diff --git a/Dockerfile b/Dockerfile
index f69a089..fb37cf5 100644
--- a/Dockerfile
+++ b/Dockerfile
@@ -7,4 +7,4 @@ RUN pip install --no-cache-dir -r requirements.txt
 
 COPY . .
 
-CMD ["python", "main.py"]
+CMD ["python", "scripts/main.py"]
diff --git a/README.md b/README.md
index 9e1c91d..71306cb 100644
--- a/README.md
+++ b/README.md
@@ -1,7 +1,72 @@
-# f1-data-analysis
-Using Pandas in Python, I analyzed massive amounts of F1 race data to figure out the team with the highest average places gained from start to finish.
+# F1 Data Analysis: Constructor Racecraft Performance
 
-## Run with Docker
+This project analyzes historical Formula 1 race result data to evaluate constructor-level racecraft through **Average Positions Gained (APG)**, defined as:
+
+`APG = grid position - final position`
+
+A higher APG indicates that, on average, a team improves more positions between the race start and finish.
+
+## Project Goals
+
+- Quantify which constructors gain the most positions during races.
+- Compare APG against sample size (total race entries) to highlight variance in small datasets.
+- Provide both static and interactive visualizations for exploratory analysis.
+
+## Repository Structure
+
+```text
+.
+├── data/
+│   ├── constructors.csv
+│   └── results.csv
+├── scripts/
+│   ├── main.py
+│   ├── scatter-plot.py
+│   └── interactive-scatter-plot.py
+├── Dockerfile
+└── requirements.txt
+```
+
+## Data Sources
+
+- `constructors.csv` and `results.csv` were acquired from the Kaggle dataset:  
+  https://www.kaggle.com/datasets/rohanrao/formula-1-world-championship-1950-2020
+
+## Analysis Walkthrough
+
+- Presentation with methodology and process:  
+  https://docs.google.com/presentation/d/1bZhT_yq7-mdMf9Ao8VaFRqklGW-47c8zOK8FQtB8lGk/edit?usp=sharing
+
+## Scripts
+
+- `scripts/main.py`  
+  Core APG analysis with console output and a bar chart of top teams.
+
+- `scripts/scatter-plot.py`  
+  Static scatter plot showing APG vs total entries (log-scaled x-axis).
+
+- `scripts/interactive-scatter-plot.py`  
+  Interactive Plotly scatter plot for deeper exploratory analysis.
+
+## How to Run
+
+### Option 1: Local Python Environment
+
+1. Install dependencies:
+   ```bash
+   pip install -r requirements.txt
+   ```
+2. Run the primary analysis:
+   ```bash
+   python scripts/main.py
+   ```
+3. Run visualization scripts:
+   ```bash
+   python scripts/scatter-plot.py
+   python scripts/interactive-scatter-plot.py
+   ```
+
+### Option 2: Docker
 
 Build the image:
 
@@ -9,15 +74,30 @@ Build the image:
 docker build -t f1-data-analysis .
 ```
 
-Run the default analysis script (`main.py`):
+Run the default analysis script:
 
 ```bash
 docker run --rm -it -v "$(pwd):/app" f1-data-analysis
 ```
 
-Run a different script if needed:
+Run specific scripts:
 
 ```bash
-docker run --rm -it -v "$(pwd):/app" f1-data-analysis python scatter-plot.py
-docker run --rm -it -v "$(pwd):/app" f1-data-analysis python interactive-scatter-plot.py
+docker run --rm -it -v "$(pwd):/app" f1-data-analysis python scripts/scatter-plot.py
+docker run --rm -it -v "$(pwd):/app" f1-data-analysis python scripts/interactive-scatter-plot.py
 ```
+
+## Methodology Summary
+
+1. Load race results and keep only constructor ID, grid position, and final classified position.
+2. Compute per-race positions gained for each constructor entry.
+3. Aggregate APG by constructor and compute total race entries.
+4. Merge constructor metadata for readable team names.
+5. Apply entry-count thresholds where appropriate to reduce small-sample distortion.
+6. Compare central trend and variance through ranked tables and scatter plots.
+
+## Key Interpretation Notes
+
+- Teams with very few race entries can produce extreme APG values that are not stable estimates.
+- Entry thresholds and log-scaled visualizations are used to make comparisons more statistically meaningful.
+- APG captures race-day progression, not overall team performance across qualifying, reliability, or points systems.
diff --git a/constructors.csv b/data/constructors.csv
similarity index 100%
rename from constructors.csv
rename to data/constructors.csv
diff --git a/results.csv b/data/results.csv
similarity index 100%
rename from results.csv
rename to data/results.csv
diff --git a/interactive-scatter-plot.py b/scripts/interactive-scatter-plot.py
similarity index 85%
rename from interactive-scatter-plot.py
rename to scripts/interactive-scatter-plot.py
index 20e459f..d0fdd60 100644
--- a/interactive-scatter-plot.py
+++ b/scripts/interactive-scatter-plot.py
@@ -1,6 +1,10 @@
 import plotly.express as px
 import pandas as pd
-f1 = pd.read_csv('results.csv')
+from pathlib import Path
+
+DATA_DIR = Path(__file__).resolve().parents[1] / "data"
+
+f1 = pd.read_csv(DATA_DIR / 'results.csv')
 f1 = f1[['constructorId', 'grid', 'positionOrder']].copy()
 f1.columns = ['teamId', 'startPos', 'endPos']
 f1['positionsGained'] = f1['startPos'] - f1['endPos']
@@ -9,7 +13,7 @@
 team_entries = f1['teamId'].value_counts().reset_index()
 team_entries.columns = ['teamId', 'totalEntries']
 plot_data = pd.merge(total_teams, team_entries, on='teamId', how='left')
-team_names = pd.read_csv('constructors.csv')
+team_names = pd.read_csv(DATA_DIR / 'constructors.csv')
 plot_data = pd.merge(plot_data, team_names[['constructorId', 'name']], left_on='teamId', right_on='constructorId', how='left')
 
 fig = px.scatter(
diff --git a/main.py b/scripts/main.py
similarity index 91%
rename from main.py
rename to scripts/main.py
index a68b0d5..aa94a6f 100644
--- a/main.py
+++ b/scripts/main.py
@@ -2,9 +2,12 @@
 import pandas as pd
 import matplotlib.pyplot as plt
 import seaborn as sns
+from pathlib import Path
+
+DATA_DIR = Path(__file__).resolve().parents[1] / "data"
 
 # APG - average positions gained
-f1 = pd.read_csv('results.csv') # reading results csv file into Pandas dataframe
+f1 = pd.read_csv(DATA_DIR / 'results.csv') # reading results csv file into Pandas dataframe
 f1 = f1[['constructorId', 'grid', 'positionOrder']] # only using the three columns we need
 f1.columns = ['teamId', 'startPos', 'endPos'] # renaming columns
 f1['positionsGained'] = f1['startPos'] - f1['endPos'] # creating new column with positions gained
@@ -19,7 +22,7 @@
 
 if min_entries_required <= total_teams['totalEntries'].max(): # checking that the requirement is within limits
   sat_teams = total_teams[total_teams['totalEntries'] >= min_entries_required].reset_index(drop = True) # creating new dataframe with only teams that satisfy the entries requirement
-  team_names = pd.read_csv('constructors.csv') # reading constructors csv file into Pandas dataframe
+  team_names = pd.read_csv(DATA_DIR / 'constructors.csv') # reading constructors csv file into Pandas dataframe
   sat_teams = pd.merge(sat_teams, team_names[['constructorId', 'name', 'nationality']], left_on = 'teamId', right_on = 'constructorId', how = 'left') # combining team names and nationalities with the main dataframe
   sat_teams = sat_teams[['teamId', 'name', 'nationality', 'APG', 'totalEntries']] # selecting and reordering columns
   ht_apg = round(sat_teams['APG'].max(), 2) # finding the APG of the team with the highest APG and rounding for ease of viewing
diff --git a/scatter-plot.py b/scripts/scatter-plot.py
similarity index 89%
rename from scatter-plot.py
rename to scripts/scatter-plot.py
index 60cf5e4..b16af88 100644
--- a/scatter-plot.py
+++ b/scripts/scatter-plot.py
@@ -1,7 +1,11 @@
 import pandas as pd
 import matplotlib.pyplot as plt
 import seaborn as sns
-f1 = pd.read_csv('results.csv')
+from pathlib import Path
+
+DATA_DIR = Path(__file__).resolve().parents[1] / "data"
+
+f1 = pd.read_csv(DATA_DIR / 'results.csv')
 f1 = f1[['constructorId', 'grid', 'positionOrder']].copy()
 f1.columns = ['teamId', 'startPos', 'endPos']
 f1['positionsGained'] = f1['startPos'] - f1['endPos']
@@ -10,7 +14,7 @@
 team_entries = f1['teamId'].value_counts().reset_index()
 team_entries.columns = ['teamId', 'totalEntries']
 plot_data = pd.merge(total_teams, team_entries, on='teamId', how='left')
-team_names = pd.read_csv('constructors.csv')
+team_names = pd.read_csv(DATA_DIR / 'constructors.csv')
 plot_data = pd.merge(plot_data, team_names[['constructorId', 'name']], left_on='teamId', right_on='constructorId', how='left')
 plt.figure(figsize=(11, 6))
 sns.set_theme(style="whitegrid")
diff --git a/tests/__pycache__/test_project_structure.cpython-312-pytest-9.1.1.pyc b/tests/__pycache__/test_project_structure.cpython-312-pytest-9.1.1.pyc
new file mode 100644
index 0000000..c7686d6
Binary files /dev/null and b/tests/__pycache__/test_project_structure.cpython-312-pytest-9.1.1.pyc differ
diff --git a/tests/test_project_structure.py b/tests/test_project_structure.py
new file mode 100644
index 0000000..540d133
--- /dev/null
+++ b/tests/test_project_structure.py
@@ -0,0 +1,18 @@
+from pathlib import Path
+
+
+REPO_ROOT = Path(__file__).resolve().parents[1]
+
+
+def test_expected_project_files_exist() -> None:
+    required_files = [
+        "data/constructors.csv",
+        "data/results.csv",
+        "scripts/main.py",
+        "scripts/scatter-plot.py",
+        "scripts/interactive-scatter-plot.py",
+        "requirements.txt",
+    ]
+
+    missing_files = [path for path in required_files if not (REPO_ROOT / path).is_file()]
+    assert not missing_files, f"Missing expected files: {missing_files}"