Wei-Yin Chen's Hot Streak and Velocity Improvement Later in Seasons

For most major-league pitchers, velocity is everything. The loss of one mile-per-hour on a hurler's fastball can punch his ticket to the minors, or an uptick of the same degree can elevate him to stardom. Pitchers therefore want to maximize this, obviously, and avoid any stretches of poor performance in this regard.

During the first five starts of the 2015 season, Wei-Yin Chen struggled significantly. He struck out only 16.8% of the batters he faced, walked 9.2%, and allowed fly balls on 48.8% of his balls in play — an unsavory combination that gave him a 132 xFIP- for those 28.2 innings. Since then, though, he's done a 180; posting a 24.5% strikeout rate, 3.9% walk rate, and 38.4% fly ball rate, he's brought his xFIP- down to 80.

An impressive turnaround, that — and one whose cause isn't hard to discern:

	Fourseam	Twoseam	Slider	Curveball	Changeup
Velocity, 4/7/15-5/3/15	90.3	89.7	81.3	72.3	81.9
Velocity, 5/9/15-6/15/15	91.5	91.2	82.6	72.9	82.3

More velo = better results. Case closed! Chen's velocity should stay high, and the good times will continue to roll.

But wait. The early part of the season still happened. Might this same thing, and the mediocre play it brings, occur in the years to come?

Let's look to the past. In 2012, Chen's first season stateside, he experienced the same initial slump, albeit a bit more extended:

	Fourseam	Twoseam	Slider	Curveball	Changeup
Velocity, 4/10/12-6/1/12	90.2	89.5	81.1	72	83.1
Velocity, 6/6/12-10/1/12	91.4	90.4	80.4	71.7	84.6

Like in 2015, he became more powerful in late summer and fall, and performed better with that clout (102 xFIP-) than without it (112 xFIP-). And moreover, the same thing transpired the following year:

	Fourseam	Twoseam	Slider	Curveball	Changeup
Velocity, 4/3/13-4/26/13	90.5	89.9	81.6	70.5	83.1
Velocity, 5/1/13-9/28/13	91.8	90.7	81.4	72.9	83.8

His April xFIP- of 130 overshadowed by a 97 xFIP- from then forward, Chen didn't look back after some early hard times.

So Chen has always packed on more velocity in the later parts of the season? Not necessarily: His 2014 campaign saw him post consistent velocity — and output — throughout. If the Orioles bring him back for 2016, will that stability reappear, or will Chen once again fare poorly at the year's start?

There's no singular, definitively correct reason for Chen's 2014 season. Oddly enough, Chen had to rehab from a knee surgery in the offseason that preceded it — the only time he's recuperated before the year starts. The extra work he put in on that may have driven up his April and May velocity. Plus, for what it's worth, he vowed after 2013 concluded that he'd "come back stronger" the next year.

Then again, most pitchers don't throw as hard when the year is still young, usually on account of rust from inactivity and colder weather. Knowing nothing else about Chen, we'd probably expect him to lose some velocity at first; knowing that he's done so in three of his four MLB seasons, we'd definitely expect the same.

As with most pitchers, velocity gives Chen success when present, and robs him of it when absent. He currently finds himself with as much stuff as ever, to his benefit; however, come the start of 2016, he might lose some of that, and could see some bumps in the road as a result. Every player strives for consistency, but despite his efforts, Chen may not attain it.

Investigating Wei-Yin Chen's Early Season Velocity

Wei-Yin Chen isn't a flamethrower. But he does throw slightly harder than the average major league starting pitcher. A pitcher's velocity certainly isn't everything, though it's worth keeping an eye on. And in his first start, Chen's fastball speeds were below his career averages and could possibly be an indication of the reversal of a recent trend.

Since entering the majors in 2012, Chen has an average velocity of 92.5 mph on his four-seam fastball and 91.9 mph with his sinker (or two-seam fastball). In 4-1/3 innings on Tuesday, Chen averaged 91 mph on his four-seamer and 90.1 mph on his sinker. By itself, that doesn't mean much; we're only talking about a few dozen pitches in his first appearance of 2015.

Let's take a look at Chen's average fastball velocities from April in his three previous seasons:

Date	FF	FT
Apr. '12	91.0	90.7
Apr. '13	92.0	90.5
Apr. '14	93.0	92.6

Chen has only ever pitched in three major league games where one of his fastballs averaged below 90 mph: on April 28, 2012; May 20, 2012; and April 26, 2013. So all occurred relatively early in the season.

In his first two seasons, Chen started out the year with lower fastball velocities. But that wasn't the case in 2014, and that improved velocity propelled Chen to his best season as a major leaguer. Chen has actually thrown his fastball harder every season in the majors (his age 26-28 seasons):

It's been nearly a decade since Chen had Tommy John surgery (in 2006), and he's throwing harder now than he has in a long time (and maybe ever). While pitching in Japan for the Chunichi Dragons, Chen's fastball velocity had actually been in decline before joining the Orioles:

2009: 90.7
2010: 90.2
2011: 88.8 (possible leg injury)

Pitchers usually don't see a consistent fastball velocity uptick in their late 20s. But whatever Chen has been doing now that he's in the majors has been working. Perhaps that includes a more disciplined workout regime, better coaching and guidance, and, except for an oblique injury in 2013, pretty good health (no arm injuries, at least).

Interestingly enough, Chen mostly throws fastballs, but he's starting to trade four-seamers for more sinkers:

He hasn't had to trade fastballs for more offspeed pitches; maybe that's something that will happen when he starts to lose velocity. But even if the gradual change in fastball usage hasn't had a dramatic effect, Chen's HR/9 dropped from 2012-2014 (1.35, to 1.12, to 1.11), and his flyball percentage has also dipped in the same span (42.1%, to 41.1%, to 37.5%). Chen's groundball percentage hasn't followed suit, exactly (37.1%, to 34.4%, to 41%), but last season was his career best in that department. Chen is not transforming into a groundball specialist or anything close to one, but it'll be intriguing to see what happens if he continues to opt for more sinkers.

Chen has previously demonstrated that a start or two with lower than average velocity (for him) doesn't mean bad things are necessarily on the horizon. But at some point his velocity is going to stop climbing and will level off and start to decline. As an impending free agent, Chen could be playing for another team before that becomes a real concern. But as the Orioles' lone lefty in the rotation, Chen bears a different kind of responsibility, and the O's would surely welcome a rebound effort in his next start.

Photo via Keith Allison

Highlighting Jon Shepherd's Best Work at Camden Depot

Camden Depot has become one of the go-to destinations to read in-depth discussion and analysis of the Baltimore Orioles. Obviously my view isn't objective, but I don't think anyone would say that's hyperbole. And all of the credit for crafting Camden Depot goes to Jon Shepherd, who founded the site in 2007. In that time, he published more than 700 articles (and plenty of others elsewhere), cultivating the readership into what it is today.

As Jon wrote a few days ago, he's leaving the site to work as a statistical analyst for Baseball Prospectus. So I thought it would be a good idea to provide links to some of his best work, both analyzing Orioles' topics and baseball in general.

• The Wieters Shift (May 2008): Jon tackled the topic of Matt Wieters potentially moving from catcher to first base.
• What is Nick Markakis Worth? (May 2008): Jon looked at the worth of Nick Markakis and accurately predicted his six-year contract.
• GM Family Tree Offseason 2010 (December 2009): Jon constructed a family tree of general managers around the majors. It's not the easiest thing to follow, but it's impressive.
• Why Baseball Players Use Human Growth Hormone? Part I: Addressing the Debate (February 2011): Jon waxed poetic about baseball players' use of performance enhancing drugs: "[T]he clubhouse is not a stickler for the scientific method."
• How Has Pitching Velocity Changed in the Past Decade? (October 2011): Jon examined the rise in velocity among starters and relievers.
• Bad Outfield Defense or a Bad Fielding Metric? (October 2011): Jon investigated the peculiarities of UZR among Orioles' outfielders at Camden Yards.
• An Interview With Yoo Jee-ho About the Orioles and South Korea (February 2012): Jon talked with Yoo Jee-ho of the Yonhap News Agency about the O's dealings with the Korean Baseball Organization and Korean Baseball Association after the Kim Seong-min controversy.
• Receivers in Name Only (May 2014): Jon broke down the superior value of defensive catchers vs. non-defensive catchers.

So congratulate Jon (unless you really don't want to, which is cool) and enjoy looking back at some of his excellent research and analysis.

Tillman Has a Need for Speed

Velocity is not all: it is not movement nor control
Nor deception nor repeatability of a smooth delivery

Velocity cannot alone confuse the batter
Nor take the place of a well thrown curve
So, one may be tempted to trade velocity
For a pitch on the black or to hit the lower third
It may well be, but I do not think I would

--A horrible bastardization of an Edna St. Vincent Millay poem

The above is a poem that was originally about love. It is about how love is not an essential thing, but it is highly valued.  Similarly, I think we could think of velocity quite similarly.  A high 90s fastball is thrilling and majestic.  It is rock star.  It is bacon.  It is what adds the zero to an amateur's signing bonus. Or as William Carlos Williams once said, "So much depends on velocity." Or maybe he did not.  That said, the point is simply that a great deal of success depends on how little time a hitter has to react to a pitch as well as how little time a pitch travels through the zone where the bat can reach.

Photo by Keith Allison

Chris Tillman is someone who has dealt with velocity issues. He has always seen a great deal of success when he is able to pound the ball in the lower mid 90s.  Tillman suffered a near collapse in 2010 and 2011 where may a start he had struggled to get above 90 miles per hour.  In 2012, a revitalized Tillman helped lead the Orioles to the players with an explosive fastball.  Last year, saw a little erosion of that initial velocity, but it played well around 92 or 93.  This year, his opening day start saw him gunning around 94 mph. Since then? 91, 89, and 90 as averages. Those last two starts were two of the three lowest average velocity he has had in a game in the past two years.

Why is that velocity drop such a concern? Eno Sarris discussed things in a Sports on Earth article about pitcher aging while referencing Mike Fast's work, which was a better study than the brief one I did the year before in relation to Aroldis Chapman. Anyway, both studies found that a loss of one mile per hour of velocity was on average likely to result in about a third of a run more given up by the pitcher.  That is pretty frightening given Tillman's regression with respect to velocity.

That all said, Tillman sports a 1.71 ERA.  He has been incredibly successful and many a person wonder if he has hit ace status.  What might be ignored is that he current enjoys the highest left on base percentage of his career, which is a metric that is not associated with any perceivable skill.  Second, his home run rate is half of what we would expect it to be, indicating that he has been lucky that fly balls have not been spilling out of the park.  Third, balls thrown into the strike zone are not missing bats as they were the past two seasons.  In those years, Tillman was about to send about 14% of baseball in the strike zone past swinging bats.  In the troubling years before his recent success he was able to only get 9% by batters.  This year he is sitting at 10%.

However, nothing bad has happened yet even though several signs point to an eventual collapse.  I think the trouble with increased contact is largely related to that decrease in velocity.  Teammate Ubaldo Jimenez also suffered a major drop in velocity in the past few years from being a 97 mph burner to a 92 mph chugger. Jimenez' velocity is not the only trouble he has faced. His high maintenance mechanics often have spelled trouble for him and have certainly done so this season. Similar, with great command, control, and sequencing Tillman might also be able to have continued success.

Still, I would be more comfortable if I saw him flash more velocity on a consistent basis. He is too hittable otherwise and his extreme flyball tendencies likely means many a ball may be destined for Eutaw Street.

Weaver on Strategy - Offense (Camden Depot Book Club)

Previously on the Camden Depot Book Club, I remarked on a few items that Weaver addressed in his first chapter of the book, Weaver on Strategy.  That chapter addressed his thoughts on Spring Training and contained a great many more thoughts than those I addressed.  This entry will address the next chapter.

Chapter 2 - The Offense
Praise be the Three Run Homer!

Home Run Offenses
During Weaver's continuous 15 year tenure, his teams were largely built upon scoring runs through home runs.  He was risk averse when it came to putting base runners in difficult positions and he deplored the bunt.  As Weaver wrote:

I've got nothing against the bunt--in its place.  But most of the time that place is the bottom of a long forgotten closet.

He then goes on length about the value of the home run versus the value of a contact hitter:

The power of the home run is so elementary that I fail to comprehend why people try to outsmart this game in other ways.  If I were to play a singles hitter in right field or left field or at third base, he'd have to hit well over .300 and get on base often to be as valuable as a twenty five homer man.

I do think Weaver overstates things here in him selling how great home runs are.  For one, offensive profiles for corner outfield and third base bats are not exactly equivalent.  Ignoring that, the statement is pretty profound in noting that he is equating a plus power hitter with a guy who not only makes successful contact, but also gets on via walks.

He then takes a little swipe at the industry perception of Charlie Lau:

As a hitting coach, Charlie Lau is known for producing .300 hitters, and he has turned out a lot of them.  But when Charlie was with a team that hit a lot of homers, they won.  When he was with a singles hitting club, they lost.

Suffice to say, Weaver was not a Juan Pierre kind of guy.  Lau is actually a pretty interesting guy.  He was a massive failure of a hitter (a below .200 hitter with no power), but bounced around due to his catching.  Upon being dealt to Baltimore in 1961, he completely re-engineered his batting stance.  He exaggerated the spread between his feet and lowered his bat in order to make more contact.  When he found was that with a higher contact rate, he hit more singles and would sometimes find himself in hitters' counts to exploit.  After he retired in 1967, he moved over to the Royals and made his name through George Brett and the Royals' athlete first movement by working on having their players focus on contact.  I imagine Weaver's comments are primarily directed to the 1977 and 1980 Royals' squads who won 102 and 97 games, respectively, with a good number of home runs to boot.

However, Weaver's words made me want to see how those words compared to the reality of the team that Baltimore constructed during his tenure.  To do that, for each year I determined the average percent of runs scored via home run.  Then I compared Baltimore's percentage to the league average.  What I wound up with was a number similar to OPS+.  For instance, a 124 value would mean that the team scored 24% more of its runs via home runs than the average team.  A 94 value would mean that the team scored 6% less via home runs than the average team.  A 100 value would be league average if the above inference was not clear.



As you can see above, only two of Weaver's 15 squads were below average in terms of how important home runs were to the team scoring runs.  It is also interesting to note that Memorial Stadium during these years tended to lean toward benefiting pitchers by being a slightly difficult place to hit a home run.  That is pretty remarkable.

Although Buck Showalter only has a shade over two years as Orioles manager under his belt, his offenses have also reflected the same tendencies as Weaver's.  In 2011, the Orioles ranked second in the AL by scoring 42% of their runs through the long ball.  Last year, that number increased to 47%.  We will see how much the Orioles will miss Mark Reynolds' power production as he has contributed 15% of the Orioles' home runs over the past two seasons (60 out of 405).

Weaver As Innovator
I mentioned velocity in the last post on Weaver on Strategy.  In that post, I noted how there was expressed wonder and amazement at Sammy Stewart throwing a 92 mile per hour fastball.  Nowadays, that sort of amazement would go for someone who throws in the upper 90s.  Heat at 92 these days needs to be well commanded with a solid secondary pitch to be of great use in the pen.  In this chapter, Weaver goes a bit more into radar guns and velocity:

The Orioles were the first team in the majors to makes extensive use of the radar gun, and I love it. It's another tool that gives a manager information.  It took me six years to convince the front office that we should have the guns in our minor league system.

Yes, Weaver was a great innovator always looking to improve himself, to improve his understanding of the game.  This really is the main idea behind Weaver and his success.  Never resting, always striving.  I think the current manager who really puts this sort of perspective out there is Joe Maddon with the Tampa Bay Rays.  I am sure other managers embrace advancement as well, but Maddon celebrates it quite vocally just like Weaver did.

Velocity
In his discussion on using radar guns, he mentioned using one on his own pitchers:

When a pitcher is throwing at 88 miles per hour most of the game and then goes to the mound in the eighth inning and is at 84 miles per hour, it is a good bet he is tiring.

What, again, impresses upon me here is that my assumption of 88 miles per hour being used as an example is that this is considered typical velocity.  In 2012, 90 pitchers qualified for ERA.  Of those 90, only 14 (16%) threw an average velocity below 88.5 mph.  That includes knuckleballer R.A. Dickey.  Again, this only shows that the game has changed a great deal from the 1970s.

-----

Next time on the Camden Depot Book Club:

Weaver on Strategy
Chapter 3
The Lineup: Pushing the Right Buttons

Pitcher Velocity and Run Environments

Last night Will Carroll tweeted wondering about whether the perception was accurate that there were more pitchers thrower with greater velocity than there were in seasons before.  I had been sitting on a couple graphs for a month or so waiting to have something more interesting to say, failing that I published the graphs last night because at least it was timely.  Andrew over at Camden Chat asked about how the change in the number of hard throwing pitchers has affected the run environment.  In this post I compared the number of hard throwing starters (>90 mph) and reliever (>93 mph) to changes in the run environment as measured by starters' or relievers' FIP.  Each point is the data from a single year (2002-2011)

Starters

Click to Enlarge

The R2 for the above graph is 0.86.  That is a tight graph, but it should be stated that correlation does not always mean causation. 

Relievers

The R2 is not as good or the relievers with an R2 of 0.62.  That is still a decent correlation, but again it does not outright say that the increase in velocity is the reason why the run environment has decreased.

Conclusion
The reasons for change in the run environment are likely to be rather numerous.  There has been greater recognition of the value of defense, which decreases run scoring in two ways: better fielding and reduced emphasis on offensive performance.  It is assumed that the reduction in effective PED use has also decreased offensive performances.  Another factor is that, as it always has, that baseball becomes more and more competitive every year.  By that I mean that the players today are better than the players who came before (Yes, the Babe Ruth is Matt Stairs argument).  It may also be that the recent upswing in pitchers who can throw hard has also resulted in dampening the run scoring environment.

Of course, the big question is why are there more harder throwing pitchers?  Here are the various explanations I have determined and collected:

• There are a greater number of highly talented players to choose from.
• Training has vastly improved.
• Preventative and reconstructive medicine has improved.
• Teams may be pushing more athletic, strong-armed players to the mound instead of the field.

On that last point, Keith Law disagreed with me on that one.  As far as I can tell, the four top throwers in the relief corps (Henry Rodriguez, Aroldis Chapman, Jordan Walden, and Daniel Bard) were never given much consideration for playing the field.  It may well be that idea sounds better than it looks when you begin looking at the data.  It may well be that advancements in training and medicine taking effect somewhere between 1995 and 2005 for amateurs may be the primary reason for hard throwing pitchers.  Or it may be that advances made in the 2000s in college and lower minors are improving the development and preserving the health of hard throwing pitchers.

How Has Pitching Velocity Changed in Past Decade?

I was reading a few books on baseball from the 1970s and found it amusing how often a pitcher throwing 90 mph was a flamethrower and that mid 80s was considered sufficient.  Nowadays, success with those velocities would be considered highly improbable.  This left me wondering if our comprehensive measurements of mean fastball velocity since 2002 (from FanGraphs) could show any differences in the number of hard throwers in the past decade.

Starting Pitcher Velocity

Click to Enlarge

Pitchers throwing a mean fastball velocity above 90 mph has increased to 188% in 2011 compared to 2002.  Pitchers throwing harder than 94 mph has doubled over that same time period.  This growth in velocity at the Major League level occurred in 2007 and continues to increase.

Reasons I can think of for this occurring:

• Teams want their pitchers with higher velocity to pitch more often and refrain from converting them to relievers.
• Starters are being trained to pitch all out for shorter outings rather than trying to dialing down to last the entire game.
• More hard throwers are available.

  • Fastball velocity is being valued more and baseball is able to out recruit other sports for these prospects.

  • Teams are more willing to convert hard throwing outfielders, middle infielders, and catchers to pitching. 

A number of these ideas could be tested by merely making a similar graph seen above for relief pitchers.  If we also see an increase in the number of hard throwers in relief then we will know that the overall population of pitchers are throwing harder.  That would run contrary to the first two reasons and point toward the third one as valid.

Relief Pitcher Velocity

Click to Enlarge

We see a similar increase in velocity as we did with starters.  However, growth is slow and gradual from 2002 to 2008.  In 2009, velocity jumped up and plateaued.  Relief pitchers with a mean fastball velocity above 93 mph increased to 221% in 2011 over what was available in 2002.  Perhaps more amazingly, pitchers throwing on average more than 96 mph increased 1,200% from 2002 to 2011.  In 2002, only one pitcher average a fastball over 96 mph: Billy Wagner (97.6).  In 2011, eleven pitchers sat above 96 mph: Henry Rodriguez (98), Aroldis Chapman (97.9), Jordan Walden (97.6), Daniel Bard (97.3), Bobby Parnell (97.2), Joel Hanrahan (97.1), Brandon League (96.5), Neftali Feliz (96.3), Craig Kimbell (96.2), Jason Motte (96.0), and Jason Lindstrom (96.0).

Conclusion
The number of pitchers capable for throwing hard has increased over time.  It does not readily appear that hard throwing reliever are being converted to starting pitching as the increase is apparent in both groups.  This leaves me thinking that there are simply more hard throwing arms available.  This could be due to natural progression or it may be an element of pitching that is now being emphasized to  greater degree.  I have talked to a few scouts who angrily mention how pitching velocity is becoming valued too much because it is a quantitative measure on pitching prospects that cannot be embellished and that qualitative measures are being relied on to a lesser degree.  I am not so sure I believe that, but it is an explanation that should be floated out there.

My personal belief is that hard throwing arms are being more heavily sought after.  Teams are willing to pay extra to draft prospects who show plus velocity over pitchers who perform well due to a polished approach.  In an earlier post, I showed in an awfully noisy graph that a mile per hour in velocity saves half a run over the course of 100 fastballs.  That is the difference between a pitcher giving up 6 runs over 12 innings vs 5 runs over 12 innings.  In terms of ERA, that means a 4.50 ERA would decrease to 3.75 ERA.  Certainly more is involved than simply velocity, but it certainly is a trait of a pitcher that is quite important.  It may be that other teams simply are valuing this more and are better at finding and getting these arms to the majors.  In other words, perhaps more athletic players with live arms are giving more consideration on the mound than trying to make their skills work in the field.

Perhaps more likely, competition is simply rising in baseball.  As it pretty much always has.  Players get stronger and the training improves over time.