2020 Fantasy: Introduction

Baseball’s back. Spring training games are starting, Opening Day is 5 weeks away on March 26^th, and my fantasy league’s draft is on March 15^th. I’m back in. Before I dive into this season, I’ll do a brief recap of 2019, my third year in the league.

My strategy was all about having elite pitching. Truthfully, I had a beauty of a starting staff but some things don’t always work out how you expect. I drafted Chris Sale with my first pick, thinking he was the closest thing to a guarantee, but alas he had the worst year of his career. Checking up on velocity concerns almost drove me crazier than his seven starts of 5+ earned runs. Soon after the draft, I managed a 2-for-2 trade revolving around flipping Noah Syndergaard for Corey Kluber, who I could buy low after he had a rough second start. After seven starts, where he had a 5.80 ERA, Kluber got injured and was done for the season. My number three pitcher ended up becoming my reliant rock for the season, Charlie Morton. He put up great value for a 105^th overall pick with his 240 strikeouts and 3.05 ERA. My number four pitcher is what breaks my heart the most and probably had the hugest impact on my team: Jack Flaherty. I drafted Flaherty 57^th overall and had big expectations for him, so you can imagine I was pretty crushed when he had a 4.90 ERA on July 2^nd. Ultimately I traded him after that, and he proceeded to pitch to a 0.93 ERA with 130 K’s to only 24 walks over his last 16 starts. I also had Rich Hill, who was great when he was healthy. Only 13 starts is disappointing but I’m not too heartbroken when his inevitable injuries happen. Lastly, I drafted Kyle Freeland, who I figured was worth a shot at 201^st overall after his 2.85 ERA season in 2018. He was awful, but for his draft position, I was okay with it.

Anyways, all that led me to ultimately not making the playoffs. The pitchers who were supposed to be making my team great were not just disappointing, but really hurting my team. I started the season 1-5, which is making me sad just to type out, and I miraculously just missed the playoffs by finishing 10-9. Morton, Lance Lynn, and just about every relief pitcher providing value were carrying my pitching. I did win $50 for achieving the most improved team, so I’ll take what I can get.

                

The first thing I like to do when preparing for a season is just to plot out the last season’s data:

The biggest takeaways are how the data is sloped and how pitchers compare to position players (who I just refer to as “players” in the charts). Only a few players are elite enough to get to the 600+ points range, and value starts dropping quick after that. The point that gets reinforced is that the elite minority of players dominate in fantasy; players become quickly replaceable after that. The drastic steepness slows down after the 400 points range.

The other thing you’ll notice is the plot starts off very blue (position-player heavy) and ends very red (pitcher heavy). The tail off for pitchers at the end isn’t as concerning, since those are pitchers who probably pitched poorly in few outings and would likely never have been on our teams. Zooming in on the plot to view just the 300 most valuable players, we get the main point about pitchers versus position players: Valuable position-players are way more common than pitchers. Another nice way to view this is in a histogram:

In all the higher point intervals, aside from pitchers Justin Verlander and Gerrit Cole being the two most valuable players last year, position-players vastly outnumber the number of pitchers. The histogram shows exactly why I try so hard to get elite pitching; it’s rare, it’s hard to replace if you can’t obtain it, and position-players are comparatively much easier to obtain. From the previous plot, it’s easy to see how quickly value drops off, and the histogram especially emphasizes it for pitchers, who show way steeper drops in counts as the intervals increase than position-players.

This thought process is what led me to start drafting my rosters thinking about “points above replacement.” The rarity and drop off in value for good pitching makes their replacement value much lower than position players. If it’s hard enough to find 300+ point pitchers, imagine how valuable a 650+ point pitcher like Verlander and Cole is. If I draft pitchers early on, I know there are still valuable position-players to pick later. One thing for me to keep in mind is that my league members aren't dumb and also surely realize the importance of elite pitching. Here's how the first round of the draft went in 2019: six of the first twelve picks were starting pitchers.

The next thing I like to do is take a projection system (my favorite is Steamer) and use it to project fantasy points based off my league’s point values. Sure, a lot of drafting in fantasy baseball comes down to sleepers, breakouts, and busts, but a good start for me is seeing which players seem to respond the best based off my league’s point values. Projection systems like Steamer also factor in age, whether it’s younger players entering their prime or older players declining, as well as previous years-worth of data with emphasis on the recent ones. 

The way I calculate replacement level may not be the most exact way, but I work with what I can wrap my head around. There are ten teams in the league this year, whereas in the past we always had twelve. Each roster has nine pitchers, so there are going to be 90 pitchers total on a roster at once. I go to my projections, see what the 90^th most valuable pitcher is projected to be, and bam, that’s what I call replacement level. We get one catcher, so the 10^th most valuable catcher is replacement level. We get five outfielders, not separated by left, center, and right, so the 50^th most valuable is replacement level. The infield positions are tricky, because in addition to a slot for each position, we also get a 1B/3B flex and a 2B/SS flex. I’m not sure how to better account for this, so I just predict that there is two of each on each team.

And that’s how I start my fantasy baseball research! You can see in the screenshot of my spreadsheet how I can view points above replacement (PAR). Mookie Betts and Gerrit Cole are projected to be worth about the same amount of points, but Cole’s PAR at 325 completely shatters Betts at 201. If I have fifth pick in the draft, and both Betts and Cole are available, I’m probably picking Cole. Also in the spreadsheet, I have their average draft position (ADP) from fantasypros.com, as well as a net rank comparing how a player ranks in ADP versus projected points. It’s just a quick number for me to see which players may be outperforming their draft position.