FanPost

Colin Kaepernick & the spatial-based analysis

The San Francisco 49ers finished the past NFL season with an 8-8 record, missed the playoffs for the first time since 2010, parted ways with coach Jim Harbaugh and had to deal with offseason surprises as linebackers Patrick Willis and Chris Borland decided to retire from pro football. Hopefully, there will be no more surprises and we can assume that the 49ers are set at quarterback position, with Colin Kaepernick poised to lead the 49ers in 2015.

Based on what we've seen over the past two seasons, odds are there won't be any surprises or dramatic changes in the way quarterback wearing number seven plays his position. We've witnessed what he can or can't do and despite encouraging reports out of Arizona, where he spent time training with quarterback coaches, I don't expect any sudden changes in his style of play. In this post I'll try to illustrate Kaepernick's tendencies when quarterbacking for the 49ers with visualizations based on spatial data. In order to do so, I broke down each of his dropback from the past two seasons, with the spotlight on the locations from where he threw a pass or took a sack. Kaepernick did finish the 2014 NFL season with career-high 3369 passing yards, 172 yards more than in his first full season in 2013, but he did it with 62 more pass attempts. In total, Kaepernick dropped back to pass 578 times if we add 48 scrambles and 52 sacks to the 478 passes he attempted in 2014. Stats displayed in the table give us a chance to compare last year's numbers with the 2013 regular season.

table.0.jpg

There's no doubt Kaepernick had a down year in 2014, with lower yards per dropback and higher sack rate. He was better in 2013 especially if we consider the fact that the 49ers faced tougher defenses in 2013. I looked at Football Outsiders' defensive DVOA numbers for the past two seasons and simply averaged the numbers for the 49ers' opponents. It turns out 2013 opponents had an average of -4.6% defensive DVOA (NFC West teams -16.0%) compared to -0.8% (NFC West teams -8.5%) in 2014. That means 49ers' average opponent in 2013 had a defense that was 4.6 percent above NFL average.

In general Kaepernick hasn't changed a lot from 2013 to 2014, at least not in the way he played quarterback position, although at times it felt like the coaching staff wanted him to be more of a pocket passer in 2014.

Last June I wrote a guest column for Football Outsiders in which I've tried to graphically present the difference between classic pocket passers and mobile quarterbacks. I've used 2013 playoff data for Peyton Manning, Russell Wilson and Colin Kaepernick and designed heat charts for the trio of quarterbacks based on collected spatial data with the focus on the locations from which they've thrown passes or took a sack during 2013 playoffs. Similar, but with a much bigger sample size I've designed charts for Kaepernick's 2013 and 2014 regular season. Before we go to the charts, a few words about how I came up with the visualizations. First, I've reviewed the games in order to gather spatial based data, then I've divided the backfield into a grid of so-called "throwing cells" defined by regular hexagons with a side length of one yard. With that in place I was then able to analyze the data and determine how many times Kaepernick threw a pass in certain area of the field. Together with coordinate data I've also collected some attribute information regarding pocket, outside pocket passes, designed rollouts and pressure plays which helped me further analyze Kaepernick's tendencies.

First, a look at heat chart for 2013 regular season in which Kaepernick attempted 416 passes, scrambled 48 times and took a sack 39 times. The heat chart ignores scrambles and sacks, but reveals from where and how often Kap threw a pass in certain area of the backfield.

2013-regular-season.0.jpg

Remember the heat chart of the classic pocket passer from the linked guest column? This one looks much different and displays a chart of the quarterback who tends to move out of the pocket, scrambling to avoid the pressure. The result of the previous play determines where the center snaps the ball on the next play so we can't quite read from the chart how many of passes were thrown outside the pocket, but throws scattered across the backfield from sideline to sideline suggest that Kap often times found himself throwing the ball outside the pocket.
A closer look reveals he most frequently attempted passes 7-to-8 yards deep in the backfield just inside of the left hashmarks, but the majority of his throws were made more from the right side of the field. Kaepernick attempted seven of his passes when located between the numbers and left sideline and 19 of them or 4.6 percent of the total attempts, between the numbers and right sideline. There were plays in which the 49ers called for a designed rollout, mainly to the quarterback's right side, 47 plays to be exact or 9.3 percent of the total number of dropbacks. But the vast majority of throws distributed around the backfield were simply Kap trying to escape the pressure.

Similar picture paints the heat chart for 2014 regular season. The biggest difference presents the number of passes thrown. During 2014 regular season, the Colin Kaepernick-led offense had to pass more because they weren't in position to just ran out the clock late in games. San Francisco entered the fourth quarter tied or trailing eight times during 2014, quite a contrast to the 2013 in which they were in that same position only three times. Kaepernick attempted 478 passes, scrambled 48 times and took 52 sacks in 2014.

2014 Regular Season Heat Chart

Just like in 2013, Kaepernick most frequently (55 times) attempted passes 7-to-8 yards deep in the backfield just inside of the left hashmarks. The vast majority of passes were thrown on the right side of the field, with 21 or 4.3 percent of the total attempts between the numbers and right sideline. The 49ers dialed in 55 designed rollouts which makes for 9.5 percent of total number of Kaepernick's dropbacks, almost exact same percentage when compared to 2013.

Both charts reveal Kaepernick's well-known tendencies. When in trouble and under duress the right-handed quarterback tends to escape the pocket and tries to extend the play by scrambling to his right side. So, in terms of throwing locations, there's not much of a difference between two charts. Although there were fair amount of plays drawn and called for Kap to roll out of the pocket, there were still plenty of standard dropbacks designed for quarterback to make the play from the pocket.
The opposing defenses, Kaepernick's indecisiveness, holding the ball to long and inability to read the defenses at times were major reasons that he found himself in the situations in which he was forced to take a sack or scramble out of the pocket and throw a pass on the move. To analyze the effect of the opposing defenses, I split the 2014 data into two three-game sets. The first set includes three games against NFC West teams, Week 9 game vs Rams, Week 15 at Seahawks and Week 17 vs Cardinals. That way we'll be able to look at Kaepernick's pass and sack locations when playing against some of the top defenses. According to Football Outsiders' advanced metric all three teams had top ten defenses in 2014. In those three games against division rivals Kaepernick attempted 78 passes, scrambled ten times and took incredibly high number of sacks (15) which resulted in whopping 14.6 percent sack rate and relatively low 5.5 yards per dropback.

2014-nfc-west-split.0.jpg

Kaepernick attempted 19 passes outside the tackle box, almost one quarter of all pass attempts. It's obvious that he felt the pressure and we can see that also on the chart above on which I've added sack locations. Three of the 15 sacks were sacks Kaepernick took outside the pocket.

With the second set of data we'll look at how Kaepernick fared against three NFC East opponents on the 2014 schedule. Dallas Cowboys, New York Giants and Washington Redskins had some of the worst defenses in the league, ranked 22nd or worst based on defensive DVOA. In Week 1 game at Cowboys, Week 11 at Giants and Week 12 vs Redskins, Kaepernick threw 81 passes, scrambled six times and took only four sacks. He gained 7.2 yards per dropback and had a much lower, 4.4 percent sack rate.

2014-nfc-east-split.0.jpg

The chart based on second set of data looks a lot more like a chart of a pocket passer with a sizeable number of passes condensed and close to the left and right hashmarks. Three NFC East teams just didn't deliver as much pressure as NFC West rivals did. I've charted only eight outside-pocket throws, which is 9.9 percent of the total pass attempts considered in this second set of data and all four sacks were so-called pocket sacks.

It's a pure speculation, but Kaepernick's regular season chart would probably look different, if he didn't have to play Seattle, Arizona and St. Louis twice a year. It's about the pressure and how quarterback deals with the pressure. We are all familiar with the recent state of the NFC West division. Teams from the West have been playing dominant defense and have been able to apply lots of pressure on the quarterback who took second most sacks in 2014 and hasn't been able to deal with the pressure within the boundaries of the pocket. It was nice to read reports about Kap making big strides, improving on his mechanics, throwing motion and touch passes in offseason workouts in Arizona. That's probably music to ears of every 49er fan out there. One thing is doing it in training session, but the question is, can he perform as well on gamedays? If ten-week workout didn't help him improve in his major area of weakness - pocket presence or if he won't be able to translate the improvements into real game situations then we'll possibly be looking at the 2015 regular season chart similar to the ones of the previous two seasons.

This is a FanPost and does not necessarily reflect the views of Niners Nation's writers or editors. It does reflect the views of this particular fan though, which is as important as the views of Niners Nation's writers or editors.