Did Expansion Dilute BBA Defensive Skills

[RonCo]

We often hear people talk about expansion causing rifts in a league’s statistics, and it’s true they can. Mostly, though, we think about hitting and pitching. As I was messing again with the script I use to capture defensive performances, it struck me that I could look at how league-wide defense has moved as a result of our latest expansion of two teams, which occurred in year 2040.

So I went back and grabbed the end of year reports for every year since 2035 (though for whatever reason, I’m missing 2036), and tabulated overall defensive Plays Above Average for every team in the league during those years, and then looked at raw worst and best, the gaps between those, and then standard deviations. I should note that I included 2043 for reference, even though 2043 is only two-thirds finished.

It’s interesting, though not massively conclusive.

Bottom line: it seems that if there’s been a dilution of defensive talents brought on by expansion, that dilution has been minimal. At least by this measure…

First, here’s the data (note the last few lines are min/max/gap/STDEV):

TEAM	2035	2037	2038	2039	2040	2041	2042	2043
Altantic City Gamblers	-11.81	-58.03	-52.73	18.50	8.19	-73.61	19.43	21.51
Boise Spuds	-55.45	-12.90	53.98	36.76	9.99	48.86	69.51	46.82
Brooklyn Robins	13.76	58.38	11.29	26.84	-28.15	-55.05	39.02	32.44
Calgary Pioneers	19.53	25.78	9.85	-38.88	19.25	15.28	-20.65	-13.99
Charlotte					-2.46	56.09	12.15	13.63
Charm City Jimmies	27.42	3.53	29.42	50.90	-21.27	3.83	16.44	3.95
Chicago Black Sox	39.49	-31.34	-19.01	-0.19	99.06	36.48	37.20	34.19
Des Moines Kernels	89.58	21.84	-1.94	-6.79	6.69	-68.28	-54.21	-85.03
Edmonton Jackrabbits	16.11	39.28	13.14	-53.35	54.38	-14.72	75.44	29.89
Hawaii Tropics	-13.22	6.63	-37.64	56.29	-19.37	-12.52	-55.74	-11.55
Jacksonville Hurricanes	-14.64	6.54	59.24	25.03	-24.96	-56.54	-87.72	24.61
Las Vegas Hustlers	-11.93	27.54	-61.75	33.90	57.70	37.04	23.90	-7.26
Long Beach Surfers	12.92	21.65	53.20	-6.06	-53.25	-69.69	-69.77	-62.07
Louisville Sluggers	11.37	12.58	50.29	-8.58	0.91	83.91	67.29	4.71
Madison Wolves	-15.11	-76.94	-55.33	-105.61	-76.12	-42.56	-43.02	-13.32
Mexico City Aztecs	-2.84	64.18	72.01	65.72	25.79	61.28	-21.29	-13.33
Montreal Blazers	-8.48	-118.27	6.24	-28.65	25.03	-4.17	28.35	15.23
Nashville Goats	9.49	12.90	6.07	68.36	4.17	4.18	32.82	-25.26
New Orleans Crawdads	-57.47	2.03	-93.46	-37.67	21.58	111.60	-7.39	39.02
Omaha Cyclones	-46.68	6.29	-9.86	-28.65	53.53	31.70	45.32	-3.58
Phoenix Talons	46.31	51.61	41.43	24.82	3.45	-33.40	-17.19	-2.83
Portland					-15.57	63.10	-31.63	14.33
Rockville Pikemen	18.36	50.96	57.24	44.66	-34.05	-47.82	-44.40	3.39
Sacramento Mad Popes	64.46	38.70	8.33	4.91	25.87	48.57	26.55	75.89
San Antonio Outlaws	15.42	69.93	11.32	-50.44	-37.01	-47.14	44.76	19.25
San Fernando Bears	-15.70	-69.75	-14.88	-3.58	-61.55	-43.82	-25.94	-10.37
Seattle Storm	-11.46	3.08	-12.06	-8.69	-9.19	-65.79	-58.83	-80.03
Twin Cities River Monsters	12.75	26.97	-96.63	-45.94	-75.48	-24.34	-15.23	-16.81
Valencia Stars	-63.33	-42.99	-25.46	-48.98	9.79	-12.33	-38.27	-50.60
Vancouver Mounties	-12.28	34.47	-6.72	-55.42	-8.09	62.44	-21.45	-15.33
Wichita Aviators	-43.81	-119.59	4.67	23.77	10.86	11.66	21.95	-4.82
Yellow Springs Nine	-12.73	-55.05	-0.26	47.02	30.27	-4.26	52.61	37.29
League	0.00	0.00	0.00	0.00	0.00	0.00	0.00	0.00
Max	89.58	69.93	72.01	68.36	99.06	111.60	75.44	75.89
Min	-63.33	-119.59	-96.63	-105.61	-76.12	-73.61	-87.72	-85.03
Gap	152.91	189.52	168.64	173.97	175.18	185.21	163.16	160.93
STDEV	35.15	50.17	42.81	43.48	39.31	49.51	44.64	35.45

Now let’s take a dive in.

To understand the data, let’s look at 2035. That season, the best defense in the league was in Des Moines, who registered 89.58 Plays Above Average. The worst was Valencia, at -63.33. The meant the gap from best to worst was about 153 plays over the season, or just under 1 per game. As a league, we played to a standard deviation of 35.15, meaning that about 70% of the league probably fell into a range between 35 and -35 PAA.

So that’s how to read the columns.


BEST AND WORST DEFENSES

Scanning across the list we see that the two best defenses (Chicago 2040, and New Orleans 2041) came immediately after our expansion. 2043, however, dropped down to Edmonton’s 75 PAA, which is still better than two of the three seasons prior to expansion, but nut massively.

I’d say you could argue that after expansion, the “best: defensive teams have been better relative to average than the best defensive teams before. In fact, the best defensive teams in years prior to expansion averaged 75 PAA, whereas the three afterward have averaged 95. With Sacramento already at 75 PAA in late July, 2043 appears to be on that same track. So, there’s more to the equation.

On the flipside, though, the worst teams prior to expansion averaged -96 PAA, whereas the worst since expansion happened have “improved” to -79 (a 17 PAA improvement).


THE GAP BETWEEN

Scanning the gap between best and worst, we find the greatest gap between best and worst happened in 2037 (when San Antonio’s 69.39 lead the league, but Wichita posted a buttalicious -119.59 PAA, the worst in recorded history). Over the years, the span between best and worst has been moderately consistent (153, 189, 169, 173, 175, 185, 163). 2043 appears to be on pace to be on the high end of the spread, but nothing far out of line.

• Average gap prior to expansion: 171
• Average gap after expansion: 174.5

By raw numbers you’d say expansion did move the gap a little, but not more than one or two percent at best on a raw PAA basis.


STANDARD DEVIATION

This is an interesting little piece of data. The bottom line is that 2040—the expansion year itself—saw the tightest Standard Deviation of all other than 2035. So it suggest defense—at least in the most general ranges, was more equal (?) across the league than other years. But then cane 2041, and the standard deviations have grown to among the greatest.

Hmm.

Math, right?

If you trendline Standard Deviation, it does show a (again gentle) expansion between the defensive “haves” and the defensive “have nots.”


WHAT DOES IT MEAN?

Well, I’d say this data does suggest a relative warping of defensive skills being fielded by teams before and after expansion.

The key word, though, is “relative.” I realized that about half way through this. This study measures defense across the league relative to the league members, but does not measure raw defensive skill in the league itself. So, yeah. Expansion probably did make a difference, but I'm not sure this little fiddling around with data really "proves" it.

Looks like I’ve got a new exercise, eh?

[HoosierVic]

That 2040 Chicago defense was due in large part to the start of the Jim Lorenzen experiment in CF (he was a 12-rated fielder at the time, which I can hardly even visualize). Manny Cleide was in RF (a 10-rated guy) and the infield was stacked with good fielders, too.

There was just one teeny, tiny problem: none of 'em (especially Lorenzen) could hit worth a damn.

And so, the team pretty much sucked (10 games under .500). They looked great in the field, though ...

[jiminyhopkins]

Great! Now do: Was there a global change made to depress offense in order to blunt the impact of the new crop of hitting superstars from the uber-drafts of the late 2030's??

[JimBob2232]

i dont think i realized just how good my defernse was in 2041.