This offseason at RotoFanatic we have been working on a number of different items to help make you a better Fantasy Baseball player. One such idea has been teased by myself on Twitter as well as on our different podcasts. This is what we are affectionately calling, The Data Monster.


What Is The Data Monster?

The first question that obviously comes to mind is what is The Data Monster? Simply put it is the home of data and tools that utilize some of the different models we have discussed and referenced in articles over the course of the season. In the first release of the tool, it will hold all of the different aspects of my StuffERA model as well as the plate discipline metrics that I have developed for hitters. These items are contained in a series of sortable and filterable tables which I will explain a bit further along in the article.

The last part of the first phase of the drop includes the results of my Fantasy Prospect Model. The important thing to remember is this model relies on 2019 data as there was no 2020 MiLB season. Also related to the prospect model is a tool that allows you to view the probability-based range of outcomes for players and a handy tool to compare these ranges for two players.


Pitchers Tab

The first tab on the tool is the Pitchers tab.

At the top, you will see these different input options. The first section allows you to choose between five different leaderboard types with the following statistics:

  • Whiffs – (Whiff, xWhiff, In_Whiff)
  • Swings – In Zone – (IZ.Swing, IZ.xSwing, IZ)
  • Swings – Out of Zone – (OOZ.Swing, OOZ.xSwing, OOZ)
  • wOBA – (wOBA, xwOBA, In_wOBA)
  • StuffERA – (Command, StuffERA)

You will then see a slider that allows you to adjust the minimum number of pitches thrown. It defaults to 500. Explanations for all of these metrics can be found here.


Hitters Tab


The second tab is the Hitters tab. This tab is extremely similar to the Pitchers one however with one small change. Instead of the StuffERA leaderboard, there is one title Plate Discipline. This contains three metrics, xwOBA_Take, xwOBA_Swing, and SAE. This takes into account the decisions made by a given hitter. As the first two metrics state, this is the location-based expected woba on the pitches a hitter decides to swing at or take. SAE. This is essentially a measure of how much better the pitches a hitter decides to swing at versus all of the pitches he actually sees. The metric is scaled on a percentage basis so a value of 110 means that the hitter swung at pitches with a 10% higher expected woba than his overall expected woba. An explanation of this metric can also be found at this link.


Prospect Tabs


Personally, these are my favorite tabs among all of those included in the Data Monster. This tab is the home of my Fantasy Prospect Model. I will be doing a more in-depth write up of this soon but the model uses a comparison algorithm to find the most similar seasons to a given player. These similar seasons are weighted based on the stats that correlate most strongly with future fantasy success. Then, these seasons are weighted for proximity to generate a singular value for expected fantasy value per 600 PAs.

Additionally, there is a value called Upside Rate that determines the weighted probability that a given player becomes a fantasy star. This essentially measures the high-end similar results for a given player. This helps to evaluate high floor prospects versus those who have a wide range of outcomes. The leaderboard also contains a value for Adjusted Value which accounts for the levels the player played at which allows us to better compare players who are far away versus those in AAA. A full write-up on the model is coming soon and I will go more in-depth into this idea.

The best part of these tabs is easily the range of outcomes comparison tool. This allows you to select two prospects and a specific level that guy played at and build a chart comparing the players.



You can see the two selectors above and then in order to update the plot you need to click the Generate Plot button. The above selections create this output.

Both Julio Rodriguez and Brennan Davis are two of my model’s favorite prospects in all of baseball. As you can see Davis has a higher chance of a zero-sum outcome which is why his mean expectation (the lines) is lower than Rodriguez’s. I am really excited about this feature and I can’t wait for you all to get your hands on this.


What’s Next

We have a bunch of other features we will be adding shortly including projections based on statcast analysis from Crosby Spencer, the results of my previously mentioned relief pitcher model, and more other really cool interactive tools.