NFL Week 11 | Coin vs. Machine
5-Model Unanimous Picks Have Won 70% of Games Against the Spread
NOV 18 2017
Week 1 - Week 2 - Week 3 - Week 4 - Week 5 - Week 6
Week 7 - Week 8 - Week 9 - Week 10 - Week 11 - Week 12
Week 13 - Week 14 - Week 15 - Week 16 - Week 17
Wild Card - Divisional - Conference - Super Bowl
5-Model Unanimous Picks Have Won 70% of Games Against the Spread
NOV 18 2017
Week 1 - Week 2 - Week 3 - Week 4 - Week 5 - Week 6
Week 7 - Week 8 - Week 9 - Week 10 - Week 11 - Week 12
Week 13 - Week 14 - Week 15 - Week 16 - Week 17
Wild Card - Divisional - Conference - Super Bowl
Welcome to Week 11 of Coin versus Machine. Let's dive into our weekly look at NFL picks and trends.
Last week served up the second-highest average margin of victory (13.7 points) of any week this season. It was the fifth-highest average margin of victory for a single week since the start of the 2015 season.
Seven of the week’s first nine games – from Thursday night through Sunday’s early afternoon games – were decided by eight points or less, and it looked like the week was off to a competitive start. But if you stopped watching after that, you were in luck. Sunday’s three late afternoon games, Sunday Night Football, and Monday Night Football were won by an average of 18.7 points. Four were total blowouts.
Three Week 10 games – New Orleans versus Buffalo (47-10), New England versus Denver (41-16), and Carolina versus Miami (45-21) – were all “high-scoring blowouts.” Those games made Week 10 the first week in exactly three years – and the fifth week since 2012 – in which three games had both 55 or more total points and a margin of victory of 24 or more. (You might recall that Week 3 featured the most “high-scoring close games” in NFL history, and Week 7 had the most “low-scoring blowouts” in NFL history.)
Last week served up the second-highest average margin of victory (13.7 points) of any week this season. It was the fifth-highest average margin of victory for a single week since the start of the 2015 season.
Seven of the week’s first nine games – from Thursday night through Sunday’s early afternoon games – were decided by eight points or less, and it looked like the week was off to a competitive start. But if you stopped watching after that, you were in luck. Sunday’s three late afternoon games, Sunday Night Football, and Monday Night Football were won by an average of 18.7 points. Four were total blowouts.
Three Week 10 games – New Orleans versus Buffalo (47-10), New England versus Denver (41-16), and Carolina versus Miami (45-21) – were all “high-scoring blowouts.” Those games made Week 10 the first week in exactly three years – and the fifth week since 2012 – in which three games had both 55 or more total points and a margin of victory of 24 or more. (You might recall that Week 3 featured the most “high-scoring close games” in NFL history, and Week 7 had the most “low-scoring blowouts” in NFL history.)
|
|
Looking Back at Week 10’s Picks
The three main models (“Machine”) again had a very pedestrian week, hovering right around or just below .500. Lady luck (“Coin”) fared a good bit better at 9-4-1 against the spread. It was the second time this season that the Coin beat the Machine by three games in a week. (The coin still lags on the year.)
The bigger story in Week 10 were the fourth and fifth models – David Wilson’s Nutshell Predictive and Jeff Sagarin’s Offense-Defense Method Predictions – which aren’t part of the “three main models” you see in the chart, but are overlaid to determine the five more nuanced pick tiers on a week to week basis.
Nutshell Predictive went 9-4-1 against the spread in Week 10 and is 26-11-3 against the spread over the past three weeks. Sagarin’s Offense-Defense Method went 10-3-1 against the spread in Week 10 and is 25-12-3 against the spread over the past three weeks. Sagarin’s Offense-Defense Method is now 81-61-4 on the season (57% win percentage) – seven wins better than any of the other four models covered here.
The big story on the season continues to be the picks that all five models agree on each week (referred to here as five-model unanimous picks or “Tier I best bets”). Those picks went 3-1 against the spread in Week 10 and are now a sensational 21-9 against the spread so far this season (70% win percentage). No other model or method comes anywhere close. It could very well be random, but it hasn’t really let up.
Tier II selections went 2-2-1 against the spread in Week 10 and are now 25-29-4 year-to-date. Tier III picks went 1-3 (9-11 YTD), Tier IV picks went 0-0 (8-7 YTD), and Tier V “toss-ups” went 0-1 (11-12 YTD).
The three main models (“Machine”) again had a very pedestrian week, hovering right around or just below .500. Lady luck (“Coin”) fared a good bit better at 9-4-1 against the spread. It was the second time this season that the Coin beat the Machine by three games in a week. (The coin still lags on the year.)
The bigger story in Week 10 were the fourth and fifth models – David Wilson’s Nutshell Predictive and Jeff Sagarin’s Offense-Defense Method Predictions – which aren’t part of the “three main models” you see in the chart, but are overlaid to determine the five more nuanced pick tiers on a week to week basis.
Nutshell Predictive went 9-4-1 against the spread in Week 10 and is 26-11-3 against the spread over the past three weeks. Sagarin’s Offense-Defense Method went 10-3-1 against the spread in Week 10 and is 25-12-3 against the spread over the past three weeks. Sagarin’s Offense-Defense Method is now 81-61-4 on the season (57% win percentage) – seven wins better than any of the other four models covered here.
The big story on the season continues to be the picks that all five models agree on each week (referred to here as five-model unanimous picks or “Tier I best bets”). Those picks went 3-1 against the spread in Week 10 and are now a sensational 21-9 against the spread so far this season (70% win percentage). No other model or method comes anywhere close. It could very well be random, but it hasn’t really let up.
Tier II selections went 2-2-1 against the spread in Week 10 and are now 25-29-4 year-to-date. Tier III picks went 1-3 (9-11 YTD), Tier IV picks went 0-0 (8-7 YTD), and Tier V “toss-ups” went 0-1 (11-12 YTD).
|
|
Looking Ahead to Week 11’s Picks
Week 11 offers up four three- and five-model unanimous picks (Tier I “best bets”): Pittsburgh (-6.5) versus Tennessee, Detroit (-3.0) at Chicago, Green Bay (+2.0) at Baltimore, and Dallas (+5.5) versus Philadelphia. The Lions, Packers, and Cowboys were Tier I last week, which makes you wonder whether the models like them more than the oddsmakers do. (Detroit and Green Bay covered. Dallas did not.)
You’ll again want to keep an eye on injuries and suspensions, which can take a while to flow through the models but are immediately reflected in weekly lines. The models love Dallas but don’t know about Ezekiel Elliott’s suspension; they love Green Bay but the Aaron Rodgers injury probably hasn’t made its full way through; and they really like Houston but don’t know that Tom Savage is under center. (Picks with noteworthy suspensions or injuries are noted with an asterisk at the bottom of the table below.)
And as always you’ll want to keep on eye on your lines. The three-model majority (“machine”) pick is Houston at -1.5 but Arizona at +2.0; Detroit at -3.0 but Chicago at +4.0; Minnesota at -2.5 but the L.A. Rams at +3.0; Denver at -2.5 but Cincinnati at +3.0; and finally Atlanta at +3.0 but Seattle at -2.0.
Week 11 offers up four three- and five-model unanimous picks (Tier I “best bets”): Pittsburgh (-6.5) versus Tennessee, Detroit (-3.0) at Chicago, Green Bay (+2.0) at Baltimore, and Dallas (+5.5) versus Philadelphia. The Lions, Packers, and Cowboys were Tier I last week, which makes you wonder whether the models like them more than the oddsmakers do. (Detroit and Green Bay covered. Dallas did not.)
You’ll again want to keep an eye on injuries and suspensions, which can take a while to flow through the models but are immediately reflected in weekly lines. The models love Dallas but don’t know about Ezekiel Elliott’s suspension; they love Green Bay but the Aaron Rodgers injury probably hasn’t made its full way through; and they really like Houston but don’t know that Tom Savage is under center. (Picks with noteworthy suspensions or injuries are noted with an asterisk at the bottom of the table below.)
And as always you’ll want to keep on eye on your lines. The three-model majority (“machine”) pick is Houston at -1.5 but Arizona at +2.0; Detroit at -3.0 but Chicago at +4.0; Minnesota at -2.5 but the L.A. Rams at +3.0; Denver at -2.5 but Cincinnati at +3.0; and finally Atlanta at +3.0 but Seattle at -2.0.
Tiered picks are based on the five models listed below (methodology)
Good luck and enjoy Week 11!
|
|
The five models referenced are created by and sourced from FiveThirtyEight ("Elo"), Jeff Sagarin via USA Today ("Rating" and "Offense-Defense Method Projections"), RP-Excel.com ("R-P Excel I"), and NutshellSports.com ("Nutshell Sports NFL Ratings"). Lines are sourced from FootballLocks.com as of Nov 18 at 12pm PT. Data was compiled and analyzed by ELDORADO. All charts and graphics herein were created by ELDORADO.
ELDORADO | Berkeley, CA | New York, NY
eldo.co | @eldo_co
ELDORADO | Berkeley, CA | New York, NY
eldo.co | @eldo_co
