There is something wrong with team performance in home v away. I simmed four seasons on full detail for the premier league. There's just simply not enough goals coming from away teams.
Over the four seasons simmed there have been:
- 2394 goals scored for home teams,
- an average of 29.9 per team
- with a standard deviation of 12.2.
Compared to the four most recent non covid affected seasons in real life there have been:
- 2362 goals scored (-30 really good!)
- an average of 29.5 per team
- and a std dev of 11.3
these are pretty much equal and a t test fails to reject any hypothesis that these are different distributions. No complaints on the number of goals scored at home. - 30 goals over 4 years is exactly the kind of randomness we'd expect.
However for away teams in FM 23:
- they have only scored 1647 goals.
- for an average of 20.6 per team
- and a standard deviation of 7.95
In real life away teams:
- 1865 goals scored (difference of ~+220 goals)
- an average of 23.3 per team
- and a standard deviation of 7.7
More importantly this results in a rejection of the null hypothesis at the 5% level. Indicating there is simply not enough goals scored by away teams in this game. Even given this small sample size of sim tests, it confirms what I've been seeing while watching games in the match engine where away teams players make significantly more "mistakes" than they should be given their talent level. Some of these the ME don't even mark as events.
To understand what's wrong I ranked teams via points from real life and the sim into 80 rows each from high to low (rank 1 is most points rank 80 is fewest points). Above zero indicates more goals were scored (and in later graphs, conceded) in real life than the sim. I did rolling average of 5 teams on either side of the ranking. There's no real pattern here. Every team seems to suffer a bit in goal scoring when playing away. The 10-20 placed teams are benefitting a bit from being grouped in with the title contending teams and would fall more in line over repeated sampling to show a more smoothed feature where they slot in between 1-10 and 20-30 more nicely in both graphs.
The bigger effect I think is seen that teams simply don't concede enough at home. The spearman rank coefficient here is .03 and not statistically significant. It should be noted though that the effect may not be linear when looking at the graph. There should be about 3 goals more conceded per team at home on average for all teams in a season. The home effect appears to peak around the mid table teams (between 30-60) where we're probably losing out ~4-5 goals against a year per team.
What does this do for actual results? Well, interestingly away points won pretty much mirrors the home goals conceded plot. Title contending squads and relegation level teams aren't winning enough games relative to their real life counterparts
My conclusion? I think playing at home is too overpowered. The obvious answer is that away mentality in the game is too low. Particularly with relegation scrapping teams. I disagree with that notion. In watching many games I've noticed that there are a lot of events where an away player will behave very, very differently than they do at home; that is certainly an understandable and desirable effect. Teams generally perform much better at home. Based on the evidence here we are bang on for home performance in terms of scoring goals. It appears that away teams are nerfed too much (particularly mid tables teams and below in the Premier League) in composure, decision making, morale/motivation or some component of technical abilities to replicate a sufficient number of goals scored by away teams.
tables for the sim and RL.
2021-22 PL RL results
2018-19 PL RL results
2017-2018
Recommended Comments
Create an account or sign in to comment
You need to be a member in order to leave a comment
Create an account
Sign up for a new account in our community. It's easy!
Register a new accountSign in
Already have an account? Sign in here.
Sign In Now