How to Do a Deep Dive: Pitchers (2020 Fantasy Baseball)
If you want to dive deeper into fantasy baseball, be sure to check out our award-winning slate of Fantasy Baseball Tools as you prepare for your draft this season. From our Cheat Sheet Creator – which allows you to combine rankings from 100+ experts into one cheat sheet – to our Draft Assistant – that optimizes your picks with expert advice – we’ve got you covered this fantasy baseball draft season.
Giddy up! After discussing how to do a deep dive on hitters, it’s time to turn our attention to guys on the bump. We’ll focus on the standard 5×5 roto categories, and how sabermetrics help us project those categories. Hyun-Jin Ryu makes for a fascinating case study. He is a groundball pitcher with a deep pitch arsenal, and based on my thorough math, is tied with Wile E. Coyote for injuries.
Let’s dig in.
Hyun-Jin Ryu (SP – TOR)
My 162-game projection: 24 games started, 150 IP, 11 W, 3.44 ERA, 1.14 WHIP, 137 K (22.5% K-rate)
The industry is souring a bit on Ryu after his career year, as some analysts think that his ratios will suffer as a result of moving from Los Angeles to Toronto. Those reasons stem from Rogers Centre being a boon for homers and doubles, plus his infield defense is significantly worse in Toronto. However, some analysts think that he has a deep enough pitch mix to accumulate innings when he is healthy to provide above-average ratios while not killing you in strikeouts.
So, you want to make this decision on your own? I’ll give you the tools necessary to do so.
Vital sites: FantasyPros, FanGraphs, Baseball Savant, Pro Sports Transactions, Baseball-Reference, Mike Podhorzer’s Projecting X, Brooks Baseball
A crucial part of the projection is determining how many innings we can expect Ryu to accumulate. Consider the following.
Looking at Ryu’s Pro Sports Transactions page, we can see all of the injuries that Ryu has dealt with. The Blue Jays will trot him out there every chance they get, but the injury risk is too high to lock Ryu in for more than 150 innings in a normal season. If you make the decision that Ryu can stay off the IL, you will be higher on him than the vast majority.
Games started/ Innings pitched: Taking a three-year average for veterans is a good foundation. Young starting pitchers typically see a 20% increase in workload year-over-year, so be cautious of projections that exceed that threshold.
To get to a total strikeout number, let’s explore Ry’s projected strikeout rate. To do this, head to FanGraphs for his previous strikeout rates.
Since there is so much volatility, we have to dive deeper. Let’s consider his plate discipline metrics from FanGraphs.
|Year||Swing %||Contact %||SwStr%|
The trend here is that batters are swinging and making contact on more of Ryu’s pitches. His swinging-strike rate (SwStr%) is relatively stable over this three-year period. Using this set of information, Ryu’s dip in strikeout rate from 2018-2019 makes sense. If you are short on time, the easiest way to approximate strikeout rate is to take SwStr% and multiply it by two. You can see in 2017 and 2019 that this happened with Ryu, and why Ryu’s 2018 strikeout rate is an outlier. I like 2019’s 22.5% K-rate as a good baseline.
To confirm that baseline, I’ll look at average fastball velocity on Baseball Savant. Using the Pitch Tracking table, we can see that Ryu’s fastball velocity has remained steady, and the rest of his pitches are +/- within one MPH. I will also review year-over-year changes in both pitch mix and pitch effectiveness – we will dive in on this in the ERA section.
Now that we’ve got Ryu’s K-rate, we need to project total batters faced (TBF). Mike Podhorzer’s “Projecting X” does a great job of calculating this in-depth, and we can also use the publicly available projections. Consider the following:
These projections differ in both strikeout rate and innings pitched. However, you can easily convert TBF onto a per-inning basis (TBF/innings), and you’ll arrive at an average of 4.15. Multiply that 4.15 by the number of innings you expect Ryu to pitch, and then multiply by your projected K-rate to arrive at total strikeouts.
Strikeouts: Using plate discipline metrics can help project strikeout rate. However, a change in pitch arsenal or velocity can change a K-rate drastically.
Next, let’s discuss walks and hits per innings pitched, or WHIP. If you are not creating your own projections, determining a range of a pitcher’s WHIP is a good compromise. Like strikeouts, walks are best determined through a rate. Here are Ryu’s walk rates and WHIP over the last three years.
Wow, Ryu’s walk rates are all over the place, and it’s tempting to say that we can expect a continued decline based on this trend alone. Based on his plate discipline metrics (more swings, more contact), a declining walk rate certainly makes sense. Further, Eno Sarris has looked at Ryu’s command metrics developed by STATS LLC and determined that Ryu’s command is 12% above league-average. Ryu ranked second in walk rate among qualified starters in 2019. While I wouldn’t project Ryu to have another career year, using a baseline of 4.5% is reasonable.
To continue the research on WHIP, let’s focus on Ryu’s BABIP and barrel rate. While homers are an important part of both WHIP and ERA, HR/FB rate is not a predictive metric. Instead, let’s focus on barrel rate, which is more predictive, and hard-hit rate. We can get hard-hit rate and BABIP from FanGraphs, and barrel rate from Baseball Savant.
|Year||BABIP||Barrel Rate||Hard-hit rate|
It makes perfect sense that Ryu’s BABIP has decreased as his barrel and hard-hit rates have also decreased. In fact, Ryu posted well-above-average barrel rates and exit velocities in 2019. If Ryu is able to put the ball where he wants to (based on his Command+), we can induce that Ryu’s ability to get weak contact is real.
However, getting a defense to field that weak contact is also important. Defensive metrics are not as reliable as hitting and pitching metrics, but no matter which way you slice the data, the Blue Jays were one of the worst defensive teams in the league last year. If you dig deeper, you will see that each projected starter for the Jays’ infield was average or below average according to UZR/150. To add a sour cherry on top, Rogers Centre is a much better offensive environment than Dodger Stadium, according to FantasyPros Park Factors.
More hits and homers at his new home park (along with a bevy of hitter’s parks in his division) translates into a higher BABIP. While it doesn’t make sense to project another career year from a 33-year-old, posting a WHIP between 1.10-1.15 is realistic.
WHIP: Don’t get lost in the weeds. Break down this category by walks and hits, using the factors above to develop a projected range.
Like WHIP, ERA can be a very labor-intensive projection, so it’s best to develop a range. At the end of the day, a good ERA is a result of keeping the ball in the yard, inducing weak contact, and keeping runners off the base paths. Being an innings-eater also helps ERA – this is why guys like Kyle Hendricks and Mike Soroka are so valuable. The best way to pitch deep into ballgames is to have a deep pitch mix. In 2019, Ryu featured five pitches that he threw at least ten percent of the time.
To understand why these pitches are effective, we can dive into spin rate, vertical/horizontal movement, and “tunneling,” the ability to make different pitches look alike until the hitter has to decide to swing. Spin rate and vertical/horizontal movement are easily found on Baseball Savant at the bottom of each player’s Statcast page. Brooks Baseball also has easy-to-navigate charts and tables to understand a pitcher’s zone profile, sabermetric outcomes, and scatter charts of vertical and horizontal movement. Looking at these stats year-over-year is also important if a pitcher had drastically different stat lines in recent seasons. While pitchers often speak a different language, each website has legends/keys to understanding how good or bad a certain pitch profile is.
Realistically, you probably don’t have the time to do a deep dive at this level. Instead, you can focus on several ERA indicators to come up with an ERA projection.
|ERA (FanGraphs)||xERA (Baseball Savant)||FIP (FanGraphs)||xFIP (FanGraphs)||SIERA (FanGraphs)|
I recommend reviewing each website’s glossary to get a deeper understanding of each ERA indicator. SIERA has proven to be the most predictive metric of ERA, so I tend to lean on SIERA for projections. FIP and xFIP remove balls in play, so SIERA can be more helpful for a groundball pitcher like Ryu. Based on Ryu consistently outperforming his indicators, projecting him for an ERA in the low-to-mid 3’s is reasonable.
ERA: For a deep dive, review the ERA indicators and understand why they differ. For a deeper dive, research each pitch’s effectiveness and understand why it’s effective.
If you’ve gotten this far – congrats! I send my best wishes because we are on to projecting wins. Frankly, there hasn’t been a great way to forecast this category. There are complex formulas you can run based on runs-per-game scored and against, but even those calculations do not predict wins well. FanGraphs, Clay Davenport, and Baseball Prospectus all have projected runs per game for each team if you dare go down this road.
Alternatively, reviewing the publicly available projections may be the most efficient route.
|THE BAT||ATC||Depth Charts||Steamer||ZiPS|
Based on the above projections, 11 wins for Ryu is the way to go.
Wins: If you are deciding against two pitchers with similar ratios and strikeouts, oftentimes the pitcher with more innings and/or on the better team will pick up more wins.
While many casual fans of baseball know that these websites exist, understanding the key metrics on these sites (including ours) will give you an edge on your teammates. The deeper the dives you can do in the offseason, the better you will be prepared for draft day.