Here’s what we’re about at RotoViz: Helping you murder your league. It’s pretty much that simple. We’re trying to give you a bunch of content from smart writers, pre-season apps that will help you think about roster construction, and then once we get to the season we’ll have a whole new suite of tools for you to use.
Because we’re fanatical about helping you win, we’re always trying to get better at what we do. That’s where today’s app comes from. It’s a Custom Cheat Sheet App and it’s the natural continuation of forecasting using Similarity Scores. One of the limitations of the Sim Score apps is that they don’t give you a global picture of how all of the players compare to each other. We could just give you the results of the Sim Score projections for all players, but that would be inferior to what we’re doing by giving you the app.
Because I know you want to start playing around with the app, let me try to explain it’s functionality briefly.
- The key tab is the Cheat Sheet tab. That contains ADP and custom projections (in standard and PPR for this year’s running backs). It also contains rankings based on those projections, as well as a Value field, which is roughly just the difference between ADP and the projections.
- But what if you want to overcome one of the obvious shortfalls of the Sim Score projections, the fact that they can’t see changes in situation or usage? How would you forecast then? I’ve made the app to allow you a few ways to change the forecast.
- The first way to change the forecast is to adjust the games sample that the comps and the projections are using to project from. For instance, let’s say that you wanted to change Adrian Peterson’s projection because you think the Sim Scores don’t correctly view the fact that he was coming back from injury (which may or may not be a valid reason to adjust them). You could adjust the sliders to throw out the early season games. That will change his projection.
- But what if just getting rid of part of the season isn’t enough? What if you think a player has an increase in usage in store, like CJ Spiller for instance? You can also just throw out games that don’t meet a certain target threshold, but using the touch filter. Doing that would change the season averages for Spiller, which would change the comps, which would change the forecast.
- But what if the player’s 2012 doesn’t inform at all what you expect from them in 2013? An example of this might be a rookie. Then you can still just throw in a Wild Ass Guess and change the forecast that way.
- Once you have either a Similarity Based Adjustment that you like (see the last tab for the new forecast) or a WAG that you like, just click the Update button for the forecast you want to use and a new cheat sheet will be calculated.
- There is a control that adjusts the penalty for games played that will be a lot more important for running backs than it is for receivers. All it does is assign a priority to Games Played as a stat in the similarity search.
- If you want to sort any of the columns of the cheat sheet, you can just click on the header. You can also page through results using the arrows below the cheat sheet.
- Once you’re done adjusting projections, you can download into an Excel file the results of your custom cheat sheet.
Because you might be wondering why we don’t just do this and feed you the answer, I’ll address that. Getting good at fantasy football is about understanding how the past relates to the future. If we just tell you who to pick, then you don’t get the benefit of doing things like going through lists of receiver comps and seeing which players were able to reproduce touchdowns. This app takes that idea a little further because it makes you the forecaster. You have to think critically about what amounts of usage are reasonable and whether your sense of value can be backed up by what has happened in the past.
In a future post I’ll go through some exercises in creating forecasts for players, but until then, feel free to play around with the app and make your own cheat sheet.