Effects of Stats So Far on Depth Charts RoS Projections — Pitchers

Previously, I looked at hitters and how their current in-season performance has changed their rest-of-season projections. Here, I do the same for pitchers.

I limited my analysis to starting pitchers. For starting pitchers, I compared their preseason Depth Charts projections to their rest-of-season Depth Charts projections, then found the five starters with the biggest positive and negative differences in their projected ERA. All statistics are from May 27^th.

The Biggest Losers

 

Adam Warren (+0.29 in ERA)

 

50.7 IP, 3.91 ERA, 4.54 FIP, 4.42 xFIP, 1.24 WHIP—current

166 IP, 3.81 ERA, 1.28 WHIP—preseason

115 IP, 4.10 ERA, 1.33 WHIP—rest-of-season

 

Used exclusively as a reliever in 2014, Adam Warren had a career-high 8.7 K/9 and career-low 0.5 HR/9. As a starting pitcher this year, Warren has seen his K/9 drop to 5.7 and his HR/9 go up to 1.1. Even though he has a 3.91 ERA, his 4.54 FIP and 4.42 xFIP explain why his rest-of-season ERA projection has increased by 0.29 from his preseason projection. He’s lost a little more than two miles per hour on his fastball, from 94.2 to 91.9 but has increased the use of the fastball at the expense of his slider and has seen his contact percentage go up from 76% in 2014 to 81% this year. Warren is a good illustration of the difficult transition from reliever to starter. As a reliever, he threw harder and used his slider more and struck out more batters in short stints. As a starter, he’s throwing his slider less, using more two-seam fastballs and generating fewer strikeouts.

Phil Hughes (+0.29)

 

64.7 IP, 4.59 ERA, 4.60 FIP, 4.16 xFIP, 1.29 WHIP—current

205 IP, 3.66 ERA, 1.15 WHIP—preseason

144 IP, 3.95 ERA, 1.19 WHIP—rest-of-season

 

Phil Hughes had an unreal 1.9% walk rate last year that was coupled with the best strikeout rate he’s had as a starting pitcher (21.8%). This year, he’s still been quite stingy with the walks (2.2% BB%) but his strikeout rate has dropped to 15.1%, his lowest mark since 2011. The biggest change in his pitch types (per PITCHf/x) has been a dramatic rise in the percentage of two-seam fastballs, from 5.2% last year to 23.9% this year, with a drop in four-seam fastballs (61.8% to 50.2%) and cutters (16.7% to 12.8%). He’s also lost about one mile per hour on his fastball. Batters are making more contact on his pitches and they are doing damage with the long ball, which had long been a problem for Hughes before last season. From 2010 to 2013, Hughes allowed 1.4 HR/9 and had a HR/FB rate of 10.8%. Last season, those numbers dropped to 0.7 HR/9 and 6.2%. This year, he’s back to his old ways, giving up 1.6 HR/9 and a 12.3% HR/FB. Looking at Hughes’ last six years as a starting pitcher, it’s clear that his 2014 season was the aberration.

Matt Garza (+0.29)

 

57 IP, 6.00 ERA, 5.31 FIP, 4.37 xFIP, 1.58 WHIP—current

185 IP, 3.96 ERA, 1.26 WHIP—preseason

124 IP, 4.25 ERA, 1.32 WHIP—rest-of-season

 

Garza came into this season with a walk rate consistently in the range of 7.5% over the previous five seasons but has seen that rate jump to 10.5% so far this year. He also may be experiencing age related decline as his strikeout rate is on pace to decline for the fourth straight year right along with his fastball velocity, which has also gradually declined over that time from 93.8 miles per hour in 2011 to 92.5 this year. Like other pitchers in the “biggest losers” category, Garza has had trouble with home runs, giving up 10 so far in 57 innings with a career-high 18.2% HR/FB rate. Last year, pitching for the same team in the same ballpark, Garza allowed just 12 home runs in 163 1/3 innings (7.0% HR/FB). He’s also had trouble stranding runners for the second year in a row (67.5% LOB% this year, 66.6% last year).

R.A. Dickey (+0.28)

 

64 IP, 5.77 ERA, 5.70 FIP, 4.80 xFIP, 1.33 WHIP—current

209 IP, 4.08 ERA, 1.28 WHIP—preseason

140 IP, 4.36 ERA, 1.33 WHIP—rest-of-season

 

Dickey’s preseason Depth Charts projection called for a 7.1 K/9, which was reasonable considering he’d been over seven strikeouts per nine in each of the last three seasons. Unfortunately, the knuckleballer has seen his strikeout rate plummet this year. Through 10 starts, he’s at 4.9 K/9. Dickey has also seen his walk rate go up in each of the last three seasons and he’s giving up more homers than he has since establishing himself as an above-average starting pitcher with the Mets in 2010. So, strikeouts down, walks and homers up, that’s a recipe for disaster.

Taijuan Walker (+0.28)

 

43 IP, 7.33 ERA, 5.48 FIP, 4.84 xFIP, 1.84 WHIP—current

158 IP, 3.98 ERA, 1.30 WHIP—preseason

116 IP, 4.26 ERA, 1.34 WHIP—rest-of-season

 

Taijuan Walker pitched well in 11 games and 53 innings over his first two partial seasons in the major leagues at 21 and 22 years old (2.89 ERA, 3.28 FIP, 1.21 WHIP). He then ramped up the hype machine with a sterling spring training this year that saw him give up just two runs in 27 innings with 26 strikeouts and five walks. He had a nice projection for a 23-year-old pitcher coming into this season. Unfortunately, he’s been lit up so far in 2015, posting a 7.33 ERA and 1.84 WHIP. His .356 BABIP and 63% LOB% are a big part of the problem, as are the eight home runs allowed in 43 innings (1.7 HR/9, 14.5% HR/FB). His FIP (5.48) and xFIP (4.84) are much better than his actual ERA, but still nothing to be excited about, which is why his rest-of-season projection for ERA is 0.28 higher than it was before the season started.

That 14.5% HR/FB should come down but Walker has also allowed many more fly balls than he did last season, up from 26.2% to 39.0% so he is likely to give up more home runs than he has in the past even with regression in his HR/FB rate. He’s also giving up a higher percentage of hard-hit balls, increasing from 23.8% last year to 34.3% this year. On the bright side, his fastball velocity has stayed at the same level since his rookie year. Walker still has the raw skills to be a good major-league pitcher but a combination of giving up more fly balls and seeing more fly over the fence, along with a below-average left-on-base percentage has really hurt him this season.

 

The Biggest Winners

 

Michael Pineda (-0.40)

 

64.3 IP, 3.36 ERA, 2.54 FIP, 2.52 xFIP, 1.14 WHIP—current

142 IP, 3.74 ERA, 1.20 WHIP—preseason

117 IP, 3.34 ERA, 1.12 WHIP—rest-of-season

 

Last year, Phil Hughes had the second-lowest walk rate of any pitcher who qualified for the ERA title since 2001. He walked just 1.9% of the batters he faced. Only Carlos Silva in 2005 (1.2%) has had a better rate over the last 15 years. This year, Michael Pineda is walking even fewer batters than Hughes did last year, allowing just 1.7% of the batters he’s faced to reach via the base on balls. Pineda has the added bonus of a 25% strikeout rate to go with that miniscule walk rate. Along with elite strikeout and walk rates, Pineda has added the ability to induce groundballs to his arsenal. He came into this year with a career groundball rate of 37.2%. This year, his groundball rate is up to 52.4%. He’s turned into Felix Hernandez Lite. Imagine if the two of them were on the same staff. Jesus (Montero) that would be nice for Mariner fans. Pineda’s 3.34 ERA is actually higher than you’d expect based on his peripherals, mainly due to a .339 BABIP and 68.3% left on base percentage.

Jake Odorizzi (-0.24)

 

66.3 IP, 2.31 ERA, 2.52 FIP, 3.61 xFIP, 0.96 WHIP—current

178 IP, 4.05 ERA, 1.31 WHIP—preseason

136 IP, 3.81 ERA, 1.27 WHIP—rest-of-season

 

Jake Odorizzi has seen his strikeout rate drop from last year’s 24.2% to 20.2% this year but he’s offset that with a walk rate that has dropped at an even greater rate, from 8.2% last year to 4.6% so far this year. According to PITCHf/x Pitch Types, he’s changed his pitch arsenal in a number of ways. Consider the chart below:

Odorizzi has dropped his usage of the four-seam fastball and almost completely eliminated his slider and increased the use of his two-seamer, cutter, and splitter. The result has been a lower strikeout rate but better control and an increase in ground ball percentage, from 29.9% last year to 41.2% this year. Last year, Odorizzi gave up the second-highest percentage of fly balls of any pitcher who qualified for the ERA title, at 48.7%, behind only Chris Young (58.7%). This year, he’s 34^th out of 109 pitchers who qualify. Fewer fly balls generally mean fewer home runs allowed and Odorizzi has seen a big drop there also. Last year, he gave up 20 homers in 168 innings. This year, he’s allowed just two long balls in 66 1/3 innings, although a very low 2.8% HR/FB rate is part of that decrease and he’s not likely to sustain a home run per fly ball rate that low for the whole season. Still, despite the lower strikeout rate, Odorizzi has been even more effective this year than previously because he’s made up for it with better control and more ground balls.

Jake Arrieta (-0.23)

 

58 IP, 2.95 ERA, 2.39 FIP, 2.68 xFIP, 1.09 WHIP—current

173 IP, 3.62 ERA, 1.25 WHIP—preseason

127 IP, 3.39 ERA, 1.21 WHIP—rest-of-season

 

Over the first four years of his career, Jake Arrieta had a 5.23 ERA (4.75 FIP) and 1.43 WHIP. He struck out 17.5% of the batters he faced, walked 10.2%, and induced ground balls on 43.3% of his balls in play. Since the beginning of 2014, Arrieta has a 2.64 ERA (2.29 FIP), and 1.02 WHIP. He’s struck out 27.1% of the batters he’s faced, walked 6.5%, and has a 49.8% ground ball rate. All of the most important things you want a pitcher to do better, he’s done better. He’s done this by changing his pitch arsenal. If you go by Baseball Info Solutions pitch types, Arrieta started throwing cutters nearly 30% of the time in 2014 and has continued to do so this season after not throwing any cutters from 2010 to 2012 and throwing it 6% of the time in 2013. According to PITCHf/x pitch types, he went from throwing sliders roughly 14% of the time prior to the 2014 season to nearly 30% of the time this year and last. I don’t know enough about the difference in classification between the two sources but Arrieta made a change of some sort and it is working. It may be more cutters, it may be more sliders. Maybe it’s a slutter. Whatever it is, this change in pitch type that has resulted in much improved numbers since the beginning of last season would suggest to me that even that improved rest-of-season projection is likely going to come in on the high side for Arrieta.

Aaron Harang (-0.22)

 

65 1/3 IP, 1.93 ERA, 2.86 FIP, 4.30 xFIP, 1.03 WHIP—current

155 IP, 4.51 ERA, 1.40 WHIP—preseason

125 IP, 4.29 ERA, 1.37 WHIP—rest-of-season

 

Based on some metrics, Aaron Harang is off to a terrific start. He’s currently seventh in WAR among all pitchers with a sub 2.00 ERA. His 2.86 FIP is also quite good but that comes with a caveat—Harang’s 2.3% HR/FB rate, which would be much better than anything he’s ever done before. In his career of over 2215 innings, Harang has allowed 9.9% of his fly balls to leave the yard, so it’s hard to believe he has somehow magically developed the ability to limit home runs on fly balls. He has improved his walk rate but is also striking out fewer batters and has a .258 BABIP, which is much lower than his career .304 BABIP. All signs point to regression from his current numbers, but his rest-of-season projection doesn’t think he’ll be as bad as his preseason projection thought he’d be. That’s still not good, of course.

Chris Archer (-0.21)

 

59.3 IP, 2.12 ERA, 2.46 FIP, 2.59 xFIP, 0.99 WHIP—current

183 IP, 3.69 ERA, 1.29 WHIP—preseason

138 IP, 3.48 ERA, 1.26 WHIP—rest-of-season

 

Chris Archer was good in 2013 and 2014 but has turned it up a notch this year. Over the first two years of his career, Archer had a 21.2% strikeout rate, 7.9% walk rate, and 46.3% ground ball rate. This year, he’s increased his strikeout rate to 30.7%, dropped his walk rate to 7.5%%, and increased his ground ball rate to 52.5%. Archer has increased his slider usage by about 10% over last year (based on BIS and PITCHf/x) and is generating more swinging strikes than he ever has, up to 12.1% this year compared to 9.3% last year.