The Understanding Statistics Episode of the Beat the Shift Podcast – a baseball podcast for fantasy baseball players.
Guest: Mike Podhorzer
Bar Mitzvah Talk
Understandings Statistics Section
Read the rest of this entry »
Podcast (beat-the-shift): Play in new window | Download
Today, I share the final review of my pre-season Pod Projections posts. This time, we shift to a starting pitcher, Zach Plesac, whose original writeup is here. Plesac did post a sub-4.00 ERA during his 2019 debut, but it wasn’t backed by his skills, as he handily outperformed his ugly 5.13 SIERA. During the short 2020, he enjoyed a true breakout as his strikeout rate surged thanks to pitch mix changes. Let’s see how he did for an encore and how it compared to the projections.
Let’s continue reviewing the Pod Projections I shared early in the year. Today, I’ll review my Trent Grisham forecast. You can find the original writeup here. Grisham enjoyed somewhat of a fantasy breakout during the short 2020 season, as he went 10/10 over 252 plate appearances, putting him on a 20+/20+ pace over a full season. We fantasy owners salivate over that power/speed potential. Let’s see how he followed up and compare it to my projections and the rest of the forecasts.
As you probably already know, I manually project player performance each and every year, and make the forecasts available on my Pod Projections page. It’s a seriously time-consuming task, but the manual process gives me some advantages versus a computer system, so I continue to create them. Early in the year, I share a couple of my Pod Projections, the individual forecasted metrics, and an explanation of the process I follow to arrive at each number. This year, the first projection I shared was that of Ha-seong Kim, who had just signed a four year contract with the Padres after spending seven seasons in the KBO (Korean Baseball Organization). Projecting veteran baseball players is challenging enough, so you can imagine the added layer of difficulty when working on a forecast for a player coming over from a foreign league. Let’s find out how Kim performed compared to my projection and the two that were published in early January.
Last week, I reviewed my hitter Pod Projections vs Steamer projections comparisons. Let’s now move along to the starting pitchers and ERA. As a reminder from my original post:
Though Steamer is the best pitching projection system out there, it struggles on pitchers that have shown consistent BABIP and HR/FB rate suppression skills and deficiencies, as what usually works for the majority of pitchers — projecting a heavy dose of regression to the MLB mean — means it misses on those uncommon exceptions. I use Statcast’s xBABIP now for my projections, so I’m not afraid to forecast a mark that strays from the league average. However, I certainly still include some regression as we don’t always have enough batted balls in a pitcher’s history for that mark to stabilize.
A whopping eight years ago, I shared the hitter xK% metric I developed using a couple of our plate discipline metrics. It was quite good, using only three variables, but still had a strong R-squared of 0.81. Since then, I haven’t discussed it all that much, but still use it to help formulate my Pod Projections. However, I have actually been using an updated version that I had never shared and it’s even better. The comments on my recent xwOBA articles inspired me to finally reveal the latest and greatest version of the hitter xK% metric.
On Monday, I shared the names of eight pitchers whose Pod Projected ERA is significantly lower than Steamer. Today, let’s flip to the ERA downside names. Remember that in aggregate, Pod ERA projections are lower than Steamer, so the gap between ERA forecasts below are a lot smaller than on the upside list. Since it’s really relative projections and calculated dollar values that matter (we care how the projections compare to the player pool, not whether the pitcher is projected for a 3.00 ERA vs a 14.00 ERA), try to ignore the small degree Pod’s ERA is higher than Steamer and remember these are the largest outliers, so if put on the same ERA scale, the difference would be greater.
This week, I finish up the Pod vs Steamer series that pits my Pod Projections against the Steamer projections. Today, we move on to pitchers, where I’ll compare the ERA forecasts from each of the systems and identify those pitchers I am projecting for significantly better ERA marks. Though Steamer is the best pitching projection system out there, it struggles on pitchers that have shown consistent BABIP and HR/FB rate suppression skills and deficiencies, as what usually works for the majority of pitchers — projecting a heavy dose of regression to the MLB mean — means it misses on those uncommon exceptions. I use Statcast’s xBABIP now for my projections, so I’m not afraid to forecast a mark that strays from the league average. However, I certainly still include some regression as we don’t always have enough batted balls in a pitcher’s history for that mark to stabilize.
Yesterday, I compared my Pod Projections in the stolen base category to Steamer and identified five hitters I am forecasting for a meaningfully higher stolen base total. Today, let’s now look at the hitters I’m projecting for fewer stolen bases than Steamer. I’ll only highlight the fantasy relevant names as there are a number projected for limited playing time that aren’t worth discussing.
The Pod Projection process sharing continues! The 2021 forecasts are now available and include nearly 600 player lines. As usual, I’ll dive into my projection methodology (detailed in Projecting X 2.0) by sharing my process on several hitters and pitchers.
2021 Pod Projection Index