Free — no account required

How Good Is Your Classifier? ROC, AUC & the Best Cutoff

Upload a CSV with a true yes/no outcome and a model score, and get the full ROC curve, AUC with a confidence interval, and the optimal decision cutoff — sensitivity, specificity, and all. Free.

24,000+ analyses run
Encrypted & deleted in 7 days
PDF & citation included

Free analyses run on up to 10,000 rows. Larger files are randomly sampled to that size — sign up to analyze your full dataset.

📊
-
Rows
-
Columns
-
Numeric

Running roc curve & classifier cutoff analysis...

Drawing the ROC curve and computing AUC...

Your report is ready

Sent to — ROC curve, AUC with a 95% confidence interval, the Youden-optimal cutoff with sensitivity and specificity, a threshold sweep, R code, and AI insights.

Analyze another file
Sample Output

Every report includes interactive charts, tables, and AI insights

Upload your data to get your own report

View all case studies See all free tools

How it works

The analysis treats one class as positive, then sweeps every candidate cutoff on your score to trace the ROC curve — true positive rate against false positive rate. The area under that curve (AUC) is the probability a random positive scores higher than a random negative; it is reported with a Hanley-McNeil 95% confidence interval and cross-checked against the curve's own trapezoidal area. Youden's J picks the cutoff that maximizes sensitivity plus specificity, and the report gives sensitivity, specificity, PPV, NPV, accuracy, and the confusion counts at that cutoff.

Use it when you have a real yes/no outcome and a continuous score or probability, and you want to know both how good the score is and where to set the decision threshold.

Not for multi-class outcomes and not a substitute for calibration analysis — AUC judges ranking, not whether predicted probabilities are numerically right.

Built for: Analysts, data scientists, and operators evaluating a classifier or diagnostic score

Typical data source: A table with a true binary label and a model score or probability per row

HealthcareFinanceMarketingRiskSaaS

What data do you need?

Any table with a true yes/no outcome and a numeric score. For example, a churn model's scored customers:

churned (categorical) churn_score (numeric)
Yes 0.82
No 0.13
No 0.41

Minimum 20 rows · Best with 200-50,000 rows with at least 10 of each class

What's in the report?

Standard-library analysis: evaluate a binary classifier or diagnostic score against the true yes/no outcome. Draws the full ROC curve, computes the AUC with a 95% confidence interval, and finds the best operating cutoff by Youden's J — with sensitivity, specificity, PPV, NPV, accuracy, and the confusion counts at that cutoff. Works on any dataset: map the true outcome and a numeric score where higher means more likely positive.

🔵

ROC Curve

The ROC curve — the higher it bows toward the top-left corner, the better the score separates the two classes; the diagonal is pure chance.

📋

Cutoff Performance

AUC, its confidence interval, and every operating-point metric at the recommended Youden cutoff, so you know exactly what that decision threshold buys.

📋

Threshold Sweep

Sensitivity and specificity at candidate cutoffs, so you can see the trade-off and move the line to suit your costs.

🤖

AI Insights

Plain-English interpretation — what the numbers mean, what's significant, and what to do next.

The Question This Answers

How good is my classifier, and where should I set the cutoff?

Map your true yes/no outcome and the model's score. The ROC curve and AUC tell you how well the score ranks positives above negatives, and the Youden-optimal cutoff tells you exactly where to draw the line — with the sensitivity and specificity you get there.

Questions?

See our FAQ for details on pricing, data privacy, and how the analysis works. Every report includes a Methodology section showing the statistical test, assumptions checked, and diagnostics run.

Your data has more stories to tell

Run any analysis on your own data — validated R analyses, interactive reports, AI insights, and PDF export.

Try Free — No Credit Card
Powered by MCP Analytics