My blog has moved!

You should be automatically redirected in 6 seconds. If not, visit
and update your bookmarks.

Tuesday, October 27, 2009

Ground Game Grab Bag: More Fun With Numbers

Usually most statistical research is done as a means of answering some core question. But I'm not sure there's an answer to Georgia's running game woes beyond a simple, sweeping statement: The Dawgs are just bad at running the ball.

So these numbers weren't meant to prove anything. Instead, I was wondering if they might create some new questions. In many respects, I think that may be the case. So while the bottom line may still be that Georgia stinks at running the ball, there is certainly more to the story.

As I pointed out in posts yesterday, there's some statistical evidence that shows that maybe Richard Samuel's lack of "instincts" isn't his problem and there's some evidence that shows that Clint Boling's potential move to left guard could be a boost for the Dawgs, too.

Here's some more of what I found in looking closer at the running game. (WARNING: If you haven't enjoyed the previous nuts-and-bolts, statistics-heavy posts we've done, you're going to hate this one.)

Breaking down the running game by quarter:

Quarter Carries YardsTB carries
TB yards
1st 49 135 41 115
2nd 42 214 32 195
3rd 54 157 43 154
4th 63 235 47 245

First, a note: TB carries and TB yards are runs by tailbacks only. My thought process was that Branden Smith's runs or Logan Gray's sneaks (can they be called a "sneak" when everyone in the stadium knows what's coming?) or the fullback runs aren't really a true indication of Georgia's running game since they're typically gadget plays or short-yardage situations. They all count in the stadings and contribute to the outcome, but they don't necessarily give us the best information about Georgia's actual success running the football. This also takes out the negative yardage from QB sacks.

In either case, there's a pretty obvious trend that develops here that is similar to what we noticed when we studied Willie Martinez's defense a few weeks ago: Georgia is much better in the second and fourth quarters than it is in the first and third quarters.

I don't know if that means that Georgia's game plan entering games is bad or that the adjustments made during the game are good. It could certainly be the case that most defenses enter a game geared to stop the run, and as the offense loosens things up throwing the ball, the running game finds more room to run.

Whatever the reason, the outcome is pretty obvious: In the first quarters of games this season, Georgia is averaging 2.8 yards per carry by its tailbacks. In the second quarter, that number jumps to 6.09 ypc. In the third quarter, the average falls again, all the way down to 3.58 ypc. But in the fourth quarter, the Dawgs' find success once again, running at a 5.21 ypc clip.

Even if we add in all the yardage on the ground along with QB sacks, the trend remains, albeit slightly less pronounced. In the first quarter, UGA averages 2.75 ypc, in the second, that jumps to 5.09 ypc. The third quarter takes another dip to 2.91 ypc, then jumps again in the fourth quarter to 3.73 ypc.

Again, it's hard to know for sure what conclusions we can draw from these numbers, but it does seem to support Mike Bobo's dedication to running the football, even when the early success isn't seen.

Of course, yards are one thing. Points are another. Does running the ball more actually lead to more points on the board for Georgia? After all, the score is really the bottom line.

Carries Offensive Points
First 49 45
Second 42 47
Third 54 28
Fourth 63 47

From these aggregate numbers, it seems pretty clear that running the ball more or less doesn't really have much of an effect on denting the scoreboard at all. In fact, the only real glaring stat here is that Georgia downright stinks in the third quarter. Other than that, the offensive output is essentially unchanged at any other point in the game, regardless of whether the Dawgs run it 42 times or 63 times.

Those are aggregate numbers, however. What about more specific instances?

The truth is, there haven't been many quarters in which Georgia has been particularly successful running the football. That probably shouldn't come as a surprise, given that the Dawgs rank 103rd nationally in rushing yardage. But really, it's almost impressive how consistently bad Georgia has been.

In 28 quarters of football this season, Georgia has run for 20 or fewer yards 19 times (68 percent). In eight instances, Georgia has gone an entire quarter with 10 or fewer rushing yards.

Even if we eliminate all the gadget plays and sack yardage, etc., and stick just with the work of the tailbacks, those numbers still look awful. The tailbacks have failed to net 20 yards in a quarter 16 times and have been held to 10 or fewer on six occasions.

And for all the bad efforts, there are very few good ones to balance them out. Georgia's tailbacks have topped 40 yards in a quarter just four times all season, and one of those was buoyed by an 80-yard run by Richard Samuel. Add back in all the gadget plays, etc., and as a team the Bulldogs topped 40 yards in a quarter just six times (21 percent).

But forget about success for a second. Let's just look at commitment. Does running the ball more translate to more points?

Here are Georgia's most prolific quarters of the year in terms of running the ball:

Points next Qtr
17 4Q vs Van
14 N/A
12 4Q vs Ark
10 N/A
11 1Q vs Ark
10 17
11 3Q vs OSU
3 0
10 3Q vs ASU
0 6
10 1Q vs OSU
7 0

So Georgia's six most prolific running quarters (i.e. the top 21 percent of its total quarters) of the season have accounted for 26 percent of the Bulldogs' total offensive points. Essentially there is no appreciable difference in scoring success when the Dawgs run the ball more often and there's little evidence to show that running the ball a great deal in one quarter "loosens up" the defense for the next quarter.

Of course, there is one other thing that running the football often does: It keeps the opposition off the field. We've discussed on this blog several times about how Georgia's inability to sustain long drives has killed the Dawgs in time of possession and left the defense reeling far too often. On 58 of Georgia 88 offensive drives (or would-be drives in the cases of special teams fumbles) the Dawgs have run five or fewer plays this season. That's a whopping 66 percent of the time. It's not wonder the Dawgs' D ranks last in the SEC in scoring defense.

The way to combat that? Running the football.

Running the ball effectively generally leads to more offensive plays which leads to a greater time of possession which leads to fewer opportunities for the opposition to score.

So, has running the ball more helped Georgia's defense? Here, again, are Georgia's most prolific running quarters in terms of carries:

Opp Pts
17 4Q vs Van 10:59 0
12 4Q vs Ark
8:50 3
11 1Q vs Ark
9:04 21
11 3Q vs OSU
6:59 7
10 3Q vs ASU
7:45 14
10 1Q vs OSU
8:34 0

It's safe to call the first quarter against Arkansas and the third quarter against Arizona State anomolies in terms of scoring. Georgia struggled with turnovers in each, and of those 35 points, 28 followed turnovers.

So set that aside, and the numbers show what intuitively should be obvious: When Georgia runs the ball often, the offense wins the time of possession battle (in this case, five out of six times) and the defense looks better.

Now let's look at the six quarters in which Georgia ran the ball the least:

#carries When T.O.P.Opp Pts
3 4Q vs. OSU
7:31 7
3 2Q vs LSU
2:50 3
4 1Q vs Tenn
5:59 0
4 2Q vs Tenn
7:26 21
4 1Q vs SC
3:41 17
4 3Q vs Ark
4:46 17

That first quarter vs. Tennessee is the clear outlier here, but remember, Georgia dinked and dunked the ball in the passing game enough that it effectively simulated a running game. While the Dawgs only allowed three points against LSU in that quarter, too, it's worth remembering that the defense essentially had to stand on its head to keep the Tigers off the scoreboard.

Overall, the lack of a running game not only resulted in points for the opposition, but left Georgia absolutely crushed in the time of possession sweepstakes.

Does this mean Georgia needs to run the ball more? Or should we really consider that lack of production and say Georgia is already probably running the ball too much?

I'm not entirely sure what the answer is, but I do know this: Senator Blutarsky makes a compelling argument that Florida's best strategy against Georgia this week is to sit back and wait for the Bulldogs to beat themselves.

If you’re Urban Meyer, it’s not hard to draw up a basic strategy for this week. It boils down to three simple words: play it safe. Let Georgia beat itself; with the Dags’ turnover problems and inability to run the football consistently, that’s not exactly a monumental task for a team with the best defense in college football.

It's probably an effective strategy, considering the number of miscues Georgia has made this year.

But running the football -- regardless of how effective those runs are -- helps kill the clock and (assuming Richard Samuel isn't coughing up the ball) limits your opportunities to make mistakes.

So perhaps there's some real merit to simply waiting for the right moment to look for the big play and spending the rest of the time chipping away on the ground.


ChicagoDawg said...

Good stuff David and certainly some interesting data. However, in doing statistical analysis and/or regressions (linear or multiple) one should always remember that a set of data may have a high degree of correlation without necessarily answering causation. This is especially so in football statistical analysis as the data sets are usually quite small (relatively speaking) and there are so many variables to consider. For example, a given team may run more in the 4th quarter, have more points in the 4th quarter, etc., because they amass these statistics in games in which they hold a large lead and thus are inclined to run the ball more and are playing mis-matched opponent. So, the bad team(s) might be the causal effect of greater 4th quarter ground and scoring success, not the commitment to running the ball in the 4th quarter.

Anonymous said...

I think a big chunck of the yardage gained in the 2nd & 4th qtrs were on late drives and the D is playing 'bend don't break' to prevent scores, thus allowing our RB to gain easy yards outside the red zone. Just a thought.

Sorry for the same comment on the next story - I wasn't paying attention.

Ant123 said...

David, Maybe what all your research shows is that the dinks in the 1Q of the TN game are the way to go. It accomplishes the same thing as far a time of possession goes, and we seem to be able to execute it effectivly.

Anonymous said...

You can't run the ball often if you run it poorly because running it results in 3 and out and the D going back on the field. If you aren't a good running team, the key to running it more is to pass enough to keep the chains moving. You run more because you get more opportunities, something your analysis seemed to largely ignore. Runnning into a stone wall is a bad strategy.

UGA69Dawg said...

I have wondered for sometime if we pipe in soft music and let the players nap during half time. It takes us a whole quarter to get going, that's why we are better in the 2nd qtr and in the 4th. We sleep walk through the 1st and 3rd.

Lane said...

David, have you considered economics. Stay out of accounting.

Boling does make sense. However is there any regret in moving Tripp back to defense?

Senseless penalities are a major factor in not running the ball, with success.

Blocks are not being finished, which should have been corrected in preseason.

Good possibility these guys are not that good. Moreno, possibly hid these weaknesses last year.

Trinton was a big loss!!

Since, Searles isn't talking will never know his opinion on the overall talent.