For the last few weeks I’ve been crunching numbers from the first half of the Mynci season. My goal has been to find out which categories are most predictable in determining head-to-head points.

In our 12-team league I used the data from the first 11 matchups, a single round-robin. At the end of the regular season I will use the data from matchups 12 – 22 to double the data set.

From this data I calculated how many points each team generated in each category, with ties worth 1/2 point. I then  calculated the win ratio for each team in each category using the formula (wins + 1/2 ties) / matchups. The formed one matrix.

In another matrix I used each team’s 11-week averages to rank from one to 12 within each category.

Then I cross-indexed category rank with win percentage.

In the charts below, the x-axis is the category rank for the 12 teams in the league. The y-axis is the win ratio corresponding to each rank.

There’s a lot of information to consider here. I’ll start with two points.

First, investment in saves and steals generates the best return in head-to-head wins (and ties). This is not just a case of haves and have-nots. There are fewer outliers even among the top half of each category.

Compare this to wins. If you dropped the bottom three teams from the wins category, the trendline would be almost flat. Teams that accumulate more wins over many weeks do not have much advantage in any given matchup.

These are the category fantasy owners hate the most. The unforgiving and the arbitrary.