Baking Consistency Into Player Draft Rankings (2020 Fantasy Baseball)
Ask any expert. They can flip open their favorite projection tool and rattle off a 5×5 rotisserie categories projection at the drop of a hat. Projection systems typically project at a player’s 50th percentile – somewhere in between their floor and ceiling. Player draft rankings are based on these projections. Thus, we have to ask ourselves the key question:
How does consistency play into these projections, and therefore rankings?
The answer, of course, is that it depends. Every player has a floor and a ceiling, and it’s up to each individual drafter to determine what he or she believes is the most accurate or depend on public projection systems or your favorite analysts’ projections. I’ll take you through the journey of how I arrive at my projections and rankings.
My projections are largely based on Mike Podhorzer’s “Projecting X” system – I highly recommend giving it a read. That projection system begins with a three-year average of tons of underlying metrics – strikeout and walk rates, doubles and triples per at-bat, etc. I use new-age stats like barrels, hard-hit rate, etc. to influence the projection. The three-year average is a great foundation, as it typically encapsulates the 50% percentile of any player. Reading through the multitude of projections, it is clear that many projection systems use this approach. For example, let’s take a look at Mike Moustakas. His three-year average underlying metrics are as follows.
I believe that Moustakas’s three-year average on these metrics is a great starting point, given that he has been in the big leagues since 2011, and he has been known as a .250 hitter with 30-homer potential. Using these metrics (and a few more) to project Moose’s 2020 season, this is how the three-year average compares to some public projections.
Regardless of the projection system, it’s safe to say that using a three-year average gets us close to other projection systems, at least for Moustakas. We can say the same thing for Zack Greinke or Kyle Hendricks. Simply put, a three-year projection makes a ton of sense for veterans and players with stable hitting/pitching profiles. Their 50th percentile projection is more consistent than a young player or someone with an unusual hitting profile (like Javier Baez). Looking at the FantasyPros rankings shows that Moustakas’s ECR and ADP are virtually the game, which is a good sign that he is extremely consistent.
Of course, most fantasy players are not making their own projections. But, knowing that veterans can be reasonably predicted by multiple projection systems indicates that you can draft that player at their ranking with confidence. Now, what happens when a three-year average isn’t available for a young player or a player has uncertainty heading into the season?
Doing the Research
A lack of consistency in player rankings typically comes from an injury, playing time concerns, or a change in identifiable trends (pitch usage for a pitcher, launch angle for a hitter, etc.).
These three variables will result in a lack of consistency in projections, and thus, player rankings. The best way to identify a lack of consistency in a player’s ranking is by looking at a player’s standard deviation in the FantasyPros’ ECR. Some of those players in the 2020 season are Christian Walker, Khris Davis, and Zac Gallen.
For a typical fantasy baseball player, my advice is to understand why they have a large variance in their rankings. For playing time questions, head over to FanGraphs’ depth charts to understand who may be taking away a certain player’s at-bats, or who that starting pitcher is battling for the back end of the rotation.
For injuries, I recommend following beat reporters on Twitter and reading the injury articles here at FantasyPros. Another site good for injuries is Inside Injuries – their work is featured on several different fantasy websites.
Finally, for changes in trends, I recommend you learn to do deep dives on players with that boom/bust variance. Baseball Savant has easy-to-follow leaderboards that can aid in this analysis, along with FanGraphs, Brooks Baseball, and Baseball-Reference.
Forming a Consistent Opinion
One key to watch out for is confirmation bias, particularly when looking at underlying metrics. If you believe that a high-strikeout pitcher with a wide range of outcomes will be a league-winner, do not simply look for trends that verify your thinking. Instead, lean towards what the data is telling you.
For playing time, research how the skills of that player compare to the player he is battling with. For injuries, I typically lean on the conservative side, and bank on the consistency of injuries when ranking players. For example, Carlos Correa has not played a full season since 2016. As a result, I have him ranked 100 spots lower than the ECR for 2020. If you think Correa will stay healthy for a full season, bump up his ranking. If you think he will get injured again, avoid him altogether. The key to understand is that Correa’s ranking will not be consistent amongst experts nor website ADPs, so you have to make the ranking determination for yourself by taking the above factors into account.
In short, identifying consistent (often called “boring”) veterans to pair with a few less consistent (often called “boom/bust”) players can make for a winning team, provided you balance your portfolio appropriately.