For as long as I’ve been in the DSL, Robbie has provided Luck Analysis on a pretty much week to week basis. I’ve always found it interesting, so I decided to go about building my own tool to do the same thing for the MiLF starting with the 2021 season.
Basically, the luck analysis compares a team’s weekly performance against all other teams in each week. This gives you a total win percentage (both for the week and the cumulative season) as if you’d played every other team every week. This is then compared to your actual win percentage to find the luck variance, which is basically how fucked or helped you are getting by your matchups.
As of the time I am writing this, the luckiest team in MiLF history was Bieber’s 2017 team that finished in 5th place (thanks to a loss in the first round of the playoffs). This team had a total win percentage of 54.88% and an actual season win percentage of 81.25%, giving it a positive variance of 26.37%.
The unluckiest team in MiLF history was Phil’s 2017 team that finished in 12th (dead last). While this team’s total win percentage was a lackluster 39.86%, it’s actual season win percentage was only 7.69%, giving the team a negative variance of 32.17%.
The luckiest championship team in MiLF history was Bane’s 2018 team, with a positive luck variance of 20.73%. This team had a total win percentage of 54.27% and an actual season win percentage of 75.00%.
The unluckiest championship team in MiLF history was Kyle’s 2020 team, with a negative luck variance of 3.14%. This one is kind of laughable. This team’s total win percentage was 83.14% (the highest in league history) but it’s actual season win percentage was only 80.00%.
In all of these cases, the luck variances discussed didn’t really matter. These were all cases of good teams getting luck they didn’t need or bad teams that good luck wasn’t going to help anyway. Or a team that was too good for bad luck to have any real effect on. When looking at the historical data on a macro scale, it’s fairly easy to go through and identify some teams who maybe weren’t the luckiest or unluckiest based just on the value of their luck variance, but for whom luck clearly played a significant role in their season. Some examples:
Bane’s 2020 team finished in 2nd place with an actual season win percentage of 60.00%. But when the season was run through the luck calculator, this team’s total win percentage was only 38.95%, giving it a positive luck variance of 21.05%. In this situation, what would probably be identified as an overall subpar team was able to fight all the way to the championship game thanks in large part to the luck of the matchups.
In contrast, Kindra’s 2019 team finished in 10th place with an actual season win percentage of 40.00%. When the season was run through the luck calculator, this team’s total win percentage was 63.64%, giving it a negative luck variance of 23.64%.
Comparing those two examples, where this particular team of Kindra’s had a total win percentage that was 24.69% higher than this particular team belonging to her husband, she finished 8 places lower.
Glass ceiling confirmed.
Looking at the historical luck data broken down by team manager, Noll is, somehow, statistically the luckiest team manager in the league with an aggregate luck variance of +5.64%.
Phil is, understandably, the unluckiest manager in the league with an aggregate luck variance of -3.92%.
Kyle is, historically, the team manager who is least affected by luck, with an aggregate luck variance of +0.05%.