If there was ever a game prepared for a kicker, it was Sunday's Oakland Raiders contest against the Chicago Bears. The Raiders brought an offense with big play potential but with a low success rate against a Bears team among the league leaders in preventing successes. Many situations in football call for the coach to go for it on fourth down more often than we see in practice. This game, however, presents one scenario in which it is often smart to take the points if they're available -- the combination of a misfiring offense and a brick wall defense could tilt the field position game beyond reversal.
Of course, getting the three points on a field goal attempt isn't so easy as just calling the kicker and his support team onto the field. Field goals are risky propositions -- anywhere past the 30-yard mark and we encounter at least a 10% chance of failure. The lack of a capable field goal kicker could tilt the equation right back to supporting repeated fourth down conversion attempts.
- Home Archives for November 2011
Sebastian Janikowski Sinks The Bears From Afar
Team Rankings: Week Thirteen
Tebow
Is Tebow the reason the Broncos are winning? Every third news item or blog post I come across seems to be pulling its collective hair out, pondering the Broncos' 5-1 run. He's a winner! No, he's no good! He can't throw! But he runs! He doesn't turn the ball over! He doesn't take sacks! He puts his team in a hole! He's clutch! On one hand this, on the other hand that. If only their were a single number, tuned precisely to tell us how much a player contributes to his team's fortunes...That would settle the question.
Fortunately, that's what Win Probability Added (WPA) does. As a starter in 2011, Tebow has a -0.18 WPA. Suppose the football gods gave us a test with the question, On net, has Tebow helped or hurt the Broncos' chances of winning since he became the starter? Show your work. I'd write down: Hurt. -0.18 WPA. And I'd be marked correct.
We could end the discussion right there if we wanted, but that's no fun.
An Unlikely & Unsung Bengals Conversion
Down 7-0, just 4 minutes into the game, the Secretary of our Trent Dilfer Club, Andy Dalton, stepped onto the field for the first time. After a 1-yard run, 10-yard holding penalty, and incomplete pass, the Bengals were backed into a 3rd-and-19 on their own 14. Dalton drops back, hits Andrew Hawkins who goes for exactly 19, converting to a new set of downs. Here's a look at how the drive unfolded using our Markov model:
Advanced Team Stat Visualization Page
I've been tinkering with these over the past few days, and they're ready to be an official feature. Just like the teaser I previewed the other day, the visualizations have two areas--an x,y plot of offense and defense stats, plus game-to-game trend lines for each team. Hovering over each team's logo selects that team's game stats and highlights the other teams in its division for easy comparison. Now, in addition to EPA stats, there are tabs for separate visualizations for WPA and SR stats.
To keep things simple, up and to the right are always 'good'. Hovering over any data-point will reveal detailed information. There are lots of other small improvements too. The numbers are updated through Sunday night's game.
Here's the Advanced NFL Stats team stat visualization page.
Sunday's Numbers Have Been Crunched
Punishment
For some reason people think this site is about football statistics, but it's really about articulating the un-articulated. It's about identifying that which flies under the our radar of intuition. When we identify our pre-conceived assumptions, our thinking becomes clearer. While I don't pretend to be a professor of criminal justice, I'll share some thoughts on what punishment is about.
In my mind, punishment has at least five possible purposes: incapacitation, restitution, rehabilitation, deterrent, and prevention of retribution.
Totally VIP Fantasy Adds: Week Twelve
It goes without saying that the typical reader of Advanced NFL Stats is what's frequently referred to as "VIP material." Power meetings abound, followed by erotic liaisons on private yachts, followed by ever more powerful meetings, followed by even more erotic liaisons on even larger private yachts. It's tiring business, indeed.
Against this backdrop of fast living, the reader attempts to maintain an elite fantasy football team. Yet, it's not always possible to make the necessary mid-week waiver-wire moves. So the reader finds himself at week's end -- perhaps even on Sunday morning -- with an injured quarterback here, an ineffective wide receiver there, and neither the time nor inclination to research all the available players duly.
This is where Advanced NFL Stats can help: below are a couple of options at each of the typical fantasy-football positions that are likely to perform better than what we'd expect from a typical a freely available player. Each player named below is owned in fewer than 50% of Yahoo leagues, making it quite possible that at least one of them is available in the reader's league.
What sort of performance might the reader expect from the following players? Obviously, it's impossible to answer this question with any certainty; however, in the interest of full transparency, I've included a list of all the picks from the first three weeks of this experiment at the bottom of this post.
Here's a summary of the average points by position over that same three-week span (using Yahoo's default scoring, minus the last four categories, which are largely random):
And here are this week's picks:
Roundup 11/26/11
Natural talent matters more than we'd like to admit.
Is Tim Tebow 'clutch'?
Keep it simple--the secret to Jim Harbaugh's success.
Analytics is the future.
What could'a been--pro basketball in a lost season.
This is exactly how I feel about the game probability model.
Why our brains make us laugh.
More EPA Visualization - Team Trends
Here's a new twist on the team Expected Points Added (EPA) visualizations. This time I've added a week-by-week plot of offensive and defensive production. When you hover over a team on the main plot, you'll see their production over time. When you hover over each point on the weekly plot, you'll see the opponent.
I've also flipped the defensive EPA values so that positive numbers are good on the graphs. I think it's a more intuitive way to do it. Most people, including myself, think of "up" as good. So the better teams will be up and to the right.
Playoff Probabilities: Week 12
Welcome to this week's edition of playoff probabilities, wherein I survey the playoff field in preparation for the annual binge of turkey, football, and beer that is Thanksgiving in America. As always, these numbers come courtesy of Chris Cox at NFL-forecast.com and are generated with the help of his NFL-Forecast software app, which uses the win probabilities generated by the team efficiency model to simulate the NFL season 5,000 times. And if you don't buy the game probabilities from Advanced NFL Stats, you can tweak them as much as you like to generate your own playoff projections. I encourage everyone to download the app and test out your own scenarios.
Things have really come into focus in the NFC. The Packers, 49ers, and Saints all have comfortable leads in their respective division races, and the wild card is mostly down to the Bears, Lions, Falcons, and whichever team winds up in second place in the NFC East.
Speaking of which, this was not a very good week for the New York Giants. Two weeks ago, they had a two-game cushion atop the East, but the combination of their loss to the Eagles last week and the Cowboys' win over the Redskins dropped the Giants' overall playoff probability down to 32%. Things this week will not be getting any easier, as the Giants travel to New Orleans to face a Saints team that is undefeated at home, followed by a game the week after against the Packers. Yikes.
Team EPA Visualizations
For some reason I gravitate to graphs and visualizations. It's just the way my brain works. Here's an interesting way to look at team Expected Points Added (EPA). On the horizontal axis is offensive EPA and on the vertical axis is defensive EPA. Negative EPA is good on defense, so the best, most balanced teams would be in the lower right quadrant. The basket cases will be in the top left quadrant.
The teams are color coded by division. Hover over a team to highlight its division and see the numbers.
WP: Graham Gano's Accuracy
This week at the Post I take a look at Graham Gano's career accuracy after his two misses contributed to the heartbreaking loss against the Cowboys.
Weekly Game Probabilities
Weekly game probabilities are available now at the nytimes.com Fifth Down. This week I take a brief look at how eerily similar the Ravens and 49ers are statistically. It's almost like looking across the field at your own brother!
Dallas Downs Washington On Third
This past weekend's renewal of the Cowboys-Redskins rivalry wasn't supposed to be that interesting. Sure, the Cowboys have shown a maddening ability to lose games this year, but since starting 3-1, the Redskins haven't shown any ability to win games whatsoever. Despite the seven game losing streak and the seemingly superior opponent, Rex Grossman and the Redskins reminded us just how difficult the road game can be in the NFL. It took four full periods of back-and-forth play and overtime for the Cowboys to finally dispatch the Redskins by a score of 27-24.
Team Rankings: Week Twelve
Slate/Deadspin Roundtable: Offensive Identities
We're conditioned to think of a gain of just two or three yards as, well, a gain. But unless those couple yards are for a first down, it's anything but a gain. About 60 percent of all running plays are setbacks, meaning that the offense is less likely to score after the play than it was before. Offenses might as well be throwing downs away. Defenses would be happy to spot an offense a third-and-four every time—the conversion rate on third-and-four (55 percent) is lower than for a series that starts with a first-and-10 (67 percent).
You can read the post at either site: Slate or Deadspin.
Sunday's Numbers Have Been Crunched
Introducing The Trent Dilfer Club
- No games above +0.50 WPA
- Win-Loss record of at least .500
- WPA below 0 in at least 40% of their games
Fantasy Adds for Week Eleven
For the third week in a row, we offer this modest attempt at fantasy analysis.
This week offers two games, Kansas City at New England and Tampa Bay at Green Bay, with promising fantasy possibilities, as the favored team in both cases (the Patriots and Packers) also has a poor defense (26th and 28th, respectively, per GWP). As such, when the Chiefs and Bucs turn to the pass in low-WP situations, they're likely to be more efficient than we might otherwise expect.
I'd also like to make a note on the timing of this piece: a number of commenters have noted that Saturday might be too late in the week to take advantage of fantasy analysis. The results from a poll I ran last Sunday show that about 20% of readers need to make waiver-wire adds before Saturday. If possible, I'll attempt to run this closer to mid-week in the future; however, for the remaining 80% of readers, there's also some advantage to having all the information that a week of practice and reporting can provide.
Finally, don't hesitate to make comments below, or take time to point and laugh at last week's picks.
Quarterback
The weekly game probabilities suggest that Kansas City has only an 8% chance of beating New England -- nor does that even account for the fact that starting quarterback Matt Cassel will miss the game after undergoing hand surgery on Monday. However, with New England likely to take an early lead, Cassel's replacement Tyler Palko (5% owned) is likely to compile a pretty substantial number of pass attempts. The fact also remains that New England has the seventh-worst defensive GWP and allows the fifth-most net yards per pass attempt (7.3) in the league. Another option at QB is Philadelphia's Vince Young (6% owned), who'll be replacing the injured Michael Vick against the Giants. Young has the capacity to gain a fair number of rushing yards, which are generally worth more than the passing kind in fantasy football.
Roundup 11/18/11
Vote for the GOAT.
Crowdsourcing might be the future of science. It has already been a fixture of fantasy sports and in sabermetrics, but it's usually just a collective vote or estimate about a player's future performance. What if there were a distributed data collection and analysis network for rich, robust football analytics?
When I did my Washington Post column on 'rebuilding' through the draft, I thought about testing the correlation between the number of picks and team wins. But then I thought, I bet someone's already done this. They did. The correlation found was 0.4. Of all the things that determine wins, including random luck, that's nothing to sneeze at.
Michael Lewis and Billy Beane talk Moneyball.
"It's a lot of things." Sure is.
"I present to you a bit of a look into the black box that is DVOA. I can't show you how I figure each baseline, or what the specific equation is to figure play value for each play." Huh? Why not? Is someone going to steal the magic formula?
Identity Crisis
Are we talking about professional football teams or teenagers trying to figure out whether to hang out with the jocks, dweebs, preppies, or wastoids?
Identity has replaced rhythm as the most meaningless word in the NFL. Remember those days when every broadcaster at one point in the game had to say, “The 49ers need a couple of completions here to get into a rhythm"?
What the hell does that even mean?
Interceptions by Targeted Receiver
We're always happy to list interceptions by quarterback or by defender, but rarely do we see interception stats by receiver. So I put one together. These are not interceptions I have qualitatively determined to be the receivers fault, but simply all interceptions when the receiver was listed as the intended target.
Playoff Probabilities: Week 11
Welcome to this week's edition of playoff probabilities, now featuring new and improved probability tables. As always, these numbers come courtesy of Chris Cox at NFL-forecast.com and are generated with the help of his NFL-Forecast software app, which uses the win probabilities generated by the team efficiency model to simulate the NFL season 5,000 times. And if you don't buy the game probabilities from Advanced NFL Stats, you can tweak them as much as you like to generate your own playoff projections. I encourage everyone to download the app and test out your own scenarios.
Regular readers of Advanced NFL Stats will know that we've had the Texans as the AFC team most likely to earn a playoff berth for several weeks now. But with starting quarterback Matt Schaub's injury, it's difficult to say where things stand in the AFC. Schaub has been a big part of Houston's success this year. His EPA ranks 5th among quarterbacks and his 6.9 AYPA is second only to Aaron Rodgers. Schaub's prognosis remains uncertain, but it's almost sure he will miss the next few games and very possible that he will be out for the season.
While losing your starting quarterback generally won't help your team's playoff chances (see the Colts, Indianapolis), even if we assume that the loss of Schaub severely reduces Houston's effectiveness, there are two primary reasons that this won't necessarily signal the death knell for the Texans' playoff hopes:
Weekly Game Probabilities
Weekly game probabilities are available now at the nytimes.com Fifth Down. The write up is a condensed version of the Tebow analysis I just posted.
How Is Tebow Winning?
Tebow is decidedly negative in both of those stats for the full season (-0.35 WPA and -13.3 EPA), but fortunately for the Broncos, the vast majority of the damage was confined to a meltdown in his single loss against the Lions. Here, I'm only going to focus on how the Broncos managed their three wins with Tebow at the helm.
There's a method to the Tebow madness. If we add his production from scrambles to his production from pass plays, he's in modest positive territory in his three wins:
WP: Rebuilding with Moneyball for Football
The Redskins need to restock the cupboard with talent. Here's how a team can build and sustain a winning roster.
Playing Moneyball in the NFL is about jettisoning expensive and under-producing veterans, rejecting the big-splash free agent, and stockpiling draft picks. There are two ways of generating those picks. First, you can trade away soon-to-be free agents to other teams in return for picks or allow restricted free agents to sign elsewhere in return for compensatory picks. For too long, the Redskins have been on the wrong end of those transactions.
The second way is to trade picks for more picks. Overconfidence and urgency run rife in personnel departments around the league, and smart teams can take advantage of this. There are always teams willing to overpay for a pick that they are so certain will immediately turn their team into a Super Bowl winner. A team can sell its first-round pick for a second-round pick this year, plus a first-round pick next year. In the next draft, that team will have an additional first-round pick that could be sold for another second-rounder, plus another future first rounder. Presuming there are enough buyers, a team could generate an additional second-round pick in perpetuity by foregoing its first-round pick in only one year.
There's one team in the league that understands this, and they've been phenomenally successful doing it:
Team Rankings: Week Eleven
Who's Ready For a New Number One?
The Texans' GWP is ridiculously high, despite losing players like Mario Williams for the year and playing without Andre Johnson for a huge chunk of the season. Now, quarterback Matt Schaub is going to miss the rest of the season, and Houston is going to be stuck with either Matt Leinart or T.J. Yates at QB. They aren't going to be able to survive this loss, but with seven wins, they may still be able to sneak into the playoffs.
San Francisco Wins With Field Position
It should come as no surprise, then, that the game was close for much of the way. The score was within a touchdown from the opening kickoff until the 12:26 mark when Kendall Hunter rolled in for a score to put the 49ers up 27-13. Neither team had a win probability greater than 75% in the first half. Outside of a few moments at 76% for the 49ers in the third quarter, the same applied to the second half until an Alex Smith touchdown to Vernon Davis gave the 49ers a 20-13 lead, the last lead change of the game.
Appearance on Fox Sports Radio
I did a spot on Fox Sports Radio this morning on the Atlanta 4th down decision. I think it turned out well. Zakk & Jack asked a lot of good questions and allowed me plenty of time to answer.
If the embedded player doesn't work for you, the link is here.
How to Talk to a Skeptic about Risky 4th Downs
I'm not sure I buy your numbers. Where do they come from anyway?
The win probability (WP) numbers are from actual teams playing actual games. They’re average win rates based on recent NFL history. They are not theoretical or simulated outcomes. They’re simply how often teams win given certain situations. In many cases where there are too few games for a reliable estimate, other similar situations are used to smooth the gaps. There are some sophisticated statistical tools that help make sure we do that right, but underneath it all are actual games and actual wins and losses.
Football is a game of momentum. If a team goes for it and fails, it will lose all its momentum and its players will be deflated.
Momentum in sports, for the most part, is an illusion. People naturally expect to see events alternate more often that they really do. Streaky outcomes are a natural part of the world, and momentum is not needed to explain it. Flip a coin a few times and you’ll see that there are streaks of the same result, and no one would ever say the coin had momentum.
Here’s a quarter. Go ahead and start flipping now. I’ll wait. Half out of every set of two flips will be the same two results in a row. One in four sets of three flips will be a streak of the same result…And one out of eight times you’ll see four straight of the same result. There are at least 20 drives in an NFL game, plenty of opportunity to witness perfectly natural streaks.
Despite this illusion, emotion can be an important part of a sport like football. No one can deny that. And I agree that large successes and failures are an important part of players’ emotional states. But the momentum/emotion argument always assumes failure. It assumes momentum/emotion can only be lost. It can be gained too, perhaps from a 4th down conversion. Besides, what happens to your opponent’s emotional state when your team successfully converts a 4th down?
But that offense was only 1 for 3 in converting short yardage situations in the game. No way the chance of conversion was close to 75%. More like 33%!
QB Sneak vs. RB Dive
QB sneaks seem more successful on inches-to-go situations than RB dives. We'd like to know if the data back this up. Unfortunately, the play descriptions don't note how long the 'and 1' is, whether it's a long yard or just inches. We'd expect to see more QB sneaks on the shorter distances and more RB dives on the longer distances, which bias the numbers because longer to go distances would naturally be tougher to convert. Still, we may be able to draw some inferences.
The table below lists the success rates for 3rd and 4th down runs with 1 yard to go. It breaks out plays by QBs, RBs, and FBs. QB scrambles on pass plays have been removed. Kneel downs and spikes are also removed. Plays inside the 10 yd line are removed due to field compression effects.
Redskins in the Red Zone
The Redskins were inside the 20 only 3 times against Miami. Out of those 3 trips, they converted 2 field goals and turned the ball over on an interception. The first red zone trip came late in the 1st quarter after a Matt Moore interception set Rex Grossman up with a 1st-and-goal from the 5. Using our Markov model, we look at the quick deflation of the drive:
The Falcons' 4th Down in OT
With 10:52 left in overtime against the Saints, the Falcons faced a 4th and inches at their own 29. Head coach Mike Smith decided to for it. Was it a smart call?
Calculating Win Probability (WP) in OT is surprisingly simple compared to regulation time. Except for the final few minutes when there is real possibility of a tie, time is not a factor and the score is always tied.
A punt would be the conventional call. A typical punt from the 29 nets 38 yards, giving the Saints a first down at their own 33, worth 0.58 WP (a 58% chance of winning). This makes intuitive sense, because teams that win the coin toss are in a similar situation and win just under 60% of the time. The Falcons would therefore have a 0.42 WP following a punt.
Sunday's Numbers Have Been Crunched
New Site Feature: The Fourthdownulator
This is something that belongs in the long-overdue category. In effect, I've automated myself, which is not something one does eagerly.
The 4th Down Calculator estimates the Expected Points (EP) and Win Probability (WP) values for all three options--go for it, punt, and field goal attempt. All you need to do is enter the game state at the time of the 4th down and click 'calculate'. A table of results will appear that lists:
- the league-average success rates for conversion attempts and FGs
- the EP and WP values for success and failure
- the total EP and WP values based on an expected utility calculation
- and lastly, a break-even conversion probability
Mid-Season All-ANS Offense
I'm not cutting my hair until Advanced NFL Stats recognizes my accomplishments. |
First, up Quarterbacks.
Zach:Aaron Rodgers. Just watch the guy. He is in complete control out there.
Josh: It's hard not to pick Aaron Rodgers--#1 in EPA and AYPA and an EPA/P that is simply ridiculous.
Jack: Aaron Rodgers. He makes me think every other quarterback is garbage. A 0.14 point EPA/P lead and a 1.6 AYPA lead are just unfathomable.
Keith: Aaron Rodgers, ‘nuff said. The guy has a 72.3% completion percentage (not to mention an absurd 0.41 EPA/P) and has thrown just 3 INTs. Only the all-mighty Alex Smith has less through 8 games. Oh, and he can act too.
Carson: At this point, Aaron Rodgers is the closest thing I've seen to an ideal quarterback in my lifetime. He possesses the quarterbacking holy trinity, as it were: he reads defenses spectacularly and his arm is both strong and accurate. He also uses his athleticism excellently -- less to gain rushing yards and more to buy extra seconds for passing. His competitiveness or intangibles or whatever you want to call them are also outstanding. At this point, the one weakness is his susceptibility to concussions.
Brian: Brady actually edges Rodgers in WPA at the moment, but that's only because he's had to compensate for such a lousy defense. Rodgers kills everybody on EPA, so I'm with everyone else: Aaron Rodgers it is. Don't overlook Brees though. He's having another great year, but a lot of his EPA was generated in that blowout vs. IND.
Next, Running Backs.
Poll: What's the Last Day for Adds in Your Fantasy League?
A couple of commenters -- in response to the fantasy pieces that have appeared here each of the past two Saturdays -- have noted that their waiver-wire deadlines occur before Saturday, greatly diminishing the relevance of the analysis therein.
So, a question for our readers-cum-fantasy-owners: what's generally the last day of the week on which you're allowed to make waiver-wire pick-ups? Feel free to add any relevant notes in the comments section.
Fantasy Adds for Week Ten
Last Saturday, we took our first (entirely modest) step towards providing fantasy advice that's also in line with the defining principles of the work here at Advanced NFL Stats.
Some early comments on that Saturday piece led to research in the area of pass/run distribution by Win Probability (WP), with the discovery that a team's play-calling is pretty seriously informed by their in-game WP -- potentially valuable information for the fantasy owner.
Further work in this area last Tuesday by Chase Stuart of the NY Times' Fifth Down blog gives us a portait of every team's "platonic" run/pass mix -- that is, run/pass mix stripped of game state.
In what follows, I've attempted to use these concepts towards the end of a More Informed Fantasy Analysis. Below, I've listed (at least) one player at each of the main fantasy positions who's (a) owned in fewer than 50% of Yahoo leagues and (b) likely to have some fantasy value this weekend. As per usual, don't hesitate either to ask questions or simply harass me in the comments
Quarterback
Matt Hasselbeck is actually available in 55% of leagues (i.e. slightly above the 50% threshold), but the next-best alternative, John Beck (8%), appears to have only a tenuous grasp on the starting job in Washington. Hasselbeck has actually been pretty efficient, having posted a 104 NY/A+ this season. With the collapse of the Titan running game (Tennessee has a league-low 30.9% run success rate), the team has skewed pass heavy -- and will likely be passing late against a favored (0.62 PROB) Carolina team.
Roundup 11/12/11
The true definition of a pass-first offense. This dovetails nicely to Carson's previous article on that very topic.
I guess we weren't the first to apply Kahneman and Tversky's cognitive psychology theories to pro sports. Some guy named Kahneman did it first. (I should note a young economist named Justin Tapp first pointed me toward prospect theory as an explanation for many coaching errors. Thanks, Justin.)
The AFC North has 3 teams with 6 wins midway through the season. How uncommon is that?
How to beat NE's passing attack.
Comparing GB to other 8-0 teams.
Measuring economic benefits of sports. Remember, financial != economic.
Is there enough parity in the MLB? And comparative parity in the NFL and MLB. My take: of course there's not enough parity in baseball. It's a joke. A lot of the debate comes from a misunderstanding and/or the misuse of statistics. The "r-squared" of payroll to wins in MLB is 0.17. And since 0.17 is a small number, as the claim goes, payroll therefore has a small effect on winning. Hogwash. An r-squared of .17 means that for every standard deviation increase in payroll, teams should expect a 0.41 SD increase in wins. That's small?
New Game-by-Game Stat Page
The site now features a player-year stat page, so you can break out player performance game by game during a season.
Let's say you want to see how Tom Brady's performance has contributed to the Patriots' two-game skid this season. You can go to the Quarterbacks page and find Tom Brady right near the top of the list. If you click on his name, you'll get his career page.
Each year now has a link behind it. Clicking the 2011 year will take you to his player-year page. You'll see he advanced stats for each game this season. Plus, the link in the game result column will take you to the game boxscore and win probability graph.
Looks like Brady has really cooled off since the first two games of the season against MIA and SD.
Are the Bengals For Real?
The Case for the Bengals
First, offense. QB Andy Dalton has a respectable 4.8 AYPA, currently 22nd in the league, and his pass SR is 47%, currently 13th in the league. Dalton has a slightly better than average interception rate at 2.6%. The team as a whole protects the ball and is tied for the league’s best fumble rate at 0.3%. As a rookie, it’s typical to expect Dalton’s performance to improve as the season goes on.
Playoff Probabilities: Week 10
Welcome to this week's edition of playoff probabilities, now complete with bonus graphics! As always, these numbers come courtesy of Chris Cox at NFL-forecast.com and are generated with the help of the NFL-Forecast software app, which uses the win probabilities generated by the team efficiency model to simulate the NFL season 5,000 times. And if you don't buy the game probabilities from Advanced NFL Stats, you can tweak them as much as you like to generate your own playoff projections. I encourage everyone to download the app and test out your own scenarios.
How 'bout them Giants? In the most exciting game of the year thus far, Eli Manning drove the Giants to victory over the Patriots, giving New York a two game lead over their nearest opponent in the NFC East. At #9, the Giants sit relatively high in the rankings, yet they're projected to win the East in only 55% of simulations—less often than both the Chargers and the Ravens, neither of whom are even in sole possession of first place in their divisions.
A lot of this comes down to the degree of difficulty of the Giants' upcoming games. Over the weekend, I heard a lot of sports chatter about how tough their remaining schedule is, and the team efficiency rankings bear this out. The chart below lists the average GWP of each team's remaining opponents:
Weekly Game Probabilities
Weekly game probabilities are available now at the nytimes.com Fifth Down. I take a detailed look at the NE-NYJ match-up. The Patriots are actually a much better running team than most think, and the Jets are better at passing than they are running so far this season.
WP: Instant Offensive Improvement
I've poured over the data, and spent the last few days doing lots of advanced math, and I think I've figured out the Redskins' problem: They don't have any good players on offense.
There isn't much that can be done about their roster at this point in the season, but here are five things the 'Skins can do to improve their offense overnight.
The Giants Make Tom Brady Look Human
Tom Brady's Patriots may be a full six seasons removed from their last Super Bowl victory, but don't blame him. Since 2007, Brady has gone from great quarterback on a fantastic team to the engine which powers otherwise mediocre squads to division titles. Over the past four-and-a-half years -- 56 games due to Brady's 2008 injury, only the second year of his career in which the Patriots missed the postseason -- the former sixth-round pick has a shocking 132-to-33 touchdown-to-interception ratio. Brady has backed that up with a stellar 6.8 adjusted yards per attempt as well, a mark which would rank only behind Aaron Rodgers this season.
This is a particularly impressive performance given Tom Brady's age. In August, Brady turned 34, an age which usually signals the end of a career for many players, much less the beginning of a decline phase. Brady's age 34 season, though, has been far from a decline year. He ranks first in WPA, third in EPA per play, and fourth in adjusted yards per attempt so far through 2011.
Of course, the Giants are familiar with facing an on-point Tom Brady, and they were up to the challenge once again in Sunday's 24-20 victory at Foxboro. Despite a final-drive fueled 0.47 WPA, Brady wasn't the infallible field general we've grown accustomed to. Brady took two sacks, threw two interceptions, and finished with a mediocre 4.7 adjusted yards per attempt at the end of the day, clearly leaving points on the board and allowing Eli Manning the chance to win the game in the fourth quarter.
Team Rankings: Week Ten
We didn't see a whole lot of movement on top of the list, but that's to be expected as the season goes on. The larger the sample, the more the stagnant the rankings, thanks to increased knowledge and less noise.
The AFC
- After being the Dolphins' first victory of the season, the Kansas City Chiefs fall to second-to-last being 4-4 on the year. Yes, you heard that right, astute readership: a team tied for the AFC West lead has been the second worst team in the league, at least according to these rankings. All but one of KC's wins were very close games, and all but one of their losses came in a blowout.
SI Article on NFL Passing
I tried my best to be quiet and learn, preferring to be thought an idiot than speaking and removing all doubt. But I couldn't hide for long. Full disclosure, the final comment attributed to me was actually me quoting Coach Leach. After the others signed off the call, I was asked what I thought was most interesting, so I cited some of Leach's earlier comments.
The roundtable was the basis for an article in a recent issue of SI, and the online version is here.
Tomlin's 4th Down Call vs. Baltimore
Reader Borat asks: With 2:34 remaining anda 4-point lead, PIT faced a 4th and 5 at the BAL 29. Tomlin is taking a lot of heat from Steelers fans about his handling of the situation. Have you done an analysis?
At that point, Tomlin should have preferred a FG attempt (0.77 total WP) to a punt (0.75 total WP). The league average for that distance is 64%, and a successful kick would have made it a 7-point game, assuring at least overtime. But either decision could be defensible depending on the particulars--most importantly the expected resulting field position of the punt and likelihood of success of a FG attempt.
But Tomlin wavered, unsure of his team's new holder, a recently signed punter. He may have been correct because a drop of only a few percentage points in FG probability makes the decision a wash. The play clock ran out, and PIT took a delay of game penalty.
At this point, felt he had no choice but to punt. And the numbers reflect this. At 4th and 10 from the BAL 34, the punt is the better option, 0.75 WP to 0.74 WP.
But, at the risk of sounding like a broken record, I think he should have gone for it. Yes, gone for it even on 4th and 10. Here's why.
Fat Albert's Impact in 2011
- 0.12 +WPA (78th)
- 1.4 +EPA (106th)
- 3 Success Plays(101st)
- 2 Tackles
- 1 Assist
- 0.19 TF (111th-last)
- 2 QB Hits
- 0 sacks, passes defended, interceptions, forced fumbles, or tackles for losses
He hasn't played a majority of snaps, but almost all if not all DTs platoon to some degree. Admittedly, his job as a 3-4 DT is to eat space and not necessarily to make plays, but there are many 3-4 DTs (and some DTs classified as 3-4 DEs) who are making much bigger impacts. And from what I understand, he was often used in a role and in alignments that let him penetrate to make plays. This is really lackluster for the guy who made such an impact in TEN.
Don't Pay Fortay!
CJ2k, or CJ0.7k as he might be better known after this season, has highlighted the folly of handing over 10% of a team’s cap space to single RB. That’s just too many eggs in one basket for a position that simply does not drive wins and losses except in rare instances.
Forte is 2nd so far this year in WPA/G, 6th in EPA per play, and 18th in SR. He’s averaging 5.4 YPC, which ties for 6th in the league for qualifying RBs. We all agree that’s really good. The question is whether the Bears are better off locking him up for 6 years and $20 million guaranteed or spending that money elsewhere.
We don’t need to look at Chris Johnson or DeAngelo Williams to see the danger of giving huge guaranteed contracts to RBs. We only have to look at Forte himself. As we’ve seen with several RBs, they can go from thermonuclear hot one year to cold as a fish the next, even without injuries or switching teams, and Forte is no exception.
Run/Pass Distribution by Win Probability
One question that came about as a result of Saturday's very modest attempt at fantasy football analysis -- a question particularly in the wheelhouse of the present site -- concerned the relationship between run/pass distribution and game situation. Because we know that a team with a lead will try to use the game clock to its advantage -- and because, with few exceptions, the run is the best way to maximize time of possession -- it follows that a team with a lead will call more run plays. This idea is nothing new.
What I haven't seen before, however -- and it could very well be from a lack of looking -- is an illustration of the precise relationship between run/pass distribution and in-game win probability (WP).
Thanks to Mr. Brian Burke, that thing now exists, right here, in table form (numbers from 2006 though Week Eight of 2011):
And here, once again, in graph form:
Sunday's Numbers Have Been Crunched
Quick Reactions to Sunday's Games
It was a tale of two time slots on Sunday. The early games averaged a very boring 2.3 Excitement Index (EI), while the late games including the night game averaged a riveting 5.8 EI.
NYJ 27 BUF 11
The Jets took control of this game in the 3rd quarter. Sanchez (-0.09 WPA/6.9 EPA/59% SR/6.1 AYPA) didn’t need to have a great day, because the NYJ defense handed Fitzpatrick his worst game of the season. Burress stepped up big (0.23 WPA/6.7 EPA/100% SR/5 tgts).
SEA 13 DAL 23
SEA hung tough until just past halftime. Romo had a good day (0.27 WPA/8.6 EPA/55% SR/9.0 AYPA) with no turnovers and no sacks. DAL handily won the battle of the trenches (0.19 –WPA, 2.5 –EPA) against SEA (0.04 –WPA, -0.7 –EPA). The two plays that turned the game were the 33 yard TD pass to Witten and the Hatcher interception.
CLE 12 HOU 30
Both QBs had AYPAs of 3.2, but Shaub had the better go if it (0.13 WPA/-3.2 EPA/46%) than McCoy (-0.10 WPA, -4.3 EPA/45% SR). HOU’s running game dominated on the backs of Foster (0.10 WPA/3.9 EPa/56%SR/6.5 YPC) and Tate (0.15 WPa/6.5 EPA/59% SR/9.6 YPC). Overall HOU had a 67% run SR, compared to only 19% for CLE. This game was over early.
ATL 31 IND 7
IND hits another low point. They never had a chance in this one. Ryan picked apart the IND defense with 9.0 AYPA. Curtis Lofton stole the show for the ATL defense (0.18 +WPA/6.2 +EPA/2.14 TF/8 SC).
Running Into Trouble on 4th Down
Tied 13-13 with 1:54 left in the game, the St. Louis Rams found themselves in a 3rd-and-1 on the Cardinals 33 yardline. Looking to just run the clock down and kick the game-winning field goal, the Rams called two straight run plays for Steven Jackson, who had 130 rushing yards on the game. Jackson was stuffed twice, turning the ball over to the Cardinals and causing a 45% drop in win probability. Typically, running the ball on 3rd or 4th-and-short is the correct decision, and the Rams were certainly right to go for it in that situation. But, it got me thinking about trends in rushing and passing frequency and efficiency on 4th down.
So far this season, teams have run the ball on 4th down just 3.6% of the time. This is down from an average of 4.7% over the last 10 years and a peak in 2008 of 5.5%. Assuming 4th down situations are roughly similar to those of year's past in terms of to go distance, field position, and other factors, this is a noteworthy decline.
Roundup 11/5/11
"The lack of big plays could be a result of limited work after contact. [Chris] Johnson is averaging 1.3 yards per rush after contact this season."
What can Woody the Woodpecker teach us about football head injuries?
Most and least penalized teams through week 8.
There's in-game win probability, and then there's World Series win probability. Very cool analysis by Jack Moore. Here is WS-WPA for batters.
PFR Blog takes a knee. It's an apt metaphor because it's what you do after you've sealed a win. I'd like to thank Doug Drinen and his crew for the inspiration and education they've provided me and the rest of the stat enthusiast community. Doug was part of the re-birth of football analytics, long before ANS or Football Outsiders, largely motivated by fantasy football at the beginning of the last decade. Chase Stuart has landed at Smart Football and Fifth Down, and JKL is still at Big Lead.
KC and the old counter trey.
One of the interesting things ESPN can do with it's new TQBR stat is to analyze the inputs. For example, swapping Tom Brady's pass pro with Jay Cutler's pass pro would cost NE 19 net points. That would cost NE at least one win, probably two, and maybe three.
Five Fantasy Adds for Week Nine
This week's edition of The Weekly League isn't The Weekly League, at all, but rather an experiment in nerd-centric fantasy analysis.
Fantasy Notes for Week Nine
Introduction
There are a lot of sites with a lot of fantasy football experts on them -- and, likely, many of those experts are good at what they do. What follows merely represents an attempt to apply some of the concepts central to Advanced NFL Stats -- especially, for example, our emphasis on separating repeatable skills from those acts which are more the product of variance -- to the world of fantasy football.
Below I've listed one player at each of the main fantasy positions who's (a) owned in fewer than 50% of Yahoo leagues and (b) likely to have some fantasy value this weekend. Don't hesitate either to ask questions or simply harass me in the comments.
Quarterback
Indy's Curtis Painter (11% owned) and Kansas City's Matt Cassel (44%) both play at home against teams (Atlanta and Miami, respectively) with poor defensive ratings per GWP -- and the pair have identical net yards per pass averages, too (at 6.0). If Painter has an edge it's that there's a non-zero possibility that he could add production via the rush. Last week he ran for 79 yards on seven attempts. That's merely one data point, yes, but rushing yards are more valuable in most fantasy formats, which makes it a data point of interest.