*Steg's Week 1 Power Rankings

**iStegosauruz** · (This post was last modified: 03-30-2021, 02:35 PM by Tesla.)

Introduction

I’m pretty upset. I checked WolfieBot the other day and saw that I’ve fallen from 5th richest to 6th richest in the league. This just will not do. I pride myself on having a ridiculous amount of fake football league money - its literally the main thing I have going for myself. I’ve got to get my spot on the hierarchy back and secure it since more people are climbing up the ladder every day.

The best way for me to essentially print money is to find media I can consistently churn out. I’m an absolutely atrocious artist so graphics are out of the question. I’m not really into the role-playing aspect of the league so I can’t just write a ton of stories about what my character does. I needed to find something I could write about multiple times a week that falls into my niche of “stats that really make no sense but look cool and make me feel smart.”

Figuring out what I wanted to focus on took a couple of days. One of the things the league is lacking currently is a consistent set of power rankings. That being said, I absolutely hate power rankings normally. They’re subjective and I really don’t think they serve much of an overall purpose. To fix this issue I set out to try to determine if there was a way to create a metric - albeit one that is probably trash and won’t work properly - to power rank teams.

Meat and Potatoes

I really want to call this new metric the “Steg Score” or something of the sort but that feels really cringeworthy. For now I’ll just refer to it as the “PR Score” since its supposed to be a power ranking metric. Its based on four inputs that each factor an equal 25% into a team’s overall ranking.

The first input is a strength-of-victory calculation. Strength of victory takes a look at the amount of wins a team’s opponents have compared to the amount of games they’ve played. It can be used to weight how strong a win is or it can be used as a strength of schedule calculation. I chose to use it for the latter for now but might adjust later on.

The second input is a pythagorean win calculation. This is something I’ve written about extensive on this site but essentially pythagorean wins take a look at a team’s point differential and try to determine how many wins that team “should” have. I’ve also used a comparable calculation - linear wins - to accomplish the same thing. For this second input I’ve taken both of those calculations and averaged them to figure out the average amount of wins that the math feels a team should have (i.e. how much a team is over or underperforming relative to their offensive and defensive output).

The third input is a simple point differential metric. I’m not floored by using this but it’ll work for now until I figure out something else I’d like to use. Instead of running it through a logarithmic equation like pythagorean wins does I only care about the raw number.

The fourth input is an elo ranking. I know my friend Ithica Hawk has used elo extensively on the site and it was the basis for the casino. I talked to him and I know he has some interest in doing more elo based media at some point and I don’t want to steal his lane so I added a few of my own flares to my calculations and used some outlandish k-values to get some crazily volatile values. Then I’m plugging it all into this metric, so hopefully I’ve differentiated myself enough from him.

All four inputs - strength of victory, pythagorean/linear wins, point differential, and elo - are all normalized onto a 100 point scale. Its a relative scale, so a team will be higher or lower on it relative to the performance of the other teams in the league. Then they’re each given a 25% weight into the final equation which ends up looking like this:

RelativeSoV*.25 + RelativePythLinWins*.25 + RelativePD*.25 + RelativeElo*.25

That spits out a normalized “PR Score” which I then sort and use to rank the teams appropriately.

I’m considering not putting elo on a relative scale. I’ve got the calculations for what it looks like if I don’t and I’ve just got to decide if I prefer it more or not. It just feels wonky to me to normalize everything else onto a scale and not normalize it.

Pre-Season Rankings

There’s an issue with using the process I just laid out in the preseason - I don’t have strength-of-victory, pythagorean win, or point differential metrics to use for teams. The only thing I have is elo. So for the purposes of transparency I’m going to provide the initial elo ranking score for each team prior to the Week 1 sim.

1. Sarasota Sailfish SAR

- 1657.75

2. Chicago Butchers CHI

- 1610.38

3. New Orleans Second Line NOLA

- 1587.78

4. Honolulu Hahalua HON

- 1572.38

5. Arizona Outlaws ARI

- 1564.37

6. Colorado Yeti COL

- 1553.56

7. Yellowknife Wraiths YKW

- 1538.99

8. Orange County Otters OCO

- 1487.43

9. San Jose Sabercats SJS

- 1479.61

10. Philadelphia Liberty PHI

- 1454.42

11. Berlin Fire Salamanders - BER

- 1447.70

12. New York Silverbacks NYS

- 1420.87

13. Austin Copperheads AUS

- 1399.44

14. Baltimore Hawks BAL

- 1307.43

These feel a bit wonky. New Orleans is still riding off the high of making the Ultimus, Berlin feels a bit low, I’m higher on Austin than them being the second worst team in the league, etc. For the start of the season, however, this is what we have,

Week 1

Now that Week 1 has been played we finally have some statistics so I can actually use the formula I concocted. There is one caveat for the Week 1 rankings, however, which is that I don’t have enough data for a true strength-of-victory calculation so I chose to use the pre-season win estimates the casino put out to determine a team’s “strength of schedule.” That was put on the relative scale I’m normalizing everything to and then used in place of the strength-of-victory input.

Tier 1 - Really Damn Good (At Least in Week 1)

1. Yellowknife Wraiths YKW

Raw Score: 72.5

Relative Score: 100

The Wraiths came out of the gates HOT yesterday, blowing out the Yeti 47-10 at home. The calculation loves them right now and thinks they’re the best team in the league by a country mile. They were middle in the pack for pre-season elo ranking, however, which makes them storming up to #1 in the rankings now really interesting to me. The incredible amount of weight the calculation puts on scoring output and point differential must’ve really jazzed this number.

2. Honolulu Hahalua HON

Raw Score: 70

Relative Score: 96

The Hahalua knocked off the Outlaws yesterday, a team that the casino set the line on in the preseason at 11.5 wins. The Outlaws and Hahalua were both in the top 5 in the pre-season elo ranking which makes it easy for Honolulu to jump up a few spots and clock in at #2 behind the Wraiths. A matchup against the Second Line - a team the metric rates highly - gives the Hahalua a chance to climb up to the pinnacle of the rankings next week.

3. New Orleans Second Line NOLA

Raw Score: 68.75

Relative Score: 92

I don’t know if the casino and public was too low on the Second Line or if we just haven’t seen enough yet. This score is partially due to their high placement in pre-season elo ranking, however they also have a tough schedule based on early season casino rankings which bumps their score up a bit. Similarly to Honolulu, their matchup next week could determine if they can climb to #1 overall in these standings.

Tier 2 - Solid Teams, Nothing to Scoff At

4. Chicago Butchers CHI

Raw Score: 66.25

Relative Score 89

The Butchers get a huge boost in their score by knocking off the pre-season favorite Sailfish. Although they don’t have a great mark for strength of schedule, getting to play a tough opponent early pads their already high elo ranking which gives them a solid slot in the rankings after Week 1. A potential spoiler matchup against the Hawks in Week 2 awaits the Butchers which could secure them in the top 5 of the rankings early in the season or send them tumbling down them.

5. San Jose Sabercats SJS

Raw Score: 66

Relative Score: 88

The Sabercats are nipping (pun intended) on the heels of the Butchers in the Week 1 ranking. They have the toughest schedule in the league based on pre-season metrics and beat a team riding a wave of success from last season. They’ll get another Week 1 darling - the Silverbacks - in Week 2 which will help the model determine which of the teams, if not both, are for real this year.

6. New York Silverbacks NYS

Raw Score: 62.5

Relative Score: 82

This is a huge climb for the Silverbacks, up from 12th in the pre-season elo rankings to sixth on the power rankings. The climb is fueled by an above average scaled strength of schedule and a pretty solid pythagorean/linear win total as well. Although they only beat the Otters by 8 in Week 1 the metric really liked that performance in comparison to some of the other, closer wins that teams experienced. Their Week 2 matchup on the road against the Sabercats will be a huge determinant of whether this Silverbacks team is worthy of being in consideration as a dark horse Ultimus contender.

Tier 3: A Drop to the Middle

7. Philadelphia Liberty PHI

Raw Score: 52.75

Relative Score: 64

Philadelphia stormed up the rankings in a similar way that the Silverbacks did. Fueled by a really tough schedule based on pre-season projections the Liberty clock in as a middle of the pack team now but with a lot of potential. Their elo ranking internally leaves a bit to be desired but they have a great chance at raising that with a matchup against the Otters on the books in Week 2.

8. Arizona Outlaws ARI

Raw Score: 48.75

Relative Score: 56

The Outlaws were a team I was really high on coming into the season. The casino apparently felt the same way - setting the pre-season line at 11.5 wins for this Arizona team. As a result of being a “favorite” they had a much easier schedule than almost any team in the ASFC which is a huge knock on them early and a reason they’re this far down the rankings after Week 1. I expect that should normalize if they perform up to reasonable expectations over the course of the season. They get a chance to correct course and move up the board with a matchup at home against the Copperheads in Week 2.

9. Berlin Fire Salamanders BER

Raw Score: 48.5

Relative Score: 56

The Salamanders are just a hair below the Outlaws in these rankings. Although they have the same raw score I gave the edge to the Outlaws because of the slightly higher raw score. The Fire Salamanders clocked in at 11th in the pre-season elo rankings and although they won their game against the Liberty it was close and the Liberty were only one spot ahead of them in those pre-season rankings. With an easier schedule on the books compared to most of the league (based on pre-season casino rankings) the Fire Salamanders didn’t get much of a boost after Week 1, however if they can bring home a win against the Wraiths in Week 2 they’ll shoot up this board expoenentially.

10. Baltimore Hawks BAL

Raw Score: 47

Relative Score: 53

I mean no slight to the Hawks when I say I’m kinda shocked to see them this high up the board this early, but its Week 1 and the rankings are still in flux. They have the second toughest schedule in the league based on pre-season projections which means that they get a lot of credit for keeping it close against the Copperheads Coming off a down season in Season 27 they had a much easier route to raising their elo score with a good matchup as well. Although they’re in the middle of a rebuild I think the Hawks showed us enough in Week 1 to convince me to drop a big bet on their O/U win total in the casino. They look like a solid team, even if I do think they’ll drop a bit in these rankings over the course of the season. We’ll get to see them in action against the Butchers in Week 2, providing them another chance to showcase their talent to the league and potentially sneak a win against a pre-season favorite.

11. TIE - Austin Copperheads AUS

and Sarasota Sailfish SAR

Raw Score: 44.5

Relative Score: 48.5

We have a tie at 11th in the rankings this week. The Copperheads are coming off a close overtime win against the Hawks in Week 1. They were 13th in pre-season elo rankings after their down year in Season 27 but were still a significant bit higher in terms of the overall elo number than the Hawks, meaning that sneaking a win doesn’t help them much in terms of projections for the season thus far. A tough matchup on the road against the Outlaws in Week 2 awaits Austin.

The Sailfish get knocked for having by far the easiest schedule in the league in terms of pre-season projections. Losing by 7 - which isn’t a ton bus is still the third biggest deficit in the league from Week 1 - also hurts their overall ranking. I fully expect them to revert more towards their mean at minimum over the course of the season, but at the conclusion of Week 1 they’re in a rough spot. Their Week 2 matchup against the Yeti in Week 2 will help determine which of those two teams gets to surge up the scoreboard toward where we expected them to be to start the year.

Tier 4: Not Calculation Errors, Just Way Too Early

13. Orange County Otters OCO

Raw Score: 38.5

Relative Score: 37

The Otters had a rough showing in Season 27 and started off Season 28 with a loss to the Silverbacks. Having the second worst loss in the league during Week 1 - although it was only 8 points - does pretty drastically impact the Otters score which is already a bit lower than some might expect because of their pre-season elo ranking. A matchup on the road against the Liberty in Week 2 gives the Otters a chance to bring their scores in line with where pre-season pundits expected them to be.

14. Colorado Yeti COL

Raw Score: 19

Relative Score: 1

This is the point in the rankings where I know I’m going to have to tack a disclaimer on this entire post. The Yeti aren’t a bad team. Over the course of the year I fully expect them to rise up these rankings. Week 1 was disastrous for them, however. Losing by 37 is never easy to handle, especially for a model that uses points scored, points allowed, and point differential in almost all its calculations. The Yeti also had a fairly easy schedule based on pre-season projections which brings their score even lower. A Week 2 matchup against the Sailfish provides them the chance to shoot up the board barring another disastrous bit of sim luck.

Overall Disclaimer

If you read this post and went “wow my team is so low. Steg you suck and are wrong” this is for you. These rankings will DEFINITELY change DRASTICALLY. We only have one week of data in. I just want to chart all of this over the course of the season so we can see how it changes and I can refine things for next season and all the eventual other ones I might want to continue this project during. So take this all with a grain of salt and wait until Week 2 before you really grab a plank and start trying to enact some vigilante justice. I’m just tryna make them fat stacks man.

Barring some big disaster in my life Week 2 results should be out Thursday. I’m also working on my HOF metric thing from last year now which is taking some time. For those of you have asked me where that it is - its coming, with lots of fixes.