When do you give Brock Purdy credit? 

In my analysis to members last week, I didn’t give him credit when writing up Seattle at San Francisco. Instead, I leaned toward a small sample size of five games for Purdy. 

However, Purdy had another great game during Wild Card weekend. To look at how he did, let’s look at success rate on passing plays.

A play is a success if the offense gets the following fraction of yards towards the next first down:

• 50% on 1st down
• 70% on 2nd down
• 100% on 3rd, 4th down

As I wrote last week in discussing Philadelphia, this is the most predictive team metric in the NFL.

Against Seattle, Purdy and the San Francisco pass offense had a 54.8% success rate. Let’s look at a few more passing success rates:

• 51.5% – San Francisco over the past 6 games with Purdy
• 42.8% – NFL average during the 2022 season
• 52.4% – Kansas City’s league best rate

Sure, Kyle Shanahan is probably making his life easy with play calling. It also helps to have Christian McCaffrey, George Kittle and Deebo Samuel.

However, I’m now leaning towards giving Purdy credit.

In addition, Purdy had excellent pocket presence against Seattle. He was able to anticipate pressure and scramble to an open part of the field.

Let’s also take a look at San Francisco’s defense. They led the NFL in points allowed per game (16.3) by a large margin during the regular season.

San Francisco is weaker in the underlying metrics though. They are 7th in pass success rate allowed, but they drop to 15th after my schedule adjustments.

The Niners defense has had the honor of facing Carolina, Denver and Arizona twice, all bottom five teams by my metrics.

Dallas is 7th in my pass offense numbers by adjusted success rate. With the exception of Kansas City, they will be the best pass offense San Francisco has faced this season. 

It should also be noted that San Francisco faced Miami with Tua Tagovailoa and Atlanta with Marcus Mariota (they actually had some decent numbers when he was the QB). The pass offense numbers of these two teams have declined since the back ups took over. The numbers most likely underestimate the San Francisco pass defense.

My member numbers have San Francisco by 2.7 points. However, I don’t see value in Dallas.

As of Thursday morning, some books have moved to San Francisco -4, and I don’t think this number comes back towards 3.

Football betting with a PhD edge

This article was sent to The Power Rank’s email newsletter on Thursday, January 19, 2023. This newsletter is a free service that strives to be:

• Valuable
• Concise
• Entertaining

The predictions are based on my PhD research from Stanford, and the content covers college football and the NFL.

To get this free service delivered straight into your inbox, enter your best email address and click on “Sign up now!”

 

 

 

Philadelphia has had an excellent season, as their 14-3 record earned them the top seed in the NFC. Circa has them at +475 to win the Super Bowl heading into the playoffs, the fourth best odds.

QB Jalen Hurts was the frontrunner for NFL Most Valuable Player until a late season injury derailed his campaign. While Hurts came back for the finale against the New York Giants, he will welcome a week of rest before his playoffs start.

However, there is one concerning factor about Philadelphia: passing efficiency. 

In my work, I’ve found success rate as the most predictive metric for an offense. A play is a success if the offense gets the following fraction of yards towards the next first down:

• 50% on 1st down
• 70% on 2nd down
• 100% on 3rd, 4th down

With the overwhelming importance passing in the NFL, let’s look at success rate on passing plays. I split the season up into the first 9 weeks and the remainder of the season (including playoffs) and looked at the correlation between these two periods.

The r-squared value for passing success rate between these two sets is 0.331. That means that passing success rate in the early part of the season explains almost a third of the variance in the later part of the season. 

In the chaotic world of football, this is as predictive as any NFL team metric gets. If this notion of predictability doesn’t make sense, I have a visual primer on linear regression and r-squared.

For the season, Philadelphia is 14th in the NFL in passing success rate. They have had success on 44.2% of pass plays compared to an NFL average of 42.7%. 

It should be noted that Philadelphia’s pass rate has decreased the past two games:

• Gardner Minshew against a strong New Orleans defense.
• Jalen Hurts coming back from a shoulder injury last week against the New York Giants.

Philadelphia is most likely better than 14th in the NFL by passing success rate.

But what about explosive plays? A team might not need success on a per play basis if they can regularly break off big plays.

To look at this, consider yards per play on successful pass plays. Philadelphia averaged 15.8 yards per play, 4th best in the NFL and significantly better than the 13.9 NFL average this season.

However, explosiveness is not predictive. If we do the same study as with success rate, we get an r-squared of 0.053. Randomness plays a large role in explosive passing plays.

This research suggests that the Eagles will not continue to break big plays in the future. Even with the talent of A.J. Brown and DeVonta Smith at WR. Even with the improvement of Hurts as a passer this season.

For additional evidence of the randomness in explosive plays, consider this: Carolina led the NFL in yards per play on successful pass plays. The Panthers. It didn’t matter that they got Baker Mayfield, P.J. Walker and Sam Darnold from the NFL morgue to play QB.

Predictions based on regression are not perfect, as the Minnesota Vikings in one score games proved this season. However, you’ll be right more often than not.

If the big plays evaporate for Philadelphia, they might find themselves in a tight game late in the 4th quarter. Then they fail to convert a 3rd and 5 late. My metrics suggest that Philadelphia is less likely than Kansas City or Buffalo to convert this play. 

A promising playoff run ends early.

Football betting with a PhD edge

This article was sent to The Power Rank’s email newsletter on Friday, January 13, 2023. This newsletter is a free service that strives to be:

• Valuable
• Concise
• Entertaining

The predictions are based on my PhD research from Stanford, and the content covers college football and the NFL.

To get this free service delivered straight into your inbox, enter your best email address and click on “Sign up now!”