Back in 2018, I introduced a game theory approach for comparing baseball projection systems. Proudly, the article was nominated for Baseball Article of the Year by the Fantasy Sports Writers Association (FSWA). The game theory methodology is now back for its third straight year.
This approach is not the standard projections comparison analysis that most others embark on. The typical comparison makes use of some type of statistical measure. The standard analysis involves calculating least square errors, performing chi-squared tests, or perhaps even hypothesis testing. My method does not use any of these capable methods.
Read the rest of this entry »
On June 23, Commissioner of Baseball Robert D. Manfred, Jr. announced that Major League Baseball would begin its 2020 regular season on July 23rd. It submitted a 60-game regular season schedule for review by the MLB Players Association. The proposed schedule featured divisional play, with the remaining games being played against their opposite league’s corresponding geographical division.
That 60-game proposal came to fruition. It was an unusual season to say the least. The St. Louis Cardinals did not play a game from July 30th through August 14th. Doubleheader games were all seven innings each, and extra innings started with a runner on 2nd base. The designated hitter was in effect for the National League, and so on, and so forth. In the end, the season came and went, and the Los Angeles Dodgers were crowned as champions of the Fall Classic.
Now we are squarely in the midst of the baseball offseason. Most fantasy baseball players are on holiday from their annual game, eagerly awaiting one of the most important ingredients to their annual draft preparation …
Projections.
For many (including myself), player projections are the backbone that form the strategies and planning for the upcoming fantasy baseball season. Understanding how player statistics are forecasted for the coming season is the essential part of fantasy preparation.
Previously, I looked at the largest auction player bargains of 2020. These were the players who were highly profitable after considering their opportunity cost of acquisition. Value should always be considered relative to cost.
We defined the bargain amount as:
$Bargain = $Value – $AAV
We defined $Value as the accumulated 5×5 full season rotisserie value of each player, and $AAV as the average auction cost to purchase the player pre-season. We made use of the July NFBC Average Auction Values, which was one of the best sources of “market” data this year.
Whereas I previously looked at the players who generated the most excess value in 2020, today’s attention will be directed to what I refer to as the value drainers. These are the largest “rip-offs” of the season – i.e., the players who earned the most negative profits for fantasy owners on a full season basis (net of their auction price).
Prior to unveiling 2020’s most unprofitable players, it is important to discuss one additional step in the analysis – the capping of values. I have previously spoken about this concept, but I will touch on it again today.
Eduardo Rodriguez was a player that I drafted on a few of my fantasy rosters this season. His NFBC average auction value during July drafts (auctions) was $7. In Tout Wars, I acquired the Boston pitcher for $10. Unfortunately, Rodriguez came down with COVID-19. He developed heart complications due to the virus, and consequently did not pitch a single inning in 2020.
The question is – what value did Rodriguez accumulate in 2020? What damage did he cost to your team’s aggregate value? Owners certainly lost their original investment on him, but how much more were they penalized? He wouldn’t have made it to one’s active roster – but how much did it cost owners for Eduardo taking up a bench spot?
The key to succeeding in fantasy baseball:
Maximize the value of your accumulated roster.
At the start of a draft, each fantasy owner is handed a set of draft picks. Each owner receives a 1st round selection, a 2nd round selection, a 3rd round selection, and so on. If your league chooses to hold an auction rather than a more traditional serpentine draft – each team is handed $260 at the auction start. Players are then purchased throughout the auction with the use of these finite funds.
The key to gainfully drafting is not to draft a 3rd round player in the 3rd round, or a 9th round player in the 9th round, etc. The key is to draft a 3rd round player in the 10th round, and a 9th round player in the 20th round.
In an auction, if you purchase every player at his projected value, you will have paid $260 of auction dollars for $260 of value. What you will have is an average team. You won’t finish last, but you won’t finish first. Instead, with your $260 – you need to buy some $290 or $300 or $310+ of total value.
The key is to make a “profit” on as many roster spots as you can. The goal is to purchase players at bargain prices.
I have asked this question before – but it is worth asking every now and again. Suppose that you competed in an NFBC fantasy baseball auction back in July this season.
Which player was the better purchase?
Bryce Harper (OF, PHI)
OR
Andrew McCutchen (OF, PHI)
Before opining on the better Philly outfield purchase of 2020, let’s take a look at their final 2020 stat lines:
On the surface, it seems like a pretty obvious answer. Harper had more HR, SB, R and a better batting average than McCutchen. He had just one fewer RBI.
The 2020 MLB regular season has now concluded. In most years, this introductory sentence would be a simple fact. One ordinarily would not pay much attention to such an evident truth. However, in 2020, the consequence of baseball completing the year without a major full stop is a sparkling achievement.
Yes, the Marlins and Cardinals did not play for the course of about a week due to team COVID infections. Yes, there were more make-up doubleheaders played in 2020 than in any season during my lifetime. Yes, there were a few teams that made the playoffs despite a losing record. Yes, the league-wide batting average of .245 was the 6th lowest full-season mark since 1900.
But baseball made it through, and now embarks on their expanded playoffs journey.
As such, it is now time to check back on how we fared in the fantasy season. For me personally, it was a rather positive one. I did not finish below 6th place in any league that I played in this year. Amazingly, I was crowned as the 2020 Tout Wars Head to Head League Champion, my very first expert league title. 2020 showed that the ATC projections work well, even in smaller sample sizes.
In today’s article, I will recap my 2020 bold predictions. To remind the reader, the goal at the outset was to predict 70th to 90th percentile events (10% to 30% likely occurrences). I don’t expect to get the majority of these correct. If I wanted to achieve a higher success rate, I would simply have predicted that Jacob deGrom would win the Cy Young award, and the like.
Now let’s recap! Read the rest of this entry »
This is the fourth article in my wPDI vs. CSW series. You can catch up by reading the first three articles – on called strikes, whiffs and residuals.
Here is a quick summary of some of the basics of wPDI & CSW from this series:
Last year, I developed the Weighted Plate Discipline Index (wPDI) framework, whereby all pitches can be classified into six different outcomes as follows:
Each outcome is then assigned a weight, or an index. A% through F% are the percent of pitches thrown in each outcome. The general formula for wPDI, the Weighted Plate Discipline Index is given as:
wPDI = IndexA * A% + IndexB * B% + IndexC * C% + IndexD * D% + IndexE * E% + IndexF * F%
wPDI can generate an all-in-one sortable metric used to evaluate pitchers. The plate discipline framework may be tailored to mimic (or to correlate to) various measures of deception or effectiveness.
In the first three articles of this series, we developed indices for wPDI to approximate the PitcherList metric, CSW. The Called Strikes + Whiffs (CSW) statistic was featured in last year’s FSWA Research Article of the Year by Alex Fast, and is defined as:
Called Strikes + Whiffs Total Pitches
We separately tacked the called strikes and whiffs components, and landed on the following wPDI equation to represent CSW: Read the rest of this entry »
This is the third article in my series – wPDI & CSW. You can catch up by reading the first two articles – on called strikes and whiffs – found here and here.
Here is a quick recap of what we have covered so far:
In this series, we are looking at the PitcherList metric, CSW and how it relates to my plate discipline framework, wPDI. Last year’s FSWA Research Article of the Year by Alex Fast featured CSW, which is defined as:
With the Weighted Plate Discipline Index (wPDI) framework, all pitches are classified into six different outcomes as follows:
This is the second article of my series – wPDI vs. CSW. For those new to either metric, I will quickly catch you up. [The opening article can be found here.]
In last year’s FSWA Research Article of the Year, CSW Rate: An Intro to an Important New Metric, Alex Fast of PitcherList examines his site’s pitching statistic, CSW. The short and simple formula for CSW is defined as follows:
Independently, I came up with the concept of Weighted Plate Discipline Index (wPDI). With wPDI, we ask just three questions, or three binary events for every pitch:
Every pitch can then be classified into 6 possible pitching outcomes based on the above. The definition of each outcome is as follows:
Each outcome is then assigned a weight, or an index. The formula for wPDI, the Weighted Plate Discipline Index is then given as:
A% through F% are the percent of pitches thrown in each outcome, and the indexes are linear multipliers to obtain the aggregated, sortable metric.
What CSW has most in common with wPDI, is that it shares the same denominator – Total Pitches. That being the case, we can attempt to use the wPDI framework to express the PitcherList metric. CSW is rooted in Baseball Savant data, while wPDI is fed by FanGraphs figures. By exploring the similarities and differences between the metrics, we can also uncover some great nuggets of understanding.
Last year’s FSWA Research Article of the Year, CSW Rate: An Intro to an Important New Metric, was awarded to Alex Fast of PitcherList. In his article, Alex presents the pitching statistic, CSW – a metric which was originally coined and created by Nick Pollack in 2018. As cited in the author’s article summary, CSW is more predictive than Swinging Strike Rate (SwStr%), and is more descriptive than Whiff Rate (Whiff%).
The short and simple formula for CSW is defined as follows:
I enjoy elegant formulae. Sure – wOBA, wRC+ and the like are extraordinary metrics in their own right, but they are not the simplest to jot down. CSW is plain, simple, easy to understand, and nicely predictive.
Coincidentally, and unknowing of CSW, I came up with the concept of wPDI back in 2018. I then published my first works of the plate discipline framework on April 2, 2019. The original article was entitled Introducing: Weighted Plate Discipline Index (wPDI) for Pitchers, and can be found here.
What jumped out to me immediately upon reading Fasts’s article – was that the two metrics have something very in common. CSW and wPDI both share the very same denominator – Total Pitches. The base of both of our metrics are identical. Both utilize the very same sample size, both stabilize just as quickly, and both describe baseball through the very same lens – the pitch.
As a quick reminder of how wPDI works, every pitch can be classified into 6 possible pitching outcomes.
We have a shortened season which brings a lot of bold predictions into play. In fact, it might actually keep them from being bold which means we’ll have to amp up the boldness. I’ve got five I think are bold enough to qualify here. Let me know what you think in the comments and include your own big time bold prediction. Again, it has to be feasible but not obvious. Franmil Reyes isn’t projected to lead baseball in homers, but it’s not terribly bold to pick him, either. My boy Frankie Montas is a legitimate AL Cy Young candidate so while he’s far from the favorite, picking him wouldn’t really catch anyone’s attention.
Ramón Laureano is the #1 OF
The 25-year old power-speed stud for the A’s broke out with a 126 wRC+, .288 AVG, 24 HR, and 13 SB in 481 PA last year, good for 32nd on the Auction Calculator at the position with volume no doubt holding him back. With health, he’ll play close to all 60 games as a stud defender in center for Oakland which will give him a real chance to be tops at the position. He’s projected to finish 32nd again in the ATC projections thanks to a major dip in AVG, but I think he can deliver something special like a .320 AVG, 18 HR, and 15 SB with 37 R and 30 RBI.