Fantasy Football: Help Your Data Find a Voice with BI Office

Author: | Category: BI Office Features | Tags: Fantasy Football, BI Office, Predictive Analysis | Published: 1/3/2017

Fantasy Football: Help Your Data Find a Voice with BI Office

You've probably heard the old adage "data speaks for itself."

That may be true, but data is usually telling a million stories at once. It's up to you to place data in the right context, to make it meaningful for a particular audience. This process is data storytelling, and it's what BI Office is designed to help you do.

For this blog post, I will discuss my methodology for creating a narrative by focusing on the three storyboards I built to help manage my fantasy football team. I followed a 5-step process for storyboard creation: problem identification, storylining, connecting to data sources, adding views/interactions, and enhancement. I will dive into detail on each one of these steps.

Problem Identification

The first step in building a storyboard is identifying the problem you wish to solve. Essentially, you're setting the objective for your project.

For instance, a sales manager might ask how he can best determine behaviors that lead to higher sales among members of his team. Or a purchasing manager might want to quickly assess sales forecasts to determine how much quantity of an input material to purchase.

In my case, I had three problems I wanted to solve: how to draft the best possible team, how to assess my team and player performance, and how to find replacements for players with injuries or on bye weeks. I also wanted to show off the potential of BI Office visualization, giving us different ways to look at the same data, to help speed understanding and decision making.

Determining a Story

Next I needed to figure out what stories I wanted to tell to best address each of my problems. This step of the process seeks to provide a broad answer to the individual problems raised in the previous step.

For the fantasy storyboards, I had the advantage of being both the producer and consumer of the information. This might be the case for small business owners as well, but often the consumers and producers of storyboards are different people, or even different groups of people.

When telling a data story, you are constrained by what information you have at your disposal. For this project, we assessed historical performance data and projections for each NFL player dating back to 2008. We also had access to our league information.

With 15 years of experience playing fantasy football, I had a pretty good idea of what stories I wanted to tell. For the draft storyboard, I wanted the story to revolve around how available players would perform given our specific league scoring system. The performance storyboard would provide a narrative for how well my players were performing compared to their projections and to other available players at the same position. Finally, the injury/bye week storyboard would tell the story of how which players needed replacing and which available players would provide the most value.

Connecting Data Sources

Once you have decided on a story, the next step is to connect it with your data. There are two basic options here: manually import the data when you need it or stream the data in real-time using an API (application programming interface). For a custom data source like FantasyData.com, streaming the data would have required us to build an application, which was outside the scope of what we wanted to accomplish. But for many common data sources, like Salesforce and other SAP applications, real-time streaming of data is just a few clicks away.

Rather than building an app, we imported the data from FantasyData.com and ESPN and warehoused it on our servers in a spreadsheet. We then designed an ETL (extract, transform, load) package to take in the sheets and model the data.

For more detail on integrating with popular data sources, click here.

Adding Views and Interactions

Deciding which views, visualizations, and interactions to use in your storyboard is driven by the story you decided to tell. If you're relatively new to using a tool like BI Office, you might want to experiment with the included 25 chart types, 9 advanced visualizations, 6 grid types, geospatial maps, and dynamic text. Eventually you will develop an instinct for which visualizations are most useful to express particular combinations of data.

You can learn more on views and visualizations in BI Office's help file or by watching our tutorials.

Part of what makes BI Office so easy to work with is that I never have to worry about what device my end user will be using to access my storyboards. BI Office responsively adjusts to the user's screen, making it effortless to access and use storyboards on any device.

Enhancement

Finally, I enhance my dashboard by adding additional information that will help address questions that arise during my analysis of the original problem. Creating a storyboard is not an end unto itself. Every data story evolves, so you will need to make adjustments to improve the product so it fits your users' needs.

Next year, I will add views to prevent me from drafting too many players with the same bye week. I will also create a view to see which players consistently beat expectations by showing historical draft projections versus actual performance on a season-by-season basis for each player.

Building storyboards for my fantasy football team has been on my bucket list for longer than I can remember, and I'm glad I finally got to take a crack at it. As mentioned in my last post, my team didn't make the playoffs, but that was mostly due to my dual role as manager and data architect. If I'd focused more on the former and spent less time tinkering with my views, my team might have fared better.

Hopefully you have found this insight into my dashboard creation process useful. Thanks for reading!


Related resources

  1. Using Performance Analysis to Determine How I Dropped the Ball
  2. Tackling Bye Weeks and Injuries with BI Office
  3. Using On-The-Fly Data Modeling to Draft the Optimal Fantasy Football Team
  4. Fantasy Football 2016 - Building a Data-Driven Draft
  5. Follow @PyramidAnalytics on Twitter
  6. Pyramid Analytics Home Page

Add your comment below

 

Be the first to comment


Stay up-to-date with product announcements

Popup Content