Analyze which AI model rank changes are statistically significant vs. random noise. Uses z-score analysis with 95% confidence intervals across 0 models with sparkline history.
Models Analyzed
0
Significant Changes
0
Noise (Not Significant)
0
Both Timeframes
0
0 models with z-scores exceeding the 95% confidence threshold (|z| > 1.96), sorted by significance strength.
No significant changes
All model score changes fall within normal statistical variance.
Cross-referencing daily (24h) and weekly (7d) rank changes to identify the strongest signals. Models significant on both timeframes represent the most reliable trend shifts.
No models currently significant on both daily and weekly timeframes.
No models with daily-only significance.
No models with weekly-only significance.
Coefficient of variation (CV%) measures score volatility relative to mean. High-CV models require larger changes to be statistically significant; low-CV models are the most predictable.
| Model | Score | CV% |
|---|
| Model | Score | CV% |
|---|
Understanding the statistical methodology behind our significance analysis helps you distinguish real performance shifts from random fluctuations.
We use z-scores with a 95% confidence threshold (|z| > 1.96). A z-score measures how many standard deviations a model's current score is from its historical baseline. Only changes exceeding 1.96 standard deviations are flagged as statistically significant.
The baseline is computed as the arithmetic mean of each model's 14-day sparkline data. This rolling average smooths out daily fluctuations and provides a stable reference point for detecting meaningful deviations.
Each model's 95% confidence interval is calculated as baseline ± 1.96 × standard deviation. Scores falling outside this range indicate a statistically meaningful change. The "Confidence" column shows the ± threshold value.
Daily (24h) and weekly (7d) rank changes are analyzed separately. Daily significance requires a rank shift of more than 3 positions; weekly requires more than 5. Models significant on both timeframes represent the strongest, most reliable signals.
The coefficient of variation (CV%) measures relative volatility. High-CV models have naturally noisy scores and require larger absolute changes to be significant. Low-CV models are more predictable, so even small deviations may represent real shifts.