical drift. The process is chaotic and unpredictable.

The Full Data Workflow: CSV Import & Export

You can analyze large datasets without ever leaving the dashboard. This is the ultimate workflow for professionals.

1. Import: Click "Choose File" and select a .csv or .txt file. A preview of your data will appear.
2. Select: Click on the header of the column you want to analyze. The data is instantly loaded into the calculator.
3. Analyze: Run your desired analysis (Standard, Sliding Window, etc.).
4. Save & Name: Give your analysis a descriptive name (e.g., "NVDA - 90 Day Stability") and click "Save".
5. Export: Click the "Export" button to download the current analysis (data, score, and notes) as a clean .csv file for your records or for use in other tools like Excel or Google Sheets.
6. Share: Click the "Share 📷" button to download a clean, white-background PNG of your chart, perfect for reports, presentations, or social media posts.

Real-World Use Cases