The Bucs lost games in the 2nd quarter

“It’s a game of inches”. “They just wanted it more”. “They didn’t pass the eye test”.

Cliches are hard to break. They give us easy narratives and simple answers to a complicated game. A way to package chaos in a neat bow. The 2017 season for the Tampa Bay Buccaneers was nothing short of complicated. Impossible preseason hype followed by two separate streaks of five consecutive losses on the way to a 5-11 record and the 7th pick in the 2018 Draft. A season to forget.

This isn’t a comprehensive autopsy of the 2017 season, but rather a peek at a slice of it that’s equal parts weird and undeniable, and maybe a little bit unexplainable. You may get to the end of this article with the same perspective you had to start with but I want you to make your own conclusions anyway, and hopefully we’ll all learn something along the way.

Simply put, the Bucs were awful the 2nd quarter of games last season and much better in all other quarters. But why? And why the 2nd quarter? Honestly, there are only guesses here, but I have a theory. First, let’s go over exactly what I meant by bad in the 2nd quarter.

The Bucs’ total score differential after the 1st quarter in 2017 was negative 15, or -15. This is the total sum of the point differential between the Bucs and their opponents at the end of the 1st quarter. For example, a 3-0 Bucs lead in Week 1 would be +3, and a 0-7 lead for an opponent in Week 2 would be -7 for a total of -4 and so on. So, over the course of the season the Bucs were outscored in the 1st quarter by 0.9375 points per game (or quarter). Not great but not bad either.

However, the total score differential margin from the 1st quarter to halftime is a whopping -46. If you subtract out the double-digit positive gains that quarter from the Chicago and Miami games the score margin for the other 14 games is a soul-crushing -79. That means opponents outscored the Bucs in the 2nd quarter by an average of 5.64 points per game, or about 6 times worse than the 1st quarter. Thankfully we can narrow down the big contributors: there were four games with double digit negative margins in the second quarter; vs. Arizona (-10), Minnesota (-14), and both games vs. Atlanta (-10, -14), accounting for 48 of the 79 points. If we further subtract those out, over the remaining 10 games the Bucs were outscored in the 2nd quarter by an average of 2.58 points per game, or about a field goal. That’s still about 2.75 times worse than the 1st and obviously less than ideal. The scoring margin of the 3rd quarter is -25. Not going great here but hey they were 5-11 and 3-7 in one-score games. Also, we can’t forget the Bucs had the worst defense in the NFL. Can we factor out the defense and just look at the offense?

Then I looked into Tampa’s EP, or Expected Points (EP) and Expected Points Added (EPA). The idea of EP is basically no matter where you are on the field, because we have historic data, we can estimate a team’s expected points from their down and distance from the end zone. So, if a team has 1st and 10 at the 50, that situation is worth 2.0 EP because historically teams in that situation score an average of 2.0 points. I’ll quote from the link to quickly explain the rest:

A 5-yard gain would set up a 2nd and 5 from the 45, which corresponds to a +2.1 EP. Therefore, that 5-yard gain in that particular situation represents a +0.1 gain in EP. This gain is called Expected Points Added (EPA). Likewise, a 5-yard loss on 1st down at midfield would create a 2nd and 15 from the offense’s own 45. That situation is worth +1.2 EP, representing a net difference of -0.8 EPA.

We can value turnovers in the same way. Suppose that on 2nd and 5 at the opponent’s 45 there was a fumble recovered by the defense. The 2nd and 5 was worth +2.2 EP, but now the opponent has a 1st and 10 on their own 45, worth +2.1 EP to them. The result of the play represents -2.1 EP for the original offense for a net loss of -4.3 EP. On average, a fumble in that situation means a net expected loss of a little more than 4 points.

You can tell where this is going.

Pro Football Reference has something similar or built off of this concept, called Expected Points Before (EPB) and Expected Points After, and a pretty reliable catalogue of all the Bucs’ plays on offense last season. What I’m interested in is the difference between the two, or the Expected Points Added. How many plays did the Bucs add to their EP (good plays) and how many plays did the Bucs have a bad play and lose EP? There’s tons of “good” and “bad” plays in every game that trade fractions of EP, so I chose the bigger more impactful ones - ones with at least 2 EP in either direction.

In the 1st quarter, the Bucs’ offense had thirteen (13) plays of +2 and 3 plays of -2.

In the 2nd quarter, the Bucs had 17 plays of +2 and 13 plays of -2, including the season’s worst play, a strip sack fumble of Winston on the Green Bay 30 yard line topped with a scoop and score where the Bucs earned a face-melting net loss of -10.59 EPA on the play. There were ten plays of -3 EP.

The 3rd quarter saw 21 plays of +2 and 9 plays of -2 and the 4th quarter/OT saw 29 plays of +2 (12 plays of +3 or better) and 13 plays of -2.

This kind of matches what we see on Sundays, right? They get behind and rally in the 2nd half. This also confirms that while Tampa’s defense was horrible last year the offense was also independently poor in the 2nd quarter.

Just for fun: the best offensive play of the season in terms of EPA was Jameis Winston’s big pass to O.J. Howard for a touchdown vs the Giants. Winston owns 16 of the top 20 highest-earning EPA plays from the offense last season. Some (but not all) of the worst or most damaging plays of the year (not limited to the 2nd quarter) besides the Packers scoop and score that I decided to pick out to share (misery loves company) were Peyton Barber’s fumble at the 5 vs the Falcons, Ryan Fitzpatrick’s interception at the one yard line vs the Cardinals, Adam Humphries’ heartbreaking fumble vs the Bills, and Winston’s strip sack fumble by Kawaan Short and subsequent personal foul penalty.

So, OK. They were bad in the 2nd. The 2nd quarter had 21 percent of all the +2 EP plays and 34 percent of all the -2 EP plays. Nearly a third of all the bad plays in the season occurred during the 2nd quarter. Winston had four interceptions and four fumbles in the 2nd quarter alone last year. In fact, in the 2nd quarter the Bucs had 16 sacks (5.8% of 2nd quarter plays ended in a sack), 5 interceptions (1.8%), and 9 fumbles (3.3%).

In all other quarters combined the Bucs had 24 sacks (3.2% of all plays in 1st, 3rd, 4th, and OT), 9 interceptions (1.2%), and 17 fumbles (2.2%). So the 2nd quarter saw 40 percent of all the season’s sacks, 36 percent of all interceptions, and 35 percent of the season’s fumbles.

OK, so for some reason everything got worse in the 2nd quarter - more sacks and more turnovers. We know turnovers had to contribute at least in part to the poor 2nd quarter team performance. We also know Winston is the biggest contributor to the team’s turnovers. And we can also infer more turnovers will lead to more possessions and opportunities for the opponent to score with and less for your team to score. But what is different about the 2nd quarter that makes him turn the ball over more?

Trying to Get to the Bottom Of It

Lets dig into Winston. There is a metric I like very much called success rate. Success rate is essentially used to gauge efficiency. Some definitions differ, but a play, run or pass, is considered successful if it gains at least 40 percent of needed yardage on 1st down, 60 percent of needed yardage on 2nd down and 100 percent of needed yardage on 3rd and 4th down. Much of the following metrics came from sharpfootballstats.com So, how did the Bucs do compared to the rest of the NFL?

Essentially, you could argue Winston (with a little help from Fitzpatrick) was a top 5 QB (or game manager?) in quarters 1, 3, 4, and OT. Because of his aggressive nature he keeps the offense on schedule and gains the yardage needed to have a successful play at a high rate (even if he can’t hit the deep ball). Since he’s been the starting QB the Bucs generally have much fewer 3rd downs than most of the teams in the league and have been top ten in avoiding 3&out drives. And yet the table above is even more reinforcement of something weird going on in the 2nd quarter. Winston, to a lesser extent Fitzpatrick, and the Bucs’ passing game was awful. How or why could their passing efficiency take such a nosedive? Winston is turnover prone in general but we’re not any closer to figuring out why he was so much worse in the 2nd quarter. Is it just bad luck? Or something else?

Here are Winston’s raw stats: in quarters 1, 3, 4, and OT are 210/319 65.8% for 2,630 yards, 19 TD 7 INT 8.2YPA and a 102 QB rating. The average yards to go for a first down for these quarters was 7.82.

Winston’s stats in quarter 2 are 73/125 58.4% for 880 yards, 0 TD 4 INT 7.0 YPA and a 66.8 QB rating. The average yards to go for a first in the 2nd quarter was 8.26.

Ok, so a worse QB will definitely cause more incompletions, which will lead to more frequent passing downs (i.e. 2nd and 3rd and long) which will up the average yards to go. These downs are more difficult to be successful in, just because you have farther to go for a first down and teams know you’re more likely to pass. Was there something the Bucs were doing specifically in the 2nd quarter that made passes more difficult in the 2nd quarter? Let’s start with where they were throwing the ball.

In the above table we can see the frequency of passes thrown to particular areas of the field. The Bucs like to suck up the linebackers with the threat of the run and throw behind them and in front of the safeties, and that can be in the 10-15 yard range so maybe this doesn’t tell us much. They don’t appear to have attempted more deep balls like I thought. Below that, we see Receiving Success Rate Over Average. Receiving success rate is based on how many yards a receiver gained based on the location and down of the pass he caught, and in this case compared to all other receivers in the NFL last season. As you can see, despite all of Winston’s deep ball issues the Bucs were practically in line with the league average in quarters 1, 3, 4, and OT past 15 yards. However, because this data doesn’t differentiate between anything deeper than 15 yards the Bucs’ extensive intermediate passing scheme could hide the struggles of passes 20+. Still, the same issue of the 2nd quarter crops up here too. The receivers were also much worse in the 2nd quarter at all areas of the field except one. In all other quarters though it’s nice to see the Bucs’ receivers so much better than the league average in passes caught under 15 yards. Humphries and tight ends Brate and Howard probably have a lot to do with this.

So what about air yards? How far was he actually throwing the ball?

Ok so now we’re getting somewhere. Winston was definitely throwing further down the field in the 2nd quarter, and the percentage of pass plays on 1st down went up 9 percent. What is this due to? Winston pushing the ball further down the field? Dirk Koetter trying to ‘open up’ the offense? I’ll leave you with this:

The Bucs’ three biggest personnel groups in 2017 were 11 (one running back, one tight end, which leaves 3 WRs), 12 (two tight ends) and 13 (three tight ends). 11 is easily the most common personnel grouping in the NFL as most teams choose to put a 3rd WR on the field instead of a second TE and despite having Brate and Howard the Bucs are no exception. So you can see the Bucs used heavier personnel packages in the 1st quarter and ran the ball more, then heavily went 11 personnel in the 2nd quarter on first down and called pass much more frequently. It’s also worth noting that in the 1st quarter Winston was under center 66 percent of the time and in shotgun 34 percent of the time. In the second quarter Winston is under center 51 percent of the time and in shotgun 49 percent of the time. In addition, when Winston was lined up in the shotgun only one team passed more frequently than the Bucs, who passed in shotgun 90 percent of the time in the 1st quarter and 93 percent of the time in the 2nd quarter. A running back who can run out of the shotgun might do wonders for this offense.

Why do you think the Bucs’ offense was so bad in the 2nd quarter? Winston? His injury causing him to be pulled from two 2nd quarters and missing 3 other games? Offensive predictability? Bad luck? Some combination of all of the above?