What is R?

• R is a statistical programming language; think of it like an incredibly flexible super calculator. In my course, I'll walk you through the download process in a matter of minutes.
• R is completely free and open-source. With 1000s of free add-on "libaries" created by some of the world's top analytics minds, you can easily harness new models and visualization techniques with only a few simple lines of code.
• Unlike many coding languages, R's syntax is intuitive and short. The most powerful blocks of code in R are rarely more than 2-3 lines. The learning curve of R is very suitable to anyone wanting to learn programming for the first time or anyone who wants to pivot to an analytics career.
• In this course, you'll learn all the fundamentals of R, from data pre-processing to visualizations to simulations to predictive modeling. Also, I'll teach several of my favorite add-on R libraries in-depth, so you can utilize cutting-edge analytics techniques.

Always keep the top button unbuttoned when holding a football. And code in R.

Syllabus

• Lesson 1​: Getting started with R
  • ​Using R as a calculator with arithmetic functions
    • ​Built-in statistical functions like mean, cumulative sum, and more
  • Vectors
  • TRUE/FALSE (Boolean) expressions and other comparison statements
  • Corresponding Worksheet 1 to practice concepts learned
• Lesson 2: Matrices and Lists
  • Creating a matrix 
  • Setting up lists
  • Introducing strings in R
  • Corresponding Worksheet 2 to practice concepts learned
• Primer: Bayes' Theorem
  • ​Learn to think like a Bayesian
  • Tackle seemingly complex probabilities step-by-step 
• Lesson 3: Bringing Data into RStudio
  • Viewing data in RStudio
  • Summary statistics for a dataset
  • Manipulating columns and data structure
  • Corresponding Worksheet 3 to practice concepts learned
• Lesson 4: Learning dplyr and Plotting
  • Advanced data aggregation, grouping, sorting, and manipulation
    • Learn dplyr's intuitive, step-by-step syntax
  • Scatterplots, histograms, and boxplots
  • Corresponding Worksheet 4 to practice concepts learned
• Lesson 1-4 Review
  • ​Tying together all concepts learned in lessons 1-4
  • Correlation plotting
  • Creating a basic player projection framework
• Lesson 5: Simulation and Loops
  • ​Random number generation in R
  • Setting up reproducible simulations
  • Loops
  • Scaling and standardizing variables
  • Corresponding Worksheet 5 to practice concepts learned
• Lesson 6: Functions and Joining Datasets
  • Create your own functions
    • Use your own functions to create new complex columns in your dataset
  • Learn several different ways to merge datasets
    • When to use each type of join
    • Learn what R does behind the scenes when joining​ datasets
  • Corresponding Worksheet 6 to practice concepts learned
• Lesson 7: Advanced Visualizations with ggplot2
  • Learn ggplot2's step-by-step, add-on approach to plotting
    • Create aging curves by position
      • Analyze recent running back contracts through this lens
  • Handling and converting dates
  • Corresponding Worksheet 7 to practice concepts learned
• Primer: Matrix Multiplication
  • ​Learn matrix multiplication to prepare for Markov Chain simulation
• Lesson 8: Markov Chains and data.table
  • ​Learning data.table package as an efficient dataset object
  • Setting up timers to measure how long code takes to run
  • Creating a unique identifier for player names with string manipulation
  • Volatility of running back yards year-over-year using Markov Chain simulation
  • Corresponding Worksheet 8 to practice concepts learned
• Primer: NFL Play-by-Play Data Analysis
  • Extracting information from 2019 NFL play-by-play data using functions learned in Lessons 1-8
    • Attributing passing, rushing, and receiving yards to each player
• Lesson 5-8 Review
  • ​Tying together all concepts learned in lessons 5-8
  • Analyzing historical NFL draft data and drafting patterns
  • Creating basic simulations for an entire NFL draft
  • Visualizing the effect of draft round on rookie production
• Lesson 9: Linear Regression (Part I)
  • ​Running and interpreting regression output
  • Understanding the math behind linear regression
  • Intuition behind model over-fitting
  • Squared terms and interaction terms
• Lesson 9: Linear Regression (Part II)
  • ​Logged terms in regression models
  • Learn the 5 assumptions of linear regression
  • Test for potential heteroskedasticity and autocorrelation 
  • Understand multicollinearity
  • Residual plots and identifying outliers
  • Corresponding Worksheet 9 to practice concepts learned
• Lesson 10: Logistic Regression
  • Running and interpreting logistic regression output
  • Converting logged odds predictions to probabilities
  • Intuition behind the variable slope of logistic regression predictions
  • Interpreting a confusion matrix to analyze model performance
  • Multinomial regression for multi-class predictions
  • Ordinal regression for ordered class predictions
  • Corresponding Worksheet 10 to practice concepts learned
• Lesson 11: Optimization
  • ​Learn integer, binary integer, and linear programming
  • Master handwritten optimization equations
    • ​Separate out and identify the objective function, the matrix of constraints, and the constraints vector to see the system of equations you're optimizing
  • Minimization and maximization
  • Corresponding Worksheet 11 to practice concepts learned
    • Optimize a DraftKings lineup
• Lesson 12: Text Analytics
  • ​Creating a "corpus" of multiple related text documents to analyze together
  • Analyzing a Document Term Matrix
  • TF-IDF for advanced term importance
  • Sentiment analysis with VADER
  • Designing unique and memorable word clouds
  • Corresponding Worksheet 12 to practice concepts learned
• Lesson 13: Making Future Predictions
  • ​Splitting your dataset into train-validation-test sets
  • Generating basic player projections for the 2020 season
  • Root-mean-square error to evaluate different models
  • Learning the k-nearest neighbors model for prediction
  • Corresponding Worksheet 13 to practice concepts learned
• Lesson 14: Clustering and Tree Models
  • k-means cluster analysis
    • Plotting and interpreting the clusters in ggplot2
  • Bootstrapped sampling and "bagging"
  • Individual decision tree models
    • Visualizing the entire "tree" and all its "branches"
  • Random forest modeling
  • Corresponding Worksheet 14 to practice concepts learned
• Primer: Imputing Missing Data
  • Strategically handling and replacing missing data values
  • Utilizing predictive modeling to overcome data deficiencies
    • Impute categorical variables and numeric variables
• Lesson 15: Analytics Reports in Rmarkdown
  • ​Transferring your code within Rstudio to an Rmarkdown document
  • Outputting clean HTML and PDF documents
  • Designing an in-depth, elegant projection system for running back yards in the coming season
  • Understanding the structure and order of an analytics report
  • Corresponding Worksheet 15 to convert a wide receiver projection script to an analytics report
• Primer: More Advanced Predictive Models
  • ​Support vector machines (SVM) for regression/classification
  • Extreme gradient boosting (xgboost) for regression/classification
  • Simple neural network for regression/classification

That LaDainian Tomlinson jersey is not just for show. Check out his career as part of the Lesson 1-4 review dataset!

Pictured here is the actual nerd and actual laptop used to create this course.

Obviously a touchdown pass. Anyone else prefer playing tackle sports in a button down shirt? No, just me...?

Standing 5'9" and weighing almost 158 lbs, he ran a 4.96 40-yard dash, couldn't bench 225 at the combine, and is unsure if he'll get drafted all... But dang, he looks good holding the football at that angle. And more importantly, he can code in R.