Starters: The Toronto Blue Jays’ Rotation is Very Good

Ray

Ray has had a phenomenal second half to the 2021 season. His performance has been so good that media members often mention him as a serious candidate to win the American League’s Cy Young Award. According to Tables 2 and 3, his HR/9 has improved from 11th percentile in the first half of the season to 79^th. Rays’ ERA, xERA, FIP and fWAR are all in the Top Fifth percentile. Ray also continued to limit the number of walks (72^nd percentile in BB/9), which plagued him in 2020. His pitch mix has changed a little during the July to September period. Ray’s usage of 4-seam fastballs is six percentage points higher than it was in the first half, offset by a similar reduction in breaking pitches.

Berrios

The second half data from Table 3 that I presented includes five starts for Minnesota and nine for Toronto. His ERA, xERA and FIP reside in a #2 starter slot, but his fWAR places him in #1 starter space. Berrios had a few sub-par outings as a Jay, but he bounced back nicely.

Manoah

Manoah burst onto the scene on May 27, and he has continued to pitch well. His May 27 to June 30 FIP was 21^st percentile, but it now resides in the 85^th percentile for the July to September segment. The improvement is due to better BB/9 and HR/9, as depicted in Tables 2 and 3.

Manoah’s pitch mix has changed significantly during the July to September period. His utilization of the 4-seam fastball has decreased by ten percentage points; the corresponding change was an 11 point increase in his sinker.

Another aspect to note is that Manoah’s Bauer Units (Spin Rate / Pitch Velocity) have declined by at least one unit for all of his pitches since June 30. Many people have associated a decrease of this magnitude with the cessation of sticky substance usage. However, Manoah has continued to pitch well, and he has been a dependable pitcher from start to start.

Matz

Matz’s first-half performance may have been a victim of bad luck: his ERA was 47th percentile, but his xERA was 69^th. In the July to September period, Lady Luck smiled upon Matz: his xERA remained at the 69^th percentile level, but his ERA was 84^th. His HR/9 and fWAR also saw improvement: HR/9 rose from 59^th percentile to 79^th, and his fWAR increased to 79^th from 64^th.

Compared to the April to June period, Matz increased his sinker usage by thirteen percentage points. He reduced the rate of breaking pitches and off-speed offerings by eleven and two percentage points, respectively. The pitch mix changes may be why Matz’s K/9 was 72^nd percentile in the first half of the season but 36^th in the second half.

Ryu

Ryu’s K/9 improved in the second half from 26^th percentile to 48^th. However, his ERA slipped from 79^th percentile to 35^th, and his xERA declined from 69^th percentile to 51^st. On the good news front, Ryu’s fWAR increased from its first half 64^th percentile to 75^th.

Ryu’s pitch mix has changed since the period ended June 30. He has increased his 4-seam usage from 33% to 40%, offset by a six percentage point reduction in off-speed offerings (from 28% to 22%) and breaking pitches (from 14% to 12%).

Ryu’s off-speed and breaking pitches have not been as effective this season as they were in 2020 and 2019. Table 5 illustrates the OPS and xOPS against his pitches for the 2019-2021 period. Compared to 2019 and 2020, batters have fared much better in 2021 against Ryu’s off-speed pitches. The OPS and xOPS marks have been a season-long challenge for Ryu: the OPS against, expressed as a percentage of the MLB average on off-speed pitches, was 111% in the first half and 108% in the second.

In locating pitches, the 2021 percentage of changeups situated in the Heart, Shadow, Chase and Waste Zones are very similar to 2020. However, the vertical movement on Ryu’s changeup has deteriorated compared to its 2020 and 2019 levels. Although the 2021 movement remains above-average, Ryu’s changeup moves relatively less than it did during the two prior seasons. Less movement may account for the higher OPS and xOPS marks concerning Ryu’s changeup.

The curveball is another matter. The movement characteristics of this pitch are very similar to those of 2020 and 2019. However, Ryu appears to have less command of the pitch. In 2020, 16% of his curveballs were in the Heart Zone; it is 26% in 2021. Also, Ryu located 35% of his curveballs in the Shadow Zone, which is less than 2020’s 41%.

Ryu’s four-seam fastball, which batters have recorded a 0.815 OPS against (0.926 in 2020), is finding the Heart Zone a little more frequently in 2021 (30%) than it did in 2020 (27%). The percentage in the Shadow Zone has risen from 2020’s 44% to 51% this season.

The cutter, an effective pitch this season (0.639 OPS), has found the Heart Zone 26% of the time, less than 2020’s 30%. The percentage in the Shadow Zone has not changed much from 2020.

In summary, Ryu’s 2021 woes pertain to the following factors:

• Changeup command is similar to 2020, but its movement has declined since last season
• Ryu’s command of the curveball is not as good as it was in 2020
• In 2019, the OPS and xOPS against Ryu’s four-seam fastball were 0.636 and 0.658, respectively. The OPS in 2021 and 2020 was 0.815 and 0.926, respectively. Respectfully, why the heck is Ryu using the four-seam fastball so much?
• The cutter has been an effective pitch in 2021 (OPS – 0.639; xOPS – 0.675). More cutters, please.

Game Score

Bill James developed the original Game Score metric as a single measure of how well a starter performed in each start. The metric includes the number of outs, strikeouts, walks, hits, runs and home runs. FanGraphs subsequently refined the original metric.

FanGraphs also developed a rule of thumb that one can use to gauge a pitcher’s performance. Readers should note that the Game Score calculation includes a constant such that the MLB average Game Score is 50. Game Score is an excellent metric to assess the quality and consistency of a starter’s performance.

My objective was to see how consistent the Jays’ starters were when compared to similar pitchers. To accomplish that goal, I did the following:

• Using data that included September 10 starts, I downloaded ERA data of MLB starters with at least 20 innings pitched during the 2021 season (139 pitchers).
• For each Blue Jay, I selected five starters with ERAs immediately ahead of them and five immediately below.
• The ERAs of Ryu, Matz and Manoah were very close. Therefore, there were fifteen pitchers in their group. There were eleven pitchers in each of the groups that included Ray, Berrios and Stripling.
• For the 48 pitchers, I downloaded their Game Scores for the season up to and including September 10 games.
• For each pitcher, I calculated the percentage of starts that fall in the ranges noted in Table 6.

Generally speaking, starters have a mix of outings that range from horrible to outstanding. A good measure of a starter’s consistency is the number of starts that received a Game Score higher than 50, and I expressed this number as a percentage of all outings.

Table 7 shows how each Jays’ pitcher fared when compared to similar starters. Expressed as a percentage of all outings, every one of Toronto’s starters has Game Scores in the 51 to 100 category that is median or higher. Concerning 2021 Game Scores by game for Ray, Berrios, Manoah, Matz and Ryu, click on the player’s name to see the chart. The charts include data up to and including September 17.

Ryu’s consistency/reliability deserves more analysis. Before his September 11 and 17 outings, Ryu ranked 5^th among Tier 3 pitchers in Game Scores in the 51 to 70 segment. Also, he was 3^rd of 15 in the 71 to 100 category. That is consistent performance. Then the last two starts happened.

When I added the Game Scores from his two most recent outings to Ryu’s total, his 51 to 100 category percentage dropped from 70% to 66%. Ryu’s Game Scores in the 0 to 30 segment rose to 20% from 15%. That 20% mark is the 14^th of 15 Tier 3 pitchers. Not good.

For the season, Ryu has six starts that garnered Game Scores less than 30. Of those, five occurred after June 30. In my opinion, until he returns to form, Ryu is currently not a dependable starter.

I examined Ryu’s performance on four, five and six or more days of rest. Table 8 shows that Ryu has generally performed better on five days rest than on four. His lowest Game Score on five days rest was 39, which occurred after June 30. Ryu’s other second-half Game Scores on five days rest are 60, 53 and 76. My original thought was that the Jays might give Ryu an extra day between starts during the remainder of the 2021 season. That idea is now moot because the Blue Jays announced on September 19 that the club placed Ryu on the Injured List.

The Last Word

The Blue Jays starting rotation has excelled during the second half of the 2021 season. Ray’s performance has landed him in the Cy Young Award discussion, and Berrios, Manoah and Matz have pitched well. Ryu has stumbled after June 30 and posted most of his worst 2021 game performances during the July-September period. Let’s hope Ryu can recover his mojo for the balance of the 2021 campaign.