Forecasting Kyle Kuzma’s Field Goal Percentage (FG%)

More and more tools are becoming available for data analysts and data scientists. These tools provide convenience and lower the bar of entry for aspiring data professional.

I’ve been looking for an easy way to get to learning predictive analysis and forecasting. Prophet provides that path. Prophet is released by Facebook’s Core Data Science Team.

“Prophet is a procedure for forecasting time series data based on an additive model where non-linear trends are fit with yearly, weekly, and daily seasonality, plus holiday effects. It works best with time series that have strong seasonal effects and several seasons of historical data. Prophet is robust to missing data and shifts in the trend, and typically handles outliers well.”

Just to dip my toes into the waters, I tried Prophet’s Quick Start Guide in R.

Let’s forecast the Field Goal Percentage (FG%) of Kyle Kuzma of the Los Angeles Lakers for the next 6 Months.

I’m using the FG% data scraped from Kuzma’s Game Log from 10/18/2018  to 12/18/2018.

Prophet uses a “normal model fitting API with its prophet function that performs fitting and returns a model object.”

#data source https://www.basketball-reference.com/players/k/kuzmaky01/gamelog/2019/
#load prophet  
library(prophet)

#create data frame
df <- read.csv('C:/Users/Marlon.Ribunal/Desktop/JE Docs/Install/Kuzma_FG_Pct.csv')

#define model
m <- prophet(df=df, growth = "linear", yearly.seasonality = TRUE)

#set horizon (preiods to forecast)
future <- make_future_dataframe(m, periods = 180)

That pretty much read our csv into a data frame, performs fitting, and define the model.

To run a simple forecast on our data:

#plot forecast
plot(m,forecast)

kyle kuzma field goal percentage

We can also break down that forecast into the trend, weekly, and yearly:

#Plot forecast breakdown
prophet_plot_components(m, forecast, uncertainty = TRUE, yearly_start = 0, weekly_start = 0, plot_cap = TRUE)

Kyle Kuzma Field Goal Forecast

If you’re looking into doing some forecasting, Prophet might be a good starting point.  By the way, it’s an open source software so it gets a lot of support from the open source community, plus it’s from Facebook. Visit Prophet’s Github page here.

Author: Marlon Ribunal

I am SQL Server Database Administrator for a software company catering to supply chain and retail industry.

Comments are closed.