30 June 2016

Hyun Soo Kim Has Been A Revelation

Depending on what you believe, Hyun Soo Kim's first year with the Orioles was close to not happening. Before he ever played an inning of regular season baseball, the Orioles were exploring how to get out of his contract (signed just a few months prior). As Roch Kubatko of MASN wrote in late March, "I checked again last night and the Orioles definitely are open to moving Kim, whose performance hasn’t met expectations." That would be his highly scrutinized spring training performance.

Kim did not hit well in spring training, and that, combined with other concerns (defense? baserunning? money?), put Kim's Orioles' career in jeopardy before it ever really started. The Orioles wanted Kim to start the year in Triple-A. He justifiably declined. Some fans foolishly called Kim selfish, and some even made the puzzling decision to boo him on opening day. Adam Jones rightfully admonished fans afterward.

That brings us to Kim's actual production during his debut season. Here's a summary: He just keeps hitting. Kim didn't make his debut until the O's sixth game, when he collected two singles in three trips to the plate. He appeared in the lineup sporadically through mid-May. Because he kept hitting and getting on base, he finally started seeing his name in the lineup regularly starting on May 25.

Overall, it's hard not to be impressed by the numbers. In 137 plate appearances, Kim has posted a 148 wRC+. That's second-best on the Orioles, behind Manny Machado (159 wRC+). Kim is almost certainly not this good -- a BABIP of .390 is propping up his numbers to some extent -- but he has been excellent so far and has given the Orioles way, way more than they thought they were getting in March. (Seriously, go through Kubatko's post above comparing Kim and Rickard. It's amusing.)

At the time Kim's signing was announced, fans were hoping that Kim could provide an on-base threat. He's done exactly that, with a .438 on-base percentage. That's the highest among all major league hitters with at least 100 plate appearances (David Ortiz is second at .433). Kim has walked exactly as many times as he's struck out (16), and his walk rate of 11.7% trails only Chris Davis (13.6%).

One of the amusing parts of the Orioles trying to get out of Kim's contract before the season is that not only did it seem like a reasonable deal at the time, but it only looks better now. The Orioles signed Kim for two years and $7 million, and based on his production (0.9 fWAR, 0.7 bWAR), he's essentially been worth his contract so far with a season and a half to go. Even if Kim performs at replacement level for the rest of his deal, his contract has already been worth it.

You probably couldn't have predicted it based on how the season started -- Kim buried on the bench; Adam Jones battling an injury; Joey Rickard receiving an extended stretch of at-bats as the team's everyday leadoff hitter -- but the Orioles have stumbled into an effective left field platoon with Kim and Rickard. This is how the two have performed against opposite-handed throwers:

Kim vs. RHP: 152 wRC+
Rickard vs. LHP: 129 wRC+

That is very strong production, especially considering how terrible O's left fielders were just a season ago. Perhaps Kim should be playing even more and given a chance to show what he can do against lefties, but it's hard to argue with the results.

Kim has exceeded all expectations, even before taking into account the drama during and after spring training, combined with being chained to the bench for a while. Kim hasn't reached (and likely won't reach) the point where it's expected to see him bat about 50% better than the league average. But at least 10-20% is certainly within play, and if anyone would have guaranteed you that before the season, you would have been thrilled. The same goes for the Orioles, who are surely just as surprised.

All I know is that I'd be just fine with more moments like this:

29 June 2016

Jonathan Schoop Is Breaking OBP

2015 is not 2016 in Baltimore. If the two seasons were made into one of those "spot the difference" games, the first thing you would notice is the homers. The O's lead all of baseball in bombs, and are close to breaking the major league record for home runs hit in a single month. But if you've been watching closely, the most glaring difference has been the men on base. The glorious men on base. The 2015 squad finished third in homers, but 24th in OBP (.307). Through almost three months, the 2016 Orioles have the second highest OBP in the AL with a .332 mark.

The Red Sox are the only AL team with a higher OBP, and the only AL team with a bigger gain in OBP from 2015 to 2016. There are many ways to get on base, and therefore many ways to raise your on-base percentage, but in general drawing more walks is the best way to go about it. And if we look at the teams that have improved in walk rate, we see a pretty strong correlation with the OBP gainers.

The Rays and the Jays are the only teams that have improved on their BB% and declined in OBP. The Royals (of course) are the only team that have posted a worse BB% but improved in OBP. So, unless you are following the Royals formula and making a ton of contact with speedy, high BABIP guys, then any increase in OBP is probably being fueled in large part thanks to a bump in your BB%. 

It's an intuitive concept to grasp, and it seems like everyone on the O's got the memo except for Jonathan Schoop. He actually has raised his BB% by a full percentage point this year. The only issue is that he raised it from 2.8% to 3.8%. For context, league average thus far is at around 8.4%. Walks have trended down over the last decade as strikeouts have risen, but any way you slice it, 3.8% is not good. 

And yet, somehow Schoop's OBP is currently sitting at .322. That isn't particularly good, but it's average, and that's all the O's need from him. To put it into perspective, Schoop's OBP beats out teammate Pedro Alvarez (.315) despite him having a BB% almost 7% lower than Alvarez. 

Schoop's inability to walk has always been troubling, but if he's found a way to get on base at a league-average rate despite his low walk rate, that eliminates arguably his biggest weakness and turns him all of a sudden into a complete player that could put up 3.0 WAR seasons at second base for years to come. 

Still, that's a big 'what if'. Without good speed, a .320 OBP is not easy to sustain with a 3.8 BB%. Looking at the underlying stats, the increase in OBP has been fed by a higher average that in turn has been supported by a .335 BABIP. While Schoop hits the ball hard enough to generate an above average BABIP, his career mark of .317 indicates a batting average regression is coming. So unless he starts getting hit by pitches regularly, the OBP should come down to around .300 by the end of the season. 

I did find two interesting adjustments that pitchers are making this season in response to Schoop's pitiful 2015 walk rate. First, Schoop is seeing 6.4% fewer fastballs according to PITCHf/x, which is a product of his .779 OPS vs. hard stuff / .635 OPS vs. soft stuff for his career. Second, Schoop has seen 6% more pitches in the strikezone (Zone%) this season compared to last, according to Baseball Info Solutions. In fact, his 48.8 Zone% is among the highest in baseball this season, which means that pitchers aren't afraid to give him a pitch to hit because they know he swings and misses a ton (his 17.5 SwStr would have led baseball last season if he had enough PA to qualify) and has poor discipline. 

But while he has continued to swing and miss a ton, he is taking the lemons and making lemonade. He's increased his contact percentage on pitches in the zone (Z-Contact%) by almost 5%, which has been a big reason behind his overall bump in Contact%. Baseball is a game of adjustments, and you can bet that pitchers will collectively dial back that Zone% as they realize that Schoop has the ability to do damage given enough pitches in the zone.   

Lastly, I thought it would be interesting to look at other players this season who have been valuable despite low walk rates. I set the WAR floor at 1.0 and the BB% ceiling at 5% and filtered to find these names.

Salvador Perez and J.T. Realmuto aren't the best comps because they get a big positional adjustment boost for being catchers. Starling Marte and Eduardo Nunez rely on their exceptional speed to get on base. Kevin Pillar fills up the WAR bucket almost entirely with his stellar center field defense. That leaves us with Adam Duvall and Rougned Odor. Both are intriguing comparisons. Odor obviously has speed but the fact that they're both second baseman who hit the ball hard brings them closer. Duvall is a more disciplined hitter but has very similar Soft/Med/Hard hit rates to Schoop. 

Regardless of which is the better comp, expect Schoop's OBP to come down to around Odor and Duvall's level (in the .290 to .300 range) as pitchers adjust and his BABIP regresses. 

27 June 2016

Orioles Rotation is Fine or The Rotation is Falling, The Rotation is Falling.

One Day a HR was hit and Oriole Little Thought the Rotation Was Failing
More money, more complaints.  That often is the way things roll.  The Orioles spent a good bit of change on Ubaldo Jimenez and the team is experiencing a typical even year performance from him (note - I cannot wait to see who is the Jimenez Whisperer in a collection of articles in October and November).  People then rejoiced that a potentially injured Yovani Gallardo signed a reduced deal and showed up with a reduced fastball.  It is enough to forget that Chris Tillman, Kevin Gausman, and Tyler *one second...checking roster* Wilson have actually performed well or well enough.

All in all, the starting rotation has neither been a problem or all that good although the common refrain from the crowd is that it has been a horrific disaster.  How does this compare to years past in terms of percentile performance in the American League?
If you want to go by ERA, then the starting rotation has been better than last year and equal to nearly playoff bound 2013 club.  If you want to strip a little luck and defense from things, then the rotation is tied with the best one since 2012.  If you want to now strip luck, defense, and normalize home run rates,  then it also comes up as tied for the best starting rotation in the past five years.

Another way to look at performance is based on a simple view on performance.  A while back, I ran a study on what first division, average, and bottom rung pitching rotations look like based on ERA+.  Amazingly, it follows a pretty even spread:

Slot 1 2 3 4 5
1st Div 125 115 105 95 85 105
Average 115 105 95 85 75 95
Bottom 105 95 85 75 65 85
Orioles 124 112 97 72 63 94

As it stands, the Orioles have an average starting rotation, which is neither good nor a problem.  The club has a first rate bullpen and lineup, so it can work with a mediocre starting rotation.

How does the club get to a first division rotation?  It needs to replace Jimenez' 63 with a 110 ERA+.  So someone like Orioles' pitching prospect Zach Davies ().  Unfortunately, the club traded him away last year for some guy wearing a Gerardo Parra jersey (that deal was panned by me at the time and it was expressed by several scouts in the game that the Orioles gave up way too much for half an unsustainable season of Parra).  Outside of Davies, Drew Pomeranz' breakthrough or high water point season fits the bill with his ERA/FIP/xFIP slash as 2.76/3.32/3.64.  Jeremy Hellickson can also fit into that group (4.23/4.42/3.81).  You can also argue that Sonny Gray is a hopeful candidate (5.03/4.49/4.18).

Regardless, the point is that as the Orioles are four games up in the AL East with a strong record that they really have no major weakness.  Their starting pitching is not good, but we need to be realistic.  When describing the club, we should not act like Chicken Littles.  The sky simply is not falling.  That said, the club's average starting rotation does have quite a bit of room for improvement.  If starting pitching improves, the club improves from a strong playoff contender to a team barreling down the AL East freeway into the post-season. 

24 June 2016

Why The Orioles Should Trade For Nolasco

The Orioles rotation is not having a very good year. Despite the Orioles having a 41-30 record, the rotation is only 23-23 with a 4.93 ERA (4th worst in the majors). More advanced statistics treat them more favorably, as they have only the 8th lowest FIP and somehow rank 18th in fWAR. While Chris Tillman is having an excellent year, the only other starters with an ERA under 5.00 are Kevin Gausman with a 0-5 record and a 4.37 ERA and Tyler Wilson with a 3-5 record and a 4.57 ERA. An upgrade probably wouldn’t hurt for the rotation.

The challenge is that the Orioles have already spent significant resources building their rotation. They signed two free agent starters, Jimenez and Gallardo, which are earning significant cash. Likewise, Tillman is arbitration-eligible and is also making a decent sum. The Orioles may not be willing to splurge on another starting pitching asset, especially given how poorly Jimenez and Gallardo have performed. In addition, there is little quality projected to be on the market this deadline. Despite the fact that there are nine teams at least eight games below .500, there is minimal quality pitching to be found. The Orioles need to decide whether they want to deplete their few remaining sources on a gamble.

Except, there’s one pitcher, Ricky Nolasco, that will probably be available for cheap. It’s for good reason, this is his first year out of the last three where his ERA has been below 5.00, and he’s flirting with the 5.00 mark this year also. In addition, he’ll cost a pro-rated portion of $12M this year, he’s under contract for $12M in 2017 and he’s got a shot of triggering a vesting option for $13M in 2018. For this post, the data is from ESPN Stats and Information.

On first glance, he looks like a pretty awful rotation piece. However, a closer look suggests he’s a viable option for the rotation. The first hint that he may have some potential is the fact that he has a 4.95 ERA but a 3.69 FIP. The second is that he pitches for a team that plays poor defense. The Twins defense ranks 27th from 2013-2016 and has been worth -114 runs. Is bad defense causing him problems?

This year, a quick glimpse of his stats shows that has BABIP is .336. But, more importantly, his BABIP in 2015 was .392 and in 2014 was .354. It seems fair to propose that his defense is hurting him. In addition, his walk rate went from 8.1% in 2015 to 4.7% in 2016. This improvement has come mostly against left handed batters, as he had an 11.1% walk rate vs them in 2015 and just a 4% rate against them this year. His walk rate against righties is well-within career norms. The table below shows his overall performance.

Batters seem to have hit him hard on pitches put into play, regardless of whether the pitches have been in the strike zone or not. Batters have a .403 wOBA against pitches hit in the strike zone and a .390 wOBA (.400/.394/.500) against pitches hit outside the strike zone this year. Batters usually don’t do well when they swing at pitches out of the strike zone, and it’s not like they’re hammering home runs against him. Out of the 97 qualified pitchers, that’s the 10th worst result, although Nationals’ aces Strasburg and Scherzer are allowing a .404 and .406 wOBA against those types of pitches this year. This problem can probably be attributed to bad luck or bad defense. He ranks 32nd out of 97 against pitches put into play that are in the strike zone. This isn’t good, but does suggest that we can expect a slight bit of improvement from him. The table below shows his performance in these situations.

In addition, he’s having dreadful luck with men on base. When the bases are empty, batters have a .327 wOBA against him with a .296 BABIP, which is a bit lower than his results in previous years. With one runner on base, opposing batters have a .347 wOBA against him with a .397 BABIP. These results are slightly better than they’ve been in past years. With two runners on base, batters have a .410 wOBA against him with a .370 BABIP. In addition, Nolasco only has a 10.3% K% and a 15.4% BB%. This is far worse than his results in other seasons with the exception of 2015. With bases loaded, opposing batters have a .440 wOBA against him (.571/.444/.714) and he has gotten 2 non-sacrifice outs in play, 1 strikeout, 2 sacrifice flies, 3 singles and a double. Simply put, batters are killing him in the clutch. Nolasco historically does struggle more with runners on base than without runners on base, but it seems that he’s gotten some bad breaks to start this season.

As discussed previously, one area where Nolasco has improved is walks against lefties. The reason for this is because he’s throwing fewer balls. 35.2% of his pitches against lefties have been called balls in 2016, compared to 43% in 2015 and roughly 38% from 2012-2014. Meanwhile, his called strike rate is around career norms and his swinging strike, fouled balls and balls put into play right are above his averages. As a result, he has allowed significantly fewer walks. See the chart below.

Part of the reason for this is that he figured out how to throw his splitter again. Last year, he couldn’t throw his splitter for a strike and batters simply didn’t swing at the pitch. Nearly three-fourths of his splitters last year against lefties resulted in a ball and only 12% resulted in either a called or swinging strike. This year, 40% of his splitters result in a called ball while 22% result in a strike. He’s fixed his splitter and that’s helped him be successful. See the chart below.

In addition, he’s gotten lucky with the curve and slider. He’s throwing the same amount in the zone as he has in previous year, but lefty batters are swinging at 56.6% of them as opposed to roughly 40-45% in previous years. Therefore 27.5% of these pitches have been called balls while 31% have been either called or swinging strikes. This either means that his curve/slider have become significantly better at fooling batters or that Nolasco has gotten lucky.

My opinion is that Nolasco probably isn’t as good as his FIP suggests, but I think he’s gotten more bad luck than good luck. Going forward, I’d expect him to have an ERA in the low 4s. I also think that being behind a good defense will help him significantly. He’s under contract for only one more year after this one, so trading for him wouldn’t be a huge risk.

I don’t think it would take so much to acquire him. He does have a limited no-trade clause, but the three teams on it are the Red Sox, Yankees and Blue Jays. The Orioles probably wouldn’t be willing to take on Nolasco’s entire salary, so the first piece in a deal would have to be either Jimenez or Gallardo. The Orioles seem to be more down on Jimenez than Gallardo, so it makes sense to try to build the deal around Jimenez.

The first thing the Orioles would need to include is enough cash to pay the difference between the two players’ salaries. This will only be a few million, but every little bit counts. In addition, the Orioles will likely need to add a few prospects. Nolasco hasn’t had good results for the past few years, so they won’t be expecting much in return. This is the quintessential case of buying low on a player. I think that offering Mancini and Scott would be enough to get a deal done. However, the Twins would likely prefer one of Bundy, Wright or Wilson in a deal as opposed to Mancini being as they already have Mauer and Park. I would think that the Twins would need to add another piece, or take on some more cash to get one of those pitchers.

This trade would be a risk, but perhaps not overly much. If Nolasco doesn’t pan out, well Jimenez is already in the bullpen. Furthermore, Mancini isn’t likely to be able to hit well enough to become a DH while Scott is at best a reliever. These may be some of the Orioles’ best prospects, but they have limited value. And if the Orioles are going to make a deal, something like this is the best they can hope to pull off.

20 June 2016

The Orioles Have a Lot of Hard Payroll Decisions to Make

At first glance, it would seem like the O’s have a lot of long term flexibility in terms of payroll.  According to Cot’s Contracts, they are at $87M for 2017, $51.5M for 2018 and then all the way down to $30M in 2019.  That, however, is simply the amount of money we know is already committed, as these numbers necessarily do not factor in arbitration raises. In some ways, this is good: Not having a lot of money committed long term should allow for greater flexibility to sign extensions with current players or to go out and get free agents.
The flip side is that much of the team is only under control for two more seasons.  Here’s a list of current players who will not be under team control in 2019 as of today:

Adam Jones
Matt Wieters
Brad Brach
Zach Britton
Mark Trumbo
Yovani Gallardo
Chris Tillman
Hyun-Soo Kim
Dylan Bundy
Ryan Flaherty
JJ Hardy
Oh, and a guy you have maybe heard of.  He’s having a pretty good year.  Young shortstop, used to play third, punched a guy in the face a little while back…
The point here is not to say that the team will look different in 2019, because that’s obviously true for many teams.  The issue is that the entire core of the current team will not be guaranteed to be in Baltimore in just three seasons.  Even in 2018, the only players who are under team control from the above list are Manny Machado, Jones, and Britton.
So, what’s a road map for the team in terms of payroll going forward?  Everything should start and stop with Machado.  He’s the biggest and most important domino and the future of the franchise is largely based on whether the Orioles can retain him.  Let’s say, for the sake of this post, that the O’s can sign Machado to an extension after the 2016 season.  Let’s also say he gets something similar to what Jon Shepard proposed: 10 years, $400M, with an opt out after year 5.  Assuming there is some money deferred, I'd estimate Machado will be around a $35M per year payroll number.
Under this scenario, a Machado extension would raise the total committed payroll in 2017 to $122M, which does not include salaries for Tillman, Britton, Gausman, Brach, Caleb Joseph, or Jonathan Schoop.  Combined, those players made just over $16M in 2016.  That number could easily double in 2017.  If it does, the payroll would then be at a record high of $150M, and that does not include other, smaller arb raises, not to mention possibly re-signing someone like Trumbo, Wieters, Alvarez, etc.  If the O’s need to make moves in free agency, just inking a big Machado extension will easily push the club into the top 6 or 7 payrolls in baseball (potentially even top 5) with little to no improvement in terms of actual players. Perhaps Machado takes a reduced salary for the first year or two which will be made up in subsequent seasons, but even so, it is very difficult to imagine the O’s not running a $150M+ payroll in 2017 that may not even include their current starting catcher, RF, and DH.  

Things calm down a bit in 2018.  Ubaldo Jimenez’s deal will be off the books, as will Hardy’s and Gallardo’s (though it will require $4M total to buy them out, which is the most likely outcome).  That represents a savings of over $35M.  The problem there is that Tillman will be a free agent and will probably command a $15-20M commitment per season.  If he leaves in free agency, the rotation could be something like Gausman and... well, the rest is unclear.  It also means that the team would likely need a third baseman, a left fielder, a right fielder, and possibly a catcher.  That’s a lot of holes when you consider that, even with some big contracts coming off the books, the payroll will be around $100M (with the theoretical Machado extension) without any of those positions being filled.  This is also not taking into account potential extensions for guys like Gausman, Schoop, Bundy, etc.  

There are, of course, players that could plug these holes that are currently in the system. By 2018, Chance Sisco should be in line to make his debut. Perhaps Trey Mancini moves to the outfield or becomes a DH. DJ Stewart has had a slow start to his professional career, but as a former first round pick there's a chance he is ready to contribute in two seasons. Jomar Reyes will still only be 21 years old, but he is consistently ranked among the team's top prospects and could be in a position to move up to the big club. Pitchers like Cody Sedlock, Chris Lee, Tanner Scott, Hunter Harvey, and David Hess could be knocking on the door for the rotation as well. All of these players, however, come with the inherent risk and unpredictability of being prospects and they will almost certainly not all pan out. The O's will need to rely on their farm system to fill some of the holes on the big league roster for sure, but they will also need to go out of the organization as well.
So, what’s the solution?  It’s honestly hard to see one with the current roster and payroll.  A Machado extension after 2016 (which seems like the most likely time for it to occur, if ever) will require a huge payroll jump in both the short and medium term if the Orioles want to keep the team together.  For a team that, prior to 2016, has traditionally been closer to the middle than the top of payroll rankings this would require a pretty significant shift in philosophy.  Depending on what happens the rest of this season, perhaps they simply throw caution to the wind and go all out for 2017 but the potential repercussions of that kind of approach could be drastic.

The alternative would be to sign Machado and simultaneously rebuild by trading guys like Britton, Tillman, Gallardo, O’Day, and maybe even Adam Jones to get younger and cheaper, and in so doing build around Machado, Davis and a new crop of prospects to go along with what they currently have in the minors.
Of course, the third and most treacherous path (one that many O’s fans don’t want to contemplate) is to trade Machado for a huge haul in prospects.  To me, though, that would only be viable if they also dealt the rest of their core players as well and went into a total rebuild.  Additionally, the longer they wait the less trade value Machado will have.  Teams will certainly line up to try and trade for Machado, but the price tag would necessarily go down the closer he is to free agency and one of the biggest contracts in history.  Essentially, the O’s need to make the decision on Machado by the end of this season in order to put a plan in place to either go for broke or to rebuild and re-tool.
To me, the most viable path forward for the team to remain competitive long term is to sign Machado and rebuild at the same time.  It seems unrealistic to assume that the O’s will spend $150M+ every year, and trying to keep the current team together will absolutely cost that much and is likely to cost more.  The front office has been surprisingly liberal (at times) regarding payroll issues, so perhaps this analysis is not giving enough credit to the idea that the Orioles will do whatever it takes to win a World Series.  It would not be entirely dissimilar to what the Tigers have done for the past five or so seasons, and with some clever contract structuring it might not have to be a long term disaster.  

In the end, the Orioles are going to have to make some very difficult decisions regarding the roster and payroll in the very near future.  I see this as a feather in the team’s cap.  Without spending on ridiculous free agent contracts they have built a team that has become so good that it is almost too expensive, but their success may be the impetus for pretty drastic roster changes.  Just another reason to win the World Series this year.

17 June 2016

How Tillman Has Gotten Lucky

Chris Tillman is having a pretty good year as he’s 9-1 with a 2.87 ERA while averaging 6.05 innings per start. His FIP isn’t as favorable as it’s only 3.90, but he still ranks 18th out of 45 qualified starting pitchers in FIP and 13th in WAR. This comes as a fortunate surprise after a brutal 2015 in which he had an ERA of 4.99. How has he improved?

In an interview with Jon Meoli, Chris Tillman argued that his improvement is due to throwing his off-speed pitches for strikes. The data I’ve seen (downloaded via ESPN Stats and Information) suggest that this is indeed a factor. He’s been able to throw his curve and slider for strikes more often and therefore has forced batters to swing more frequently. As a result, his curve and slider have a combined swing-strike rate of 13.1% compared to his 2013-2015 rate of 6.6%. In addition, his changeup has been deadly against left handed batters. This helps explain why his strikeout rate is at 24.2% compared to last years’ 16.2%. More strikeouts typically result in better performance.  This chart shows his basic performance.

Furthermore, batters are doing roughly the same amount of damage on balls put into play as they’ve done in previous years. When putting the ball into play, opposing batters have a .266 BABIP against him, a .355 wOBA and a 4.5 HR% in 2016 compared to an .277 BABIP, .355 wOBA and 4.1 HR% from 2013-2015. In addition, his walk rate has increased from 8% in 2013-2015 to 9.1% in 2016. The sole area where he’s improved seems to be strikeouts. In part, I definitely agree with his assessment.

However, the change in his performance comes into better focus when we look at his performance based on whether runners are on base. His strikeout rate with the bases empty is 24.4% in 2016 compared to 17.6% from 2013-2015. This year’s rate is easily the best of the four. However, his wOBA is .334 in 2016 and just .320 in 2013-2015, suggesting that his performance is worse this year than in previous years. This can be explained by the fact that his BABIP from 2013-2015 with the bases empty was .269 compared to this years’ .297.

Tillman has improved with men on base. When there’s only a runner at 1B, opposing batters have a .247 wOBA against him in 2016 compared to a .338 wOBA from 2013-2015. This is in part due to the fact that he has an 28.8 K% and 5.1 BB% in 2016 compared to a 16.3 K% and 7.2 BB% from 2013-2015. But it’s also in part due to a .222 BABIP in 2016 compared to a .293 BABIP from 2013-2015. This explains why his wOBA in play for these situations was .373 from 2013-2015 but .310 in 2016. Certainly, he’s doing better, but he’s also getting lucky.

However, the area where Tillman has really improved is when there are runners in scoring position. His walk and strikeout rates have actually gotten worse. He has a 20 K% and a 10.7 BB% in 2016 with RISP compared to 21.8% and 7.9% in 2013-2015. But he also has a HR% of 1.30% and a BABIP of .220 in 2016 compared to a HR% of 2.2% and a BABIP of .287 from 2013-2015. As a result, his OPS from 2013-2015 with RISP was .678 and his wOBA was .295 but in 2016 his OPS was .557 while his wOBA was .258.

His wOBA in play with RISP was .257 in 2016 and .344 from 2013-2015. It’s possible that his HR% indicates improvement from 2013-2015 rather than random chance. After all, his 2013-2015 home run rate suggests that he should have allowed slightly roughly 1.7 home runs instead of the 1 home run that he actually has allowed. If so, his lower HR% accounts for 40% of his improvement while the other 60% is a result of his abnormally low BABIP.

Given that his K% and BB% aren’t particularly good, it’s reasonable to expect some regression. His OPS with RISP ranks 20th out of 99 qualified pitchers, suggesting that he isn’t unique in this regard. There were other pitchers that also have been successful in these situations despite poor strikeout and walk numbers which should be expected given the small sample sizes we’re dealing with.

It does appear that Tillman has gotten lucky with pitches put into play with RISP. He’s given up slightly fewer line drives and fly balls in 2016 compared to 2013-2015 while allowing slightly more ground balls and popups. But this dataset seems to think that Tillman has given up a .892 wOBA against popups in 2016, so it seems that there isn’t a drastic change in the type of contact he’s allowing. The drastic change is that the average FB and LD in 2016 is only doing 85% of the damage that it did from 2013-2015 while the average GB is doing only 62.5%.

Likewise, it’s worth noting that Tillman is having more success throwing all of his pitches with RISP in 2016 than he has previously from 2013-2015 and that batters are doing less damage putting his pitches in play with RISP than in normal situations.

All of this seems to indicate one of two things. The first possibility is that Tillman has discovered a way to not just improve his off-speed pitches so that they are harder to hit, but also that they’re more likely to result in outs. In addition, he’s found a way to be super-efficient with RISP and will be able to maintain a .220 BABIP with RISP. If so, then this will be a strong season for him.

The second possibility is that he’s improved his command of his off-speed pitches and thus will be able to get more strikeouts but has also been fortunate with RISP.  If this is the case, I would expect his results and thus his ERA to regress towards the mean. This probably does explain why his FIP is so much worse than his ERA and that his FIP probably more accurately describes his performance than ERA.

Of course, and especially after 2015, most fans would be happy with a 3.90 ERA from Tillman. This would certainly be an improvement and would make him a solid member of any rotation. He’ll provide some stability to a rotation that sorely needs it and is on pace to earning an all-star bid. But he also probably isn’t the ace that the Orioles need.  The Orioles shouldn't overweight his current performance when deciding to give him an extension. It looks like he’s just back to his usual of being a solid but not great starter. 

15 June 2016

2016 Orioles Draft: Sinkers, Strong Arms, and Good Contact

The Orioles went hard after pitching in the draft this year, which makes sense because this draft was dominant with pitching prospects.  Based on the word of several people in the industry, the Orioles' draft went hard after pitchers who effectively use two seam fastballs.  The glory of the two seam fastball was preached in Moneyball and is generally well respected and acknowledged to this day.  It is also known that on level swings with these pitches results in a good bit of homeruns.  In short, everyone knows about sinkers and everyone knows they are not a silver bullet in a repertoire.

What may not be known is whether there is a greater probability of advancement or production for pitchers with established two seamers.  That might be where the Orioles might think they have an advantage.  A movement toward being an organization catering to two seam pitchers may be real.  The club has focused on drafting strong armed infielders when available.  It also appears to have some notion to which catchers pitch frame well.  Framing helps a two seam pitcher use a strike zone that is slightly extended below the knees.  Frankly, it would not be surprising to see an average framer like Matt Wieters being let go and having excellent framers like current Oriole Francisco Pena or FA-to-be Jason Castro being paired up with Caleb Joseph.

For the few position players the Orioles drafted, there is one commonality: arm strength.  Nearly all of the draftees who play the field have a strong arm and good contact skills.  One minor contention I had was the selection of Austin Hays over Heath Quinn.  Both are valued about the same with Quinn possessing plus raw power, but with Hays having a more polished and playable swing. Prep SS Alex Torres can be described similarly: strong arm and good contact.  Preston Palmeiro's arm is not bad, but is largely irrelevant at first base.  His swing though is a good contact swing.  Position players chosen outside of the top ten also share those attributes.  It may be that the powers that be in the Orioles system want position player arm strength (as players like Schoop show it can make up for poor range) and contact skills (which appear to soon be reaching their high point as a value skill in the current main stream iteration of Moneyball).

Back to pitchers, below is a table summarizing the pitchers the Orioles selected in the first ten rounds.  I was only able to verify two of the top 10 pitches as two seam fastball pitchers, but the others fit the stereotypical slider accompaniment and scouting report descriptions about heavy fastballs.  The pitches column should be read as "Established.Unestablished" pitches.  What I mean by unestablished is that a pitcher has shown the ability to throw that pitch and it has the potential to be an average or better offering, but it is not there yet.  For a prep arm like Hanifee, that is pretty common and the expectation is that the second pitch will become established while the third pitch hopefully becomes established.  When we are discussing later picks like Humpal, then that silver lining is a bit thinner.

Name Hand FB
Sedlock R 2S
Akin L 2S*
Dietz R 2S*
Hanifee R 2S*
Myers R 2S*
Moseley R 2S
Humpal R 2S*
Dube R 2S*
* 2S is expected to be main fastball based on indirect evidence.

Depot Shadow Draft
Each year since 2007, we have gone off and done our own shadow draft.  Briefly, our farm system has been more productive than the Orioles system.  However, that might well change because we did not choose Manny Machado.  He alone will likely offset whatever advantage we currently enjoy.

Anyway, a few years back, Nick Faleris departed and I became the sole maker of decisions.  Whereas he was able to directly scout the players, I use more analytics and reports to try to assess talent.  My focus typically is on starting caliber college position prospects, injured high projection HS talent, projectable two way HS talent, and then up-the-middle players or guys who can light up a radar gun.  In years past, I tend to be more in line with where players are taken, but this season saw me open with selecting four players who ultimately went in the third round.  Moving forward, I think I might consider the presumed Orioles approach of finding position players with strong arms.  That intuitively makes sense to me.

1:27 - RF Heath Quinn (3:95, Giants)
Analytic choice. Quinn went in the third round to the Giants, a few picks behind the Orioles who selected Austin Hays.  Hays is a more all around player who no loud tools, while Quinn has plus raw power and carries a little more risk.
2:54 - C Sean Murphy (3:83, Athletics)
Analytic choice. Murphy is a solid catcher and catcher defense is one of those skills that is difficult to acquire as a pro.  He has lately come into his body while dealing with a broken hamate, so his offensive talent might be obscured a bit.  Regardless, my model thinks he is a solid average prospect.
2:69 - LHP Jesus Luzardo (3:94, Nationals)
The analytics no longer project any prominent college player as a starter, so I switch gears to try to find damaged goods plus talent.  He is currently recovering from Tommy John surgery, but in the past has worked in the mid 90s with a heavy fastball and a decent changeup.  He will likely be an overslot selection, meaning I will have to lean heavily on college signings.
3:91 - LHP/OF Khalil Lee (3:103, Royals)
Lee is a two way prep prospect.  As a southpaw, he edged into the mid-90s this year.  Good speed and defense, I would probably try him out as a position player first and then work backwards into a pitching career.  Lee will also likely be an overslot signing.
4:121 - RHP Ryan Hendrix (5:138, Reds)
Hendrix has a strong arm, topping out at 100 mph.  He also has shown a plus curveball.  However, he has incredibly poor control, which was not present in former years.  The thought here is to acquire a live arm with past success and put faith in the developmental system to make this arm useful.  Sitting in the upper 90s, several mph could be lost and Hendrix would still be useful.
5:151 - RHP Bryan Garcia (6:175, Tigers)
Garcia throws a mid 90s fastball and a plus slider.  He could move quickly as a reliever or be tried out as a starter.  I think this is a high floor for a fifth round selection.
6:181 - SS Brandon Lopez (10:303, Twins)
Projects more as a utility player with limited range, but has performed well at the plate this year.  The key for me here is to simply bring in more athletic position players.  Quantity in hopes for hidden quality.
7:211 - SS Daniel Pinero (9:265, Tigers)
Pinero has average tools across the board with a decent arm.  A Canadian player, he growth curve may be a little behind others and might more fully blossom in the professional ranks.  The backup plan would be to try him out as a reliever.
8:241 - LHP Alec Bostic (10:305, Giants)
Sinker/slider pitcher who could start out in a rotation roll until proven otherwise.
9:271 - LHP Connor Jones (11:338, Yankees)
Mid 90s fastball from the left side provides a decent foundation to work with.
10:301 - UTL Weston Wilson (17:501, Brewers)
Wilson shows good power, but it has yet to play.  He has experience at both 2B and 3B.  Squint and perhaps a Ty Wigginton-like player might be in there.  Again, I am trying to find flexible players.

09 June 2016

CRAP: Collegiate Regression Analysis Projection (2016 Draft)

Live look at how CRAP data is processed.
Baseball is a game for the math obsessed whether you ponder on runs batted in or the spin rate of a slider.  In the past decade, the number of data points have exploded in the public forum.  That data are almost exclusively recorded from Major League Baseball games.  That has trickled slightly down to the Minor League sphere.  Amateur baseball though has simply raised itself up to where we were a couple decades ago with MLB.  We know the basic counting statistics.

Projecting performance with high resolution data can be troublesome.  Replace that with the low resolution collegiate data, the process becomes more problematic.  Combine that with twenty year old players whose distance between their current package and their final one can be rather vast, you have a difficult task in front of you.  This led me to experiment with minimal expectations to try to see if we can glean anything of use from a projection model for collegiate players.  The resulting model is named Collegiate Regression Analysis Projection (CRAP).

CRAP is a rather limited model with great uncertainty.  I used player, park, and team data from collegiate seasons 2012-2014 and paired it with professional data from 2013-2015.  Professional data was further translated so that everything was measured to the league average of the Carolina League (High A ball).  That is all this model knows.  It knows players who were drafted, signed, and played in a professional league the year after their career ended.  It does not know what happens beyond the following season.  All it knows is the current year and the next, trying to connect all of those performance data points for each individual player.

Some limitations include never considering players who were unable to play professionally for whatever reason.  There is also some limitation on conference data.  This system leans heavily on the ACC, Pac10, SEC, and the Big 12 or whatever they call themselves this minute.  I also have no knowledge of any player's defensive ability.  The model assumes they perform as an average defender at those positions.  Finally, there is no consideration of how a player performs with a wood bat.  I did a secondary adjustment for performance at the Cape and for Team USA.  That adjustment is a bit weak as we are getting pretty far down the rabbit hole from the original data.

Anyway, here is the collegiate position player board, which considered players in Baseball America's top 100 from a few weeks back (it has changed a bit).  If you want a player not mentioned here, let me know and I will try to post an addendum.

CRAP 2016 Draft Board (College Position Players)
Avg/OBP/Slg are 2017 HiA Projections
20-80 scale is unadjusted for wood
20-80+ is adjusted for wood

PLAYER  POS  AVG OBP SLG 20-80 20-80+
Kyle Lewis OF .290 .424 .473 61 69
Will Craig 3B .291 .413 .483 63 59
Zack Collins C .294 .446 .467 69 59
Heath Quinn OF .278 .376 .459 51 58
Sean Murphy C .273 .378 .414 51 54
Matt Thaiss C .295 .392 .428 54 54
Nick Senzel 3B .281 .374 .421 47 53
Corey Ray OF .276 .360 .425 46 51
Chris Okey C .270 .368 .419 51 48
Bryan Reynolds OF .268 .369 .429 48 45
Anfernee Grier OF .265 .349 .403 41 39
Nick Banks OF .255 .324 .391 35 35
Jake Fraley OF .261 .349 .367 34 33
Ryan Boldt OF .249 .310 .362 29 31
Stephen Wrenn OF .246 .314 .361 29 31
Bryson Brigman SS .257 .323 .343 30 30
Buddy Reed OF .233 .318 .354 31 30
Bobby Dalbec 3B .214 .292 .346 28 29
Errol Robinson SS .245 .309 .339 28 28

Below are the rankings for CRAP and BA's current rankings of college level position players.  I did not include some of the other players who have moved into the top 100 since I scraped the list a few weeks ago.

Kyle Lewis
Will Craig
Zack Collins
Heath Quinn
Sean Murphy
Matt Thaiss
Nick Senzel
Corey Ray
Chris Okey
Bryan Reynolds
Anfernee Grier
Nick Banks
Jake Fraley
Ryan Boldt
Stephen Wrenn
Bryson Brigman
Buddy Reed
Bobby Dalbec
Errol Robinson

As it stands, the Orioles are largely connected with Buddy Reed who stands high on the Baseball America rankings, but who is highly suspect on the CRAP rankings.