By Owen Tedford
How do you measure the best? Is it quantifiable at all or is it the intangibles that could never be measured that make a team better than another one? This question has been an issue for many years in college football and has been complicated in recent years with the creation of a four-team playoff. The problem is selecting which metric should be given the most weight, which has led to the creation of a number of new metrics for measuring the best.
One that I find most intriguing is the idea of strength of record, created by ESPN this year. It measures the probability of an average Top 25 team having the same record against the same schedule. To me, this seems like the best metric that is out there that I know of. But what I find interesting about ESPN’s use of this metric is why they don’t calculate it for the NFL, which leads me to my next question of why do we not question the NFL playoffs as much as the college football playoffs? We accept record as the metric of who is best without taking into account strength of schedule or all of the other factors that can lead to a better or worse record. With this in mind, I set out to create my own metric, inspired by strength of record, comparing strength of schedule and team’s records.
To come up with the exact metric, I graphed strength of schedule against a team’s winning percentage. Next, I created a line of best fit, which predicts what a team’s record should be given their strength of schedule based on the current trend in the NFL season. Then, by measuring the vertical distance between the line of best fit and a given point of a team, I could tell how much they were under or over performing for their schedule with negative values indicating an underperforming team and positive values indicating an overperforming team. For example, the Patriots have a winning percentage of 0.8 and a predicted winning percentage, based on their strength of schedule at 19th, of 0.5237. Then, the difference between these is 0.2763, meaning that the Patriots are outperforming their schedule by that amount. This number alone indicates how much higher the Patriots win percentage is than it should be, however it is more valuable when used with the rest of the teams for comparison as this allows for ranking. An example of this would be how the Cowboys have a difference of 0.258, which, when compared to the Patriots, means the Patriots are the better team. Below are the full graph and table through Week 11 of the NFL.
So what does this tell us? Who is the best? Well, by this metric, the Raiders are the best team in the NFL through Week 11. Who is the worst? No surprise, the winless Browns, who according to the model should have almost four wins. Now, are there any surprises in this in terms of what perhaps people perceive and what this says? Well, the Cowboys aren’t quite as good as their record indicates, though they’re still part of a clear top tier of the Raiders, Seahawks, Patriots, Cowboys, Falcons, and Broncos, which is separated by 0.1 over the Giants. As for other surprises, the Lions, despite leading their division, are still underperforming their schedule, as they are the worst team with a winning record, which would mean they could see a collapse as the season goes on.
What else does this say we can look for going forward, in particular, playoff considerations? The Steelers should win the AFC North despite trailing the Ravens at present. The Texans, who for a time were in a tight race for the AFC South, should walk away pretty easily with it. The Patriots should win the AFC East but could face late pressure from the Dolphins and Bills down the stretch who are both starting to come into their stride, perhaps that early season loss to the Bills wasn’t as fluky as it may have seemed. Then, the AFC West should be a wild ride to the finish, starting with the Chiefs against the Broncos on November 27. Based on this metric, the Broncos are 71% favorites, but it should be a fun game to watch nonetheless. Wildcard seeds will likely go to at least one AFC West team, if not two AFC West teams. Pressure on the AFC West could come from the Bills or the Dolphins in the AFC East and on an outside chance the runner-up in the AFC North, either the Steelers or Ravens, likely. But if I had to guess today, both wildcard seeds will go to AFC West teams. Whether these teams are truly the best, we will have to wait for the end of the season to see.
In the NFC, the South seems fairly secure for the Falcons and the West should be a safe bet for the Seahawks. The two messy races are in the North and the East. The North is a two-team race between the Lions and the Vikings, with this metric giving a slight edge to the Vikings, but it should be close. A key game to highlight in this divisional matchup is Thanksgiving’s game between the Lions and the Vikings, especially with the Lions having won before, and looking to complete the season sweep and gain the head-to-head tiebreaker. Meanwhile, the East has a three, possibly even four, team race with the Giants and Redskins both applying pressure on the Cowboys. No doubt, the Cowboys are in the driver’s seat for the division, but it won’t be for a few weeks until the division is decided. The wildcards in the NFC will probably go to two teams from the East, likely the Giants and Redskins, but the Eagles will certainly put pressure on, and the North runner-up could as well.
For now, we sit back and enjoy the football, and just try to think, who is the best, and how do you prove that?
Table of NFL Strength of Schedule scores and records by team in 2016. Note that the Distance column represents how much better a team performed over the strength of schedule expected winning percentage trendline
|Team||Strength of Schedule (easier is higher)||Record (%)||Distance|