The Value Of #4/#5 Starters Has Skyrocketed

Most fans hope for the best for their team. They hope that their major league players will show improvement from their past performance and that their top prospects (regardless of overall rank) will end up being successful in the majors. This divide between optimism and reality becomes clearer when looking at starting pitching. People hope that their top pitching prospects can become successful in the majors at the same time that starting pitching is becoming hard to find. As a result, fans undervalue legitimate backend starters and overvalue unranked pitching prospects. This came to light last week when Jon talked about the value of Kevin Gausman.

For starters, the performance of starting pitching has changed significantly recently. This chart shows the count of qualified pitchers, their average ERA (not waited by innings pitched), their average FIP and their average WAR.

It’s pretty simple, there are 30 teams in the majors and each team historically has five starters in the rotation, meaning there are 150 starters that have a shot to be qualified. From 2010-2014, roughly 90 starters threw over 160 innings, or on average each team had three qualified starters. In 2017 that dropped to 56 starters, or on average each team had only two qualified starters. However, despite the drop in qualified starting pitchers, their performance hasn’t improved. The average ERA and FIP have gotten worse over time, suggesting that finding qualified pitchers is harder in this new age. That’s one reason why only 25% of qualified starters were worth 2 WAR or less in 2017. Starters that can give their team 160 innings with a decent ERA and FIP have become much more valuable than they were even two years ago.

Unsurprisingly, the number of starters used in a year has gone from 273 in 2010 to 315 in 2017. Part of this is because teams received on average 970 innings from their starters in 2010 but only 890 innings from their starters in 2017. But part of it is that the average starter has gone from throwing 106 innings in 2010 to only 85 in 2017. As a result, teams have gone from using 9 starters on average to using 10.5 starters on average. As more starters are used, the average ERA and FIP has also gotten worse. Things are somewhat better this year, but not by much. According to TruMedia, there are 79 qualified starters in 2018 compared to 73 at this point in 2017. Expect a small increase of qualified pitchers from last year, but probably not a large one. Here's how the numbers look for all starters.

The value of backend starters that can give you a large amount of innings without having terrible results has skyrocketed due to their scarcity. Teams only have so much starting pitching depth. The more starters that they’re forced to use, the more likely that they’re going to get an atrocious performance from somebody. The teams with the best starting pitching are those like the Indians who made it through 2017 using just seven starting pitchers. Having guys like Bundy and Gausman on your roster helps keep the bullpen fresh and ensure teams don’t need to use their AAAA guys as starters.

It’s possible to use FIP to rank starting pitchers from 2010-2017. For each year, we know how many starters each team used on average, so it makes sense to put starters in groups based on the average number of starters used by a team. For example, in 2010, teams used 9.1 starters per year, so we can rank starters from 1 (best) to 10 (worst) based on their FIP. In 2017, teams used 10.5 starters per year, so we can rank starters from 1 (best) to 11 (worst).

When using this method, it becomes pretty clear that there’s a big difference between aces (average FIP of 2.59) and #2 starters (average FIP of 3.28), the second worst starters (average FIP of 5.27) and the worst starters (average FIP of 8.12) and the third worst starters (average FIP of 4.72) and the second worst starters. Aside from those groups there’s roughly a .2 or .3 FIP difference between ranks. Over 180 innings, this is equivalent to roughly 5 runs or half a win. A decent-sized distinction, but not a huge one. Here's how the groupings look.

Using this method, Kevin Gausman consistently (2015-2017) appears to be a #5-6 starter while Dylan Bundy looks to be an outright #6 or even worse. That stated, FIP probably isn’t particularly fair to Orioles starters. FIP presumes that pitchers are fully responsible for all home runs that they allow, but it’s a lot easier to hit a home run in Camden Yards than in the Oakland Coliseum. Fangraphs WAR uses a park factor to take this into account, but FIP does not. So, it probably makes sense to consider Gausman a #4-5 starter. Likewise, Sonny Gray was a 3-4 starter in 2014 and 2015 using this metric, while he dropped to a #8 starter in 2016. But due to pitching in a pitcher friendly stadium, it’s likely he should also have been treated as a #4-5 starter. In other words, these two pitchers are probably closer in value than just looking at their FIPs or ERAs would indicate.

At any given time, there are typically around 40 pitching prospects on top 100 prospect lists. Not all of these pitchers graduate in a given year, but if top prospects had a high success rate, then there would be a lot more than 60 qualified starters. The fact is that the likelihood of a top prospect being successful isn’t great, and therefore the value of a prospect that is successful is high. If top pitching prospects that are ranked struggle to be successful, then pitching prospects that aren’t ranked struggle even more often. It’s highly unlikely that unranked pitching prospects will be successful in the majors. They’ll get a shot because teams need to rely on their minor league system for starters, but they’re probably not going to succeed. Unfortunately, fans don’t remember failed prospects.

Kevin Gausman and Dylan Bundy are likely going to be nothing more than #4-5 starters on the Orioles. It’s possible that another team could successfully develop them and turn them into top of the rotation pitchers. But even #4-5 starters that can pitch a full season have significant value. Their performance may not be great, but these guys can solidify a rotation, ensure that teams don’t need to rely on minor league pitchers with minimal talent and preserve a bullpen. The value of that has skyrocketed over the past few years.

Orioles Fifth Starter Competition Should Last All Season

To some extent the Orioles have a blessing and a curse. Their fifth starter position in their rotation looks to be rather undecided. On top of that, Chris Tillman looked very shaky on Tuesday. Plus, Andrew Cashner is no one's idea of a sure thing. This should give the Orioles an opportunity to test out their young arms by using a swingman. What I mean is that the Orioles could keep to four days off and one day on for their starting pitchers. With off days, this increases the frequency that the top of the rotation pitchers will enjoy.

Slot	Strict 5	Priority 5
1	33	36
2	33	36
3	32	35
4	32	31
5	32	24

Standard practice these days tends to give starters as much rest as possible. What you typically see is that everyone takes their place in the rotation, which is why you wind up with about 32 starts per pitcher with the first two slots looking at 33. What a swingman role will do is give greater opportunity to your best pitchers. Gausman, Bundy, and Cashner will potentially get more opportunities. Tillman would get about the same and the fifth starter will get far fewer. Yes, rainouts, injuries, and performance will change the distribution, but this is the general idea.

So what would this mean?  Below I tabulated the expected number of total runs for each pitcher.  Data shows that ERA does not significantly differ between four days of rest and five days of rest, so we will assume that run rates will not change.  Gabriel Ynoa (boot and all) is standing in as our generic fifth starter.

	Runs
	Strict 5	Priority 5
Kevin Gausman	91	99
Dylan Bundy	100	109
Andrew Cashner	111	122
Chris Tillman	111	73
Gabriel Ynoa	103	77
	516	480

What this suggests is that bringing back a true swingman and prioritizing starts for pitchers with at least four days rest would result in a decrease of 36 runs scored.  If such an approach could be taken and if these numbers hold out, it would mean a bump of four wins.

For fun, what if the club signed Alex Cobb.  We would slot him in Cashner's place and bump Cashner down to the fourth slot.  Tillman would then enter the fifth starter abyss.  The priority five run total would drop from 480 to 454 runs score, another 26 runs.

This means that changing from a straight five approach to a priority five approach plus Cobb would net the club 62 runs, which is about seven wins.

But, wait, there is more.  Going back to the swingman, the club can get creative here.  You could imagine Chris Tillman getting a few starts to open the seasons or be placed on the DL with some starts later on. Nestor Cortes could be given a shot.  Mike Wright could try to put it all back together again.  Miguel Castro might have a whirl or two.  In the minors, the likes of Alec Asher, Michael Kelly, Yefry Ramirez, Tanner Scott, Jayson Aquino, and Eddie Gamboa could be run through.

The swingman approach could rekindle what was beneficial with the 2012 team, which was the ability to give opportunities to pitchers in order to find out whether Chris Tillman or Miguel Gonzalez could put it all together.  The swingman approach also reduces the importance of finding those guys.  Time spans between fifth pitcher starts could be as long as 14 and 17 days between, which allows the club to beef up the pen to make up for the starter's tendencies to not go deep into games.

The opportunity to find useful starters or, at worst, provide more depth for the bullpen or bench could well be worth another win.  If Fangraphs is right that this is a 75 win team, then this approach plus Cobb would get them to a projected 83 wins.  That club would be playing meaningful baseball into August, at least.

How Pitch Sequencing Can Affect Outcomes

The 2017 Baltimore Orioles starting rotation has been a bit of a mess. Maybe with some data and some advance planning, the starting pitchers can become more effective. If we can identify the two-pitch sequences to end plate attempts and compare outcomes to those sequences, we may find that some pitches work well together, with one setting up the next. This can be considered a proof of concept; the same sequencing can be extended to sequences of three and four pitches for more granularity (at the expense of sample sizes).

To put together these sequences, I first needed pitch-by-pitch data. PitchFX is great for that, if you can get your hands on it. I set up a database of PitchFX records from 2015 to today (and counting!) using the MLBAM Gameday records stored in XML for what seems like perpetuity. From there, finding pitches that end plate attempts is fairly straightforward, as is finding the pitch that preceded it. Because my database only goes back two and a half seasons, data is a little limited, especially for the youngest members of the Orioles rotation.

Consider Dylan Bundy, whose slider is identified as a high-quality pitch by BrooksBaseball.net, noting its high swing-and-miss rate. Fangraphs gives Bundy's slider a much higher value rating than any of his other pitches (his next highest rated pitch per Fangraphs' weighted values is his Changeup). Bundy's slider appears most effective when paired with his four-seam fastball given the triple-slash outcomes of pitch sequences involving a slider (and that end within one pitch of a slider being thrown):

Limited to the 10 most frequent two-pitch combinations to end plate attempts

Bundy has used a four-seam and slider combo nearly 50 times in some order, and the results have been great. Batters are hitting 0.100 off of either pitch when paired with the other. The SL-FF combination produces a slightly higher on-base percentage, which leads me to believe that the fastball is occasionally thrown after the slider goes for a ball and results in some walks.

However, Bundy's slider isn't entirely unhittable. When he goes back-to-back with sliders and the second one ends a plate attempt, they're getting hit well. Batters are slashing roughly .350/.400/.400 when putting a second consecutive slider in play. While this doesn't account for all the times a SL-SL sequence did not result in the plate appearance being terminated, it definitely looks like Bundy should be cautious to throw his slider multiple times in a row - which makes sense, and isn't unique to Dylan Bundy. A batter is expected to get better at identifying a pitch and tracking its movement the more often he sees it.

Kevin Gausman likes to throw heat - and he's good at it. But with sequencing in mind, Gausman might find that the pitch he throws most isn't his best pitch, and that his most frequent pitch sequence isn't doing him any favors. Gausman throws four-seam fastballs over 60% of the time, but they tend to get hit well when batters see them back-to-back. Gausman's split-finger fastball, however, is frequently involved in positive outcomes.

The FS-FS sequence to end a plate attempt has yielded a .180/.260/.270 slash line since the beginning of the 2015 season, and a FF-FS or FS-FF sequence has resulted in a lower batting average and on-base percentage than the more common FF-FF sequence. If Gausman wants to throw hard, and it looks like he does, the pitching staff should encourage him to mix more split-finger fastballs to the mix.

Gausman's full stats show that he throws the FF-FF combination significantly more often than any other pitch sequence, which we should expect given how frequently he throws the four-seam fastball. The positive outcomes on a FS-FS combination come on just 65 instances of that sequence ending a plate attempt, and may be subject to a small sample size. That number of plate attempts is far fewer than the point at which most pitching statistics stabilize. Since the outcomes are so positive, it probably wouldn't hurt for Gausman to test throwing FS-FS sequences more frequently, or just to throw his split-finger fastball instead of his four-seam fastball in some instances.

This type of sequencing can be extended to more pitches to map the value of mixing pitches within a plate attempt, or it can be used to find two-pitch sequences throughout plate attempts that don't result in outcomes. For instance, Gausman's FF-FF sequence might not have the best outcomes when a batter connects with the pitch, but it might coerce more swings and misses or swings outside the zone during an at bat. At the very least, it's helpful for pitchers, catchers, and coaches to understand the value and risk in throwing specific pitches back to back.

Starting Pitching Was Not (Really) a Strength of 2016 Orioles

Joe Reisel's Archives

Jon Shepherd's recent Camden Depot article, Orioles Starting Pitching Was Not a Weak Part of 2016 Club, made a case that the 2016 Orioles starting pitching was, well, not a weak part of the 2016 team. Some commenters, including myself, observed that while the rotation as a whole might have been average, the front half (Gausman, Tillman, Bundy, and Worley (4 starts) performed well while the back half (Jimenez, Gallardo, Wilson, Wright, and Miley) performed poorly. This started a lively exchange in the comments.

I probably misunderstood what Jon was saying. Digging beyond the headline, he was seeking to correct the narrative that the 2016 Orioles starting pitching was a "train wreck" (among other narratives.) The title and the article itself contain another source of potential misinterpretation - that if starting pitching wasn't a weakness, then there is no need to improve it for 2017. Jon didn't actually say that in the article and, in subsequent comments, made it clear that he wasn't trying to imply that the starting pitching was just fine the way it was.

Nevertheless, I thought I would explore the idea that the Orioles' starting pitching was either very good or very bad in 2016, and see what (if anything) I could discover. The simplest approach was to look at each team, and count the number of games started by good, average, and poor starting pitchers. As this is an initial exploration, I decided to keep things really simple, using ERA+ as the measurement. ERA+ is a context-adjusted measurement of a pitcher's ERA, so that 100 is league-average (adjusted for context), above 100 is better than average, and below 100 is worse than average. Because starting pitchers have higher ERAs than relief pitchers, I decided to define an ERA+ of 101 or better as above average, 90-100 as average, and 89 or below as below average.

The below table shows the counts. For each team, the number of games started by a pitcher with a season ERA+ at the appropriate level for that team is counted. These values are retrieved from baseball-reference.com. Two points - this doesn't distinguish a pitcher's ERA+ as a starter as opposed to his ERA as a relief pitcher, so a few pitchers who were primarily relief pitchers but made a couple of starts would be listed in a category heavily influenced by their relief usage. And, it uses the pitcher's ERA+ with the team only, so James Shields is considered average for San Diego and below average for the Chicago White Sox.

	101+	90-100	89-			101+	90-100	89-
BAL	78	0	84		ARI	38	0	124
BOS	92	54	16		ATL	38	32	91
CHW	87	28	47		CHC	160	0	2
CLE	142	0	19		CIN	100	4	58
DET	73	19	69		COL	120	3	39
HOU	29	43	90		LAD	84	14	64
KAN	63	46	53		MIA	61	33	67
LAA	57	32	73		MIL	52	81	29
MIN	30	0	132		NYM	124	0	38
NYY	66	70	26		PHI	74	24	64
OAK	43	43	76		PIT	50	30	82
SEA	60	66	36		STL	36	0	126
TAM	87	40	35		SDG	40	11	111
TEX	101	20	41		SFO	112	12	38
TOR	99	61	2		WAS	110	33	19

Regarding the 2016 Orioles, it does seem that they didn't have a truly terrible starting rotation, although it was the worst among playoff teams. Their 84 games started by poor pitchers was the 24th most among all teams and by far the most among playoff teams. The next highest-total was 64 (the Dodgers.) On the other hand, their 78 games started by good pitchers was the 17th most among all teams, and last among the playoff teams. Their starting pitching wasn't bad enough to keep them out of the playoffs.

Unlike two other contending teams. While most of the teams with bad starting rotations were bad teams, Houston and St. Louis were in wild-card contention until late in the season (the final day for St. Louis.) And most observers concluded that their lack of starting pitching is why those two teams didn't make the postseason.

To some extent, this chart understates how bad the Orioles' second-line starting pitching was. I used a cutoff ERA+ of 89 to define a "bad" starting pitcher; none of the five Orioles in that category - Jimenez, Gallardo, Wilson, Wright, and Miley - came particularly close to that level. (Wilson was the best at 84.) So the Orioles could improve their second-line starting pitching and still not have it show up in this measurement.

Although the Orioles did make the postseason in 2016, they probably need to improve their second-line starting pitching if they want to make the postseason in 2017. To make the playoffs, you have to survive the 162-game regular season - and it's very tough to do so if you have below-average pitchers starting half your games. They don't necessarily need more good starting pitching; they just need less bad starting pitching.

A few arbitrary observations:

• There seems to be a "floor" for starts by bad pitchers, below which it doesn't seem to affect a team's chances. Almost every team seems to have a couple of guys who pitch badly before it's decided to replace them, and every team seems to have a few spot / September starts by pitchers who pitch poorly. A team should expect to have around forty games started by pitchers who don't pitch well.
• The big surprises in a positive way were the Reds and the Phillies; I wasn't expecting them to be this good. 
• In the 2016 American League, the Minnesota Twins were so awful at run prevention that "average" loses much of its meaning. The Twins gave up 889 runs; the next-worst team, Oakland, was closer to the best team (761 runs for Oakland, 666 for Toronto) than to the Twins. The American League average was 724 runs allowed - but if you exclude the Twins, the average shrinks to 712. The Twins themselves are responsible for five teams being above-league-average in run prevention as opposed to being below average.
• This doesn't have anything to do with the subject, but the leader in innings pitched for the Pirates was Jeff Locke - with 127 1/3. Not only did the Pirates not have anyone who qualified for the ERA title, they didn't have anyone who would have qualified for a full-season minor-league ERA title if the minors leagues used the one-inning-per-scheduled game qualification.

First-Place Orioles Obviously Aren't Sellers, But Should They Buy?

If you're a glass-half-full type of person, then, at best, you may view the Orioles' trade deadline deals of the past few years as a mixed bag. If you're not, then surely you hated many of them. There are a couple of players who helped the O's -- Andrew Miller, Bud Norris, Scott Feldman -- and also a bunch of guys who didn't. And when you look back at the price to acquire them, it doesn't look so great.

July Trade Players Of Note In The Duquette Era

O’s get: Jim Thome, Omar Quintanilla, Scott Feldman, Steve Clevenger, Francisco Rodriguez, Bud Norris, Andrew Miller, Gerardo Parra, Junior Lake

O’s gave up: Jake Arrieta, Pedro Strop, Nick Delmonico, Josh Hader, L.J. Hoes, 2014 comp. balance pick, Eduardo Rodriguez, Zach Davies, Tommy Hunter, Gabriel Lino, Kyle Simon

I might have left a couple out, but it's clear that the O's didn't get many bargains. The O's pulled off a cream-of-the-crop addition one time: Andrew Miller. Regardless of what you think of the Miller trade, he was very good in an O's uniform.

The rest of the time, Duquette aimed for the middle, or for relatively minor or moderate upgrades, and those moves mostly did not pay off. If the Orioles wanted to go after another reliever -- clearly Buck Showalter craves another left-hander in the bullpen -- they could put together a trade package for that. But many teams desperately need starting pitching upgrades, and nearly all of them will be able to assemble more enticing groups of prospects than what the O's can cobble together. That could lead the Orioles again shooting for a fringe starter.

So how should the Orioles proceed? It's not an easy question to answer. The way I see it, the Orioles could do one of three things. They could do nothing, rolling the dice with what they have and hoping it somehow works out. Second, they could target a few low-risk, high-variance players who wouldn't cost much. That could keep the likes of Ubaldo Jimenez, Mike Wright, and/or Tyler Wilson out of the starting rotation, even if there's no guarantee the new additions pitch any better. Or third, they could go after a legitimate starting rotation upgrade, whether that's Julio Teheran, Jake Odorizzi, Sonny Gray, Matt Moore, Rich Hill, Drew Pomeranz, Jeremy Hellickson, Ervin Santana, or someone else. Maybe some of those guys aren't on the market. Most likely, the O's would struggle to put together an enticing package for many of those names. And maybe they'd struggle to outbid other teams in the hunt for rotation upgrades. But at least some of those names are within reach.

I've been going back and forth about this. For a while I was in the do-nothing camp, but it's extremely difficult to talk yourself into riding things out with this group of starters. It's just too painful for a core that has a chance to win something significant right now.

The Orioles clearly want to add a left-handed starter. My guess is they would prefer to add two starters to go along with Chris Tillman, Kevin Gausman, and Yovani Gallardo. Maybe it would be easier to do that by adding someone like Jon Niese instead of Rich Hill. Either way, you're rolling the dice. These guys aren't aces.

You'd figure it wouldn't take a whole lot to acquire Niese. And again, Niese is just one example. There don't appear to be any slam-dunk starters in this market, or anything similar to the skill level the O's acquired when they picked up Miller to improve their bullpen a couple years ago. Maybe the O's could similarly acquire an add-on to a powerhouse bullpen to aid a bad rotation, but it's unlikely they go for a top-shelf reliever again. The O's also probably view Darren O'Day's impending return as the shot in the arm the bullpen needs.

At the minimum, I think the O's need to pursue a starting pitcher of some kind and a left-handed reliever. But no matter what, I won't feel great about it. The O's have the opportunity to make a decent move or two and improve their chances of getting into the playoffs, but in order to do that, they'd hurt an already weak farm system. Just because the Orioles have a low-rated minor league system doesn't mean it's completely devoid of talent. Whether the O's would be able to get the most out of those prospects is a fair question to ask, but that that's even a question is pretty sad for the state of the O's prospect development system currently in place.

This is what can happen when you both refuse to pay top dollar for starting pitching but are also terrible at developing your own pitching prospects. And yet, the O's are still, somehow, two games up in the American League East.

Being a GM is hard. The Orioles need starting rotation help. They also routinely need prospect help. Who's ready for the trade deadline?

Yovani Gallardo Has Two Sliders

During his first start for the Orioles on April 11, Gallardo threw 96 pitches - 67 of which PITCHf/x determined to be sliders. A look at the movement of his 96 pitches (with gravity - this is what the batter saw) shows two very distinct clusters of sliders. One looks the way you might expect a slider to look: dropping lower in the zone, traveling from left to right across the plate. The other cluster stayed up, dropping about 10 inches with the help of Mother Nature, and never actually slid. Instead, the second bunch of sliders stayed straight. They look the way you would expect a fastball to look.

PITCHf/x classifications are determined today by algorithm, although it has been done by hand in the past. It's certainly possible that these are simple misclassifications as a result of that estimation model. Gallardo's fastball was only thrown about 4 miles per hour faster on average than his slider - a major difference, but one that can easily be spanned by arm fatigue. Gallardo's fastest slider on April 11 was thrown at 87.5 miles per hour. It's entirely conceivable that an algorithm could label an 87.5 mile per hour slider that doesn't slide as an 88 mile per hour fastball.

At the same time, Gallardo has a short bit of a history with these two types of sliders. The vast majority of Gallardo's straight sliders is from 2015 onward, indicating that he could have added a slider with a different spin and a different movement path to his arsenal. According to PITCHf/x, Gallardo is also increasing the number of sliders he throws every year. If there were a sudden spike in slider usage, I might be more apt to consider it a fluke pointing to an issue in the algorithm's ability to classify Gallardo's pitches.

This leaves three possibilities (or some combination thereof, I suppose) to explain why PITCHf/x thinks Gallardo is throwing so many sliders, and in two distinct types. The first is that PITCHf/x is simply misclassifying Gallardo's fastballs as sliders for some reason and with some regularity. The second is that Gallardo actually did throw an obscene amount of sliders on April 11 and that they have always been more common for him than fastballs, he just isn't getting the break on them the way we would want to see. The final possibility is that Gallardo actually has two sliders that he throws in two different ways, on purpose, for different effect.

John Walsh at the Hardball Times put together a helpful pitch identification primer in 2009, in which he describes sliders as generally falling between fastballs and curveballs in terms of both speed and movement. He also identifies sliders that don't have any horizontal spin or break at all, pitches with essentially no movement. Here's how he describes the specific subset of sliders that we are possibly seeing from Gallardo:

[I]f the axis of spin is aligned with the direction of the pitch (like a football toss, or a … gyroball), there will be no break. I believe this is what we are seeing when we see a pitch with no movement. I wouldn’t call these gyroballs, by the way. Or maybe I should—in any case, a number of different pitchers throw sliders that fit the description, and I don’t believe it’s anything new.

While it's pretty cool to throw the same pitch two ways for two different effects (which, uh, might make them two pitches?) - if that is the case - it hasn't necessarily helped Gallardo. His batting average allowed on decisive sliders - ones that ended the plate appearance - from 2012 through September 2015 is 0.239, while his slugging percentage on decisive sliders is 0.370. When it's broken into the two categories of sliders, it's clear that one is more helpful than the other.

Pitch Type	Times Thrown	Batting Average	Slugging Percentage
Traditional Slider	3,987	0.233	0.359
Straight Slider	61	0.328	0.361

Pitches in red are what I'm calling Straight Sliders.

Notice that Gallardo hasn't thrown what I called a straight slider very often over his career. However, 49 of those came in 2015 alone, and he looks to have thrown about 20 in his first 2016 start as well. While the straight slider hasn't burned Gallardo for extra bases yet, it's worth keeping an eye on. If he means to get movement across the plate, it's a major issue that Gallardo is suddenly less and less capable of doing so. If he is using a straight slider as a pseudo-fastball for some reason, batters seem able to hit it pretty well. Gallardo's next few starts for the O's may shed more light on whether he's actually throwing this many sliders and whether he's throwing straight ones on purpose. If the latter is true, we're probably in the process of watching Gallardo reinvent his pitch arsenal, and arguably to one that gives up more hits.