Let’s all welcome CJ Kaltenbach to the Elite Mafia! His “hot taeks” will lead to copious amounts of winning… For his debut article, Seige discusses MLB DFS Strategy for the unique 2020 season.
First, I want to say, I appreciate all the warm words on social media over the last 24 hours. I’m thrilled to be a part of the Elite Mafia. We are going to have fun and do some really cool things in the weeks and months ahead so stay tuned.
This MLB season is going to be the craziest DFS season we’ve ever had. For starters, it’s only 60 games instead of the normal 162 game season, but we also have expanded rosters that change size during the season. We have a 60-man taxi squad, we have a new three-batter rule for relievers, and we have time-zone based scheduling. I wouldn’t blame you if you were saying to yourself, how the hell does any of these things affect my DFS lineups, and it’s a good question to ask.
Over the next few weeks, I’m going to tackle these topics and show how they can affect the way we build lineups. Am I redefining how we play MLB DFS? Of course not. Come Opening Day, we’ll be stacking and taking mini-stacks heavy in power and low ownership, but I wanted to show how nitty and gritty I get when analyzing baseball. Hope you enjoy!
MLB DFS Strategies Part I: Schedule Should Change Our Expectations on Player Performance
Ownership percentages in MLB DFS are based on projections and expectations. Factors that lead to higher or lower ownership are BvP (don’t use this one), history vs. a team, hot/cold, or a player playing better or worse than expected. Expectations can be good; it’s accurate to expect the Baltimore Orioles are bad and the Los Angeles Dodgers are good.
How do I know that’s a good expectation to have? I went back and looked at what the team/player has done historically and looked at the changes to the team to see if I need to change that expectation.
For the 2020 season, I started doing the same thing for season-long baseball. It’s almost draft season, and as I was going through my projections and having wildly different projections than normal, this led me to ask: Why am I expecting players to perform in 2020 season as they did in 2019 season or across their careers?
Over the course of a normal baseball season, we get 162 games and most starting hitters get over 500-600 ABs. The teams play good teams, bad teams, hot teams, cold teams across the grueling six-month MLB schedule. Over that large of a sample, players get closer to their true talent baseline statistics.
In a 162-game season, a team would have the following breakdown of games played:
19 games vs. 4 opponents within division (76 games)
6 games vs. 4 Teams in your league (24 games)
7 games vs. the other 6 teams in your league (42 games)
20 games of Interleague Play (20 games)
However, this year, we don’t have 162 games, and we don’t even have teams playing all the teams in their league! We have a 60-game season that is 37% in length and contains only regional match-ups.
In the 60-game season, a team will have this breakdown:
10 games vs. 4 opponents within division (40 games)
4 games vs. 5 opponents in cross-league – AL vs. NL Central, etc. (20 games)
The differences here are massive. In a normal season, interleague play is 12.3% of the season; this year, it accounts for 33%. Also, those tough games vs. the Astros, A’s, Yankees and Red Sox for the Indians/Twins? No longer on the schedule. Why should we have the same baseline expectation of Twins/Indians players as in past years when, instead of those tough match-ups, they get a higher percentage of games against the Tigers, Royals and Pirates?
While this will apply to every player on every team, let’s show an example of Mike Trout. Mike Trout is the best player in baseball, so we would expect he is similarly good in a short season to a long season. Mike Trout is a career .305 hitter with 39 HRs and 102 RBIs as his 162-game average. Most people will say for 2020, Mike Trout is a .305 hitter with 14.5 HRs and 38 RBIs, as that’s what his 162 average prorates to in a 60-game schedule.
This would be true if he was playing every team at the same rate as a normal 162-game season, but with regional-only scheduling, this doesn’t work. I took his career averages against all the teams on his schedule and weighted them based on how many times he’s gonna play them, and the results are shocking.
Career Mike Trout Baseline: .305/14.5/38 RBIs
2020 Mike Trout Baseline: .344/15/38.5 RBI’s
So, while the power numbers and RBIs are expected to be similar in this schedule vs. a normal one, his expected batting average jumps a ton. This makes sense, as he doesn’t have to face the grueling rotations of CLE, BOS, NYY he would in a normal year.
So, I’ve done a lot of math talk and still haven’t gotten to the point of why this matters for DFS. Let’s take a look at a couple situations we will face across a DFS season.
It’s Game 1 of the MLB season. Mike Trout has a neutral match-up and is priced at the same $6k as he was at the end of last season. The perception is that Mike Trout is going to perform close to the historical baseline and will carry an ownership to reflect that. However, by adjusting our baseline expectation, we know he should be outperforming his historical expectations by around 10%, and therefore, can get Mike Trout at a relative value.
Let’s now fast-forward to Game 31 of the MLB Season, and Mike Trout has that same neutral match-up against the same pitcher. He’s posted a .344 average with 8 HRs and 19 RBIs through the first 30 games. His price will now be higher than Game 1 because of his raw statistics.
If you look at these stats in a general form, you’d think, “wow, Trout is performing so much better than in a normal season.” That creates a perception Mike Trout is playing better, and therefore, will cause him to be higher owned than in Game 1.
Why is the ownership higher tho? All Mike Trout has done in the 30 games is perform at the level we expected him to, based on the schedule. By calibrating our expectations to the schedule, we can take advantage of incorrect perceptions and pivot to a lower owned player with similar or even higher ceiling the second time, while playing Mike Trout the first time.
Anyway, I hope you all enjoyed this. I promise the rest of the previews will not be this dense, but I think it’s really important we all think about these points consistently throughout the 2020 season. I know, for a fact, I will fall into this trap at least once in 2020. It’s hard to not look at the numbers and go, “WOW, he’s doing so much better this year” without checking to make sure that’s true and just not the result of a friendly schedule (or saying, “wow, this player is having a rough year” when we should have expected that based on the schedule).
Next Week: Why the three-batter rule could mean less PH’ing of Joc Pederson!
Come peruse our entire library of MLB & KBO DFS content!