≡ Menu

Implied SRS Ratings for NFL in 2017

Back in May, CG Technology released point spreads for all NFL games during the first 16 weeks of the 2016 season. We can use these lines to generate implied NFL ratings — as of May 10, 2017 — for this upcoming season.

Basically, we take the point spread in each game, adjust for home field, and then determine how by many points Vegas thinks Team A is better than Team B.  When the Seahawks are favored by 13.5 points in a home game against the Rams, we can take this to mean that Vegas thinks the Seahawks are 10.5 points better than Los Angeles.  When Seattle is a 6-point road favorite in Los Angeles against the Rams, that tells us that Vegas thinks the Seahawks are 9 points better than the Rams.  That’s just two games, of course: Using the iterative SRS process, we can generate season ratings based on the 240 point spreads involved. Here are those ratings, again as of May 10, 2017.

Here’s how to read the table below. After adjusting for home field, the Patriots are expected to beat their average opponent by 6.6 points. On average, New England’s opponents (after adjusting for *their* strength of schedule) are 0.3 points better than average, which means the Patriots are expected to be 6.8 points better than average (difference due to rounding). That’s the best in the league, far ahead of the Seahawks, Cowboys, and Packers (the only other teams that are 4 points better than average). [click to continue…]

{ 11 comments }

The Patriots and the Spread, Part I

Since 2001, the Patriots have been favored to win in a whopping 79% of all games, including postseason (giving half-credit as a favorite in games where the spread is zero). The Steelers are second at 73%, the Packers and Colts are next at 69%, the Eagles are at 68%, the Broncos at 67%, and the Saints at 61% are the only other team over 60%. In other words, the Patriots have been in a class by themselves when it comes to being favored.

But even that kind of underrates New England. The Patriots weren’t favored in any of the first 8 games of the 2001 season; the team was only favored in one of its first 12 games, at which point in time New England had a 7-5 record (and an 8-4 mark against the spread). There have also been 19 games since 2001 where Tom Brady was not the starting quarterback, and the Patriots were underdogs in 4 of those games (and a pick’em in a fifth). And there were meaningless week 17 games in 2006 and 2009 that the Patriots were underdogs because they were projected to rest their starters.

The graph below shows how many points the Patriots were expected to win in each game, regular and post-season, since 2001. I have included as red dots games not started by Brady or during meaningless week 17 games: [click to continue…]

{ 1 comment }

Why Haven’t We Improved At Making NFL Predictions?

Yesterday, we looked at the biggest “covers” in NFL history: those games where the final score was farthest from the projected margin of victory. In a 2010 game in Denver, the Raiders were 7-point underdogs, but beat the Broncos by 45 points. That means the point spread was off by 52 points, the most in any single game.

The first year we have historical point spread data was in 1978. That year, the average point spread was off by (or the average amount of points by which the favorite covered by was) 9.9 points. That number probably doesn’t mean much to you in the abstract, so let’s give it some context. From 1978 to 1982, the average point spread was off by 10.4 points. Over the last five years, the average point spread has been off by… 10.3 points.

Now I’m not quite sure what you expected, but isn’t that weird? In 1978, Vegas bookmakers were using the most rudimentary of models. Think of how farther along we are when it comes to football analytics than we were four decades ago. All of that work, of course, has to have made us *better* at predicting football games, right?

But don’t these results suggest that we are not any better at predicting games? If Vegas was missing games by about 10 points forty years ago, why are they still missing games by about 10 points? One explanation is that the NFL is harder to predict now, which… well, I’m not so sure about that. After all, even if you think free agency and the salary cap bring about parity (which is a debatable position regardless), it’s not like the lines are more accurate later in the season once we know more information. Games are also slightly higher scoring, and you could make the argument that we should be measuring how far games are off by as a percentage of the projected over/under?

Let’s look at the data. The graph below shows in blue the average “cover” in each game for each year since 1978.  As it turns out, 2016 was a really good year for Vegas — the average cover was just 9.0 points, which ranks as the most accurate season ever.  However, there’s no evidence that this was anything more than a one season blip: 2013 and 2015 were average years, and 2014 was the least accurate season ever.  It’s not like our prediction models just started getting sophisticated last season.

For reference, in the orange line, I have also shown the average point spread for each game.  That line has also been pretty consistent over time, with the average spread usually being just above 5 points. [click to continue…]

{ 15 comments }

Biggest Covers In Vegas History (1978-2016)

Last week, I noted that the Colts/Vikings game was the least-conforming game of the 2016 season. Here’s what I wrote then:

The Colts were 0.2 points per game better than average last year, as measured by the Simple Rating System (which takes the points scored and allowed in each game, and adjusts for opponent strength and home field advantage).

The Vikings were 0.9. points per game better than average in 2016, and hosted the Colts in week 15. Given those facts, we would expect Minnesota to have won by 3.7 points. Instead, Indianapolis upset the Vikings, 34-6, beating the expected line by 31.7 points. That was the least-conforming game of 2016.

Well, it wasn’t just the SRS that found that game to be pretty odd. Our friends in the desert expected the Vikings to win by 5 points, which means the Colts covered the point spread by a whopping 33 points.  Two weeks earlier, Indianapolis was actually an underdog in a Monday Night Football game that you would have had to been an idiot to attend in person.  The Colts were 1-point underdogs, but won by 31 points, giving Indianapolis a 32-point cover.  Those were two of the three biggest covers of the year, with the Eagles 34-3 win over Pittsburgh as 3.5 point underdogs (+34.5) being the biggest cover of 2016.

At Pro-Football-Reference.com, we have points spread data going back to 1978. Below are the biggest covers in history: [click to continue…]

{ 2 comments }

Super Bowl Streaks And Conference Affiliation

The NFL and the Lombardi Packers won the first two Super Bowls. Then, each conference went on a long streak:

  • The AFL/AAFC won 11 of the next 13 Super Bowls (1968-1980): the Jets and Chiefs closed out the AFL with Super Bowl upsets, while the Steelers, Dolphins, and Raiders carried the AFC.
  • Then, from 1981 to 1996, the NFC won 15 of the next 16 Super Bowls, with the 49ers and the NFC East teams (well, not all of them) carrying the conference to 13 of those titles.
  • The balance shifted then to the AFC, as the conference won 8 of the next 10 Super Bowls (1997 to 2006).  The Patriots won three of those, but perhaps most surprising was that the run ending with 18-0 New England losing as heavy favorite to the Giants.

Since then? The NFC went on a mini-run, winning 6 of 8 Super Bowls from 2007 to 2014.   The AFC has responded by winning the last two Super Bowls, and the conference is again a favorite in Super Bowl LI. Here are the results in graphic form, with NFL/NFC wins in blue, and AFL/AFC wins in red: [click to continue…]

{ 9 comments }

The median amount of points scored in Falcons games this year is 57 points; in Packers games, that number is 53 points. So why is the over/under 60 points?

The average isn’t much different: there have been 58.9 points scored in Atlanta games, and 52.0 points scored in Green Bay games. You may be surprised to learn that during Green Bay’s 8-game winning streak, there have been 51.6 points scored per game: 32.1 by the Packers, and 19.5 by Packers opponents.

Of course, what’s really driving these numbers is not the points scored by both teams in these games, but by both offenses. The Packers are averaging 28.0 points per game this year over 18 games, 32.1 points per game during this 8-game winning streak, and 34.8 over the team’s last 5 games. Atlanta is averaging 33.9 points per game over 17 games, and 38.0 points over their current 5-game winning streak.

So by that line of thinking, a 60-point over/under probably feels low. But it is currently (the line may change) tied for the 2nd highest over/under of any game since 1978, with the only other playoff game on the list:

WinnerLoserYearWeekBoxscoreLineOver/UnderPFPATotal Pts
STLSFO20009Boxscore-763342458
NORDET2011WCBoxscore-10.560452873
KANOAK200416Boxscore-9.560313061
CARSTL200010Boxscore13.559.5272451
DENWAS20138Boxscore-1158.5452166
STLIND200116Boxscore-1358.5421759
STLATL20007Boxscore-1858.5452974
GNBNWE201413Boxscore-358262147
INDMIN20049Boxscore-758312859
CARSTL200014Boxscore85816319
DENPHI20134Boxscore-11.557.5522072

What do you think? Over or Under?

{ 5 comments }

Browns Continue To Lose … Against the Spread, Too

Cleveland is 0-14 this year, but that’s maybe not the saddest stat of the Browns season. Everyone expected Cleveland to be bad, but Cleveland has also been really bad relative to expectations. After losing against on Sunday to the Bills, and again failing to cover against the spread, the Browns are now just 2-12 against the spread this season.

Cleveland covered in games against the Dolphins and Titans early in the year, but the Browns have now failed to cover the spread in eight straight games. The graph below shows the number of points Cleveland was expected to lose by in black, and the actual points differential in orange. Since the Browns have been underdogs and lost every game, the range goes from 0 to -30: [click to continue…]

{ 2 comments }

Final 2016 NFL Division Odds

I never quite know what to post on the first Sunday of the season, so here are the final NFL 2016 Division Odds (as of prior to the Thursday night game). In each case, I’ll be showing the odds for each team to win its division as of May 23rd and as of September 8th, along with their vig-adjusted percentage based on those odds. Let’s start with one of three divisions with a clear favorite, and one of just two divisions with only one team having at least a 15% chance of winning the crown.

AFC East

Team5/23 Odds9/8 Odds5/23 Perc9/8 Perc
New England Patriots     5/115/1261.5%64.6%
Buffalo Bills            21/46/114.3%13.1%
New York Jets            21/413/214.3%12.2%
Miami Dolphins           8/18/19.9%10.2%

The Patriots raw odds actually went up over the last few months, even with the Tom Brady suspension becoming official. But because the Jets and Bills have seen their percentage drop, New England’s odds of winning another AFC East crown have increased. Buffalo’s had a rough offseason due to injuries, but it’s a little harder to explain why the Jets odds have gone down. New York has a very difficult schedule, which may play a part in that.

AFC North

Team5/23 Odds9/8 Odds5/23 Perc9/8 Perc
Pittsburgh Steelers      5/46/539.6%41.8%
Cincinnati Bengals       7/42/132.4%30.7%
Baltimore Ravens         11/47/223.8%20.4%
Cleveland Browns         20/112/14.2%7.1%

These odds have been pretty static, but a (very) small Cleveland hype train has emerged. Vegas sees the top 3 teams as very strong, but does have a clear 1-2-3 order, too. [click to continue…]

{ 0 comments }

Super Bowl 51 Odds: Top 6 Teams Or The Field?

Fun article over at SB Nation on national championship odds in college football for this season. The question the folks were trying to answer was what is the smallest number of teams you could group together to give you a 50/50 (or better) shot of containing the eventual champion?  As it turns out, most people thought “taking the top five or six teams presents close to a fair wagering opportunity.”

What about in the NFL? Well, despite the presence of nearly 100 fewer teams, the answer is about the same.  The Patriots, Packers, Panthers, Steelers, Seahawks, and Cardinals form the upper crust of the NFL, at least according to Vegas odds.  Together, that group has about a 50% chance of containing the Super Bowl 51 champion.

Take a look at the odds from Football Locks, which is pretty similar to the odds at other places.  Here’s how to read the table below: The Patriots have 15/2 odds, which translates to 1 out of 8.5, or 11.8%. That includes a vig, tho, and if we remove the vig from each team, that drops the Patriots odds to 9.9%, which is a better approximation of New England’s real odds. I then sorted the teams in the NFL by that number, and calculated the cumulative Super Bowl percentage — after six teams, it’s pretty close to a 50/50 proposition. [click to continue…]

{ 5 comments }
Coaches of two of the top 3 teams in college football... again.

Coaches of two of the top 3 teams in college football… again.

Meet the new boss, Nick Saban as always.

The Golden Nugget released the point spreads for 100 games this season, and Johnny Detroit was kind enough to pass along that data for purposes of this post.  With only data for 100 games, how am I able to conclude that Vegas views Alabama as the best team (or, at least, one of the top 2 teams)  in college football? Consider:

  • Alabama is a 6-point road favorite at Ole Miss this year. That is the only game this year (of the seven we have lines for) where Mississippi is an underdog, and the Rebels are an 8-point home favorite against Auburn and a 4.5-point home favorite against Georgia.  The Rebels finished 10th in the polls last year and are projected to be the 10th-best team this year, so this line says all you need to know about Alabama.
  • Against Auburn, Alabama is a 15-point home favorite (that’s a touchdown better than Ole Miss is against Auburn).   The Tigers were not great last year, but are still projected at #20 this year.
  • In Arkansas, the Crimson Tide are 8.5-point favorites.  In the other 3 home games for Arkansas, the Razorbacks are 7.5-point dogs to LSU (the #3 team by this methodology), 1-point underdogs to Mississippi, and a 2.5-point favorite against Florida.
  • Alabama is a 15-point favorite at home against Mississippi State and a 14-point home favorite against Texas A&M.  Both of those teams are projected to be, by Vegas, top 30 teams this year.
  • In Tennessee, Alabama is a 1-point dog, but the Vols are projected as the 6th best team this year! Tennessee is a pick’em in Georgia, a 5-point favorite in College Station, an 11-point favorite at home against Florida, and a 13-point favorite in a neutral site game against Virginia Tech.
  • LSU is projected to be the 3rd best team in college football. The Tigers are an 11-point favorite at home against MSU, a 9.5-point home favorite against Ole Miss, 7.5-point road favorites in Florida and Arkansas, a touchdown favorite in Auburn, a 6-point favorite in College Station, and – only – a 2.5-point home favorite against Alabama.

You may be wondering, how do we know how good Alabama’s opponents are? Well, we can imply the ratings of each team in college football based on these points spreads.  I explained how to do this last year, but here is the refresher:

The system is pretty simple: I took the point spread for each game and turned it into a margin of victory, after assigning 3 points to the road team in each game. Do this for every game, iterate the results hundreds of times ala the Simple Rating System, and you end up with a set of power ratings.

Two quick notes about the rankings.

1) These are not intended to be surprise. The methodology may be somewhat complicated, but all these ratings are intended to do is quantify public perception.

2) These are not “my” ratings. These are simply the implied ratings based on the Vegas (or, more specifically, the Golden Nugget) points spreads; nothing more, nothing less.

Below are the ratings for 51 college football teams. In the table below, I’ve included the number of games for which we have point spreads for each team on the far left. The “MOV” column shows the home field-adjusted average margin of victory for that team, the “SOS” column shows the average rating of each team’s opponents (for only the number of games for which we have lines), and the “SRS” column shows the school’s implied SRS rating. As you can see, Alabama is projected to be the strongest team in college football, but Oklahoma is just a hair behind: [click to continue…]

{ 6 comments }

A Look at 2016 Vegas Futures Win Totals

Bovada has released futures wins totals for the 2016 season. Five teams are set at 10.5 wins, but not all teams with X numbers of wins are equal. For example, if you want to bet on the Packers going over 10.5 wins, you need to put down $165 to win $100, which translates to a 62.3% chance of success. If you want to bet against Green Bay, an Under bet of $100 brings back $135, implying a 42.6% chance. Those odds will always add up to over 100% because of the vig of about five percent. Remove that, and these lines have Green Bay pegged at about a 59% chance of going over 10.5 wins. Conversely, Pittsburgh is given a true 50/50 chance at going over 10.5 wins: you have to bet $115 to win $100 on the Steelers either going over or under 10.5 wins.

RkTeamWinsOverUnderOver %
1Green Bay Packers       10.5-16513559%
2New England Patriots 10.5-15012057%
2Seattle Seahawks10.5-15012057%
4Carolina Panthers10.5-130EVEN53%
5Pittsburgh Steelers      10.5-115-11550%
6Arizona Cardinals9.5-16013059%
7Cincinnati Bengals       9.5-14011055%
8Kansas City Chiefs        9.5-130EVEN53%
8Minnesota Vikings       9.5-130EVEN53%
10Dallas Cowboys    9.5EVEN-13047%
11Indianapolis Colts        9.5110-14045%
12Denver Broncos    9-115-11550%
13Houston Texans   8.5-13510554%
14Oakland Raiders   8.5-115-11550%
15Baltimore Ravens8.5110-14045%
16New York Giants  8-16013059%
17Buffalo Bills    8-115-11550%
18New York Jets8EVEN-13047%
19Jacksonville Jaguars    7.5-15012057%
20Chicago Bears 7.5-115-11550%
20Washington Redskins 7.5-115-11550%
22Atlanta Falcons     7.5120-15043%
22Los Angeles Rams7.5120-15043%
22Tampa Bay Buccaneers      7.5120-15043%
25Detroit Lions   7-130EVEN53%
26Miami Dolphins    7-115-11550%
27New Orleans Saints     7EVEN-13047%
27Philadelphia Eagles     7EVEN-13047%
29San Diego Chargers      7105-13546%
30Tennessee Titans5.5-16013059%
31San Francisco 49ers      5.5-115-11550%
32Cleveland Browns        4.5-130EVEN53%

[click to continue…]

{ 1 comment }

It’s safe to say that no team has exceeded expectations through two weeks quite like the Jets. In week 1, New York was a 3.5-point home favorite against the Browns, but won by 21 points (a 17.5-point cover). In week 2, the Jets won 20-7 in Indianapolis, despite being 7-point underdogs (a 20-point cover). The Jets are the only team to cover by 17+ points in each of the first two weeks; in fact, Arizona (+10 against New Orleans, +23 against Chicago) is the only other team to even cover by at least five points in both games so far.

The last team to pull off this feat? The 2007 Patriots. Yes, another day, another Tom Brady/Ryan Fitzpatrick comparison. From 1978 to 2014, there were 19 teams that covered by at least 17 points in each of their first two games. How did those teams do the prior year, and during the rest of that season?

I’ve included the relevant data for each team in the table below. Here’s how to read the line of the ’06 Chargers. San Diego covered by 24 points in week 1, and 21 points in week 2. The Chargers won 9 games in 2005, but the hot start in ’06 was a sign of things to come, as San Diego won 14 games. That was an improvement of 5 wins, although the Chargers season ended in the Division round of the playoffs. [click to continue…]

{ 0 comments }

What Can We Learn About The 49ers Defense From Week 1?

Yesterday, we looked at what Tennessee’s offensive explosion in week 1 might mean for the rest of the year. Today, let’s do the same but for the 49ers defense. The 49ers were 2.5-point underdogs against Minnesota in week one, and the Over/Under in the game was 41.5 points. This translates to a projected a final score of 22-19.5 in favor of Minnesota. As it turns out, San Francisco won the game, 20-3, which means the Vikings were held 19 points below their expected total. That’s the 4th best performance by a team by this methodology since 2002.

The most impressive game? That came in 2003, in the Lawyer Milloy game. The Bills shut out New England, 31-0, while the pre-game spread projected New England to score 21.75 points. That wasn’t a sign that Buffalo was about to break through (the team finished 6-10), but it did provide some insight into a Bills defense that jumped from 27th (in 2002) to 5th (in 2003) in points allowed. [click to continue…]

{ 1 comment }

The Titans were 3-point underdogs against Tampa Bay in week one, and the Over/Under in the game was 41 points.  This translates to a projected a final score of 22-19 in favor of Tampa Bay. Of course, Tennessee scored 42 points, outscoring its projection by a whopping 23 points, tied for the fourth biggest number in all week 1 games since 2002.  In the graph below, I’ve plotted each team’s expected points scored in week 1 on the X-axis, and their actual week 1 score on the Y-axis. [click to continue…]

{ 4 comments }

Implied SRS NFL Ratings For 2015

In each of the last two years, I’ve derived implied SRS ratings for NFL teams based on the Vegas point spreads (I also do the same for college football teams). Well, in late April, CG Technology released lines for 238 NFL games. Things have changed since late April, of course, but for now, let’s work with that data.

For the third straight year, Seattle, Denver, New England, and Green Bay are ranked among the top five teams in the NFL. And before you ask, yes, we will get to the Tom Brady issue in a few moments. The Seahawks are underdogs in just one game this year, and even in that game, Seattle is a just 1-point underdog in Green Bay. The Packers are underdogs in just one game, too: Green Bay is a 1.5-point underdog during a week 8 trip to Denver. On the other side, the Raiders aren’t favored in a game all year: the closest is a pick’em when the Jets come to Oakland.

As a reminder, we can use the Simple Rating System to take all 238 point spreads and derive ratings. But as a sign of how good Vegas viewed Seattle, consider these four Seahawks road lines: [click to continue…]

{ 15 comments }

The Golden Nugget has released point spreads for a large number of college football games.  And these spreads can tell us a lot about how Vegas views these teams.  That’s because, for the most part, the spreads are consistent.

Let’s look at Ohio State, the defending national champions and a team the Golden Nugget released lines for four games. The Buckeyes are 14-point home favorites against Michigan State, 16-point road favorites against Michigan, 19-point home favorites against Penn State, and 16-point road favorites against Virginia Tech. So how good is Ohio State? Well, that depends on how good Michigan State, Michigan, Penn State, and Virginia Tech are. As it turns out, those teams aren’t half bad, so Ohio State must be really, really good. Let’s ignore the games where two of Michigan State, Michigan, and Penn State play each other (since that won’t tell us much about Ohio State), and look at the rest:

  • Michigan State is a 6-point road favorite in Nebraska and a 1-point home favorite against Oregon. This would imply that Ohio State is about 9 points better than the Ducks1, an annual college football contender.
  • The only non-Big 10 game for Penn State where a line was released was Penn State -28 against Army.
  • Michigan is a 33-point home favorite against UNVL, a 4-point road dog against Utah, a 14-point home favorite against Oregon State, and a 7-point home favorite against BYU. The Wolverines aren’t great, but remember that Ohio State is favored by 16 against them in Ann Arbor.
  • Virginia Tech is a 9-point home favorite against Pittsburgh, a 4-point road favorite against virginia, a 9.5-point road dog against Georgia Tech, and a 6-point road dog against Miami. And, remember, a 16-point home dog against Ohio State.

But we don’t need to strain our brains trying to piece together these ratings. As I showed last year and in 2013, we can take the point spreads from each game to determine what Vegas’ implied ratings are for 70 college football teams. [click to continue…]

  1. Michigan State would be viewed as 2 points worse on a neutral field than Oregon, while being 11 points worse than Ohio State on a neutral field. []
{ 11 comments }

An Early Look at 2015 Vegas Win Totals

Like last year, CG Technology (formerly Cantor Gaming) is the first Las Vegas book to release win totals. For your convenience, I have produced them below, and sorted the list by the difference between 2015 Vegas wins and 2014 wins. [click to continue…]

{ 3 comments }

The physicist Werner Heisenberg (this guy, not this guy) found that observers affect the systems they attempt to measure, something that is related to but actually separate from his Uncertainty Principle. Even if Heisenberg was thinking about submicroscopic particles whizzing around, his ideas can still apply to writing about NFL betting. Writing about my bets could change the sequence of events that follow, at least in theory, just like all the other actions people take everywhere that put the world on a different course. The NFL season that just unfolded was just one of an infinite number of potential seasons that could have happened. In what share of the possible seasons did my pick for the NFL’s worst team start the season 9-1? Am I just the worst predictor ever, someone dumb enough to underestimate the great Arians and the new great HC of the NYJ? Or was I tempting fate by writing about real bets?

Since I am supposed to be a coldly-rational, data-driven guy, I am going to chance it and review my NFL betting this year. This is risky since my betting year could still be saved by events yet to be determined. Before I get to all that, I am hoping that maybe my writing about football can influence something much more plausible, namely whether I attend the Super Bowl next week. Apologies for this distraction, but I could really use some help.

***HUMBLE REQUEST BEGIN***

If you have read any of my stuff here or on Football Outsiders, you may know that I am a Patriots fan. Sufficiently dedicated to have flown from Los Angeles to Boston for the Ravens game, then back to LA for the first week at Loyola Marymount, before flying back to Boston for the Colts game. Now I am hoping to obtain two tickets to the Super Bowl. Here is what I can offer: [click to continue…]

{ 6 comments }

Records Against the Spread

The Titans lost to the Jaguars last night, dropping Tennessee’s record to a woeful 2-13. The 2014 season started off nicely for the Titans, who upset the Chiefs in Kansas City, 26-10, on opening day. Since then, not only has Tennessee gone just 1-13 (the sole win being a 2-point home victory against Jacksonville), but the team is a mind-bogglingly poor 2-11-1 against the spread.

Points spread data is not official, of course, and some sources of data are better than others. Using what is available at Pro-Football-Reference, I calculated the worst teams against the spread since 1978. If the Titans fail to cover next week against the Colts, they will end the year at 3-12-1 against the spread. That would make them one of just 13 teams since 1978 to post such a poor ATS record. On the other hand, it would only tie them with another AFC South team from the past two years:

TeamYearWLTwin%ATS WATS LATS TPerc
BAL200751100.31331300.188
NWE198121400.12531300.188
PIT19809700.56331300.188
CIN198741100.26731200.2
HOU201321400.12531210.219
STL201121400.12531210.219
NYG200341200.2531210.219
OAK200341200.2531210.219
DAL199761000.37531210.219
HOU199421400.12531210.219
BAL198121400.12531210.219
SFO197821400.12531210.219
HOU19821800.1112700.222
PHI201241200.2541200.25
TAM201141200.2541200.25
CAR201021400.12541200.25
JAX200851100.31341200.25
STL20027900.43841200.25
CIN200221400.12541200.25
ARI200031300.18841200.25
OAK199741200.2541200.25
CIN199131300.18841200.25
RAM199131300.18841200.25
NWE199011500.06341200.25
NYJ198941200.2541200.25
NOR198551100.31341200.25
ATL198441200.2541200.25
HOU198431300.18841200.25
DEN20088800.541110.281
PHI200561000.37541110.281
SFO200210600.62541110.281
NOR199931300.18841110.281
CIN199831300.18841110.281
NYJ199241200.2541110.281
DEN199051100.31341110.281
MIA198861000.37541110.281
DET197921400.12541110.281
CHI20138800.541020.313
WAS201331300.18851100.313
OAK201241200.2551100.313
KAN201221400.12551100.313
CLE201051100.31351100.313
ARI201051100.31351100.313
DEN201041200.2551100.313
JAX20097900.43851100.313
DET200921400.12541020.313
DEN20077900.43851100.313
STL200731300.18851100.313
DEN20069700.56351100.313
STL200561000.37551100.313
NOR200531300.18851100.313
SEA20049700.56351100.313
TEN200451100.31351100.313
CHI200241200.2551100.313
CLE200031300.18851100.313
MIN199910600.62541020.313
SFO199941200.2551100.313
DET199851100.31351100.313
STL199841200.2551100.313
DET199651100.31351100.313
DEN19947900.43841020.313
PHI19947900.43851100.313
RAM199351100.31351100.313
IND199341200.2551100.313
NYG199261000.37551100.313
CHI199251100.31351100.313
NYG19918800.551100.313
IND199111500.06351100.313
CHI198961000.37551100.313
WAS19887900.43851100.313
STL198551100.31351100.313
MIN198431300.18851100.313
GNB19838800.541020.313
SDG198361000.37551100.313
NYG198331210.21951100.313
NYJ198041200.2551100.313
DAL197911500.68851100.313

The 2007 Ravens went 5-11 overall and 3-13 against the spread, making them the worst team in recent history when it comes to covering the point spread. That year marked the end of the Brian Billick, Steve McNair, and Kyle Boller eras in Baltimore. And while first-year head coach Ken Whisenhunt is probably safe, Titans fans can rest easy knowing that the Jake Locker era is almost certainly over. As for Zach Mettenberger and Charlie Whitehurst? The door may be about to close on them as well. After losing to the Jets and Jaguars, Tennessee looks to be in great shape once the music stops to land Marcus Mariota or Jameis Winston.

{ 6 comments }

There's been a long drought in Cleveland

There’s been a long drought in Cleveland

October 27, 1991. The 4-3 Browns were hosting the 3-4 Steelers, and Vegas oddsmakers set the Browns as 1.5-point favorites. Bernie Kosar would complete 21 of 29 passes for 179 yards and a score, while Kevin Mack would lead the team with 54 yards rushing on 19 carries. It was not a great offensive day for the Browns, but the team managed to pick off Neil O’Donnell two times, and held Merrill Hoge to just 48 yards on 12 carries (the factor back chipped in with 56 receiving yards, too). Clay Matthews — the middle one — had one sack, Louis Lipps led all players with 69 receiving yards, and the only thing that would trick you into thinking that this game didn’t take place generations ago was that Matt Stover started the scoring with a 34-yard field goal. [click to continue…]

{ 10 comments }

In the third quarter on Monday night, I texted my Patriots fan buddy Matt, “Is it possible that we suck? Maybe the run is finally over.” Bill Barnwell mused on this, and Aaron Schatz also wrote about it. It was hard not to think that, given the way the Patriots were manhandled by a mediocre team playing without several key players. It looked every bit as bad as the 41-14 score and maybe worse.

I remember the last time I wondered if the Pats were done. In a 34-14 loss to the Browns in 2010, the Patriots looked pretty impotent. In that game, as in the Chiefs one, the Pats had just under 300 yards of offense. Peyton Hillis ran over the Patriots. Of course, that wasn’t the end. Maybe this time is different, though. If anything the Chiefs game was even worse, so it’s possible this time really is the end.1

Will the Patriots offense be good later this year? To provide a little insight into this, I went back and looked at performance trends for quarterbacks who have had long careers. The first table looks at quarterbacks since 1969 who have the biggest single-season drops in adjusted net yards per attempt (ANY/A) from the previous five year trend. I look just at quarterbacks with at least 100 attempts in a season and I weight by the number of attempts when calculating the average ANY/A over the previous five years.

[click to continue…]

  1. And those Pats were 6-1 at the time of the loss to the Browns. []
{ 20 comments }

I am getting some well-deserved crap from people about just how bad my predictions have been so far. The Arizona Cardinals have already somehow outperformed the number of wins I gave them. The Jacksonville Jaguars, my pick to win the AFC South at 8-8, at one point in the game against the Colts had been outscored 112-13 over a stretch of about nine quarters. And my pick to win the NFC North at 14-2 could be 0-3 if Marty Mornhinweg let his head coach call the timeouts.1

But I did win my first Stone-Cold Mega-Lock of the Week with my very comfortable tease of the Bengals and Falcons. So things are looking up and I’m taking that as license to check out some historical betting data for anything that might seem appealing after three weeks.

Last year’s Carolina Panthers are the inspiration for the analysis here. After three weeks, they were a 1-2 team with a big positive point differential. The Panthers last year lost 12-7 to Seattle and 24-23 to Buffalo before annihilating the Giants 38-0. Despite VOA liking the Panthers even after just three games, the betting market came around later in at least one way. The Panthers were at 3/1 to make the playoffs last year after three weeks, even though Football Outsiders had their playoff odds at over 50% at that time.

Is it possible that teams like the 2013 Panthers have historically been undervalued? It seems likely that Carolina was a little undervalued last year after three weeks. By looking at point spread data, we can see if teams that have likely been better than their records have been good bets in the early part of the season. Specifically, I’m going to look at whether betting on early-season underachievers (teams with deceptively poor records) or against overachievers has been profitable now and in the past.

Data and Methods

Feel free to skip this part, but here’s the background for those interested. I have put together Pro Football Reference’s point spread data for all games from 1979 to 2012. This sample is good enough for the tests of long-term and recent betting strategies that I want to do.

I’m going to look at betting outcomes in games 4-8 for teams that are either losing teams (winning percentage below 0.5) with strong Pythagorean records or winning teams with weak Pythagorean records. I will keep things simple and define Pythagorean wins here as:

Pythagorean Wins = (Previous Points Scored ^2.53)/(Previous Points Scored^2.53 + Previous Points Allowed^2.53)

In a continuing effort to avoid unnecessary complications, I’m just going to split the data up over time, looking separately at results before and after 2000.

Betting On and Against Pythagorean Outliers

Below is how you would have done over time if you bet on or against two kinds of teams:

  • Overachievers: Teams with winning records with bad point differentials for their records
  • Underachievers: Teams with losing records with good point differentials for their records

An overachiever is more specifically a team with a winning record that has a Pythagorean winning percentage at least 25 percentage points worse than their actual winning percentage. An underachiever has a Pythagorean winning percentage at least 25 percentage points better than actual.

YearsOverachieving TeamsUnderachieving Teams
1979-1999174-142-11 (55.1%)141-146-5 (50.9%)
2000-2012109-99-4 (52.4%)108-100-8 (52.0%)

The results show that, before 2000, you would have won most of the time betting on overachieving teams, teams that were not as good as their records would suggest. I was surprised by that and it even made me wonder if I made a coding mistake. I certainly expected that any tendency away from an even split would have been in favor of betting against teams with good records and relatively poor point differentials. Note that the even split occurred in the past for the underachievers, the teams with good point differentials and poor records.

More recently, the data come pretty close to an even split for betting both on the overachievers and the underachievers. Betting on the overachievers and the underachievers has been successful about 52% of the time since 1999.

So the overall message is that there is little value now or in the past in identifying Pythagorean outliers and either riding the teams with deceptively poor records or fading the teams with misleadingly good ones. In fact, the only pattern from the past suggested it was a good idea to ride the teams with misleadingly good records. I tried to check this out a bit to just see if it was just betting on teams with good records that was profitable, but betting on all teams with winning percentages over .750 has gone almost exactly dead even over time. It would be great to hear any thoughts you might have in the comments for this pattern. I feel like I’m missing something.

Overall, the message here is the one that we get most of the time if we try to find patterns that might lead to a consistently profitable and simple betting strategy. It just ain’t there. That doesn’t make this a bad post, though: as Chase once noted, an answer of “not useful” is often just as meaningful as any other answers.

The Stone-Cold (I Think There May Be a 60% Chance This Bet Will Win) Mega-Lock of the Week

So I am now 33% on my Stone-Cold Mega Locks of the Week. If I get the next two, I will be at 60%. If I get the next two after that, I’ll be at 71%. I kind of think I should be able to claim extra points already, Chris Berman-style, for my tease last week, since the Falcons and Bengals won by a combined 89-21 score that wasn’t that close. But I will instead put my faith in the always reliable larger sample size that will bear out these predictions living up to their title.2

Two-team teaser: Pittsburgh down to -1.5 and Indianapolis down to -1.5

This week, I like another two-team teaser of two home teams, this time down to 1.5 points. I particularly like the Steelers down to 1.5 points. I do not understand how they could be the same offense for quarters 3-8 of this season as they were for the other high-efficiency ones. Still, I like the Steelers (#10 in DVOA) at home against the Buccaneers (#32).

I’m a little less sure about the other side of the tease, where I have Indianapolis (#21) over Tennessee (#25). In fact, I mainly just wanted to get the Pittsburgh end of the tease. I may be getting that queasy-knees feeling come Sunday. It’s hard to feel that way about Andrew Luck, but I didn’t imagine I’d ever be going into the water tethered to a Ryan Grigson-led team.

Season record: 1-2

  1. We are talking about the Jets here so they probably would have blown that game, anyway. []
  2. Note that no mega-lock promises were made on the season predictions. []
{ 7 comments }

Betting Bad: Thinking About Uncertainty in Prediction

Barack Obama was not the only winner in the 2012 presidential election. Nate Silver, now founder and editor in chief of Five Thirty Eight, and other stats-y election forecasters basked in the praise that came when the returns matched their predictions.

But part of the praise was overstated. At the very end, Silver’s models essentially called Florida a toss-up, with the probability of an Obama win going just a few tenths of a percentage point above 50%. But because his model gave Obama the slightest of edges in Florida, his forecast in most of the media essentially became a predicted Obama win there. In addition to accurately forecasting the national popular vote, Silver then received credit for predicting all fifty states correctly.

I am all in favor of stats winning, but the flip side of this is the problem. If Obama had not won Florida, Silver’s prediction―which, like that of other forecasters such as Sam Wang of the Princeton Election Consortium, was excellent―would have been no less good.1 And if stats folks bask too much in the glow when everything comes up on the side where the probabilities leaned, what happens the next time when people see a 25% event happening and say that it invalidates the model?2

Lots of people have made this point before — heck, Silver wrote about this in his launch post at the new 538 — but it is really useful to think carefully about the uncertainty in our predictions. Neil has done that with his graphs depicting the distribution of team win totals at 538, and Chase did so in this post last Saturday. Football Outsiders does this in its Almanac every year, with probabilities on different ranges of win totals. [click to continue…]

  1. This is a column about football, but you might want to check out some of the stuff through that link on the differences between Silver and Wang on the upcoming midterm elections. They both know way more than I do, but for the small amount that it is worth, I lean more towards Wang on this one. []
  2. Of course, maybe Football Outsiders has already run into that with the 2007 Super Bowl prediction. Perhaps sports people are ahead of politics on this stuff. []
{ 12 comments }

Prop Joe’s Favorite NFL Prop Bets

I think I’m one of eight billion people who love “The Wire” and “Breaking Bad.” Those are the two best TV shows I’ve seen and it isn’t particularly close.1 “The Wire” had an amazing volume of unforgettably vivid characters. Below is my list of memorable “Wire” characters. To be a real test of unforgettableness, it’s got to be off the top of my head, so I’m sure I’m going to forget somebody, but here goes and I’ll include the first thought that jumps to mind:

Omar (“man’s gotta have a code”), Bunk (“f***”), McNulty (“f***”), DeAngelo (library), Stringer (mastermind), Avon (winner), Brother Mouzone (bow tie), Cedric (good posture), Garcetti (that’s actually the mayor of LA, I mean the Baltimore mayor), Clay Davis (“sheeeeeet”), Bunny (“New Hamsterdam”), Keisha (car chase scene), Lester (wood carving), Bodie (corner), Prop Joe (large), …

Ah, Prop Joe. Prop Joe was a very large and very reasonable drug kingpin. His name apparently stemmed from saying “I’ve got a proposition for you,” so we could certainly see him getting into prop bets. So, in honor of Prop Joe, I’ll cover some intriguing season prop bets.2 Most of these bets are only available online, which continues to be a legal gray area. Like Prop Joe, I would never directly touch anything slightly questionable, so I will be referring to bets made by my good friend Rawls.3 We’ll start with his favorite prop bet for 2014 and go from there in descending order.

Rawls’s Prop Bet #1: $76 On Any Team To Win at Least 14 Games (Odds: 3/1)

At first glance this bet seems to have a lot of merit. Since the 1987 strike, at least one team won 14 games 15 out of 26 times (57.7%). In the last 15 years, it’s even better, hitting 10 out of 15 times (66.7%). The bet only needs to win 25% of the time to break even, so this looks fantastic.

But Chase brought up a point that Rawls missed: schedule strength. The years without a 14-game winner in the last 15 years include 2012 and 2013. Rawls dismissed that as a blip, but it comes in part from two of the best teams in football playing in the same division. Moreover, the last run of years without a 14-game winner (1993-1997) also happened during a time of NFC dominance, at least until ‘97. The Cowboys played the Packers and Niners every year during that span, for example. This season, the best teams in football may have it even tougher. The Niners and Seahawks have to play each other twice, and each has one of the four hardest schedules in football this year. The Broncos get the NFC West, the Saints have the sixth-hardest schedule, and the Packers have above-average schedule strength. Only the Patriots have an easy schedule amongst the main threats to win 14 games. [click to continue…]

  1. “Seinfeld” is all alone in third with a pretty big gap after that, too. []
  2. The actor who played Prop Joe, Robert F. Chew, sadly passed away in January 2013. []
  3. Definitely not this Rawls who is the enemy of all that is good. []
{ 10 comments }

Updated: Vegas Futures Wins Totals

Some background links:

Today I want to look at the latest odds from Vegas on NFL futures, this time courtesy of Bovada.  While we often focus on the number of wins a team is projected to have, the payouts associated with each bet are also key sources of information. Consider the Bears and the Panthers, two teams Bovada has pegged at 8.5 wins. You might think Chicago projects as a better team than Carolina this year; as it turns out, so does Bovada.

If you want to bet on Chicago winning more than 8.5 games this year, Bovada is requiring you bet $155 just to win $100 in the event the Bears win nine games. Of course, if you’re brave enough to suggest that the Bears will win eight or fewer games, Bovada would pay you $125 for your $100 bet. While Chicago is at -155(o)/+125(u), the Panthers are at +145(o), -175(u). So if you think the Panthers are overvalued at 8.5 wins, well, you need to bet $175 on the under just to win $100 if Carolina falls short of that number. On the other hand, Bovada would pay you $145 if you want to take the Panthers winning nine or more games.

Based on those numbers, we can conclude that Vegas thinks Chicago has a 58.2% chance of going over 8.5 wins1, while Carolina has just a 38.6% chance of going over 8.5 wins.2 The table below shows the number of projected wins for each team in the NFL this year, along with the lines associated with their over and under bets. The final column shows the implied likelihood (by the over/under lines) of the team going over their win total; that column was used to break ties between teams with the same number of projected wins.

[click to continue…]

  1. The -155 implies a 60.8% chance of going over 8.5 wins (155/255), while the +125 on the under implies a 55.5% chance of going over 8.5 wins (1 – [100/225]).  The average of 0.555 and 0.608 is .582. []
  2. An over line of +145 implies a 40.8% chance of going over (100/245), while an under of -175 implies just a 36.4% chance of going over (1 – [175/275] []
{ 13 comments }

2014 MVP Odds and Historical QB MVP Performance

On July 8th, Bovada released some early MVP odds, so I figured it would be fun to take a few minutes and examine which players seem like the best and worst bets. Bovada listed odds for 40 players. For example, Peyton Manning has odds of “3/1” which implies that he has a 25% chance of winning the MVP (if you bet $10 on Manning, you get your $10 back plus $30 from the casino). The odds for all 40 players sum to about 140%, which means there’s a healthy house cushion built into these odds. And it’s even worse than that, as Bovada did not include a “Field” category, so the 140% doesn’t even include all possibilities. In any event, I divided each player’s implied odds by 140% to get “adjusted” percentages (or vigorish-adjusted odds) of winning the MVP. Take a look: [click to continue…]

{ 26 comments }

Not-Entirely-Awful NFL Futures Bets

In the 1990s, there was a hedge fund called Long Term Capital Management that almost brought down the world economy. LTCM made enormous bets on very arcane things such as the spread between two kinds of bonds. Their whole reason-for-being was that they would find small inefficiencies in prices and borrow like crazy to take advantage of those brief opportunities. Other hedge funds did similar things, but these guys thought that they were smarter than everyone else. And, to some extent, they may have even been right. But they were also a little contradictory. What made this hedge fund interesting was not just that it employed two Noble Prize-winning economists, but two who made their name arguing that markets for financial assets were efficient. If their research was right, the inefficiencies on which they were betting should not have existed in the first place.

Now, these guys are way smarter than me, but you may have noticed that I recently wrote about how the NFL betting market appears to be pretty efficient. If that’s right, there shouldn’t be any chances to make profitable NFL bets. If the prices are right, all I’m doing is paying the commission every time I make a bet. Like the guys at LTCM, however, I think I’m smart enough to find bets that are mispriced and that offer some opportunity to make money. I’m probably overconfident and wrong about that, but it’s too much fun to try. And if I fail, which the LTCM guys spectacularly did when some of their billion dollar bets went wrong, at least the implications will not send shock waves to central bankers in Peru.

Bets I Sort of Liked But Decided Not to Pull the Trigger On

There were a series of bets on season win totals that I liked but decided did not quite make sense in the end. Some of the reasons were hard-headedly analytical and others were more visceral. Most notably, I couldn’t commit actual dollars to betting on Ryan Fitzpatrick, even though I came pretty close. [click to continue…]

{ 11 comments }

A few days ago, I was in Vegas with friends and without a car. So I took the chance to shop NFL futures odds to the extent that I felt it was worth it to walk to a given sportsbook. I decided the 3+ mile walk to the Superbook was not worth the opportunity cost in the 105 degree heat, so I didn’t get their numbers. But I did get numbers from three of the major oddsmakers: William Hill, Cantor Gaming, and MGM. Tomorrow, I’ll talk about some bets that seem potentially attractive. As I described recently, the numbers are pretty good now and don’t leave obvious opportunities for the most part, I think.

Yes, I still did like some bets. I only found one season win total I really wanted to bet on, and it’s not one of the ones I would have bet back in March. I made a few bets at the William Hill sportsbook, just a little hole in the wall at the Hooters’ Casino a little ways off the Strip, which could just as easily have been in Nevada towns forgotten by time like Laughlin or Mesquite as in Las Vegas. Then I made a few bets not too far from the beautiful people at the Cantor book in the Cosmopolitan. I spent way too much time thinking about all this stuff, which might not have been necessary if I only had that time machine and could have bet on the initial lines. But there’s also some cool stuff by looking at the teams’ odds that have changed the most in both directions.

Season Win Totals

Some interesting movements have happened in the numbers that Cantor Gaming released in March. Those changes reflect everything that happened in free agency and the draft, but also maybe some numbers that people would have bet on anyway even if nothing had changed personnel-wise.  Below are the opening numbers along with the numbers I gathered during the last week. The Cantor numbers are mostly from 6/18 because their books that I went to would only give me the numbers one at a time. I gathered about eight of those numbers because I was at least considering them as wagers. For those teams, the most recent line is the one that I posted. The other companies’ books gave me complete printouts of all their season win-total lines.

A note on the odds: Lines like -140 mean that you would wager $140 to win $100. Lines like +130 mean you wager $100 to win $130. The numbers are usually split by 20 on either side, which represents the vig, or Vegas’s commission. For example, Denver being at -115 for the over would usually go with -105 on the under. For bigger odds, the over and under can be split by more. Also, the MGM has a slightly larger vig, with a 30 split between the over and under. [click to continue…]

{ 5 comments }

The Smarter (Sigh) Football Betting Market

Economists (I am one) have historically been trained to believe in the efficiency of markets. The simplest way to think of this is that market prices capture all relevant information. Of course, this is sometimes not quite right, or even close to right. All the mortgage-backed securities that helped bring down our economy were horrendously mispriced, for example, despite lots of people seeing the warning signs. Even then, people betting against those securities provided information about their true value. They were just drowned out for too long by people clamoring to buy that worthless stuff.

The sports betting market, though, is a case that we might actually expect to work better. Unlike mortgage-backed securities, everyone making a wager in Las Vegas is incentivized to get the price right. There’s nobody who’s pushing a bad wager on their clients, for example.1 Therefore, we might expect efficient markets to mostly work in Vegas and that the odds would converge to the correct number.

Mostly, it seems like that’s what’s going on. Whatever information is not contained in the initial odds may be quickly corrected as people swoop in to take advantage. I’ve experienced this first-hand. Last year, I went to Vegas about a week after the first season win-totals for 2013 came out. I found the numbers online and came up with this list of wagers I was interested in. [click to continue…]

  1. These perverse incentives have been going on a long time, too. Check out Michael Lewis’s Liar’s Poker for fascinating stories of investment bankers pushing junk on their clients. []
{ 23 comments }
FSU is a heavy favorite to wind up in the national title game again

FSU is a heavy favorite to wind up in the national title game again.

The Simple Rating System is a set of computer rankings that is focused on only two variables: strength of schedule and margin of victory. I publish weekly college football SRS ratings each season, and you can read more about the SRS there. Last year, I took the Las Vegas point spreads for over 200 college football games to come up with a set of power rankings. By taking every data point, and using Excel to iterate the ratings hundreds of times, I was able to generate a set of implied team ratings.

Well on Friday, the Golden Nugget released the point spreads for 200 games (h/t to RJ Bell). You might not think we can do much with just a couple hundred games, but by using an SRS-style process, those point spreads can help us determine the implied ratings that Las Vegas has assigned to each team.

We don’t have a full slate of games, but we do have at least 1 game for 77 different teams. Theoretically, this is different than using actual game results: one game can be enough to come up with Vegas’ implied rating for the team. Purdue may only have a spread for one game, but that’s enough. Why? Because Purdue is a 21-point underdog at a neutral field (Lucas Oil) against Notre Dame, and we have point spreads for the Fighting Irish in ten other games. Since we can be reasonably confident in Notre Dame’s rating, that makes us able to be pretty confident about Purdue’s rating, too.

The system is pretty simple: I took the point spread for each game and turned it into a marvin of victory, after assigning 3 points to the road team in each game. For example, Alabama is a 6-point home favorite against Auburn. So for that game, we assume Vegas believes the Tide are three points better than the Tigers; if we do this for each of the other 199 games, and then iterate the results hundreds of times, we can come up with a set of power ratings. [click to continue…]

{ 11 comments }
Previous Posts