Preparing for a New League Type

It’s been a wild week in the Bell household! My wife and I welcomed our third child into the world last Friday… A healthy baby boy to go along with our two lovely daughters. On top of that my oldest daughter played in her first tee-ball game. My heartstrings were pulled a lot last weekend.

With all of that going on, I also received an invitation to play in a test league for a new fantasy baseball format. You may recall that Ron Shandler created and ran a monthly game format during the 2015 season. The game was intended to be a middle ground between the ultra-fast, yet time consuming, DFS format and the sluggish grind of the six-month long rotisserie game. That original incarnation of the game was closed down but league hosting site Fantrax.com is attempting to bring back a slightly different version. They set up two test leagues composed of various industry analysts (and me, somehow!) in order to garner feedback and suggestions on how to make the monthly format work.

What follows is an outline of the rules, the participants, and the strategy I used in selected my roster. After I share my process, I’d love your feedback and thoughts on what you would have done differently.

The Rules

While we are testing a monthly format, this specific test run of the contest is set up to run eight weeks. Traditional rotisserie categories and scoring are used.

The league is composed of ten teams. Each team selects 32 players that have been assigned salaries. Those 32 players will be used to submit a starting lineup. The lineup is also consistent with the traditional rotisserie setup (14 hitters, nine pitchers, two catchers, middle and corner infield, etc.). This implies a bench of nine players. The total salary for each team must not exceed $300. Each team gets only ONE transaction once the contest begins. Roster changes (starting and benching players) can be done twice a week.

The Participants

Here are the participants in my league (I say “my league” because there’s another league with Ron Shandler, Todd Zola, Lawr Michaels, etc.):

My Process

The process I followed is admittedly not the “optimal” way to approach the problem. I would have liked to calculated “standings gain points per dollar of salary” or “z score per dollar of salary” and then run that through a solver calculation to produce the optimal lineup. Similar to what some do to generate DFS lineups.

While ideal, that’s not quite as fun and interesting. And being honest, I didn’t have the time to jump through all those hoops. There are holes in such a process as well. Relying purely on a “rest of season” projection would surely exclude players like Trevor Story, Michael Conforto, or Drew Pomeranz from my lineup. No projection believes in them. But I might.

So instead of a purely mechanical approach, I set out to use a manual approximation. Here are the tenets of my approach.

Exploit Low Salaries

You can see a full list of the salaries for the contest here. I didn’t have time to run a full blown solver solution to optimize my team, but there are so many players involved that I did run the Fangraphs rest of season projections through my dollar value calculation process and then calculated a basic “Cost per Standings Gain Point” measure.

There are a few flaws in this approach. First, it’s difficult to determine “replacement level” in this format because player selection is not mutually exclusive. Theoretically we all could have drafted Bryce Harper. The other issue is that cheaper players will surely end up atop this “Cost per Standings Gain Point” list. I could fill the entire team with $1 players and have a very high “Cost per Standings Gain Point” ratio… But I’d finish in last.

With those weaknesses in mind, this is still a great approach to identify players that are mispriced. Following this methodology, I was able to quickly identify that hitters like Jarrod Dyson, Chris Carter, Ben Revere, Lucas Duda, Steven Souza, Joc Pederson, Billy Burns, Khris Davis, and Adam Eaton had $1 price tags. Jonathan Villar and Eugenio Suarez could fill the middle infield spots for $2 each. Likewise, Francisco Rodriguez, Carlos Martinez, and Vincent Velasquez had $1 salaries.

Squeeze in High- and Obviously Under-Priced Stars

Players didn’t have to only be valued at $1 for me to see them as a target. Players like Yasiel Puig ($11), Jose Abreu ($15), and Miguel Cabrera ($18) also jumped out as bargains, regardless of their slow starts. The pricing in the game is designed to take early season performance into account, and it seems to me like these guys should still be thought of as $20 – $30 players.

I was feeling pretty good at this point. A team with Cabrera, Abreu, Eaton, Davis, Burns, Souza, and Revere (I elected to go for a $1 OF instead of Puig at $11) for a salary of less than $40? I could spend on some bigger names.

I could have splurged on a $40 Bryce Harper, but a $25 Giancarlo Stanton seemed like a better alternative. When Stanton is healthy and producing (he is!), he’s worth more than $25. I still had a hole at third base, and a $21 Kris Bryant seemed like a reasonable price. My last big purchase was a $34 Jose Altuve. He’s playing out of his mind and I’ll double-down that this breakout is for real.

Think About Games Played

I was prepared to spend a lot of time analyzing schedules and games played, but this being eight week competition instead of a four week one minimized that dynamic for me. I imagine in a 30-game window the difference in games played could be 10% advantage or more for some teams (e.g. some MLB team plays 30 games in the month and another plays 27). But in the eight-week window that difference is only about 3%-6% for most teams. I did enough analysis to know that the Dodgers and Indians play 53 games during the eight week window and most teams play 50, 51, or 52 games. Only the Rockies and Cardinals are scheduled for 49.

In analyzing the schedule, I did create a nifty Google Sheet that calculates the number of games a team plays between two specific dates. You are welcome to make a copy of the spreadsheet and use it yourself (to do this, click on the link above and select the “File” menu then the “Make a copy…” menu option).

Maintain Some Roster Flexibility

We are allowed one transaction during the game, but I wanted to build in some insurance from having to use that move unnecessarily, so I also rostered Brock Holt for $1. I should be able to withstand an injury in the OF, 3B, or middle infield by employing him and his 2B, 3B, and OF eligibility.

Focus on Pitching (Aces and Bench Arms)

I mentioned Francisco Rodriguez, Carlos Martinez, and Vincent Velasquez as clear $1 pitcher buys. I also think I built a pretty diverse offense with players that do a little bit of everything (Altuve, Bryant, Souza) and a diverse group of specialists (Burns, Eaton, Davis, Revere, Chris Carter).

My main strategy was to spend big on pitching. I hope to delve more into this topic in the near future, but I think the 67/33 or 70/30 split we hear oh so much about might be misguided in the season long game and especially misguided in a monthly game. I have a hunch that many will try to build bargain staffs and try to maximize starts. So I planned to zig here and invested heavily on some of the hottest pitching arms.

I plugged in a $41 Clayton Kershaw, $30 Jake Arrieta, and $25 Noah Syndergaard. I supplemented those with $14 purchases of both Chris Archer and Zack Greinke. I also hope to play some matchups and maximize starts by plugging in some other cheap gambles ($1 Jose Berrios, Nate Karns, Erasmo Ramirez, and Jon Gray, and a $2 Michael Fulmer).

For relievers I noted that the Mariners play 30 home games while the average team plays only 25 or 26 games. Steve Cishek has surprised and looked decent. He’s also one of the cheaper full-time closer options at only $15. I also chose Zach Britton for $16 (many of the elite closers are $20+). Finally, I selected Dellin Betances to use if I don’t have a nice starter option or if I find that I need to battle teams in ratios late in the contest.

My Roster

Here’s my complete roster:

So, What Do You Think?

What would you have done differently? What approach would you take? Did you play the monthly game last season? Have any suggestions?