FIX THE PITCH

×

Fix the Pitch

How to craft compelling, dazzling pitchbooks. Thoughts, ideas, and inspiration to help construct advanced financial analysis, build stunning data visualizations and tips for mastering client meetings.

Popular tags


Ranking Chart: Who is most likely to win Super Bowl 50?

Eric RattnerEric Rattner

The holiday season is over, which in my opinion, means the actual most wonderful time of the year is here. The NFL playoffs.

Soon every sport talking head will be dissecting records, points scored, total yards, and which team has the hot quarterback—analyzing who is most likely to make it to the big game. But we think traditional metrics miss some important facets of playoff success: how efficient is a team at moving the chains? Who is likely to benefit most from playing at home? And what does it really mean if DeAngelo Williams’ ankle doesn’t recover in time for next week?

So, we devised some metrics of our own and added them to our favorite (lesser-known) existing stats in an attempt to tease out which of this year’s playoff field is well-situated to go the distance. While untested from a statistical standpoint, the metrics are grounded in fan intuition, achieved only by dedicating seven hours to the sport each Sunday.

At the time of publication, the Cardinals are considered by many to be Super Bowl favorites. Indeed, Vegas gives them the best odds of bringing home the Lombardi trophy. But our analysis suggests Arizona won’t be showing off new silverware anytime soon, given offensive production inconsistency over the course of the season and middling at-home performance—which is where most of their games will probably be held. Time will tell…

Behold, our 2015 NFL playoff team Ranking Chart.

How it was calculated

Super Bowl win probability

Source: FiveThirtyEight
How it was calculated: Adapted from the Elo computer rating system—a type of power ranking originally developed for chess players. FiveThirtyEight extended the Elo system to assess a win probability for individual games then used this in a simulation model to calculate each team’s probability of winning the Super Bowl.

Authentic Games index

Data source: Gregg Easterbrook
How it was calculated: A back of the napkin approach that, once the playoffs are determined, has successfully predicted the Super Bowl contestants since 2013. The metric is based on the qualitative estimation of the number of big games each team has played, and won, during the season. The idea is that the greater exposure, and success, a team has in a “playoff-like” atmosphere, the better it will perform in the high stakes of a playoff match.

Team efficiency

Data source: Football Outsiders
How it was calculated: The DVOA (Defense-adjusted Value Over Average) methodology measures the play-by-play success of each team compared to a league average baseline. The baseline takes into account numerous state factors such as down and distance to go, and strength of opposition. It treats each NFL game as a collection of individual successes and failures (a good first down run, followed by a sack on second down) with the team efficiency metric being the aggregate result of each micro-outcome.

Offensive success rate

Data source: NFLsavant.com
How it was calculated: Another Pellucid creation, this metric reflects the play-by-play success rate of a team in advancing its static goal of converting first downs. At the start of every offensive series a team has three downs before it punts (except in some rare four-down territory situations) to try and get a first down. Therefore, offensive success is measured as the ratio of yards gained divided by the quotient of additional yards required for first down and remaining downs to go. For example, a run play that gains four yards on first down would be considered a success, while a four-yard pass play on third and five would not. Taking fewer downs to convert more first downs should lead to more offensive and hence overall team success. This measure of performance can be applied to either offenses and defenses or be aggregated as a way to evaluate overall team efficiency.

Defensive Hog Index

Data source: Cold Hard Football Facts
How it was calculated: Produced since 2007, it estimates the defensive strength of teams by combining their ability to limit yardage per play, create negative pass plays (sack, interception), and limit the third down conversion rate of opposing offenses. Defensive Hog Index is simply the average team rank across these three metrics. If you believe defense stills wins championships, then this metric should help identify the teams you should bet on.

Injury leakage

Data source: Pro-Football-Reference.com
How it was calculated: We devised this methodology to account for how much a team’s performance may be impacted by its key players sitting on the bench due to injury. It is calculated by estimating the per-snap value of all the players on the field using an approximate value formula. Approximate value uses a combination of individual player stats, team stats, and postseason awards to give an apples-to-apples value estimate for all players across position groups. This means the relative importance of a left tackle and a third string cornerback can be compared. Injury leakage is the aggregate value of all of the currently injured players divided by the aggregate value of the team if it were at full strength. It estimates the proportion of each team’s regular season “production” that is likely to be unavailable for all or part of the upcoming playoffs due to injuries.

Scoring volatility

Data source: Pro-Football-Reference.com
How it was calculated: The simplest Pellucid-derived metric, scoring volatility rewards teams with consistent scoring success. Measured as the standard deviation of the points generated by teams in each of their games, this metric provides a barometer of how stable the scoring prowess is of any one team. While the longer regular season can absorb a team’s game-to-game variability, in the bracket format of the playoffs, teams that consistently score points will be favored over boom or bust performers.

Expected venue advantage

Data source: Pro-Football-Reference.com
How it was calculated: A Pellucid invention, expected venue advantage gives insight into how, on the margin, the expected location of a team’s games across the playoffs may impact its outcomes. First, each team is assigned a spread reflecting the average points difference between its home and away games over the 2015 season. Probabilities are then assigned to each possible path the playoff brackets could take, based on the Elo methodology. The expected venue advantage is the total number of excess points a team is expected to receive based on where it will likely play its games. High seeds (those likelier to play at home) who outperformed in home games are expected to have an advantage in the playoffs, all else equal.

So, if your team made the playoffs (I’m looking at you, Jets!) how are they looking? If you have any questions about the methodology, email me at eric@pellucid.com.

Learn more about Pellucid Analytics’ custom data analysis services. Visit www.pellucid.com.

Eric Rattner
Author

Eric Rattner

Investment banking lifer and native New Yorker. Broadway and movie buff and aspiring soccer star. Building innovative, beautiful charts filled with really smart data analysis.

Comments