The last two weeks, I've revealed the results of a pet project analyzing UCF's offensive and defensive production by over the past five seasons by possession. The results were interesting.

Let's go back and look at the offensive and defensive numbers:

Offense

Year	W/L	Possessions	TD	FG	Punts	Turnovers	Def/ST TD	Half	Poss./ Game	Off. Points
2017	13-0	176	77	13/18	46	25	7	10	13.5	576
2016	6-7	201	39	17/22	81	45	7	14	15.5	323
2015	0-12	162	17	13/17	75	49	1	6	13.6	165
2014	9-4	176	42	21/29	56	42	1	11	14.7	356
2013	12-1	163	54	21/24	43	26	1	18	12.5	441

Year	W/L	Scoring Rate	TD Rate	Failed Poss. Rate	Giveaway Rate	Points/Poss.
2017	13-0	54.0%	43.8%	40.3%	14.2%	3.27
2016	6-7	30.3%	19.4%	62.7%	22.4%	1.61
2015	0-12	20.4%	10.5%	75.3%	30.2%	1.02
2014	9-4	39.2%	23.9%	53.4%	23.9%	2.02
2013	12-1	47.2%	33.1%	41.1%	16.0%	2.71

Defense

Year	W/L	Def. Poss.	TD	FG	Punts	Turnovers	Def/ST TDs	Half	Def Pts.
2017	13-0	181	41	11/18	65	47	3	10	310
2016	6-7	203	38	14/22	89	44	2	10	306
2015	0-12	167	56	17/18	58	19	2	16	434
2014	9-4	177	27	16/20	77	43	2	10	242
2013	12-1	155	34	8/12	66	37	2	6	263

Year	W/L	Stop Rate	Takeaway Rate	Opp. Scoring Rate	Opp. TD Rate	Pts./Poss. Allowed
2017	13-0	61.9%	26.0%	32.6%	22.7%	1.71
2016	6-7	65.5%	21.7%	29.6%	18.7%	1.51
2015	0-12	46.1%	11.4%	44.3%	33.5%	2.60
2014	9-4	67.8%	24.3%	26.6%	15.3%	1.37
2013	12-1	66.5%	23.9%	29.7%	21.9%	1.70

Now let's break these down further by asking some questions:

How can we measure UCF's Total Efficiency by Possession?

The job of the offense is to put points on the board - or at least, give itself the opportunity to put points on the board. Conversely, the job of the defense is to prevent the other team from getting the opportunity to put points on the board. We can get a pretty good measure of these things by looking at the Scoring Rate on offense and the Stop Rate on defense.

I like to think of these as the football version of on-base percentage in baseball because they show us how often you convert your given opportunities.

So to get a total picture of the by-possession success of the team as a whole, let's add the Scoring Rate and Stop Rate, to produce a football version of OPS, if you will:

Year	W/L	Off. Scoring Rate	Def. Stop Rate	Total Efficiency
2017	13-0	54.0%	61.9%	115.9
2016	6-7	30.3%	65.5%	95.9
2015	0-12	20.4%	46.1%	66.5
2014	9-4	39.2%	67.8%	107.0
2013	12-1	47.2%	66.5%	113.7

WHOA. Interesting.

If you're a numbers geek like me, one thing jumps out: When UCF's record was above .500, its Efficiency Rate was over 100, and vice versa. It almost reminds me of Passer Efficiency. So it looks like we have a magic number. I'll have to process more data to see if there's a true correlation, but I suspect there might be.

What I like about this is that there's more than one way to show a team had a good season. For example, we know that 2017 was a great season for both sides of the ball. It doesn't take two years of calculus to figure that out.

However, 2014 was a pretty good year too. UCF won a share of the conference title, after all. But that team doesn't seem to get enough respect due to the offenses' inconsistency. While that is true (their Scoring Rate was 8 points lower than 2013 and 15 points lower than 2017), the 2014 team's defense was excellent - the best of this five-year period. That more than made up for the offense's shortcomings that year, and the number bear that out.

Scoring Drive +/-

I love the five-man unit +/- stats in basketball because I think that shows us some real insight into who is playing really well as a team together.

So a football version of that would be a version of this per possession. I had my misgivings about this stat, because I thought it might not tell us much more than, say, plain old scoring margin. But I still wanted to find out, because why the hell not:

Year	Scoring Margin	Offensive Pts/Poss	Defensive Pts/Poss	Scoring Drive +/-
2017	+21.5 (1)	3.27	1.71	+1.56 (1)
2016	+1.3 (4)	1.61	1.51	+0.10 (4)
2015	-25.8 (5)	1.02	2.60	-1.58 (5)
2014	+6.8 (3)	2.02	1.37	+0.65 (3)
2013	+13.3 (2)	2.71	1.70	+1.01 (2)

Meh. I didn't learn anything new here.

True Turnover +/-

I always thought that turnovers were more than just interceptions and lost fumbles. So when analyzing possessions, I counted the other plays I considered turnovers - events that end possessions with no opportunity for points and may also put you at disadvantageous field position:

• Lost Fumbles
• Interceptions
• Turnovers on Downs
• Safeties Taken
• Blocked Punts
• Blocked Field Goals

So I want to see if there's some kind of correlation between their true turnover rate and turnover margin and/or additional insight to be gained by adding in that extra data.

Year	Giveaway Rate	Takeaway Rate	True Turnover +/-	Turnover Margin
2017	14.2%	26.0%	+11.8	+17
2016	22.4%	21.7%	-0.7	+1
2015	30.2%	11.4%	-18.8	-20
2014	23.9%	24.3%	+0.4	-1
2013	16.0%	23.9%	+7.9	+5

Well, OK. This just confirms something we already knew. No big differences here.

Conclusions (?)

I think I'm on to something with the Total Efficiency stat. Here we can synthesize a team's offensive and defensive performance into a single number that shows how good a team was at both capitalizing on its possessions and preventing the opponent from capitalizing on theirs.

The other numbers tell us, well, nothing new. I thought we might find out something interesting, but in the end, they just validate what we already know from readily available numbers like turnover margin and scoring margin.

Of course, I need a lot more data before I can verify the results of the Total Efficiency stat. My hope is to look at the entire AAC for 2017, and perhaps some other leagues as well, especially in the Power 6.

I also want to add a Special Teams component to this, although it's going to take quite a bit of work to synthesize field goals, punts, punt coverage, kickoffs and kickoff coverage together into one number that also fits that scale. But, it's worth a try.

If you've got any suggestions or recommendations on how to do that or what to emphasize, don't be afraid to reach out on Twitter at @Jeff_Sharon or @UCF_Banneret. Also, if you're a coach or scout, or someone who's good at statistical analyis and/or Excel, drop me a line and let me know what you think.

For now, we'll have to chew on these numbers and keep track of them closely once 2018 comes around.