Before getting into previewing this season, I thought I would take a look back at last year. Specifically, I want to isolate offenses and defenses within the ACC, to see exactly how they stack up. For this purpose, I’ve looked at only conference games. I have done this for 2 reasons. First, I think it gives us a better comparison because it keeps the competition more consistent. Clemson played Auburn and South Carolina while VT played 4 nobodies out of conference, for example. Those kind of schedule strength disparities unfairly skew any comparison based on the whole season. The schedules played in conference are not perfectly equal, but they are better. Second, I’ve heard way too much all offseason (mostly from GT fans) that our offense wasn’t really that good last year, and only padded its stats against Western Carolina, MTSU and Kansas. The popular belief goes that, in conference play, our offense was rather average, and failed us in our 3 losses. As I’ll show in a second, our offense was not actually all that average. Anyway, let’s start with the conventional measures of offense and defense, points per game and yards per game.
Offense:
Team
|
Points Per Game
|
Clemson
|
34.2
|
FSU
|
31.3
|
UNC
|
28.1
|
Georgia Tech
|
28.0
|
Wake Forest
|
27.5
|
Miami
|
27.4
|
Virginia Tech
|
25.8
|
NC State
|
25.8
|
Maryland
|
23.8
|
UVA
|
19.9
|
Duke
|
19.5
|
Boston College
|
17.4
|
Team
|
Yards Per Game
|
Clemson
|
445.2
|
Virginia Tech
|
409.3
|
UNC
|
396.6
|
FSU
|
393.6
|
Maryland
|
392.5
|
Georgia Tech
|
391.8
|
Miami
|
388.5
|
UVA
|
379.5
|
Wake Forest
|
376
|
Duke
|
351.6
|
NC State
|
339.5
|
Boston College
|
293
|
OK so a couple of things jump out at me. Clemson is clearly the best offense, in both categories by a good margin. VT and FSU are both better than I would have guessed. And wow Boston College was terrible. As for GT, we are 4th in points and 6th in yards, but pretty close to the average. Even though we were 4th in scoring, the margin between us and 2nd place was bigger than that between us and 8th place. Pretty much middle of the pack.
However, as I’ve argued on here a few times, I think what really matters is efficiency. Per Game stats are very overrated. The reason is simple. The more efficient you are, the better your odds of winning. As an extreme example, imagine an offense that held the ball for 20 plays and scored a TD every drive. Say you get the ball to start the game, and scored, then scored on your 2nd drive, which ends the half. Then you kickoff to start the 2nd half, and when you get the ball you go on another 20 play scoring drive. By now its the 4th quarter, you have 21 points, and your opponent is just now getting the ball for the third time. If they have scored a TD every time they had it, then you will be tied. If not, you will be ahead. When you get the ball back, most likely you will run out the clock for a victory. Its mathematically impossible for you to lose this game in regulation. Of course, you’ve only scored 21 points. This would be a “bad” game by Per Game standards, but you offense has played so well that you really can’t lose.
Now, that example is not very realistic, but I use it to make the point. If you run a hurry up offense and go for a lot of big plays, you’ll tend to score more points per game. But you’ll also be putting your defense out there more often, because your games are going to have a lot of possessions. This is not necessarily “good” offense. 28 points on 10 possessions is better than 28 points on 15 possessions, if you are trying to win the game.
Offenses that move the ball steadily (rather than a feast or famine, big play or punt style) and those that run the ball well will tend to eat a lot of clock and shorten the game in terms of possessions. Defenses that don’t get many stops but also give up few big plays will also tend to shorten the game. Of course, GT falls into both of these categories. Our offense eats up a lot of clock every possession, because we run the ball so much and rarely go 3 and out. When we get the ball, we are usually going to hold it for at least 4 or 5 minutes. That makes it pretty tough to have very many possessions. Similarly, our defense does not force many 3 and outs. We tend to let our opponents move the ball slowly and steadily, and also eat a lot of clock.
So how big of a difference does this really make? How many possessions per game do we really lose? Well, let’s look. (Also of note, in calculating these stats, I went game by game, and removed non-offensive touchdowns, and did not count any possession where a team simply ran out the clock to end the half. I’ve also included the average starting field position, which the offense has little control over.)
Team
|
Points Per Possession
|
Average Starting Field Position
|
Average Yards Per Possession
|
Total Number of Possessions
|
FSU
|
2.65
|
32.18
|
35.38
|
89
|
Miami
|
2.62
|
28.54
|
38.37
|
81
|
Georgia Tech
|
2.53
|
28.20
|
37.76
|
83
|
Clemson
|
2.51
|
30.85
|
35.15
|
114
|
Virginia Tech
|
2.30
|
29.18
|
36.48
|
101
|
UNC
|
2.22
|
30.07
|
32.38
|
98
|
Wake Forest
|
2.22
|
28.97
|
31.33
|
96
|
NC State
|
1.83
|
30.70
|
33.53
|
101
|
Duke
|
1.71
|
29.86
|
30.91
|
91
|
UVA
|
1.63
|
27.53
|
32.65
|
93
|
Maryland
|
1.57
|
28.97
|
31.72
|
99
|
Boston College
|
1.47
|
31.97
|
26.04
|
90
|
So you can see, we now move to 3rd place, but more importantly, we are now far closer to 1st than we are to the average. Rather than being “middle of the pack” with a bunch of teams, we are one of the best offenses in the conference. Only Miami is clearly better. FSU averages .12 more points per possession, but also enjoys better starting field position by more than 4 yards per possession. I’m not sure how significant that is, but I would guess it helps. We are 2nd in the conference in average yards per possession, and again much closer to first place than to the average. Our 83 total possessions are fully 11 possessions fewer than the average. The average points per possession for the conference is 2.1, so 11 possessions is equivalent to about 23 points on the season. Or nearly 3 points per game. And this is just comparing us to the average. Half the offenses you might compare us to, on a per game basis, will enjoy an even bigger advantage in terms of possessions. Clemson, for example, had 114 possessions on the season. (They also played an extra game, the ACC Champ Game). Clemson is exactly the team I was thinking of when I described an offense that makes a game longer in terms of possessions. They run a hurry up offense all the time (Chad Morris stated goal is to run as many plays as possible) and they also hit a lot of big plays. They average 12.6 possessions per game. We average only 10.3. This is an enormous difference.
The bottom line for me is that, even if you look at conference games only, our offense was quite good. This notion that we only looked good against weak non conference teams is nonsense. I mean, we did pad out stats in those games. But our stats, fairly judged, was still very good. Our offense should not be punished for moving the ball, eating clock, and playing with a defense that lets other teams move the ball.
Yes, our offense was awful against Miami. However, outside of that game, the offense was rarely a problem. Many might think “now wait a second, what about UVA?”. Well, we scored 14 points on 8 possessions that game. (don’t forget the defense scored 7). UVA’s defense allowed only 1.79 points per possession. So our 1.75 average that game was basically an average performance. When your 2nd worst game of the season is scoring what your opponent averages allowing, you’ve got a pretty good offense. The real problem this game was allowing 24 points on only 9 possessions, against one of the conference’s worst offenses. UVA scored 1.63 points per possession on the season. And yet we allowed them to score 2.67. And we let them eat clock, including the final 6 minutes of the game. The offense did not play well, but the defense was clearly more of a problem. However, if you look only to per game stats, you’ll reach the opposite conclusion - ok game for the defense, bad for the offense. Similarly, in the loss to VT, we scored 2.6 points per possession on a defense that allowed only 1.8, and allowed 3.7 to an offense that scored only 2.3. Our offense was responsible for NC State’s worst defensive game all year, arguably Clemson and UNC’s worst games all year, and VT’s 3rd worst game. That’s not a bad record.
Now, here is a look at just how bad the defense was. Once again, we will start with the conventional measures, points per game and yards per game.
Team
|
Points Per Game
|
FSU
|
17.8
|
Virginia Tech
|
20.4
|
UVA
|
22.9
|
Miami
|
23.8
|
Boston College
|
24.4
|
NC State
|
24.4
|
Clemson
|
26.2
|
Georgia Tech
|
26.5
|
UNC
|
27.9
|
Wake Forest
|
28.4
|
Duke
|
31.6
|
Maryland
|
36.0
|
Team
|
Yards Per Game
|
FSU
|
315.8
|
NC State
|
325.1
|
Virginia Tech
|
350.7
|
Clemson
|
361.7
|
Georgia Tech
|
368.6
|
UVA
|
370
|
UNC
|
377.1
|
Miami
|
380.1
|
Boston College
|
394.5
|
Duke
|
428.4
|
Wake Forest
|
441.1
|
Maryland
|
461.9
|
This time Maryland is the team that is randomly awful. FSU is clearly the best. VT is also good, as expected. Clemson and NC State are a little curious. Both are much better in yards allowed than points. And GT is basically middle of the pack. Lets see how it changes when we break it down per possession:
Team
|
Points Allowed Per Possession
|
Average Starting Field Position
|
Average Yards Allowed Per Possession
|
Total Number of Possessions
|
FSU
|
1.51
|
25.91
|
26.87
|
94
|
NC State
|
1.74
|
32.12
|
26.01
|
100
|
UVA
|
1.80
|
29.15
|
31.49
|
94
|
VT
|
1.82
|
20.01
|
31.25
|
101
|
Clemson
|
1.87
|
29.80
|
29.32
|
111
|
UNC
|
2.11
|
33.65
|
30.47
|
99
|
Boston College
|
2.12
|
26.77
|
34.30
|
92
|
Miami
|
2.14
|
25.96
|
40.01
|
76
|
Wake Forest
|
2.25
|
30.55
|
34.94
|
101
|
Georgia Tech
|
2.39
|
32.30
|
35.53
|
83
|
Duke
|
2.80
|
28.67
|
38.94
|
88
|
Maryland
|
2.82
|
32.46
|
38.09
|
97
|
One of the first things that jumps out at me is exactly how good NC States defense was. Surprising. I never considered them very good defensively. The field position column is interesting to me. I included it because, obviously, it has a big effect on what a defense (or offense) should do. I find it incredible that NC State was 2nd in points allowed per possession despite being 9th in average starting field position. They were given an average starting position a full 7 yards worse than FSU. I’ve never tracked this stat before, but my guess is that 7 yards per possession represents an enormous disadvantage.
Without any research to confirm this, my suspicion is that a defense (or an offense) really has no control over where it starts each possession. If the defense forces a 3 and out, in theory that makes it more likely they will have better field position next drive. But the offense could still turn the ball over, or the special teams could allow a big return. Even if the defense forces a stop at the opponents 20 and gives its offense the ball at midfield, that only translates into very good field position if the offense punts (and the punt unit does a good job). If the offense scores, then the next starting field position will be wherever the kickoff unit can get a stop (typically the 25 or 30). So, my guess is that a defense does not have much influence over average starting field position. The defense is dependent upon a combination of special teams, the offense, and the coach’s game management.
I think its noteworthy that GT is terrible in average starting field position of both our offense and defense. We rank 10th for the defense and 11th for the offense. We give up an average of 4 yards per possession to our opponents in field position. Which in my opinion shows how poor our special teams have been. Hope the new special teams coach gets that fixed.
Anyway, several things become apparent in the “per possession stats”. Most important for us is that GT is a lot worse than the “per game” stats make it appear. We were 8th in the per game numbers, but not very far below the mean. In per possession, we drop to 10th, and 0.3 points below the mean, or about 15%. Over the course of a regular game, that equates to about 3.5 points below the mean, instead of the 1 point or so we are below average if you look at the per game stats.
FSU is still the best defense in the conference (although I think a case can be made that State is as good, given the tremendous discrepancy in starting field position). Miami wins the award for doing the least with the most (irony... the defense per possession numbers being a microcosm of the program itself) as they have excellent starting field position but are dead last in yards allowed per possession. Also of note, Clemson is fairly undervalued by the per game stats (the opposite of GT). This is somewhat hard to believe given the bowl game, but in ACC play, Clemson’s defense was actually quite good.
And that is the bottom line for me. When you look at possessions, the middle 60 or 80% of teams are probably pretty close in how many they average per game. Thus, per game stats probably gives a reasonable measure for these teams. The team with the 20th most possessions nationally per game might be just under 12, and the team with the 100th most might be just over 11 per game. I don’t know, I’m guessing. But my gut is that there is not a huge difference in the middle, which is why most people just use per game stats. If you pick two random teams, the per game stats are probably a pretty good comparison.
But when you run into teams like GT and Clemson, per game stats can be very misleading. GT is at one extreme (10.3 possessions per game) while Clemson is at the other (12.6 possessions per game). Clemson’s offense ran hurry up and passed all the time. Clemson’s defense got more stops and turnovers. GT’s offense runs a methodical run game that is hard to get off the field. GT’s defense does not do a good job getting anybody off the field.
Believe it or not, GT’s defense was actually overrated last year. And Clemson’s defense, which was bad enough to get its DC fired, was actually one of the best units in the conference. Meanwhile, Clemson’s offense was overrated, and GT’s offense was very slightly better (or, really, basically the same). I bet you most fans don’t have that opinion based upon last year. Most fans would probably tell you GT and Clemson had similar defenses (both bad, neither terrible) and Clemson won the conference on the strength of its offense.
Well, says here that Clemson’s defense was just as good as its offense, and Clemson won the ACC by being balanced. No idea though how FSU managed to take the conference’s best offense AND defense and finish 5-3. Not sure Jimbo is a very good in game coach. They won a lot of blowouts and lost every close game. They seemingly could not wait to choke. Usually a sign of poor coaching.
Anyway, in the next few days I’ll do a little more analysis with the per possession numbers before I get into projecting next season.
No comments:
Post a Comment