Everyone loves a winner and Oracle Analytics Cloud is no exception. We have a perfect score when it comes to using data visualization to predict outcomes. In fact, we predicted the amazing comeback by the New England Patriots over the Atlanta Falcons in 2017.
That said, we are ready to show how using the "Explain" feature in Oracle Analytics can allow you to take a broad swath of data from various sources to gain relevant insight. In a business context, savvy executives use this technology to improve their strategies with facts and generate action around their ideas.
This year's American football classic is between the New England Patriots and the Los Angeles Rams. And while we've seen Patriots quarterback Tom Brady dominate the postseason, our data is ready to hand the Vince Lombardi trophy over to the Rams.
Using publicly available data, we compiled information from both teams’ regular season and factored in their offensive drive and defensive drive statistics. The data is rated on both the net advance from the Line of Scrimmage and the Time of Possession, which typically create the best conditions of the winning team. It's a lot of data, so let's walk through this.
Oracle Analytics has an "Explain" feature within its data visualization capabilities that allow you to gather the values of the data and review how they relate to each other. Then Oracle Analytics automatically discovers the key drivers of results. It asks, "What elements in this data best explain the values?" The feature then segments the results so that hidden groups in the data can predict outcomes. Finally, the Explain feature identifies the anomalies, so that you can see what groups in the data exhibit unexpected results.
The first graphic describes the process of exploring the statistics between the two teams.
The next screen breaks down the basic information. They look evenly matched in many categories, although New England has a higher rating on defense.
From here, we add the interesting visualizations to the canvas to kick-start further analysis.
Next we decided to do further analysis into Time of Possession. We run the Explain feature again and can see in the next graphic that the Time of Possession favors Los Angeles in the regular season.
The result shows us that the National Football Conference (NFC) teams have a high percentage of Time of Possession overall. This also favors Los Angeles' playing style of lots of offense. New England does have a high scoring rate as well, but this year they advanced from behind the original Line of Scrimmage more often than Los Angeles did.
Based on these factors, we're using a two-classification model script, as seen below.
Finally, we are ready to compare the values and predict a winner.
And as you can see, Los Angeles is expected to score more points than New England. Predicted winner: NFC team, or Rams.
The important thing here is that using data visualization can help us see information differently and make better decisions in a way that spreadsheets or traditional static business intelligence cannot.
If you want to see how easy it is to use the data visualization tools in Oracle Analytics, visit our website for a tour or a trial.
(Special thanks to John Hung for the data visualization and screenshots.)