Fooch's Note: Fantastic work here.
Summary of Findings
(note: the original post follows this summary)
After sorting people’s thoughts in the comments, there were several general ideas that stood out. I took a moment to organize them so I'd know what kind of conclusions we came to.
Just as a reminder, this whole process has been book-ended by two great articles. The first, by MysteryMeat, was crucial to getting the ball rolling.
Without his initial article, the rest of this would not have happened.
The second article was posted recently by liberty_JAC, in which he took a lot of these thoughts, and some good info from Fred’s play-by-play’s, and synthesized them into a compelling framework for what’s been going on the last couple of years. If you haven’t already, I’d urge you to go check it out after reading through this summary. Link posted at the bottom of this summary.
How much of slow starts is due to execution? (McTee)
Slow starts due to dropped passes (versus poor throws or poorly run routes)? (mrg80)
How much is poor ball handling skills versus great defensive plays?
Are there specific receivers with this problem?
Penalties. What causes them and how can we address them. (mgr80)
Recommended mandatory workouts. Good outline. Improves pass blocking, pass rushing, and dropped passes. All three are major points of improvements that will be needed for 2013. What should we do for decreasing penalties? (49ersFanSince1950)
Are slow starts due to shooting yourself in the foot (penalties, sacks, mis-cues) and those are mostly concentrated in the first quarter? (rick.mcbrick)
Kaep under center problems. (Karl Cuba)
"The Game is Won in the Trenches" highlights the importance of line play and pass blocking. (moiamania)
Slow starts due to game planning? Burning the clock to keep games close. (whistlingmountain)
Slow starts due to Roman using the first quarter to probe the defense? Bonus points for thinking outside the box. (djww)
Play-calling more aggressive under Kaep leading to more 3 and outs in the first quarter. (CK)
Implementing too much each week? Too much game planning leading to poor execution. (Brother Girth, rick.mcbrick)
Need less ‘gimmick plays’. Simplify. More "meat-n-potatoes" play-calling that allows players to execute what they've mastered. Execution suffers when play-calling uses complex deception so need to find balance. (Brother Girth, rick.mcbrick, Daniel McGarry) Harbaugh and Roman’s play-calling tendancies at Stanford? What were they and is there any relevance to the Niner’s play-calling? (Larushka)
How much does the defense play into slow starts? Did the defensive collapse at the end of the year distort some of these numbers? (Ray Mysterio, AKinferno)
How much is just that defenses are ‘fresh’ in the first quarter, and wear out over time? (29spYder, mrg80)
How much did the lack of rotation contribute to injuries on the defense? (Larushka)
Regarding coaching, are we looking at an issue with game plan, coaching up the players, or making in-game adjustments? (Brother Girth)
Was the coaching staff already aware of the slow start issue, and is that part of the reason Mangini was brought in? (socalisteph, rick.mcbrick)
Cautions and concerns
Reaffirming that there’s likely more than one cause. (Brother Girth)
Much of this will be fixed by Harbaugh and co. ‘self-scouting’. Course correcting. (rick.mcbrick)
Beware of mean statistics. Poorly processed data can hide information. Need a tutorial post with examples. Also, sometimes patterns are just coincidence. (oldfoggy, CK)
Regarding the game plan. We need to remember that the Niners call two plays and the quarterback adjusts at the line of scrimmage. Just looking at the film doesn’t always tell us Roman’s intent. (Karl Cuba)
DVOA stat discrepancy analyzed. (Karl Cuba)
Much of the number film review and number crunching has already been done by Football Outsiders, Pro Football Focus, and Advanced NFL Stats. Any others? (liberty_JAC, CK, Poldarn)
Video database of examples. How to implement (youtube, google?). (mgr80)
You could have categories with examples for each kind of play. Dropped balls, penalties, poor throws, etc. Could help with analyzing Kaeps weaknesses and strengths (plus other players of course).
"Mistake by Quarter" chart needed.
Need some tutorial posts with examples on DVOA and other advanced stats. (CK, Karl Cuba)
Someone needs to look into the Win, Win, Loss pattern from last year. Should make for a good post. (Kaepernick7)
Expand analysis of "failed conversions on 3rd down" to all downs (all quarters?). General analysis on catagories of failed plays to identify Niner’s weaknesses last year. This would probably tie into analyzing Kaep’s weaknesses as well. (Brother Girth)
League slow start data. Teams similar to the Niners? (ak4niner, McTee, BruinMW)
Kaep’s 3rd quarter play was divergently high (he kicks butt). (Karl Cuba)
The "Simplest Explanation" award goes to CK. Are improved 2nd and 4th quarter points due to drives extending over from the 1st and 3rd quarters? In other words, if you put all (or even half?) the points for drives in the quarters they started, instead of what quarter they ended, would some of that slow start difference disappear? (CK)
Beer is good. (Everyone)
With the recent interest in the 49ers' slow starts, I thought I’d take a look at the record to see what’s up. I compiled several charts showing various views of San Francisco’s performance last year in the first quarter. The first chart attempts to determine if the idea that the Niner’s start out slow is valid. The second chart lists all of the 1st quarter drives that stalled out, and why. A third chart summarizes what we learn from the first two.
The first chart is a list of all the game’s scores, including post-season. It’s broken out by quarter, and by quarterback.
Looking at this chart, it becomes clear that the slow start notion is true. What’s striking is just how clear the trend is. Regardless of the quarterback, regardless of conditions, almost regardless of the opponent, the Forty Niners consistently scored half the points in the first quarter that they did in the remaining quarters. This is something fundamental about the Forty Niners. Was it game plan? Was it personnel? Was it something as basic as deferring the kickoff to start the game?
Two quick notes.
1. Kaep was solid. I’d argue that his strength of schedule was harder than Alex’s, with half his games coming against playoff teams. And yet the Niners scored considerably more under him. Note how consistent his 2nd, 3rd, and 4th quarter totals were.
2. Just looking at an overview of our losses, it looks like our toughest games were against teams that looked like us. Minnesota, NY Giants, Seattle, St Louis, Baltimore and to a lesser extent, New England. Balanced offense. Good defense. Special teams. The teams with the most balance have the most tools.
The second chart lists all of San Francisco’s 1st quarter drives that failed to score a touchdown. I included drives even if we scored a field goal, because the drive still failed to convert a new set of downs. I wanted to know why. Why are we flailing around in the first quarter?
The first thing to note is that we received the ball to start the game far more often than I thought we had. Our failure to score in the 1st quarter was not because we were not getting the ball. We were getting the ball, and yet, out of all of those opportunities, only twice did we score a touchdown on the first drive of the game.
There’s an interesting difference in the drive details between Alex and Kaep. Alex started his games with a lot of long drives that failed to punch it in at the end. Kaep had a lot more 3 and outs. This is something else to look at, but my gut feeling is that Roman put more passing responsibility on Kaep. I remember a number of 3 and outs where Gore never even touched the ball. Yeah, I was shaking my fist at the TV a lot. How could you have one of the best RBs in the game and not be using him? Frustrating.
In the last column, I’m seeing a lot of sacks and penalties. Those are drive killers. I think there were only a couple of times were a sack or a penalty didn’t immediately prevent converting a new set of downs. And one of those drives did eventually stall, anyways. It looks like about half our drives stalled due to penalties and sacks. A good portion of stalls were due to incomplete passes on 3rd and whatnot. I don’t know if that’s game plan, or execution? Probably a little of both.
The third chart is a summary of information from the last column detailing why a drive stalled. What I want to do is determine where the Forty Niners should be concentrating to improve the most.
The ‘Runs’ and ‘Passes’ rows are those drives that failed to convert not due to the other four causes. There was no obvious cause. They just failed to convert. Regardless, how do we fix that? Is it game plan and/or execution. Is that just something that will come from more experience? Or is it something they need to concentrate on in practice?
The pass to run ratio in the first quarter is a little frustrating, and perplexing. "Why, Roman, why?!" (shaking my fist). Would more balance there help? In the back of my mind, a small part of me is worried that Roman wants to move us to more of a passing team? That he is just itching to let it loose 60 times a game. I hope not. I really don’t want to become Green Bay. The overall numbers don’t support that, so why just the 1st quarter? Something’s going on there, but I’m not seeing it.
The fumbles, bad snaps and interceptions are only a little concerning. They were due mostly to inclement weather and a ‘rookie’ quarterback playing under center for the first time in his career. Seattle has a surprise coming if they think Kaep will be the same quarterback they saw at the end of last year.
Sacks and penalties. For me, this is the key. Fixing this is something that mostly comes down to line play. We have one of the best run blocking units in the NFL. But our pass blocking was not so good last year. Improving pass blocking fundamentals is the next step in the evolution of our offensive line. Improved line play will decrease penalties, sacks and failed 3rd down attempts in general. Fully 75% of our stalled drives could be extended, just from this one improvement. There would be a ripple effect out through the rest of the offense. Receivers look better because better pass blocking means more time for receivers to run routes and shake defenders. Kaep will have more completions and fewer interceptions. Our running backs become even more effective, because defenses will need to respect the passing game. Improved line play translates to more 3rd down conversions, more time on the field, and more rest for our defense. More rest for the defense means they become even more dominant and get off the field sooner. That means your offense gets back on the field sooner, wearing out the opponents defense and getting more scoring opportunities.
In the same way a good pass rush makes the secondary better, improved line play will ripple out through the rest of the offense, and the defense, "fixing" multiple issues on both sides of the ball. Just from what I’m seeing here, pass blocking is the single most important improvement we need to see this next year.
Ak4niner and others were wondering about league totals. Thanks to McTee for the team rankings link. Below are the compiled charts for the entire league.
The first chart shows ‘points per game’ totals per quarter, and total ppg. League totals are at the bottom.
The second chart is a percentage view of each individual team. I did this chart in an attempt to answer AK’s thoughts about other teams that might have the same distribution as the Niners.
The totals for the first chart show what you’d expect. Defenses start out strong. Then offenses pick it up towards the end of each half. CK pin
Regarding the 2nd chart, It seems like there’s a lot more variety for individual teams. There’s not easily defined categories to put teams in. At least, not that I’m seeing quickly. There were a few teams that had a 1st quarter percentage as poor as the Niners, but the rest of their quarters are all over the place, so I don’t see that they compare. I included a ‘variance’ column that shows the teams closest to the Niners, but I’m not getting any vibe from it.
Per McTee, check out Dallas. Just for fun.
There is one conclusion that I can draw from these charts. There are some very good teams that have figured out how to start out fast. We can too.