Welcome to the “How To” post regarding the NFL Expected Points Added (EPA) web app that I recently created and deployed for all to use! I wanted to put together a quick introduction and explanation as to how to use it – so that you may use it for your own analysis/debate fueled needs.
What exactly is Expected Points (EP) and Expected Points Added (EPA)? Brian Burke, now of ESPN, brought the idea of EP and EPA to the mainstream in 2010 in this post…but I won’t make you read it all. Essentially, not all yards are created equal! Each down, yard to go, and distance to the end zone combination has a net point value for the next team to score from this situation – which is the Expected Points metric. This metric is generated from looking at historical NFL data and is slightly changing (ever so slightly) on a year to year basis, but that is neither here nor there. Intuitively, it makes sense. Expected Points will be higher when a team has a more favorable down, yards to go, and yard line situation. Here’s a simple graphic that captures this relationship and note that the negative values mean that the team on defense (not necessarily the defense itself) scores next most often given that situation:
(0 = own end zone, 100 = opponents’ end zone)
EPA is a measurement to show the value gained, or lost, in a play. To generate the EPA, you take the down, yards to go, and yard line of the play before it happened (take it’s EP) and compare it to the down, yards to go, and yard line of the play after it happened (new EP). Example:
As you can see, the play resulted in +0.08 EPA for gaining the five yards on first down!
Okay, so now that we have a basic idea of what EP/EPA is and how it’s calculated, let me explain how to use the web app that I built! To calculate the EPA from a play, put in the previous down, yards to go, yard line, and indicate which side of the field the play took place on. Next, do the same thing, but with the down, yards to go, yard line, and side of the field for the resulting play. Using the example above, you’d enter: 1, 10, 20, and own in the top portion of the tool and 2, 5, 25, and own in the bottom portion of the tool. Simply press the “Calculate Expected Points Added!” button to see the outcome.
You may notice lower on the page that an additional calculator is present. This idea came from a video on the Advanced Football Analytics YouTube channel and will allow you to generate the confidence level (in percentage form) needed for it to be worth gambling going for fourth down. Caveat here is there is more than simply plugging in EP values for this decision, but is meant to be a fun tool for users to reference in game when their favorite team maddeningly decides to punt when it’s 4th-and-3 from their opponents’ 40 yard line. For this tool, determine what the “safe” outcome would be, what the most likely failed outcome would be, and outcome that is the bare minimum needed for a first down. Plug these values in (use the calculator above to figure out those Expected Point values) and see how confident you’d need to be to go for it on 4th and short!
As always, feel free to comment with questions! The link to the GitHub repository can be found here or on the Portfolio tab.