07-09-2020, 03:23 PM
(This post was last modified: 07-10-2020, 12:37 PM by Fordhammer.)
A few weeks ago, I did a deep dive into if week one really matters in the NSFL and DSFL. It ended with the conclusion that statistically week one is a large predictor of who may win the Ultimus in particular, but that these stats are made to be broken just as all predictive statistics in sports. The biggest, best, and most memorable moments in sports happen when the statistics that are assumed to be unbreakable are, in fact, broken – I’m looking at you, 21-3.
This morning I began to wonder how similar our simulation league is to the National Football League in the correlation of week one success to participation and/or victory in the championship game. Obviously simulation football runs on algorithms, rules, and is very different from the NFL, but I wondered how closely our competitive results might correlate in this particular realm. Thus began a deep dive into the history of Super Bowl winners and losers with a similar mapping of their initial winning or losing streak. As with the NSFL article, I have listed the Super Bowl winner and loser, the final score of that game (just for fun), and the initial streak that each team went on to start the season. If they won each game until a week 5 loss, their streak is (4-0). If they lost weeks 1 and 2 but won week 3, their streak is (0-2). Ties do not end streaks as they are silly and irrelevant in this writer’s opinion. I also included notations of when new teams entered the league, increasing the spread of talent and size of the league to see if that played any significant statistical role.
Super Bowl 1 – The Green Bay Packers (4-0) beat the Kansas City Chiefs (3-0). Final Score: 35-10
Super Bowl 2 – The Green Bay Packers (3-0-1) beat the Oakland Raiders (3-0). Final Score: 33-14
Super Bowl 3 – The New York Jets (2-0) beat the Baltimore Colts (5-0). Final Score: 16-7
Super Bowl 4 – The Kansas City Chiefs (2-0) beat the Minnesota Vikings (0-1). Final Score: 23-7
NFL MERGER
Super Bowl 5 – The Baltimore Colts (1-0) beat the Dallas Cowboys (2-0). Final Score: 16-13
Super Bowl 6 – The Dallas Cowboys (2-0) beat the Miami Dolphins (1-0-1). Final Score: 24-3
Super Bowl 7 – The Miami Dolphins (17-0) beat the Washington Redskins (2-0). Final Score: 14-7
Super Bowl 8 – The Miami Dolphins (1-0) beat the Minnesota Vikings (9-0). Final Score: 24-7
THE SEATTLE SEAHAWKS AND TAMPA BAY BUCCANEERS ENTER THE NFL
Super Bowl 9 – The Pittsburgh Steelers (1-0-1) beat the Minnesota Vikings (5-0). Final Score: 16-6
Super Bowl 10 – The Pittsburgh Steelers (1-0) beat the Dallas Cowboys (4-0). Final Score: 21-17
Super Bowl 11 – The Oakland Raiders (3-0) beat the Minnesota Vikings (6-0-1). Final Score: 32-14
Super Bowl 12 – The Dallas Cowboys (8-0) beat the Denver Broncos (6-0). Final Score: 27-10
Super Bowl 13 – The Pittsburgh Steelers (7-0) beat the Dallas Cowboys (2-0). Final Score 35-31
Super Bowl 14 – The Pittsburgh Steelers (4-0) beat the Los Angeles Rams (0-1). Final Score: 31-19
Super Bowl 15 – The Oakland Raiders (1-0) beat the Philadelphia Eagles (3-0). Final Score: 27-10
Super Bowl 16 – The San Francisco 49ers (0-1) beat the Cincinnati Bengals (2-0). Final Score: 26-
21
Super Bowl 17 – The Washington Redskins (4-0) beat the Miami Dolphins (3-0). Final Score: 27-17
Super Bowl 18 – The Los Angeles Raiders (4-0) beat the Washington Redskins (0-1). Final Score:
38-9
Super Bowl 19 – The San Francisco 49ers (6-0) beat the Miami Dolphins (11-0). Final Score: 38-16
Super Bowl 20 – The Chicago Bears (12-0) beat the New England Patriots (1-0). Final Score: 46-10
Super Bowl 21 – The New York Giants (0-1) beat the Denver Broncos (6-0). Final Score: 39-20
Super Bowl 22 – The Washington Redskins (1-0) beat the Denver Broncos (1-0-1). Final Score: 42-
10
Super Bowl 23 – The San Francisco 49ers (2-0) beat the Cincinnati Bengals (6-0). Final Score: 20-
16
Super Bowl 24 – The San Francisco 49ers (3-0) beat the Denver Broncos (3-0). Final Score: 55-10
Super Bowl 25 – The New York Giants (10-0) beat the Buffalo Bills (1-0). Final Score: 20-19
Super Bowl 26 – The Washington Redskins (11-0) beat the Buffalo Bills (5-0). Final Score: 37-24
Super Bowl 27 – The Dallas Cowboys (3-0) beat the Buffalo Bills (4-0). Final Score: 52-17
Super Bowl 28 – The Dallas Cowboys (0-2) beat the Buffalo Bills (2-0). Final Score: 30-13
Super Bowl 29 – The San Francisco 49ers (1-0) beat the San Diego Chargers (6-0). Final Score: 49-
26
THE CAROLINA PANTHERS AND JACKSONVILLE JAGUARS ENTER THE NFL
Super Bowl 30 – The Dallas Cowboys (4-0) beat the Pittsburgh Steelers (2-0). Final Score: 27-17
Super Bowl 31 – The Green Bay Packers (3-0) beat the New England Patriots (0-2). Final Score: 35-
21
THE BALTIMORE RAVENS ENTER THE NFL
Super Bowl 32 – The Denver Broncos (6-0) beat the Green Bay Packers (1-0). Final Score: 31-24
Super Bowl 33 – The Denver Broncos (13-0) beat the Atlanta Falcons (2-0). Final Score: 34-19
Super Bowl 34 – The St. Louis Rams (6-0) beat the Tennessee Titans (3-0). Final Score: 23-16
Super Bowl 35 – The Baltimore Ravens (2-0) beat the New York Giants (3-0). Final Score: 34-7
Super Bowl 36 – The New England Patriots (0-2) beat the St. Louis Rams (6-0). Final Score: 20-17
THE HOUSTON TEXANS ENTER THE NFL
Super Bowl 37 – The Tampa Bay Buccaneers (0-1) beat the Oakland Raiders (4-0). Final Score: 48-
21
Super Bowl 38 – The New England Patriots (0-1) beat the Carolina Panthers (5-0). Final Score: 32-
29
Super Bowl 39 – The New England Patriots (6-0) beat the Philadelphia Eagles (7-0). Final Score: 24-
21
Super Bowl 40 – The Pittsburgh Steelers (2-0) beat the Seattle Seahawks (0-1). Final Score: 21-10
Super Bowl 41 – The Indianapolis Colts (9-0) beat the Chicago Bears (7-0). Final Score: 29-17
Super Bowl 42 – The New York Giants (0-2) beat the New England Patriots (18-0). Final Score: 17-
14
Super Bowl 43 – The Pittsburgh Steelers (2-0) beat the Arizona Cardinals (2-0). Final Score: 27-23
Super Bowl 44 – The New Orleans Saints (13-0) beat the Indianapolis Colts (14-0). Final Score: 31-
17
Super Bowl 45 – The Green Bay Packers (2-0) beat the Pittsburgh Steelers (3-0). Final Score: 31-25
Super Bowl 46 – The New York Giants (0-1) beat the New England Patriots (2-0). Final Score: 21-17
Super Bowl 47 – The Baltimore Ravens (1-0) beat the San Francisco 49ers (2-0). Final Score: 34-31
Super Bowl 48 – The Seattle Seahawks (4-0) beat the Denver Broncos (6-0). Final Score: 43-8
Super Bowl 49 – The New England Patriots (0-1) beat the Seattle Seahawks (1-0). Final Score: 28-
24
Super Bowl 50 – The Denver Broncos (7-0) beat the Carolina Panthers (14-0). Final Score: 24-10
Super Bowl 51 – The New England Patriots (3-0) beat the Atlanta Falcons (0-1). Final Score: 34-28
Super Bowl 52 – The Philadelphia Eagles (1-0) beat the New England Patriots (0-1). Final Score: 41-
33
Super Bowl 53 – The New England Patriots (1-0) beat the Los Angeles Rams (8-0). Final Score: 13-3
Super Bowl 54 – The Kansas City Chiefs (4-0) beat the San Francisco 49ers (8-0). Final Score 31-20
So there you have it. Big shoutout to pro-football-reference.com because this would have been nearly impossible without it. If you’re a football nerd like me, that website is truly fascinating to help break down your favorite seasons, players, or just learn about the history of the NFL. Now back to the topic at hand.
The first thing to do in compiling this data was to look at the week 1 records of our collective Super
Bowl teams, just as we did with the NSFL and DSFL previously. You’ll find this below:
Super Bowl Winners’ Collective Week 1 Records: 45-9 – (83.3%)
Super Bowl Losers’ Collective Week 1 Records: 47-7 – (87%)
Super Bowl Participants’ Collective Week 1 Records: 92-16 – (85.2%)
Ultimus Winners’ Collective Week 1 Records: 18-4 – (82%)
Ultimus Losers’ Collective Week 1 Records: 12-10 – (55%)
Ultimus Participants’ Collective Week 1 Records: 30-14 – (68%)
Ultimini Winners’ Collective Week 1 Records: 14-6 – (70%)
Ultimini Losers’ Collective Week 1 Records: 12-8 – (60%)
Ultimini Participants’ Collective Week 1 Records: 26-14 – (65%)
Right off the bat a few things jumped out at me in the above statistics. First, the overall correlation of Super Bowl Winners’ Collective Week 1 Records and Ultimus Winners’ Collective Week 1 Records is remarkably similar – within 2%. However, the overall correlation of these statistics is honestly not very close at all. It is about 17% more likely that a team that makes the Ultimus lost in week 1 than the teams that made the Super bowl, and about 20% more likely that a team that makes the Ultimini lost in week 1! For the comparison to be approximately 1/5 off is a significant statistical gap for sure.
However, I did notice in setting this up that the early years of the NFL seemed far more predictable start to finish than the later years have been in a similar fashion to how, as the NSFL has expanded, the winners have been less extreme favorites and the underdogs have taken home a significantly greater number of victories. I wondered if the same might be the case for the NFL. Therefore, I also compiled the records for “modern” Super Bowl participants, in which I included the last 18 seasons since the last team (the Houston Texans) was added to the NFL. The results are below, once more compared to the Ultimus and Ultimini results.
Since all teams joined the NFL (Last 18 seasons)
Super Bowl Winners’ Collective Week 1 Records: 13-5 – (72.2%)
Super Bowl Losers’ Collective Week 1 Records: 15-3 – (83.3%)
Super Bowl Participants’ Collective Week 1 Records: 28-8 – (77.8%)
Ultimus Winners’ Collective Week 1 Records: 18-4 – (82%)
Ultimus Losers’ Collective Week 1 Records: 12-10 – (55%)
Ultimus Participants’ Collective Week 1 Records: 30-14 – (68%)
Ultimini Winners’ Collective Week 1 Records: 14-6 – (70%)
Ultimini Losers’ Collective Week 1 Records: 12-8 – (60%)
Ultimini Participants’ Collective Week 1 Records: 26-14 – (65%)
I expected there to be a change in percentage here, maybe something like 5%, but I was amazed to see the level of difference that the modern NFL has seen! The odds of a team that loses week one winning the super bowl increased by over 10%! And the odds to make the super bowl after a week one loss increased by nearly 8%. When you look at these modern NFL odds to win or make the Super Bowl based on week 1 results, they are significantly more similar to the NSFL and DSFL odds to win or make their respective championship games. Overall, the odds of a conference champion having lost week 1 are right around 10% or closer to our simulations as they are in the modern NFL.
The major statistical difference that I noticed is that no large streak is a guarantee in the NFL. The 18-0 Patriots stand as the mammoth example of this truth, but we also have seen the 9-0 Vikings, 11-0 Dolphins, 7-0 Eagles and Bears, 14-0 Colts and Panthers, and the last two runners up – the 8-0 Rams and 49ers – all lose in the Super Bowl. In the DSFL no team that has started better than 5-0 and made the Ultimini has lost and in the NSFL the only time a team went 6-0 or better and lost in the Ultimus was the Hawks who lost to the Outlaws after they started 11-0. There is an assumed guarantee that if a team that started the season really hot makes the championship, they will win it, and this simply is not the case in the NFL. All of these hot starting teams have actually swayed the winning streaks in many ways to reflect higher on the NFL runner up than on the Super Bowl Champion. For some statistical examples of this, see below:
Mean (average) Initial Streak for Super Bowl Champions: 3.54 wins
Mean (average) Initial Streak for Super Bowl Losers: 3.93 wins
Mean (average) Initial Streak for Super Bowl Participants: 3.73 wins
Mean (average) Initial Streak for Ultimus Champions: 3.32 wins
Mean (average) Initial Streak for Ultimus Losers: 1.23 wins
Mean (average) Initial Streak for Ultimus Participants: 2.27 wins
Mean (average) Initial Streak for Ultimini Champions: 1.35 wins
Mean (average) Initial Streak for Ultimini Losers: 0.95 wins
Mean (average) Initial Streak for Ultimini Participants: 1.15 wins
First I took the mean or average of the initial records of the winners, losers, and participants of the Super Bowl, Ultimus, and Ultimini. In these statistics, it is clear that the DSFL in particular is not conducive to having large streaks, as it is the 18-0’s, 14-0’s and other large starts that really bump the Super Bowl means in addition to many 6-0 and a couple even larger starts for NSFL teams that help to keep them on that same kind of pace. The DSFL has seen only a handful of 4-0 or greater starts in Ultimini matchups and so their mean is much lower than that of the other leagues. However, it is confirming in my opinion that our Ultimus champions are on the same kind of trajectory as Super Bowl champions in this category, as it reflects well on the realism of our simulations. Clearly there are still less hot starts in our simulations than in the NFL, but to still be fairly close in this category is a win for our simulations.
Median (middle) Initial Streak for Super Bowl Champions: 3-0
Median (middle) Initial Streak for Super Bowl Losers: 3-0
Median (middle) Initial Streak for Super Bowl Participants: 3-0
Median (middle) Initial Streak for Ultimus Champions: 3-0
Median (middle) Initial Streak for Ultimus Losers: 1-0
Median (middle) Initial Streak for Ultimus Participants: 2-0
Median (middle) Initial Streak for Ultimini Champions: 2-0
Median (middle) Initial Streak for Ultimini Losers: 1-0
Median (middle) Initial Streak for Ultimini Participants: 1-0
Next I looked at the Median (middle) results of these same data points in order to reduce the effect of the outliers (0-4 or 18-0 does a lot to stretch out a mean). With this data, we see similarity once again in the Ultimus and Super Bowl champions, as the middle result for each category was a 3-0 start. I think this is right about where it should be. I was surprised by how low the Ultimus losers fell here to 1-0, but in looking at the number of teams that have lost the Ultimus after losing week 1, this makes a lot of sense as well. The DSFL again looks like a minor league here, and that’s okay! It’s not intended that a team run away with this league each year, but that there be even competition across the board. In my opinion, this is right where it should be hovering around 1-2 wins to start the year being enough to start on the way to a championship appearance.
Mode (most common) Initial Streak for Super Bowl Champions: 1-0 (ten times)
Mode (most common) Initial Streak for Super Bowl Losers: 2-0 (ten times)
Mode (most common) Initial Streak for Super Bowl Participants: 2-0 (eighteen times)
Mode (most common) Initial Streak for Ultimus Champions: 1-0 & 6-0 (four times each)
Mode (most common) Initial Streak for Ultimus Losers: 0-1 (ten times)
Mode (most common) Initial Streak for Ultimus Participants: 0-1 (thirteen times)
Mode (most common) Initial Streak for Ultimini Champions: 2-0 (six times)
Mode (most common) Initial Streak for Ultimini Losers: 0-1 (six times)
Mode (most common) Initial Streak for Ultimini Participants: 2-0 & 0-1 (nine times each)
Finally, I looked at the mode, or most common result for each of these categories (anybody else getting 6th grade math flashbacks?). This simply tells us what the most common start has been for participants and champions in the Super Bowl, Ultimus, and Ultimini. The most surprising thing in this data for me was seeing that in both the Ultimus and the Ultimini an 0-1 start was the most (or tied for the most) common record for participants. While it’s certainly more common for a team to have won than to have lost week one, the amount of teams who bounced back from a week 1 loss to have a winning season really impacted these statistics. In the NFL, it is much more common to have that first win (or first couple of wins) under your belt right away than any other result. However, those hot streaks are not particularly common, as most teams stopped at 2-0 before meeting an early season loss. There is, after all, a lot to learn in putting a team together each year – whether it’s a simulation or in person.
This was a LOT of data, I’m well aware, and I’m a little light headed just looking at it all. But my overall conclusion is that, while our simulation is not extremely close to the NFL in these particular data points, the overall trends are close enough to be realistic. When a team starts off hot, they are more likely to win. When a team loses an initial game or two, they are by no means damned, but better put on their hustle pants and change something quickly before their chances are too far gone. I personally know nearly nothing about putting together a simulation, but in this very limited evaluation, I’d say we are in good hands. Sure, our sims are quirky as anything, and sometimes a certain stat or result seems really strange, but the meat of the program is good. The spirit of competitive football is being respected here and that says a lot about the people who are running this league.
One more fun fact that I discovered before I finish this piece: First time organizations in the Super Bowl are 9-19, with most of those 9 wins coming against other first time Super Bowl participants, so if your first trip to the Ultimus or Ultimini doesn’t go well, know that you are in good company. There is a LOT to be said for experience, from the top of an organization all the way to the bottom. So keep on simming, keep on training, and keep on learning so you can lift that trophy one day! I know I for one am preparing my pose for some future banners…
Please keep this discussion going in the comments! I want to hear your thoughts about the importance of starting off well – either in our sims or in the NFL if you’re the fan of a particular organization. Recount a season, game, or story for us all to get to relive it with you! Also, if there’s another piece you’re wanting done but you don’t want to weed through a ton of data to do it, that’s something I really enjoy doing if you can’t tell, so comment below or send me a message – I’m always looking for ideas to write about and maybe we can team up on something if you want to help. And good luck in the second half of the season all you crazy football fans!
This morning I began to wonder how similar our simulation league is to the National Football League in the correlation of week one success to participation and/or victory in the championship game. Obviously simulation football runs on algorithms, rules, and is very different from the NFL, but I wondered how closely our competitive results might correlate in this particular realm. Thus began a deep dive into the history of Super Bowl winners and losers with a similar mapping of their initial winning or losing streak. As with the NSFL article, I have listed the Super Bowl winner and loser, the final score of that game (just for fun), and the initial streak that each team went on to start the season. If they won each game until a week 5 loss, their streak is (4-0). If they lost weeks 1 and 2 but won week 3, their streak is (0-2). Ties do not end streaks as they are silly and irrelevant in this writer’s opinion. I also included notations of when new teams entered the league, increasing the spread of talent and size of the league to see if that played any significant statistical role.
Super Bowl 1 – The Green Bay Packers (4-0) beat the Kansas City Chiefs (3-0). Final Score: 35-10
Super Bowl 2 – The Green Bay Packers (3-0-1) beat the Oakland Raiders (3-0). Final Score: 33-14
Super Bowl 3 – The New York Jets (2-0) beat the Baltimore Colts (5-0). Final Score: 16-7
Super Bowl 4 – The Kansas City Chiefs (2-0) beat the Minnesota Vikings (0-1). Final Score: 23-7
NFL MERGER
Super Bowl 5 – The Baltimore Colts (1-0) beat the Dallas Cowboys (2-0). Final Score: 16-13
Super Bowl 6 – The Dallas Cowboys (2-0) beat the Miami Dolphins (1-0-1). Final Score: 24-3
Super Bowl 7 – The Miami Dolphins (17-0) beat the Washington Redskins (2-0). Final Score: 14-7
Super Bowl 8 – The Miami Dolphins (1-0) beat the Minnesota Vikings (9-0). Final Score: 24-7
THE SEATTLE SEAHAWKS AND TAMPA BAY BUCCANEERS ENTER THE NFL
Super Bowl 9 – The Pittsburgh Steelers (1-0-1) beat the Minnesota Vikings (5-0). Final Score: 16-6
Super Bowl 10 – The Pittsburgh Steelers (1-0) beat the Dallas Cowboys (4-0). Final Score: 21-17
Super Bowl 11 – The Oakland Raiders (3-0) beat the Minnesota Vikings (6-0-1). Final Score: 32-14
Super Bowl 12 – The Dallas Cowboys (8-0) beat the Denver Broncos (6-0). Final Score: 27-10
Super Bowl 13 – The Pittsburgh Steelers (7-0) beat the Dallas Cowboys (2-0). Final Score 35-31
Super Bowl 14 – The Pittsburgh Steelers (4-0) beat the Los Angeles Rams (0-1). Final Score: 31-19
Super Bowl 15 – The Oakland Raiders (1-0) beat the Philadelphia Eagles (3-0). Final Score: 27-10
Super Bowl 16 – The San Francisco 49ers (0-1) beat the Cincinnati Bengals (2-0). Final Score: 26-
21
Super Bowl 17 – The Washington Redskins (4-0) beat the Miami Dolphins (3-0). Final Score: 27-17
Super Bowl 18 – The Los Angeles Raiders (4-0) beat the Washington Redskins (0-1). Final Score:
38-9
Super Bowl 19 – The San Francisco 49ers (6-0) beat the Miami Dolphins (11-0). Final Score: 38-16
Super Bowl 20 – The Chicago Bears (12-0) beat the New England Patriots (1-0). Final Score: 46-10
Super Bowl 21 – The New York Giants (0-1) beat the Denver Broncos (6-0). Final Score: 39-20
Super Bowl 22 – The Washington Redskins (1-0) beat the Denver Broncos (1-0-1). Final Score: 42-
10
Super Bowl 23 – The San Francisco 49ers (2-0) beat the Cincinnati Bengals (6-0). Final Score: 20-
16
Super Bowl 24 – The San Francisco 49ers (3-0) beat the Denver Broncos (3-0). Final Score: 55-10
Super Bowl 25 – The New York Giants (10-0) beat the Buffalo Bills (1-0). Final Score: 20-19
Super Bowl 26 – The Washington Redskins (11-0) beat the Buffalo Bills (5-0). Final Score: 37-24
Super Bowl 27 – The Dallas Cowboys (3-0) beat the Buffalo Bills (4-0). Final Score: 52-17
Super Bowl 28 – The Dallas Cowboys (0-2) beat the Buffalo Bills (2-0). Final Score: 30-13
Super Bowl 29 – The San Francisco 49ers (1-0) beat the San Diego Chargers (6-0). Final Score: 49-
26
THE CAROLINA PANTHERS AND JACKSONVILLE JAGUARS ENTER THE NFL
Super Bowl 30 – The Dallas Cowboys (4-0) beat the Pittsburgh Steelers (2-0). Final Score: 27-17
Super Bowl 31 – The Green Bay Packers (3-0) beat the New England Patriots (0-2). Final Score: 35-
21
THE BALTIMORE RAVENS ENTER THE NFL
Super Bowl 32 – The Denver Broncos (6-0) beat the Green Bay Packers (1-0). Final Score: 31-24
Super Bowl 33 – The Denver Broncos (13-0) beat the Atlanta Falcons (2-0). Final Score: 34-19
Super Bowl 34 – The St. Louis Rams (6-0) beat the Tennessee Titans (3-0). Final Score: 23-16
Super Bowl 35 – The Baltimore Ravens (2-0) beat the New York Giants (3-0). Final Score: 34-7
Super Bowl 36 – The New England Patriots (0-2) beat the St. Louis Rams (6-0). Final Score: 20-17
THE HOUSTON TEXANS ENTER THE NFL
Super Bowl 37 – The Tampa Bay Buccaneers (0-1) beat the Oakland Raiders (4-0). Final Score: 48-
21
Super Bowl 38 – The New England Patriots (0-1) beat the Carolina Panthers (5-0). Final Score: 32-
29
Super Bowl 39 – The New England Patriots (6-0) beat the Philadelphia Eagles (7-0). Final Score: 24-
21
Super Bowl 40 – The Pittsburgh Steelers (2-0) beat the Seattle Seahawks (0-1). Final Score: 21-10
Super Bowl 41 – The Indianapolis Colts (9-0) beat the Chicago Bears (7-0). Final Score: 29-17
Super Bowl 42 – The New York Giants (0-2) beat the New England Patriots (18-0). Final Score: 17-
14
Super Bowl 43 – The Pittsburgh Steelers (2-0) beat the Arizona Cardinals (2-0). Final Score: 27-23
Super Bowl 44 – The New Orleans Saints (13-0) beat the Indianapolis Colts (14-0). Final Score: 31-
17
Super Bowl 45 – The Green Bay Packers (2-0) beat the Pittsburgh Steelers (3-0). Final Score: 31-25
Super Bowl 46 – The New York Giants (0-1) beat the New England Patriots (2-0). Final Score: 21-17
Super Bowl 47 – The Baltimore Ravens (1-0) beat the San Francisco 49ers (2-0). Final Score: 34-31
Super Bowl 48 – The Seattle Seahawks (4-0) beat the Denver Broncos (6-0). Final Score: 43-8
Super Bowl 49 – The New England Patriots (0-1) beat the Seattle Seahawks (1-0). Final Score: 28-
24
Super Bowl 50 – The Denver Broncos (7-0) beat the Carolina Panthers (14-0). Final Score: 24-10
Super Bowl 51 – The New England Patriots (3-0) beat the Atlanta Falcons (0-1). Final Score: 34-28
Super Bowl 52 – The Philadelphia Eagles (1-0) beat the New England Patriots (0-1). Final Score: 41-
33
Super Bowl 53 – The New England Patriots (1-0) beat the Los Angeles Rams (8-0). Final Score: 13-3
Super Bowl 54 – The Kansas City Chiefs (4-0) beat the San Francisco 49ers (8-0). Final Score 31-20
So there you have it. Big shoutout to pro-football-reference.com because this would have been nearly impossible without it. If you’re a football nerd like me, that website is truly fascinating to help break down your favorite seasons, players, or just learn about the history of the NFL. Now back to the topic at hand.
The first thing to do in compiling this data was to look at the week 1 records of our collective Super
Bowl teams, just as we did with the NSFL and DSFL previously. You’ll find this below:
Super Bowl Winners’ Collective Week 1 Records: 45-9 – (83.3%)
Super Bowl Losers’ Collective Week 1 Records: 47-7 – (87%)
Super Bowl Participants’ Collective Week 1 Records: 92-16 – (85.2%)
Ultimus Winners’ Collective Week 1 Records: 18-4 – (82%)
Ultimus Losers’ Collective Week 1 Records: 12-10 – (55%)
Ultimus Participants’ Collective Week 1 Records: 30-14 – (68%)
Ultimini Winners’ Collective Week 1 Records: 14-6 – (70%)
Ultimini Losers’ Collective Week 1 Records: 12-8 – (60%)
Ultimini Participants’ Collective Week 1 Records: 26-14 – (65%)
Right off the bat a few things jumped out at me in the above statistics. First, the overall correlation of Super Bowl Winners’ Collective Week 1 Records and Ultimus Winners’ Collective Week 1 Records is remarkably similar – within 2%. However, the overall correlation of these statistics is honestly not very close at all. It is about 17% more likely that a team that makes the Ultimus lost in week 1 than the teams that made the Super bowl, and about 20% more likely that a team that makes the Ultimini lost in week 1! For the comparison to be approximately 1/5 off is a significant statistical gap for sure.
However, I did notice in setting this up that the early years of the NFL seemed far more predictable start to finish than the later years have been in a similar fashion to how, as the NSFL has expanded, the winners have been less extreme favorites and the underdogs have taken home a significantly greater number of victories. I wondered if the same might be the case for the NFL. Therefore, I also compiled the records for “modern” Super Bowl participants, in which I included the last 18 seasons since the last team (the Houston Texans) was added to the NFL. The results are below, once more compared to the Ultimus and Ultimini results.
Since all teams joined the NFL (Last 18 seasons)
Super Bowl Winners’ Collective Week 1 Records: 13-5 – (72.2%)
Super Bowl Losers’ Collective Week 1 Records: 15-3 – (83.3%)
Super Bowl Participants’ Collective Week 1 Records: 28-8 – (77.8%)
Ultimus Winners’ Collective Week 1 Records: 18-4 – (82%)
Ultimus Losers’ Collective Week 1 Records: 12-10 – (55%)
Ultimus Participants’ Collective Week 1 Records: 30-14 – (68%)
Ultimini Winners’ Collective Week 1 Records: 14-6 – (70%)
Ultimini Losers’ Collective Week 1 Records: 12-8 – (60%)
Ultimini Participants’ Collective Week 1 Records: 26-14 – (65%)
I expected there to be a change in percentage here, maybe something like 5%, but I was amazed to see the level of difference that the modern NFL has seen! The odds of a team that loses week one winning the super bowl increased by over 10%! And the odds to make the super bowl after a week one loss increased by nearly 8%. When you look at these modern NFL odds to win or make the Super Bowl based on week 1 results, they are significantly more similar to the NSFL and DSFL odds to win or make their respective championship games. Overall, the odds of a conference champion having lost week 1 are right around 10% or closer to our simulations as they are in the modern NFL.
The major statistical difference that I noticed is that no large streak is a guarantee in the NFL. The 18-0 Patriots stand as the mammoth example of this truth, but we also have seen the 9-0 Vikings, 11-0 Dolphins, 7-0 Eagles and Bears, 14-0 Colts and Panthers, and the last two runners up – the 8-0 Rams and 49ers – all lose in the Super Bowl. In the DSFL no team that has started better than 5-0 and made the Ultimini has lost and in the NSFL the only time a team went 6-0 or better and lost in the Ultimus was the Hawks who lost to the Outlaws after they started 11-0. There is an assumed guarantee that if a team that started the season really hot makes the championship, they will win it, and this simply is not the case in the NFL. All of these hot starting teams have actually swayed the winning streaks in many ways to reflect higher on the NFL runner up than on the Super Bowl Champion. For some statistical examples of this, see below:
Mean (average) Initial Streak for Super Bowl Champions: 3.54 wins
Mean (average) Initial Streak for Super Bowl Losers: 3.93 wins
Mean (average) Initial Streak for Super Bowl Participants: 3.73 wins
Mean (average) Initial Streak for Ultimus Champions: 3.32 wins
Mean (average) Initial Streak for Ultimus Losers: 1.23 wins
Mean (average) Initial Streak for Ultimus Participants: 2.27 wins
Mean (average) Initial Streak for Ultimini Champions: 1.35 wins
Mean (average) Initial Streak for Ultimini Losers: 0.95 wins
Mean (average) Initial Streak for Ultimini Participants: 1.15 wins
First I took the mean or average of the initial records of the winners, losers, and participants of the Super Bowl, Ultimus, and Ultimini. In these statistics, it is clear that the DSFL in particular is not conducive to having large streaks, as it is the 18-0’s, 14-0’s and other large starts that really bump the Super Bowl means in addition to many 6-0 and a couple even larger starts for NSFL teams that help to keep them on that same kind of pace. The DSFL has seen only a handful of 4-0 or greater starts in Ultimini matchups and so their mean is much lower than that of the other leagues. However, it is confirming in my opinion that our Ultimus champions are on the same kind of trajectory as Super Bowl champions in this category, as it reflects well on the realism of our simulations. Clearly there are still less hot starts in our simulations than in the NFL, but to still be fairly close in this category is a win for our simulations.
Median (middle) Initial Streak for Super Bowl Champions: 3-0
Median (middle) Initial Streak for Super Bowl Losers: 3-0
Median (middle) Initial Streak for Super Bowl Participants: 3-0
Median (middle) Initial Streak for Ultimus Champions: 3-0
Median (middle) Initial Streak for Ultimus Losers: 1-0
Median (middle) Initial Streak for Ultimus Participants: 2-0
Median (middle) Initial Streak for Ultimini Champions: 2-0
Median (middle) Initial Streak for Ultimini Losers: 1-0
Median (middle) Initial Streak for Ultimini Participants: 1-0
Next I looked at the Median (middle) results of these same data points in order to reduce the effect of the outliers (0-4 or 18-0 does a lot to stretch out a mean). With this data, we see similarity once again in the Ultimus and Super Bowl champions, as the middle result for each category was a 3-0 start. I think this is right about where it should be. I was surprised by how low the Ultimus losers fell here to 1-0, but in looking at the number of teams that have lost the Ultimus after losing week 1, this makes a lot of sense as well. The DSFL again looks like a minor league here, and that’s okay! It’s not intended that a team run away with this league each year, but that there be even competition across the board. In my opinion, this is right where it should be hovering around 1-2 wins to start the year being enough to start on the way to a championship appearance.
Mode (most common) Initial Streak for Super Bowl Champions: 1-0 (ten times)
Mode (most common) Initial Streak for Super Bowl Losers: 2-0 (ten times)
Mode (most common) Initial Streak for Super Bowl Participants: 2-0 (eighteen times)
Mode (most common) Initial Streak for Ultimus Champions: 1-0 & 6-0 (four times each)
Mode (most common) Initial Streak for Ultimus Losers: 0-1 (ten times)
Mode (most common) Initial Streak for Ultimus Participants: 0-1 (thirteen times)
Mode (most common) Initial Streak for Ultimini Champions: 2-0 (six times)
Mode (most common) Initial Streak for Ultimini Losers: 0-1 (six times)
Mode (most common) Initial Streak for Ultimini Participants: 2-0 & 0-1 (nine times each)
Finally, I looked at the mode, or most common result for each of these categories (anybody else getting 6th grade math flashbacks?). This simply tells us what the most common start has been for participants and champions in the Super Bowl, Ultimus, and Ultimini. The most surprising thing in this data for me was seeing that in both the Ultimus and the Ultimini an 0-1 start was the most (or tied for the most) common record for participants. While it’s certainly more common for a team to have won than to have lost week one, the amount of teams who bounced back from a week 1 loss to have a winning season really impacted these statistics. In the NFL, it is much more common to have that first win (or first couple of wins) under your belt right away than any other result. However, those hot streaks are not particularly common, as most teams stopped at 2-0 before meeting an early season loss. There is, after all, a lot to learn in putting a team together each year – whether it’s a simulation or in person.
This was a LOT of data, I’m well aware, and I’m a little light headed just looking at it all. But my overall conclusion is that, while our simulation is not extremely close to the NFL in these particular data points, the overall trends are close enough to be realistic. When a team starts off hot, they are more likely to win. When a team loses an initial game or two, they are by no means damned, but better put on their hustle pants and change something quickly before their chances are too far gone. I personally know nearly nothing about putting together a simulation, but in this very limited evaluation, I’d say we are in good hands. Sure, our sims are quirky as anything, and sometimes a certain stat or result seems really strange, but the meat of the program is good. The spirit of competitive football is being respected here and that says a lot about the people who are running this league.
One more fun fact that I discovered before I finish this piece: First time organizations in the Super Bowl are 9-19, with most of those 9 wins coming against other first time Super Bowl participants, so if your first trip to the Ultimus or Ultimini doesn’t go well, know that you are in good company. There is a LOT to be said for experience, from the top of an organization all the way to the bottom. So keep on simming, keep on training, and keep on learning so you can lift that trophy one day! I know I for one am preparing my pose for some future banners…
Please keep this discussion going in the comments! I want to hear your thoughts about the importance of starting off well – either in our sims or in the NFL if you’re the fan of a particular organization. Recount a season, game, or story for us all to get to relive it with you! Also, if there’s another piece you’re wanting done but you don’t want to weed through a ton of data to do it, that’s something I really enjoy doing if you can’t tell, so comment below or send me a message – I’m always looking for ideas to write about and maybe we can team up on something if you want to help. And good luck in the second half of the season all you crazy football fans!
![[Image: Fq9564iX0AAzNOH?format=png&name=small]](https://pbs.twimg.com/media/Fq9564iX0AAzNOH?format=png&name=small)
![[Image: IgoYKoh.gif]](https://i.imgur.com/IgoYKoh.gif)
