*NSFL Division Parity Analysis

bovovovo · (This post was last modified: 08-04-2018, 07:11 PM by bovovovo.)

There's usually some debate over parity in the league... including which conference has better parity, the NSFC or the ASFC?

There's some different ways we can break that down, so let's start with the easiest way.

This first list/table thing (I really wish jcink had a table function) shows the winner and last place teams each year by conference.

*For this first part I am excluding Season 1 because it was pre-expansion teams which skews things

NSFC WINNERS // LOSERS
S2: BAL

//

S3:

//

S4:

//

S5:

//

S6:

//

S7:

//

S8:

//

ASFC WINNERS // LOSERS
S2: ARI

//

(Legion)
S3: ARI

//

(Legion)
S4: ARI

//

(Legion)
S5: OCO

//

(Legion)
S6: OCO

//

S7:

//

S8:

//

So right off the bat we can quickly see a few things.

~The Yeti and NOLA/Legion were terrible for a very long time
~In the NSFC there's been a few different winners that flip back and forth, but the ASFC has two pretty clear periods of one team always coming out on top

That's not really the entire story though, so now we're going to look at where each team actually placed each season. Then we're going to average them all together so we can see each team's average end-of-year ranking (the lower the better).

[Image: unknown.png]

So for example, the Hawks have an average conference rank of 2.14, and the Wraiths at 1.57.

We can start to see there's more of a "split" within each conference... the NSFC has three teams that, on average, finish in at least 2nd place and one team that averages coming in 4th. The ASFC has two teams that average coming in 1st or 2nd and the other two teams have each only ever done better than 3rd place one time.

A way to quantify the amount of parity in each conference is to find the average of the variation between the conference averages. This lets us see the average "distance" between every team in each conference.

For my excel peeps / people who want to see the equation it was =SUM(I11-I10,I11-I8,I11-I9,I10-I9,I10-I8,I8-I9)/4

Below is the "parity score" for each conference. The lower the number is, the more parity there is. (Because there is less variation between each team, on average)

NSFC: 1.79
ASFC: 1.64

So at first glance it looks like the ASFC enjoys more parity than the NSFC does. This is actually misleading I think... The problem is that both conferences have a team that is/was a perpetual bottom feeder. In the NSFC that's the Yeti and in the ASFC it's NOLA/the Legion. These teams skew the data by making each conference look like there's less parity than there really is, and neither team (with the exception of NOLA in S8) was ever actually seriously competitive.

So we're going to remove these teams completely from the equation and do the same exact thing looking only at the remaining three teams.

NSFC: 0.57
ASFC: 1.05

That's better. It also kind of tells us partly what we already know... for a good portion of history the NSFC has been a battle between the Hawks, Wraiths, and Liberty, while the ASFC typically has been split between just the Otters and Outlaws for a long time. NOLA and the Sabercats have only ever made the playoffs once since S2. This shows that the NSFC has more parity as far as end-of-season rankings are concerned.

But it goes deeper than that. We can look at end-of-season rankings, and it's still useful, but it doesn't paint the entire picture. There have been seasons where a 5-9 team made the playoffs, and where three teams in one conference all had the same record at 8-6 but only two made the playoffs.

The way we can account for these discrepancies in conference competitiveness is by looking at each team's historical win % against teams within their own conference. We only want to look at inner-conference win %'s because those are the teams that you are directly competing against in order to make the playoffs.

The below chart shows each team's historical win % against other teams within their conference, and the rightmost column shows the same thing except removing Yeti and NOLA from the equation.

[Image: unknown.png]

There's actually some cool little conclusions we can get from this.

~The Outlaws still have the highest Division Win % at 70%, thanks to their previous dynasty
~The Liberty really beat up on the Yeti. Their Division Win % drops by 14% when you take away all of their Yeti games.
~On the flip side, the Wraiths Division Win % only drops by 5% when you take away their Yeti games
~NOLA has the lowest Division Win %, only winning 21% of their inner-division games
~The Sabercats have a very low Division Win % despite beating up on the Legion/NOLA, indicating that historically they were pretty far behind the Otters and Outlaws.

So now we do the same thing we did before, where we average together all of the variances between the Division Win %'s to get a Parity Score, both with and without the Yeti/NOLA

Parity Scores
NSFC: 0.33
ASFC: 0.45

Parity Scores Without Yeti/NOLA
NSFC: 0.16
ASFC: 0.26

So there you have it. Quantitatively showing what most of us know anyways: The NSFC has historically had more parity than the ASFC.

Another method you could do is look at the variance of end-of-season rankings / records on a season-by-season basis but that's a lot of work for another time. Maybe one day. I think this is a pretty accurate picture nonetheless.

It'll be cool to revisit this in the future a few seasons from now now that NOLA and Yeti aren't perpetual dumpster fires anymore.