As you all have come to expect by now, today’s article will once again function as both a breakdown of today’s slate and an evergreen style article that you can come back to any time. Today’s focus, however, isn’t on DFS theory per se, but rather a detailed, step-by-step guide of how I use the model to build MLB lineups. Let’s jump right in, beginning with pitching.
Step 1 – A First Look at Pitching
My first step is always to sort pitchers by projected value. This does not mean that I will select my pitchers before my hitters, I’m just taking a look:
In the Discord chat, Shazzam posed the question, “What do you do when there’s one clear and obvious guy (Joe Ryan)?” So let’s start with him. He has the highest DK point projection and the highest K projection, but his $10,400 price tag seems to account for that and more, as he’s only seventh in projected value.
As for my personal takeaway, the top two projected values are both under 10% projected ownership, giving them excellent GPP Scores. These are typically the slates I like the most, where our best-projected pitcher values are contrarian.
Next, I sort by projected strikeouts to see if anyone truly separates or to see if any of the high GPP Score guys are too far down the list to truly have the slate-breaking potential their GPP Score suggests:
Here, you can see why Ryan isn’t such a great value. Yes, he’s first in projected Ks, but he’s not separating much at all from the rest of the pack. This isn’t to say he’s a terrible play – just that playing him foregoes an opportunity to leverage the field based on Ryan’s outlook. Nick Lodolo, on the other hand, seems like a terrible play.
Griffin Canning still seems really good here, and Matt Waldron is fine, but this view also puts Luis Severino in a much better light. I’m likely going to use Garrett Crochet, and my SP2 will depend on how much money I have remaining after my stacks. Severino and Canning are both good options.
One final note here from the first look – Crochet and Ryan are each making their second straight start against a specific opponent. There is data that shows the hitters have a slight advantage when this happens, but it’s slight. It’s the kind of thing that if Ryan projected as the best value but 40%+ ownership, I would potentially use it as a reason to fade him. It’s not something I care about when chasing contrarian upside in a guy like Crochet whose underlying numbers paint the picture of a potentially dominant pitcher.
Step 2 – A First Look at Hitters
Here, I filter by position and sort by GPP Scores, looking for players who separate in GPP Score and have genuine upside. A high GPP Score is rare without upside, but it can happen for cheap players when no one at the position projects particularly well (whose median outcome is fine but they will never shatter their value with a huge score). Shea Langeliers looks great at catcher, with the position’s highest HR odds on the slate:
I make no decisions at this point, just a note that Oakland and Seattle are potentially in play. The infield positions are primarily led by Diamondbacks and Brewers. Second base and third base are once again the problem childs of the slate. It’s too bad we can’t kick them out of class like I was throughout middle school.
We get more of the same in the outfield, though Marcell Ozuna is hard to ignore:
So while other teams aren’t going to be completely excluded, I now have a basic list of:
- OAK
- ARI
- MIL
- ATL
- NYM (Pete Alonso and JD Martinez are both near the top of their positions)
Oakland could be a cheap secondary stack, but nothing more:
The Mets could be a high-upside secondary stack, but it gets somewhat ugly after this trio:
The Braves are super expensive, but if we have the money, this is super enticing (four guys over 20% to homer!):
The most universally under-owned stack appears to be Arizona, and like I said, it helps a ton that they also cover all of the problem-child positions:
There’s even a good chance that I would fade one of the top four GPP Scores in order to fit all of 2B, 3B and SS.
Finally, we have Milwaukee:
The fact that we can’t use Jake Bauers in the outfield anymore is a bummer, but there’s still plenty to like here. It’s time to start thinking about how these various offenses could fit together.
Step 3 – Mapping How Stacks Fit Together
Some days, this can be extremely difficult. Today, it’s not as tough since Arizona takes care of all of the toughest positions to fill. Arizona and Oakland fit perfectly together, and we can get Arizona with the Mets or Braves if we make some small sacrifices in the D-backs stack to get Gabriel Moreno in there at catcher.
I don’t believe I can post full lineups, so here are the basic structures of the best builds I’ve found so far:
- Crochet & Severino/Canning | ARI – OAK
- Severino & Canning | ARI – NYM
- Crochet & Canning | ARI – MIL
- Crochet & Severino/Canning | MIL – ARI
I’m having an awful lot of trouble making Atlanta work, but that could mean the Braves end up even more contrarian than currently projected, so we’ll want to revisit them later once lineups come out.
Final Thoughts
You might notice one thing missing above that has become a staple of my content – the idea of balancing contrarian upside with high-probability plays. This isn’t by accident, but rather because my process does this automatically. The GPP Scores guarantee a little bit of both, but especially contrarian upside, and my emphasis on value and K upside for pitchers guarantees the high-probability aspect.
Lastly, notice how simple this process is! I’m not doing anything special, I’m just taking a disciplined approach that eliminates bias and favors concepts like positional scarcity. Anyone and everyone is capable of executing the same process. On smaller slates, we will also want to consider correlated leverage, but that’s a topic for another Monday.