Brian Hight’s NFL Picks – Week Eight

Advanced Metrics are Catching On

The use of Defensive-adjusted Value Over Average (DVOA) is spreading beyond the website Football Outsiders and the advanced metrics enthusiasts like the sports team at FiveThirtyEight to mainstream ESPN articles such as Why the Cowboys are Overrated.  You’re even likely to hear on air analysts refer to “efficiency” as a measure of the ranking of any given team’s offense, defense, or special teams. Slowly, the understanding that DVOA can be a more accurate measurement of how good or bad a team is than by using raw numbers is taking hold.

So, it should come as no surprise that, despite the stunning upsets from week 7 in the NFL, the more efficient teams, as ranked by DVOA, owned a 10-5 record straight up. However, as I observed in last week’s column, it’s a little more difficult to use DVOA to predict the point spread. The more efficient teams were 6-9 against the spread last week.

How I Did in Week 7?

My hypothesis from last week was that the greater the absolute difference in DVOA, the greater the point differential should be. Not exactly the case, at least last week.

The New England Patriots had a 30.4% efficiency advantage over the New York Jets and failed to cover. The Seattle Seahawks had a 47.4% efficiency advantage over the St. Louis Rams and lost. And, the Cleveland Browns, with a 52.8% efficiency advantage over the then winless Jaguars, became Jacksonville’s first conquest of the season.

So, the good news was that I went 3-1 from last week’s column, losing only the Seahawks game. The bad news, I took a huge hit in all the other games and fell to 42% accurate on the year.

It’s worth noting that the statistical gurus over at FiveThirtyEight have about a 48% accuracy rate using the Elo rating system (named after inventor Arpad Elo and widely used in soccer).

This Week’s Picks

This week I’ll be looking at three games with NFC West teams (the 49ers are on a bye), the Sunday night game, and the Monday night game. If you’d like to see all of my picks or pick against me, you can find me as Brian Hight’s OSN Picks in the Fans of the Seattle Seahawks (spread) group on ESPN’s Pigskin Pick’em.

Also, for a more in depth explanation of the advanced metrics I’m using, check out Football Outsiders.

St. Louis (2-4) at Kansas City (3-3) -6.5

Even despite the Rams’ victory over Seattle last week, St. Louis continues to rank near the bottom of the league with an overall DVOA of -15.7%. The only bright spot for the Rams is on special teams, largely as a result of the two defining plays against the Seahawks, where they have a 1.1% efficiency. Otherwise, they have a -4.6% offensive efficiency and a 12.2% defensive efficiency (remember negative numbers are better on defense).

Kansas City ranks 9^th in overall DVOA with 9.9%, but also has a less than optimum defense with a 1.3% efficiency.

Whether St. Louis can match the Chiefs may come down to getting a good performance out of running back Zack Stacy. While the Rams offensive line ranks first in run blocking, they rarely run the ball, ranking 28^th, as Stacy is a below average back with a -10.5% DVOA. Kansas City ranks 25^th in run defense efficiency.

I like the Chiefs, with an absolute efficiency advantage of 25.6%, to win the game, but I think the continued maturation of Austin Davis at QB and the Rams’ O line should keep this game closer than 6.5 points.

Pick – St. Louis Rams

Seattle (3-3) -5 at Carolina (3-3-1)

The Seahawks appear to be reeling after losing two in a row, but it’s only recently that people have started to notice they’ve played the toughest schedule in the NFL so far. The Broncos, Packers, Cowboys, and Chargers all rank in the top 8 in the NFL in overall efficiency and all currently project as playoff teams. Seattle is 2-2 against these opponents. It’s the 1-1 record against the 24^th and 28^th ranked Washington Professional Football Team and the St. Louis Rams that’s a bit concerning.

But, still ranked 4^th overall in DVOA at 22.7%, the Seahawks are favored again on the road against a 25^th ranked Carolina Panthers team that struggles on defense with a 14.0% efficiency.

Carolina and St. Louis have similarly ranked defenses overall, but are much different against the run. Where the Rams rank 12^th against the run, the Panthers rank 29^th.

I look for Seattle to re-establish their game plan, pound the rock in Beast Mode, to both win this game and cover.

Pick – Seattle Seahawks

Philadelphia (5-1) at Arizona (5-1) -2.5

This is the only game of the week where the less efficient team is favored. Philadelphia ranks 6^th overall in DVOA at 14.5%, while Arizona ranks 15^th with 0.6%.

Much of the Cardinals mediocrity in overall efficiency can be attributed to an offense that has a -11.2% DVOA.  Even with the return of Carson Palmer at QB, don’t expect Arizona to improve that much on offense, as their offensive line ranks 25^th overall, 29^th in run blocking and 25^th in pass blocking.

The Eagles are actually better on defense than on offense with a -6.3% efficiency.

Given the explosive potential of the Eagles, despite a -2.9% offensive efficiency and the favorable defensive matchup against the Cardinals offense, I’m going to stick with the overall more efficient team, despite being on the road, and take the Eagles.

Pick – Philadelphia Eagles

Green Bay (5-2) at New Orleans (2-4) EVEN

Is the Superdome really worth what must be at least 5 points, if not a full touchdown? Probably. The Saints have won their last 10 games at the Superdome and are 7-3 ATS in that same span.

On a neutral field, the Packers should be the clear favorites. The absolute difference in overall efficiency between Green Bay and New Orleans is 30.7%, with the Packers holding an edge on offense, defense, and special teams.

Last week there were four games where the team with a 30% or greater advantage either lost or failed to cover. With that in mind, I’m going against the DVOA numbers and with the Saints at home.

Pick – New Orleans Saints

Washington (2-5) at Dallas (6-1) -10

The 10 point line is a little intimidating, especially when you factor in that the Cowboys are 3-7 against the spread the last 10 times they’ve been favored by more than 8 points.

I’m torn because Washington stinks with a 24^th ranked -10.3% overall DVOA, but also usually plays the Cowboys close in this NFC East rivalry.

Dallas ranks 7^th in overall DVOA with 13.7% for an absolute advantage over Washington of 24.0%, but Washington actually matches up pretty well against Dallas’ offense with a 8^th ranked run defense of -21.8%.

I’m going to take the Cowboys to win, but Washington to cover.

Pick – Washington Professional Football Team

Still a Lot of Football

After this week, we’ll be at the half way point and all DVOA numbers will factor in less than 8% of the pre-season projections. Remember that the overall more efficient team is winning the last two weeks at a 66% clip. Use DVOA for point spreads at your discretion.

Enjoy the games and we’ll do this again next week.