It’s just past the halfway point in the season, and it’s a good opportunity to take a look at which players are making the biggest impacts. I’ll look at offensive players in terms of their total contribution to their team’s fortunes using Win Probability Added (WPA). Although not perfect, there is no other metric better suited for judging MVP performance .
Starting with QBs, none other than Drew Brees leads the list with +2.67 WPA . Although the party line is that Brees is having a down year so far, primarily due to his 12 interceptions, WPA disagrees. Brees has led multiple comeback drives, even in games that were ultimately lost. Some of his biggest plays came in the overtime loss to the Falcons. Brees led the Saints into potentially game-winning FG position, but lost the game thanks to a missed kick. Brees isn’t simply lucky either. He’s 3rd in Success Rate (SR) and 3rd in EPA. Brees’ interception rate will almost certainly regress to a more normal level, and I'd bet we'll see the Saints offense to return to form in the second half of the season.
Joe Flacco is the runner up with +2.43 WPA, but he actually ties Brees with +0.30 WPA per game. The Ravens passing game has been the best component of Baltimore’s team so far. WPA is capturing things that other stats can’t, like deep passes that draw pass interference calls or good decisions to throw balls away or even take a sack. Flacco’s other numbers are rather average. His WPA is so high because of his comeback wins against the Jets, Browns, Bills, and Steelers.
Mid-Season WPA All Stars: Offense
Washington Post: Who Is Making an Impact?
Today's post at the Washington Post's Redskins Insider takes a look at which Redskins players on offense have made the biggest impact through the first half of the 2010 season.
Efficiency Rankings - Week 10
The team rankings below are in terms of generic win probability. The GWP is the probability a team would beat the league average team at a neutral site. Each team's opponent's average GWP is also listed, which can be considered to-date strength of schedule, and all ratings include adjustments for opponent strength.
Offensive rank (ORANK) is offensive generic win probability, which is based on each team's offensive efficiency stats only. In other words, it's the team's GWP assuming it had a league-average defense. DRANK is is a team's generic win probability rank assuming it had a league-average offense.
GWP is based on a logistic regression model applied to current team stats. The model includes offensive and defensive passing and running efficiency, offensive turnover rates, defensive interception rates, and team penalty rates. If you're scratching your head wondering why a team is ranked where it is, just scroll down to the second table to see the stats of all 32 teams.
The Randomness of Win-Loss Records
Bill Parcells famously said "You are what your record says you are." Although that's undeniably true with regard to how the NFL selects playoff teams, and I wholeheartedly believe a leader needs to think that way, Parcells is only 58% correct. That's not a joke. It's 58%, and here's how something like that can be measured.
One staple of statistical analysis in sports is estimating how much of a given process is the result of skill and how much is the result of randomness or 'luck'. By luck, I’m not referring to leprechauns, fate, or anything superstitious. Real randomness is far more boring. Imagine flipping a perfectly fair coin 10 times. It would actually be uncommon for the coin to come out 5 heads and 5 tails. (In fact, it would only happen 24% of the time). But if you flipped the coin an infinite number of times, the rate of heads would be certain to approach 50%. The difference between what we actually observe over the short-run and what we would observe over an infinite number of trials is known as sample error. No matter how many times you actually flip the coin, it’s only a sample of the infinitely possible times the coin could be flipped.
As a prime example, the NFL's short 16-game regular season schedule produces a great deal of sample error. To figure out how much randomness is involved in any one season, we can calculate the variance in team winning percentage that we would expect from a random binomial process, like coin flips. Then we can calculate the variance from the team records we actually observe. The difference is the variance due to true team ability.
Ndamukong Suh's XP
David writes in to ask, "It seems to me that Detroit would have been better off going for a two-point conversion after their kicker got injured, rather than having Ndamukong Suh attempt the PAT. What kind of confidence should Detroit have had in Suh's leg to make kicking make more sense than going for two?"
With Lions' place kicker Jason Hanson out due to injury, rookie defensive tackle Ndamukong Suh was called upon to attempt an extra point. Suh was chosen because he had won a place kicking competition in training camp to be the back-up kicker. Suh hit the right upright and missed the extra point, which turned out make a crucial difference in the game.
The Lions had scored 6 points to take a 3-point lead, 13-10, with 9:04 left in the 3rd quarter. A successful extra point would make the lead 4 points, and there can be a world of difference between a 3- and 4-point lead.
Roundup 11/6/10
Can wheat production estimates, candy bar weights, and attendance at Wimbledon improve your fantasy football team? Stein's Paradox says you can, but is it truly useful?
Wrong. It's actually 96.8%.
Neil Paine at PFR estimates how many wins Philip Rivers should have based on his insanely high Adjusted Yards Per Attempt. Neil also looks at how the Patriots are winning this year despite pedestrian yardage stats.
I have to say, I'm really impressed by the NFL Network's top 100 players. I thought I'd disagree far more than I do. I learned a lot from the series. Together with their series America's Game, NFL Network is doing some great work. But that should be no surprise, as both series are products of the indispensable NFL Films. One interesting thing is the comparison the expert panel's ranking with the fans' ranking.
The only issue I'll raise is that the #1 and #4 players had each other. Half of Jerry Rice's 10 all-pro seasons were partially thanks to Joe Montana, and all of Joe Montana's all-pro seasons were with Jerry Rice as his top receiver. And of course, they both benefited from Bill Walsh's visionary passing offense. Not that they don't belong at the top of the list, but no two other players at the very top of the list are as directly connected as Rice and Montana.
The Weekly League: Notes for Week Nine
This week's edition of The Weekly League features:
1. Game previews for Indianapolis-Philadelphia, Dallas-Green Bay, and Pittsburgh-Cincinnati.
2. An untinentionally glowing review of Ben Roethlisberger.
and
3. Equal parts vim and vigor.
The Four Factors you see for each game represent each team's raw performance thus far in four important categories (pass and rush efficiency, pass and rush efficiency against) relative to league average (where 100 is league average and anything above is good).
Along with the Four Factors, you'll see two other numbers: Generic Win Probability (GWP) and Game Probability (PROB). The GWP is the probability a team would beat the league average team at a neutral site. It can be found for all teams here. The PROB is each respective team's chance of winning this particular contest. Your host, Brian Burke, provides PROBs to the New York Times each week, and those numbers (along with methodology) can be found here.
The following games have been chosen as they'll be available to the greatest portion of the network-watching audience, per the NFL maps at the506.com.
Finally, a glossary of all unfamiliar terms can be found here.
Indianapolis at Philadelphia | Sunday, November 07 | 4:15pm ET
Four Factors
Notes
• Per the Interweb, DeSean Jackson returns this week after nearly getting killt by Dunta Robinson in Week Six.
• Per the Interweb, Michael Vick returns this week after getting sandwiched by some Washingtonians in Week Four.
• In Weeks One through Four -- i.e. the ones where Vick played -- DeSean Jackson was targeted 33 times, an average of just over eight per game. In Week Five, started by Kevin Kolb, Jackson was targeted only three times.
• Eight targets is kinda a lot to average. Consider: only 35 WRs were targeted 100+ times last season, and that only requires 6.25 targets per game.
• What this suggests -- but, of course, does not prove -- is that Vick looks for Jackson more than Kolb does.
Stories vs. Statistics
That's the title of an interesting essay by author of Innumeracy John Allen Paulos. Some highlights:
"...there is a tension between stories and statistics, and one under-appreciated contrast between them is simply the mindset with which we approach them. In listening to stories we tend to suspend disbelief in order to be entertained, whereas in evaluating statistics we generally have an opposite inclination to suspend belief in order not to be beguiled."
-and-
"Of course, the contrasts between stories and statistics don’t end here. Another example is the role of coincidences, which loom large in narratives, where they too frequently are invested with a significance that they don’t warrant probabilistically. The birthday paradox, small world links between people, psychics’ vaguely correct pronouncements, the sports pundit Paul the Octopus, and the various bible codes are all examples. In fact, if one considers any sufficiently large data set, such meaningless coincidences will naturally arise..."