It's finally happening. Stats are coming in. Even Pitching WAR which is based on Deserved Run Average (DRA) is beginning to normalize. DRA typically needs a week or two to get started each year but is up and running for good by May, and often well before.
I've added new functionality to sort out your Top 12 Studs and Top 12 Duds based on WAR. That's right. Fun with Arrays and sortable data! The following is a screenshot of the current winners and losers, followed by an embedded link to the live stats. Studs and Duds by April 26th
Studs and Duds, Live stats
Early standout Studs include Eric Thames, known as 'The God' from his Korean gameplay and popular prospect Mitch Haniger. After Hitting MLB-Leading 11th HR, Thames has just been subjected to 2nd Random Drug Test.
The early Dud list is well-represented by Pitching and may be demonstrative of the time it takes for PWAR stat to normalize. Hitters include Leonys Martin and Kevin Siegrist who are off to abysmal starts to the year. Neither is ownable in a standard fantasy baseball league and are seemingly bad picks so far. Sorry Altuven Throat Singers and Cleveland Bombers! As always, let me know if you have an idea for another feature.
0 Comments
The WAR data streams running our game have been further optimized. You can find a comparison chart of all the different types of WAR and how they are calculated here.
We're using Baseball Prospectus' stat service so the stats you'll want to be familiar with are "Batter Wins Above Replacement Player" and "Pitcher Wins Above Replacement Player" or BWARP and PWARP. Wins Above Replacement Player is Prospectus' attempt at capturing a player' total value. This means considering playing time, position, batting, baserunning, and defense for batters, and role, innings pitched, and quality of performance for pitchers. Here is an example of the Wins Above Replacement Player spectrum based on the 2016 season: Excellent - Mookie Betts 10.3 Great - Jonathan Villar 4.75 Average -Scooter Gennett 2.03 Poor - Stephen Vogt 0.57 Horrendous - A.j. Pierzynski -1.78 Perhaps no sabermetric theory is more abstract than that of the replacement-level player. Essentially, replacement-level players are of a caliber so low that they are always available in the minor leagues because the players are well below major-league average. Prospectus' definition of replacement level contends that a team full of such players would win a little over 50 games. This is a notable increase in replacement level from previous editions of Wins Above Replacement Player. WARP components can be found in this article, which also describes 2015 changes to FRAA. We'll be using BWARP and PWARP. BWARP is merely the battter's WAR which they earn as while hitting and fielding. PWARP is the Pitchign side which relies on Deserved Run Average (DRA). DRA needs a week or two to get started each year but is up and running for good by May, and often well before. Deserved Run Average (DRA) uses a collection of mixed models to tease out the most likely contributions of pitchers to the run-scoring that occurs around them. Unlike other component metrics, DRA considers (and adjusts) home runs and balls in play, and achieves significant improved reliability over the raw versions of those and other statistics.
So what exactly does that LOOK LIKE?! If you want to see BWARP and PWARP in action, check out the following 2016 datastreams and how your favorite players stacked up:
PWARP BWARP Final Note: The game now uses scores a players contributions in both. We wanted to ensure that you get Madison Bumgarner's hitting contribution (an additional .87 WAR last year) if you draft him. Unfortunately, it also means when your position player gets called up to pitch, you'll get his likely negative impacts as well. For example, I was baffled to see Miguel Montero earned a -.1 in PWARP last year. Turns out, he got the call out of the pen. "The Citi Field scoreboard registered Miguel Montero's pitches as changeups. Actually, they were two-seam fastballs. " MORE FROM Fantasy Baseball WAR:
WAR is really just an idea of how to value players. Once you have the idea in place, there are dozens of implementation details that have multiple reasonable approaches to take. Baseball Reference has summarized the major frameworks below for easy comparison.
FanGraphs also has an explanation of their WAR and our previous WAR framework. Baseball Prospectus has their framework, WARP. I'm happy to add more WAR frameworks like Baseball Gauge's WAR if folks want to help me fill in the details below.
The BACKEND is UP!
After getting the proof of concept proven using Microsoft Excel, I uploaded it all to Google Sheets to ensure it worked in the cloud. It didn't. Google Sheets doesn't like many of the functions Excel used so we adjusted quite a bit. In the end, Sheets will be the way to go since Player's individual WAR and salary will need to be scraped daily from open source. As for now, I expect I'll just set up different leagues on my own. It's still a bit too easy to have things go squirrely if you don't know what to change on the backend. The solution will be to accept league's draft boards and input the Team Rosters manually. We could allow people to use the draft board we've used in the past (below), but it doesn't automatically populate the Team Rosters anyway. It's there if anyone wants it. Or at least until we program it into the backend and protect any fields that unwitting users might accidentally change and blow up the world. |
AuthorYour Commish! Archives
April 2017
|