GridIronMine appears in <em>Patriots Football Weekly</em>

» Posted by GridIronMine on September 5, 2007«

Original article appears in Patriots Football Weekly 13(13):34-5, Sept 5,2007 and online (requires subscription). This article is re-posted with permission of PFW, who reserve all rights.

Get inside the numbers for PFW's latest feature

Ever scream at the TV when the coach of your team decides to punt on fourth down rather than go for the first down? If so, now you'll be able to find out if your reasoning was correct.

Each week during the upcoming season, PFW will feature a new wrinkle in our game coverage.

GridIronMine.com will provide a unique analysis of each Patriots game by determining how the team's chances of winning vary based on the decisions and results that take place on a play-by-play basis. The following is a look at how a pair of fanatical football fans decided to stop yelling at their TVs and start applying some facts to their logic. We hope you enjoy ...

GridIronMine.com is an innovative website that provides unique analysis of football by digging deep into the game and the statistics. It will be rolling out several novel online tools this season. These interactive tools provide a fresh, objective look at the game of football. Starting with next week's opener, PFW will offer unique insights from GridIronMine.Com every week of the 2007 season.

GridIronMine.com was founded by two computer science professors who are also avid football fans. The founders, Vince Freeh and Dave Lowenthal, met 16 years ago in graduate school at the University of Arizona, where they obtained their doctorates in computer science. (They were classmates of Tedy Bruschi, though they are quite sure Tedy is unaware of that.) Vince is now on the faculty at North Carolina State, and Dave at the University of Georgia.

They felt that too much football analysis is based on faulty reasoning and emotion. In order to confirm truths, dispel myths and dig deeper into the game, they wanted an objective, unemotional, understanding of football. They found plenty of statistics and statisticians that could give answers to amazingly detailed questions. But no one was answering, to their satisfaction, the most important question: What is a team's chance for victory? In the end, victory is the only statistic that matters.

GridIronMine.com has examined more than 180,000 plays from the last five NFL seasons. Using computer modeling and modern statistical methods, GridIronMine.com has developed software that estimates the likelihood of victory for any game situation. It determines the odds of victory, called the Victory Forecast. For example, an average NFL team with a 7-point lead, the ball on their own 20, with five minutes to go in the fourth quarter, has a 92 percent probability of winning - a VF (Victory Forecast) of 92 percent. GridIronMine.com also provides the Whether Station, an interactive online tool, where you can get a Victory Forecast for any situation.

This NFL average VF is useful when answering general questions. However, because every NFL team is unique, NFL average VF will not capture the true nature of a particular game. For example, if the Patriots have the ball in the scenario described above, they have a better than 92 percent chance of winning if they are playing the Texans. Therefore, the Victory Forecast software developed by GridIronMine.com has been built to objectively evaluate the strengths and weaknesses of each team for a given opponent. As a result, GridIronMine.com provides game-specific Victory Forecasts.

As an example of the kind of analysis that GridIronMine.com will bring this season, consider the Patriots divisional playoff win over the Chargers last season.

For example, one interesting play occurred with 5:21 remaining in the first quarter, when the Chargers went for the first down on fourth-and-11 at the 30-yard line, passing up a 47-yard Nate Kaeding field goal attempt. The result of the play was a sack, fumble and recovery by Tully Banta-Cain. The Patriots Victory Forecast increased by 9 percent on this play - which is not a surprise, as a fourth-down defensive stop is typically a big play.

Using their Victory Forecast system, GridIronMine.com analyzed the decision by Chargers coach Marty Schottenheimer to go for the first down. Their VF system shows that the average kicker makes a 47-yard field goal about 60 percent of the time, and if good, the field goal would have increased San Diego's VF by about 10 percent. On the other hand, converting a fourth-and-11 is about 20 percent likely for the average team and would not have increased San Diego's VF much more than would making the field goal. Therefore, GridIronMine.com suggests that the correct decision would have been to attempt the field goal. Certainly, Schottenheimer also had to take into account whether Kaeding is a better than average kicker, as well as the quality of the Chargers offense and the Patriots defense.

Another interesting play was the one where the Chargers stopped the Patriots on third down by sacking Tom Brady (causing a fumble recovered by Matt Light). That set up a fourth-and-17 at the 33-yard line. However, Drayton Florence provided the Pats with a free first down with his now infamous head-butt. The Chargers had forced the Patriots into fourth-and-17 at the 33, but the penalty gave the Patriots a first down on the 18. This play is not talked about as much because of the more famous Marlon McCree interception and subsequent fumble, but it resulted in a whopping 13 percent increase in VF for the Patriots, making it an extremely critical penalty. The magnitude of the increase is due to the automatic first down and the fact that a 50-yard field goal (which would have been the likely choice with the strong-legged Stephen Gostkowski) is a double-edged sword: if it's missed, the Chargers would have a first down in great field position.

GridIronMine.com has used its Victory Forecast software to investigate several questions that fans want answers to, such as: Are the Patriots employing the right strategy by going for it on fourth down as much as they do (yes), is a fullback dive for 1 yard on second-and-one a good idea (usually not), and is kicking a field goal down by 11 points late in the fourth quarter a good strategy (often not). GridIronMine.com will address such general questions each week in PFW.