Case Study 2: Diagnosing Offensive Efficiency Decline
Introduction
Midway through the season, State University's offense has declined significantly. After averaging 38 points per game in their first four games (3-1 record), they've averaged only 21 points in games 5-8 (1-3 record). The coaching staff needs to understand what changed.
Traditional analysis points to several surface-level issues—fewer rushing yards, lower completion percentage, more turnovers. But these are symptoms, not diagnoses. This case study uses EPA and success rate analysis to identify the root causes of the offensive decline and provide actionable recommendations.
Background
Team Profile: State University
Offensive Personnel: - Returning starting quarterback (senior) - New starting running back (sophomore transfer) - Experienced offensive line (4 returning starters) - New offensive coordinator (first year)
Early Season Performance (Games 1-4): - Record: 3-1 - Points per game: 38.0 - Yards per game: 485 - Plays per game: 72
Recent Performance (Games 5-8): - Record: 1-3 - Points per game: 21.0 - Yards per game: 345 - Plays per game: 65
Data Collection
Play-by-Play Summary
| Period | Plays | Pass Plays | Rush Plays | Pass % |
|---|---|---|---|---|
| Games 1-4 | 288 | 162 | 126 | 56.3% |
| Games 5-8 | 260 | 135 | 125 | 51.9% |
The team has become more run-heavy, but is that the problem or a response to it?
EPA Analysis: Before and After
Overall Offensive EPA
| Period | Total EPA | Plays | EPA/Play |
|---|---|---|---|
| Games 1-4 | +62.4 | 288 | +0.217 |
| Games 5-8 | +14.8 | 260 | +0.057 |
Critical Finding: EPA per play dropped by 74%, from elite (+0.217) to below average (+0.057). This is a dramatic decline that explains the scoring drop.
EPA by Play Type
Passing Game:
| Period | Pass EPA | Attempts | EPA/Attempt |
|---|---|---|---|
| Games 1-4 | +52.8 | 162 | +0.326 |
| Games 5-8 | +18.2 | 135 | +0.135 |
Passing EPA per attempt dropped by 59%.
Rushing Game:
| Period | Rush EPA | Carries | EPA/Carry |
|---|---|---|---|
| Games 1-4 | +9.6 | 126 | +0.076 |
| Games 5-8 | -3.4 | 125 | -0.027 |
Rushing went from positive to negative EPA.
Key Insight
Both phases of the offense declined, but the passing game decline is more significant in absolute terms. Let's dig deeper.
Passing Game Diagnosis
Completion Percentage Analysis
| Period | Comp % | EPA/Completion | EPA/Incompletion |
|---|---|---|---|
| Games 1-4 | 68.5% | +0.52 | -0.12 |
| Games 5-8 | 62.2% | +0.38 | -0.18 |
Completions are worth less (+0.52 → +0.38), and incompletions hurt more (-0.12 → -0.18).
Depth of Target Analysis
Average Air Yards per Attempt:
| Period | aDOT |
|---|---|
| Games 1-4 | 9.2 |
| Games 5-8 | 6.8 |
The offense is throwing shorter passes, reducing the value of completions.
EPA by Throw Depth:
| Depth | EPA/Att (G1-4) | EPA/Att (G5-8) | Change |
|---|---|---|---|
| Behind LOS | +0.08 | +0.04 | -50% |
| Short (1-9) | +0.18 | +0.12 | -33% |
| Medium (10-19) | +0.45 | +0.22 | -51% |
| Deep (20+) | +0.62 | +0.18 | -71% |
Critical Finding: Deep passing EPA collapsed from +0.62 to +0.18. This is the most dramatic single decline.
Why Did Deep Passing Decline?
Deep Pass Analysis:
| Metric | Games 1-4 | Games 5-8 |
|---|---|---|
| Deep attempts | 28 | 15 |
| Deep completions | 12 (43%) | 4 (27%) |
| Deep TDs | 6 | 1 |
| Deep INTs | 1 | 2 |
The team is attempting fewer deep passes AND completing them at a lower rate.
Possible Causes: 1. Quarterback regression (accuracy, confidence) 2. Offensive line decline (less time to throw) 3. Receiver issues (injuries, separation) 4. Play-calling changes (conservative) 5. Opponent adjustments (better coverage)
Time to Throw Analysis
| Period | Avg Time to Throw | Pressure Rate |
|---|---|---|
| Games 1-4 | 2.65 seconds | 28% |
| Games 5-8 | 2.38 seconds | 36% |
Finding: The quarterback is being pressured more frequently (28% → 36%) and throwing faster (2.65 → 2.38 seconds). This suggests offensive line issues are forcing quicker, shorter throws.
EPA Under Pressure
| Period | Clean EPA/Att | Pressured EPA/Att |
|---|---|---|
| Games 1-4 | +0.42 | +0.05 |
| Games 5-8 | +0.28 | -0.22 |
When pressured, the offense has gone from barely positive to significantly negative EPA.
Rushing Game Diagnosis
Overall Rushing Analysis
| Period | Yards/Carry | Success Rate | EPA/Carry |
|---|---|---|---|
| Games 1-4 | 4.8 | 46% | +0.076 |
| Games 5-8 | 3.8 | 38% | -0.027 |
All rushing metrics have declined, but by how much is situational?
Rushing by Box Count
| Period | Light Box EPA | Stacked Box EPA |
|---|---|---|
| Games 1-4 | +0.15 | +0.02 |
| Games 5-8 | +0.08 | -0.12 |
Finding: The team is seeing more stacked boxes (defenders committed to stopping the run), and their performance against stacked boxes has deteriorated significantly.
Box Count Distribution:
| Period | Light Box % | Neutral % | Stacked % |
|---|---|---|---|
| Games 1-4 | 35% | 42% | 23% |
| Games 5-8 | 22% | 38% | 40% |
Opponents are loading the box more frequently, suggesting they're less concerned about the passing game.
Running Back Performance
| Period | Carries | YAC | Broken Tackles | EPA/Carry |
|---|---|---|---|---|
| Games 1-4 | 98 | 2.8 | 18 | +0.09 |
| Games 5-8 | 102 | 2.1 | 11 | -0.04 |
The running back's yards after contact and broken tackles have declined. This could be: - Adaptation to college speed - Nagging injury - Improved game film allowing better tackling angles - Coinciding offensive line issues
Success Rate Analysis
Overall Success Rate
| Period | Overall | 1st Down | 2nd Down | 3rd Down |
|---|---|---|---|---|
| Games 1-4 | 48% | 52% | 47% | 44% |
| Games 5-8 | 40% | 43% | 39% | 35% |
Significant decline across all downs, but third down is most concerning (44% → 35%).
Third Down Deep Dive
| Period | 3rd & Short | 3rd & Medium | 3rd & Long |
|---|---|---|---|
| Games 1-4 | 62% | 45% | 28% |
| Games 5-8 | 48% | 32% | 18% |
Third and Medium (4-7 yards) conversion rate dropped from 45% to 32%. This is particularly problematic because it represents the most common third-down situation.
Third Down Play Calling
| Period | Pass % on 3rd | 3rd Down Pass EPA | 3rd Down Rush EPA |
|---|---|---|---|
| Games 1-4 | 72% | +0.28 | -0.05 |
| Games 5-8 | 68% | +0.02 | -0.35 |
Third-down passing has become nearly worthless (+0.02), and third-down rushing is now actively harmful (-0.35).
Field Position Analysis
EPA by Zone
| Zone | Games 1-4 EPA/Play | Games 5-8 EPA/Play |
|---|---|---|
| Own 0-25 | +0.12 | +0.04 |
| Own 25-50 | +0.22 | +0.08 |
| Opp 50-25 | +0.28 | +0.06 |
| Red Zone | +0.35 | +0.10 |
Red Zone Decline: The biggest drop is in the red zone (+0.35 → +0.10). The offense is struggling to finish drives.
Red Zone Detailed Analysis
| Period | RZ Trips | RZ TDs | RZ FGs | RZ Failures | TD Rate |
|---|---|---|---|---|---|
| Games 1-4 | 18 | 12 | 4 | 2 | 67% |
| Games 5-8 | 14 | 5 | 6 | 3 | 36% |
Touchdown rate dropped from 67% to 36%. The offense is settling for field goals or failing entirely.
Root Cause Analysis
Synthesizing the Evidence
Primary Issue: Offensive Line Decline - Pressure rate increased 28% → 36% - Time to throw decreased 2.65 → 2.38 seconds - Performance under pressure collapsed - Stacked boxes became more common (teams not respecting pass) - Running game YPC dropped significantly
Secondary Issue: Deep Passing Confidence/Ability - Deep attempts down 46% - Deep completion rate down 16 percentage points - Deep EPA per attempt down 71%
Tertiary Issue: Third Down Execution - Overall third down conversion dropped 9 percentage points - Third and medium conversion dropped 13 percentage points - Third down efficiency became near-zero
The Causal Chain
Offensive Line Struggles
↓
More Pressure, Less Time to Throw
↓
Short/Quick Passes Replace Deep Shots
↓
Opponents Load Box (Less Passing Threat)
↓
Running Game Struggles Against Stacked Boxes
↓
More Negative or Short-Gain Plays
↓
Longer Third Down Distances
↓
Lower Third Down Conversion Rate
↓
Fewer Points, More Punts
Recommendations
Immediate Actions
-
Diagnose Offensive Line Issues - Review film for technique breakdowns - Assess any injury impacts - Consider personnel changes
-
Scheme Adjustments for Quicker Passing - Hot routes and sight adjustments - RPOs to punish blitzes - Quick screens to get ball in space
-
Restore Deep Passing Threat - Max protection packages - Play action from under center - Target deep routes to force safety respect
-
Third Down Packages - New formations to create confusion - Better pre-snap motion - More misdirection
Medium-Term Solutions
-
Running Game Changes - More outside zone to avoid loaded boxes - Jet sweeps and reverses to stretch defense - Quarterback designed runs
-
Red Zone Approach - Heavier personnel for goal-line situations - More fade routes to larger receivers - Better play-action design
Tracking Progress
Key Metrics to Monitor:
| Metric | Current | Target | Elite |
|---|---|---|---|
| EPA/Play | +0.057 | +0.15 | +0.20+ |
| Success Rate | 40% | 45% | 48%+ |
| Pressure Rate | 36% | 30% | <28% |
| 3rd Down Conv | 35% | 42% | 45%+ |
| RZ TD Rate | 36% | 55% | 65%+ |
Expected Impact
If the team can: - Reduce pressure rate to 30% - Restore deep passing attempts to early-season levels - Improve third-down conversion to 42%
Projected EPA improvement: +0.057 → +0.14 (approximately doubling current efficiency)
Projected scoring: 21 PPG → 30+ PPG
Conclusion
The offensive decline is primarily an offensive line problem that has cascaded through the entire system. The solution isn't simply to "run more" or "be more aggressive"—it's to address the protection issues that are forcing conservative play-calling.
Without fixing the root cause, schematic adjustments will only provide marginal improvement. The passing game ceiling is limited by protection, and the running game is handicapped by opponents' ability to load the box without consequences.
This case study demonstrates how EPA analysis moves beyond "what" is happening to "why" it's happening, enabling targeted solutions rather than generic advice.
Code Implementation
The complete code for this analysis is available in code/case-study-code.py, including:
- OffensiveDeclineAnalyzer class
- Period comparison framework
- EPA decomposition by situation
- Root cause identification algorithm
Discussion Questions
-
How would you validate that offensive line play is truly the root cause rather than quarterback regression?
-
If the offensive line issues can't be immediately fixed, what schematic changes offer the best EPA improvement?
-
How much of the running back's decline is attributable to box count vs individual performance decline?
-
Should the team accept lower EPA per play but higher success rate to minimize variance late in close games?
-
What additional data (tracking, personnel groupings) would strengthen this analysis?