There is no shortage of baseball projections available to you each year for your fantasy drafts. Deciding on which projections that you use in your draft may seem like a minor deal but it’s the backbone of your entire decision-making. If you do all of your calculations based on a flawed projection system then you’re going to have a bad draft even if you did great calculations. Basically, there’s no escaping that your choice of projections means a lot for your fantasy baseball success.
In the past, I’ve annually analyzed the various projections to show their strengths and weaknesses. That analysis always interests me because the results vary a bit each year but you can see that certain systems are far more accurate than others over time. For instance, there’s no denying that Steamer is a mighty fine and powerful projection system on its own, and Jared Cross does an amazing job with it. There’s always going to something powerful about aggregating the projections though. One of my heroes in the baseball community is Tom Tango, who developed the Marcel projection system many years ago, and he ackonwledged the power of aggregation in a Twitter exchange I had with him a couple years ago.
“We’ve known aggregation is the best since the very beginning. I showed as much 10-15 years ago when I’ve been testing these things.”
That’s where my system comes in.
I’m not doing the hard work of the actual projection systems themselves but I’ve done enough research over the years to know which systems are the best at what. Without getting into the gritty details, the basic idea of my projections is that I assign a certain weight to the various projection systems when aggregating them and those weights vary for each statistic. The weights are based on the historical analysis that I’ve done for a decade now and shift a bit each year as I get more data to analyze. For some of the stats, I also found that expected statistics (xStats) or previous year statistics can help increase the accuracy of the aggregation.
When all is said and done, these are the sources that get used in my aggregation:
- Clay Davenport
- Fangraphs Depth Charts
- THE BAT
- 2021 xStats that I calculated
- 2020 actual stats
These various sources all got thrown into a big pot and my projections emerge from them.
The 2021 Projections
Though I’m posting the projections below for you to browse through, you can better access them by viewing them on this Google Sheet. Within the Google Sheet, you can make a copy for yourself or download the projections for your own use case. You’ll see in the projections that I also included the players’ projected WERTH value for a standard league format. The WERTH value is essentially the cumulated z-scores of the five standard roto categories for that player and a way of showing how valuable a player is in a rotisserie league.
As always, you can use these projections within my Excel cheatsheets as you conduct your draft.
There will be minor updates and tweaks to these projections before the season starts. I’ll keep everyone updated on Twitter about the updates that occur.
Projections last updated on 03/07/2021