Fantasy: House Rules - The Quest for an IDP Scoring System, Part 1

I played a lot of beer pong in college. My roommates and I loved the game. We played it in living rooms and basements, on driveways and grass, and on just about every shape of table you can imagine. Though the principle of the game is always the same, each location had their own house rules. Some allowed formations after each make, some allowed only one per game, and the most extreme didn’t allow any formations. There were houses that let you blow a ball out of a cup before it hit beer, and others who had a derogatory name for you if you executed said maneuver. A few places added a unique specialty rule like bounce throws only, or my all time favorite, when both partners make the same exact cup on the same turn – game over.

In many ways, IDP scoring resembles the myriad of beer pong house rules. While offensive scoring has been fairly normalized in standard, PPR, and TD-only formats, IDP scoring is all over the place. Some leagues lean towards a big-play format that mainly reward TDs. Other leagues are extremely tackle-heavy, which favors the linebackers while devaluing defensive backs and linemen. I’ve seen a number of these leagues score sacks and interceptions the same as solo tackles. Are any of these systems optimal? Is there a better way to score IDPs?

During the regular season, I had an email exchange with my PFF Fantasy IDP co-rankers, Ross Miles and Kevin Greenstein, about this very issue. We talked back and forth about the challenges of creating an IDP scoring system that is fair, balanced, and reflective of actual on-field performance. Out of this conversation was born the idea that we take a critical look at the defensive statistics to explore how IDPs are scored. Using this data, we can then experiment and tweak IDP scoring to create an ideal system. So ultimately, our goal is quite simply to establish a PFF Fantasy “house rules” IDP scoring system.

In the coming months, we will elaborate in greater detail on this project, our findings, and ultimately our recommended system. But in this first phase of the project, I gathered key defensive statistical data from the last three seasons, and then performed some detailed analysis. I chose the last three seasons because the PFF premium data goes back to the 2008 season. Ross, Kevin, and I agree that we should score the following statistics – solo tackles, assisted tackles, tackles for a loss, sacks, interceptions, passes defensed, forced fumbles, fumble recoveries, blocked kicks, touch downs, and safeties. I compiled team-by-team totals for these categories, and also established total defensive snap counts for each of the last three regular seasons:

Taken together, this data really helps us establish a lay of the land. The first thing you’ll likely notice is the consistency from one season to the next. Though 2008 shows a down-year for interceptions and an up-year for safeties, the rest of the data is either similar or identical from year to year. This regularity is actually quite useful in that it will assure that our house rules scoring system can be standardized and used from one season to the next without modification.

This data is very interesting in its raw form, but is a bit too cumbersome to translate into an optimal IDP scoring system. Hence, I suggest we convert the raw statistical data into frequency data. This will give us a format that is much easier to evaluate and will ultimately allow us to get a better sense of how to attach a fantasy value to each stat. To do so, I’m actually going to use two measurements. I calculated both the percentage of snaps on which each stat occurs, and the snaps per each stat:

In this form, that data now gives us a more user-friendly picture of how often each stat occurs. For example, solo tackles are far and away the most commonly occurring statistic. A solo tackle was recorded on 79.4% of all snaps from 2008-2010, or one tackle was recorded for every 1.3 snaps. On the other hand, safeties were extremely rare. They occurred on just 0.05% of all snaps, which translates to a safety being recorded every 2127.9 snaps.

We can now use this data to compare statistics that may often be valued the same in IDP scoring systems. For example, the aforementioned sacks and interceptions. I often see these stats worth the same in IDP scoring systems, but is this really reflective of actual frequency? A quick glance at the above chart shows that you’re more than twice as likely to see a sack recorded than you would be to see an interception. This means that many leagues out there are undervaluing interceptions compared to their actual on-field occurrence.

I think we’re starting to get somewhere.

The last form of analysis I would like to perform in the first phase of this project deals specifically with solo tackles. Ross, Kevin, and I agreed that we need a baseline in our scoring system, and suggested solo tackles serve as this baseline because they are the most commonly occurring defensive statistic. We can then compare the other defensive statistics and their frequency to that of solo tackles to gain an even better sense of proper fantasy value. In the following chart, I give you what I call the “Solo Ratio.” By this, I mean how often each statistic is recorded compared solo tackles. Let’s take a look:

So here, we see for example, that a solo tackle is recorded 24.8 times more often than a sack. Again, this chart gives us insight into frequency, and now compares this frequency to our baseline statistic. With this foundational data in place, we now can move into the next phase of the project where we can start to formulate an actual scoring system. I’ll report back when we get some preliminary results together.

Until then, make sure your elbows are behind the table. House rules.

Feel free to hit Jeff up on Twitter with any questions or comments. You can follow him at – @JeffRatcliffe