The 60-Minute Football
This morning I came across an interesting(and controversial) news from the sports section. Apparently, FIFA is experimenting with some new changes to the rules of football, which includes:
- 30 minutes per half
- Unlimited substitutions
- 5-minute suspension for yellow cards
- Kick-ins replacing throw-ins
They are all bold updates to the rule, and unsurprisingly many fans are protesting against it. The reduction in playing time piqued my interest, I always thought the sport is ridiculously short at 90 minutes, considering it’s a team sport with 11 men on the field. Sports like basketball, hockey, and tennis has a higher average playing time per match.
While reading the news, a thought popped up: “What would be the impact to the match results and leaderboard, in a world where football is only played for 60 mins.” My initial guess was: “Probably quite a bit, considering the amount of drama and stoppage-time winners the sport produces every season.”
Coincidentally, it’s a national public holiday today, so I decided to answer this hypothesis via a one-day project, with the following sources and assumption:
- Data Source: Web-scraped from WorldFootball
- Assumption: An average of 4 minute additional time per half in English Premier League games [Source]
- Logic: Only goals between 0–34th minutes and goals between 45–79th minutes are qualified as goals in the 60-minute model
For full project and scraped datasets for your projects, check out my repo!
Case Study: 2020–2021 Premier League Season
What if the latest premier league season was played in 60 minutes instead of the traditional 90? Let’s look at the leaderboard comparison:
Leaderboard
As expected from a dominant Manchester City season, the rule change wouldn’t be enough for us to see a different champion. Likewise, at the very bottom of the table, the relegated teams are the same. However, one thing to note is that the relegation battle intensified with the rule change, the bottom 3 teams were able to score more wins in a 60 minute model. This is expected as a shorter playing time would have higher variance to the model, making each match less predictable, and increase the probability of weaker teams scoring an upset victory.
Let’s go back up to the qualification zone. In the original model, Manchester City, Liverpool, Manchester United and Chelsea won the top 4 place and qualified for the Champions League. In the 60 minute model, Manchester United and Chelsea were replaced by Tottenham Hotspur and West Ham United, respectively. Perhaps Manchester United and Chelsea both performed better in the latter period of each halves?
In the 60 minute model, the point distribution between teams are slightly less spread out (Standard Deviation = 14.08), as compared to the original model (Standard Deviation = 16.89). This is explained by the higher varianced introduced by having less playing time. So, if you’re rooting for a strong team, you should hope that the rules don’t change, but if you’re rooting for a bottom table team, the reduction in playing time actually works in your favor and greatly increases the odds of your team staying above the relegation zone.
Goals
Enough with the boring statistics stuff, what about goals? Goals are the essence of football right? It’s what makes the highlight reels. Stating the obvious, but sadly, less playing time = less goals. Playing by the normal rules, the 2020–2021 season has seen a total of 1024 goals, an average of 2.7 goals per match. In a 60-minute model however, we would have only 749 goals this season, an average of 1.97 goals per match. That’s a 27% reduction in goals. Let’s also see how that affects the top scorer leaderboard:
Top Scorer
Well, the number of goals scored definitely took a hit, but it doesn’t have a significant impact to the top goalscorer list. 8 of the top 10 strikers remain in the hypothetical model, with Harry Kane winning the Golden Boot in both.
That concludes the high level overview of our 2020–2021 case study. As I was scraping the data from as early as the 2000–2001 season, I managed to get 21 years worth of data to play with. Which team do you think would have lost or won a championship title due to this rule change?
In the past 21 seasons,
- Manchester United won 7 titles
- Chelsea won 5 titles
- Manchester City won 5 titles
- Arsenal won 2 titles
- Liverpool won 1 title
- Leicester City won 1 title
If football was played in 60 minutes, we would get:
- Manchester United would win 7 titles (+0)
- Chelsea would win 4 titles (-1)
- Manchester City would win 4 titles (-1)
- Arsenal would win 4 titles (+2)
- Liverpool would win 1 title (+0)
- Totenham Hotspur would win 1 title (+1)
Hmm, seems like Arsenal would be the biggest beneficiaries.
Closing Thoughts
Experimenting with new changes are always a good thing, new technologies like VAR and GLT certainly helped make the game more fair. However, changes should ultimately be providing value for the fans, not taking it away.
The argument is that the average audience has a much shorter attention span than before. While that may be true, by reducing the game time you would be effectively stripping away 33% of the entertainment value for the die-hard football fans watching the sport week in, week out.
If you’re interested in this mini project or interested in using the dataset I scraped for your own projects, feel free to get it from here.
