Defense Wins Championships

Folks,

Over the years, I have hired numerous outside consultants and advisors to try to figure out a) why my team is so futile and b) how I can end the curse, just like the Cubs and Red Sox have.

Invariably, the consultants' opening briefing starts with them saying something like "we cannot stress enough" how important it is for my team to actually score more points than my opponent in the majority of weeks, and then for ALL weeks during the playoffs. This is always pretty unsatisfying news because it is sounds so elementary, and leaves me wondering what I am paying all that money for. But I will say that some of the consultants have provided some analysis that has been somewhat enlightening.

The first point, not exactly advanced stats but interesting, nonetheless, has been them helping me gain a better understanding of just how many points it takes to win one of these damn games. For the sake of this analysis, I'll focus mostly on the era dating to 2011, when we started to have 12 teams. (The fewer number of teams in the first year of the ESPN era, 2010, skews the numbers a bit and, also, 2010 is a season that I do not like to dwell on too much.) This has been an extremely consistent era for our league, as 11 owners have played all of the years. In addition, we have had the ~AirMex~ franchise, which is a little like the Philadelphia A's, turned Kansas City A's, turned Oakland A's, now owned by Matt Lindholm.

(I should note that this analysis, similar to the methodology our commissioner uses, adjusts scores by chopping off 3.5 points per game starting in 2016 to account for decimal scoring, which assumes that the decimals are worth about half a point per player.)

From 2011 until 2015, the average number of points per team per game, rounding to the nearest whole number, was stunningly consistent: 63, 63, 64, 63, and 63. In 2010, the average was 63, as well, which is a bit surprising because the fewer teams should have led to higher scoring. Things got wild in 2016 and it went all the way up to 65.  So, that is pretty remarkable consistency, and counsels in favor of figuring out a way to score about 64 points (adjusted) to have a decent chance of winning any given week. (The median score is what really matters but my consultants tell me that our median is about the same as our average.)

Now look at this. Average scoring in 2017 fell to 61.8 ppg. That was the lowest in the ESPN era by more than a full point. Even if the decimals are only worth 2.5 per game, it was still a record low.

Now, on to 2018. Let's call this "the Year of the Mahomes." Average scoring this year was ... 69 points per game! It's not entirely surprising that we broke a record given that scoring in the underlying league was way up, but this was an all-time record by almost 4 points a game when the ceiling and floor from the previous eight years differed by only 3 points a game.

Out of 96 team-seasons between 2011 and 2018 (12 teams x 8 years), here are some of the all-time rankings in average adjusted score among 2018 teams: AirMex, 1st all time; Ninjas, 2nd all time; Grahams (yes, Grahams) 8th all time; Atomonamtion (at 4-9) 13th all time; and Harbaughs, 23rd all time (and that's with starting a wide receiver or running back on injured reserve half the time!). An additional twist, AirMex and Ninjas not only scored the 1st and 2nd most all-time, but both exceeded the league average by more than any team had done in any previous year (next in line are the 2016 Ninjas, 2012 Air Apparent, 2017 Leviathans, 2013 Ninjas (again!) and 2017 Air Apparent.)

Now, as we begin the playoffs and I prepare for my team's inevitable demise, I will relay another interesting finding my consultants have reached. This relates to the fact that how many points you allow matters Exactly as much as how many points you score. We've all been vaguely aware that points allowed matters, as we have witnessed the phenomenon of Diablos slinking into the playoffs by mastering the art of 38 to 37 victories. But we probably thought that our opponents' scoring all balances out in the long run. It turns out that It Does Not. Some teams, for unexplained reasons, get a lot more points scored against them than others. Consistently.

Here are the total "Points Against" in the modern ESPN era. (actual points)

Air Apparent	7447.3
Team Harbaughs	7278.3
Merrill's Marauders	7276.6
Screaming Eagles	7227.9
Strike Team	7185.9
Psychotic Ninjas	7157.2
Los Diablos	7108.7
The Leviathans	7014.5
Golden Graham's	7013.1
Orion River Horses	6952.4
Suspended Atomnation	6927.9
~ AirMex ~	6793.7

As I suspected.

Just for the heck of it, I asked my consultant last year to run the numbers on who actually scored the most points over the years. Here are the partial results, in sort of a gold, silver, bronze scheme. (This is in actual points through 2017 only, as including 2018 sort of ruins the symmetry as compared to the chart above.)

Air Apparent	6601.7
Psychotic Ninjas	6454.1
Suspended Atomnation	6401.9

That’s more like it.

This Points Against thing turns out to matter even more than one might suspect. We all know that defense wins championships. Well, here's proof. It turns out that in every single year, our champion has given up fewer points than the league average. Like, Suspended is always up there, but it wasn't until 2016, when he didn't give up any points, that he finally broke through. Aaron is always bouncing from top to bottom (and two championships!). See a pattern below? Levi had one of the top scoring teams all time last year, but look what else -- he gave up only 58 and change a game. Since 2014, Maruaders' scoring ranks in the top 30 all time in every year but one. But he has given up top 30 points in every year since 2015. No 'ships.

The 2011 Horses only averaged 63 points a game, but that was only half the story. The 2016 Eagles scored about the same as the 11 Horses (adusted) but finished 3-10. Why? They gave up a Rich Rod-like 82 points a game! That's more than anyone ever scored per game until this gravity defying year, when Max and Ninjas eclipsed 84. The 2014 Diablos? Well, of course. And the 2010 Harbaughs? They were like the 2018 Wolverines before the Ohio State game.

Year	Champion		PA (actual)	League Ave PA (actual)
2010		Team Harbaughs	55.7	62.9
2011		Orion River Horses	60.6	63.2
2012		Strike Team	59.9	63.2
2013		Psychotic Ninjas	60.4	64.2
2014		Los Diablos	58.1	63.1
2015		Strike Team	62.9	63.1
2016		Atomnation	57.6	68.7
2017		Leviathans	58.8	65.3

What does all of this add up to as we head into the playoffs? Well, Merrill and Air Apparent have given up the most points of anyone in the playoffs, so they are toast.

Ninjas, Golden Grahams and Horses are all safely giving up less than the league average, so they are in contention.

But consider this: AirMex is giving up an adjusted 57.3 points a game. That is 11.6 points a game less than the league average. That is by far the biggest differential in the ESPN era.

That means, as compared to other teams, AirMex this year scored the all-time most and gave up the all-time fewest. Maybe he ate up all the points.

So I think we know our champion for this year. Not because of that white lightning offense but because of that Steel Curtain D.

--AA

All efforts were made to get the facts right in this post. Please direct complaints or fact-checking queries to my consultants.