December 2011


The fans were all nestled, all snug in their beds, while visions of clutch quarterbacks all danced in their heads.

Tim Tebow has managed to capture the imaginations of many announcers, fans, and analysts, including the eye of one Benjamin Morris. Ben posits, among other things,  that Tebow is being held back by his own conservatism,  that an inability to take passing risks in the first three quarters of the game is tossed aside in the fourth and some more true representation of his passing skill emerges.

This isn’t the first time that Ben has speculated on the nature of young quarterbacks and interceptions (This link is the most important, but also see here and here). One contradictory notion  that has come out of his analyses is that a lot of interceptions early in the career of a quarterback tends to be a good thing. It suggests a quarterback with exceptional skills testing those skills out — the idea that a talented cook has to get burned by his own grease to learn his chops spills over into the quarterbacking world.

A related question, important to NFC East fans, is Eli Manning clutch? This question was raised this year by Eli Manning’s exceptionally high ESPN QBR ratings relative to other metrics. People really got upset, claimed that the ESPN QBR was “busted”. But perhaps the ‘clutch’ factor actually saw something in Eli.

It’s almost a theme with the Giants that they fall behind and Eli either scores a couple late to win the game, or scores late to tie the game and then (win/lose) in overtime, or he puts on this furious rally that almost wins the game. They beat teams they shouldn’t, based on their Pythagoreans, and then lose to football patzers.

What to make of it? My gut unchecked feeling is yes, Eli is clutch, but  his team is another question altogether. It’s difficult to know with fans, emotions get the best of them. Donovan McNabb becomes Donovan McFlabb, good analysts try to prove that Jon Kitna is a better quarterback than Tony Romo, etc.

Thinking without benefit of numbers a bit further, Eli just doesn’t get ruffled. His play doesn’t suffer any effects of pressure. And that means, no matter how inadequate the team around him becomes, he’s still dangerous.

~~~

Kindle notes: just bought a Kindle Fire, and like it a great deal. It’s a better email platform than many web based email services, so it is  useful to forward  mails from those services to this device. I wish I could plug my  camera into the Kindle and upload photos, but  that will probably have to wait until Android 4 becomes a common base OS for these kinds of portable devices.

~~~~

Twitter notes: For those familiar with Smart Football, he tweets well, and is a useful feed if you’re at all interested. Trent Dilfer does quite a bit of good analysis via tweets. Surprisingly good is Doug Farrar, whose player analyses I tend to respect. I haven’t read much of Doug’s blog, Shutdown Corner, but given the character of his tweets, it might be worth a gander.

Because of the holidays, I’m very likely to skip week 16 of this series. On week 17, after all the games are played, we’ll make playoff predictions based on the data we’ve calculated.

~~~

To explain the columns below, Median is a median point spread, and can be used to get a feel for how good a team is without overly weighting a blowout win or blowout loss. HS is Brian Burke’s Homemade Sagarin, as implemented in Maggie Xiong’s PDL::Stats. Pred is the predicted Pythagorean expectation. The exponent for this measure is fitted to the data set itself. SOS, SRS, and MOV are the simple ranking components. MOV is margin of victory, or point spread divided by games played. SOS is strength of schedule. SRS is the simple ranking.

Some notes: there appear to be 6, perhaps 7 elite teams, ones with real chances to go deep into the playoffs. That said, the history  of the past 10 Super Bowls is that a dark horse has made it to the finals 4 of the last 10 times (New England in 2001, Carolina in 2003, New York Giants in 2007, Arizona in 2008) and further, the dark horse has either won or made the game quite interesting. In the playoffs, while home field advantage counts, regular season records, or offensive stats of any kind, are not statistically predictive.

If one were to create a “dangerous team” stat, subtracting the current record of a team (as a percentage) from their Pythagorean, then the most dangerous team presently must be the Miami Dolphins, with the Eagles close behind. Such a measure though, applied to the Denver Broncos, doesn’t adequately capture the Broncos winning streak, nor the  fascination with this team. I’ve long wondered how well scoring analysis captures the kinds of teams that win by a little and lose by a lot. Another team  in that category would be Kansas City, capable of some impressive wins, but also embarrassing losses.

If you need a case study in a statistically anomalous team that won, an interesting one would be the 1976 Oakland Raiders.

 

To explain the columns below, Median is a median point spread, and can be used to get a feel for how good a team is without overly weighting a blowout win or blowout loss. HS is Brian Burke’s Homemade Sagarin, as implemented in Maggie Xiong’s PDL::Stats. Pred is the predicted Pythagorean expectation. The exponent for this measure is fitted to the data set itself. SOS, SRS, and MOV are the simple ranking components. MOV is margin of victory, or point spread divided by games played. SOS is strength of schedule. SRS is the simple ranking.

Green Bay, Baltimore, and New England have the largest median point spreads currently. San Francisco has the largest Pythagorean expectation. Green Bay is the overwhelming leader in margin of victory and thus SRS. The team that’s played the toughest schedule with a chance of making the playoffs now is the New York Giants.

To explain the columns below, Median is a median point spread, and can be used to get a feel for how good a team is without overly weighting a blowout win or blowout loss. HS is Brian Burke’s Homemade Sagarin, as implemented in Maggie Xiong’s PDL::Stats. Pred is the predicted Pythagorean expectation. The exponent for this measure is fitted to the data set itself. SOS, SRS, and MOV are the simple ranking components. MOV is margin of victory, or point spread divided by games played. SOS is strength of schedule. SRS is the simple ranking.

At this time of year, there are two groups of aspiring teams, the ones with 3 losses or fewer, and the ones with 5 losses or more. Houston has an effective replacement for their starting quarterback, and they  won this week. Chicago has yet to find such a replacement. Injuries, and injury replacements become critical this  time of year.

Once again, different teams  top different metrics. New England leads the medians, San Francisco leads the Pythagoreans (Green Bay is merely third in this metric), and Green Bay has a substantial lead in SRS.

Airwaves in Atlanta are full of the possibility of Atlanta playing in host Dallas (or New York) for the playoffs. I suspect  the notion of playing New Orleans (a distinct possibility at this point) doesn’t appeal to Atlanta fans. If you calculate playoff odds based on home field advantage, playoff experience, and current SOS, then Atlanta has a 46% chance of beating Dallas and a 56% chance of beating New York. In  the playoffs, using currents SOSs, Atlanta has a 47% chance of beating New Orleans.

Looking at playoff odds in this way, one potential upset would be Detroit playing a host San Francisco. Now the season hasn’t ended today and to get to a San Francisco, Detroit would have to win one game. But Detroit has played a tough schedule, and if its schedule advantages continue through the season, Detroit would be favored against San Francisco (61%) even though the 49ers would have home field advantage. Detroit would have to get there, though. Using current SOS values, I calculate Detroit’s odds against New Orleans as 40% and against Dallas, 33%.

Follow

Get every new post delivered to your Inbox.

Join 243 other followers