The Royals and Expected Strikeout Rate

          The Royals experienced their first sweep of the season at the hand of the Yankees over the last weekend.  It is crazy that the Royals went this long into the season (midway through May!) without being swept.  After last year’s 12 game losing streak at the beginning of the season, this is welcome.  However, the Royals have lost 6 of their last 7 games.  I don’t think the Royals will sustain this kind of losing for long, though.

            

          Today’s post will be about strikeouts.  The Royals currently have a K% of 17.6%, which is good for 6^th in the MLB.  As I have written before, the Royals depend on contact for offense, not walks and home runs (like the aforementioned Yankees do).  I have already written about the strategic value of this strategy, but today I want to look at whether or not the Royals can be expected to continue their low K% ways.

            

          Mike Podhorzer over at FanGraphs published an article detailing a formula for a stat called “xK%”.  This stat stands for “expected K%”.  Podhorzer wanted to create an equation that given a player’s plate discipline statistics would produce an expected strikeout rate.  I applied his formula to the 9 lineup regulars of the Royals, and the results are here.

Name	Z-Swing%	Contact%	Zone%	K%	xK%	Difference
Alex Gordon	0.636	0.795	0.460	21.6%	19.5%	-2.1%
Lorenzo Cain	0.633	0.847	0.499	15.0%	14.7%	-0.3%
Billy Butler	0.623	0.855	0.424	16.1%	14.8%	-1.3%
Eric Hosmer	0.749	0.796	0.451	18.4%	16.7%	-1.7%
Salvador Perez	0.609	0.874	0.500	16.8%	12.9%	-3.9%
Mike Moustakas	0.702	0.808	0.400	13.6%	17.2%	3.6%
Alcides Escobar	0.646	0.861	0.472	9.3%	13.3%	4.0%
Jeff Francoeur	0.732	0.709	0.467	23.4%	24.7%	1.3%
Chris Getz	0.647	0.887	0.502	14.3%	10.7%	-3.6%

             

          The three variables in the equation are Z-swing%, contact%, and zone% (the three input variables are in decimal form for Excel calculation purposes).  I have previously written about contact%.  Z-swing%, in case I haven’t written about it before, is the percentage of pitches in the strike zone at which a player swings.  Zone% is the percentage of pitches a player sees inside the strike zone.  The equation that Podhorzer created has an R^2 value of 0.81, so the model is quite good, even with a small sample size.

            

          According to this model, several Royals players are actually striking out more than they should be.  Given that this research is still fairly new, explaining the differences between xK% and K% is difficult.  It is possible that random fluctuation is causing the difference, or it’s possible that the player in question has the ability to beat out the xK% consistently.  I think if these 9 players were closer to their xK% levels, the Royals might have an even lower K%.  The extra observed K% could partially explain why the offense hasn’t quite performed up to expectations so far.  

Alcides Escobar Deserves Your Attention

            Amid the overall admirable performance of the starting rotation, one position player having a fantastic season of note is Alcides Escobar.  You might think to yourself “Hey, he’s not Alex Gordon, Billy Butler, or Frenchy, so I don’t know him”.  Or, you might know him because he’s our every day shortstop.  If you don’t know him, you will.

            After 98 plate appearances, Escobar’s triple slash line thus far is .303/.351/.472.  His career triple slash is .267/.309/.362.  Escobar’s current BB% is 7.1%, while his career BB% is 4.9%.  Escobar’s current K% is 7.1% as well, while his career K% is 13.2%.  Therefore, offensive numbers indicate that Escobar is on pace for a career best year in offensive performance.  It’s only April, though, so how much of his performance might stick is another issue.

            When a player is having an offensive performance like Escobar, there are a few indicators to look for in determining if the performance is “real” or not.  First, BABIP.  Surprisingly, Escobar’s BABIP is at .300 for this season, which is in line with his career BABIP of .303.  Escobar’s elevated offensive performance does not appear to be due to blind luck.

Related to BABIP is batted ball data.  Escobar’s line drive rate is a little higher than his career value, but slightly less than in 2012.  Escobar’s GB% is lower than 2012 and his FB% is higher than in 2012.  These three items explain his lower BABIP in 2013, but his 2013 rates are in line with his career rates.  Escobar’s elevated offensive performance doesn’t appear to be related to a shift in batted balls*.

*One thing to note is that he has hit 3 home runs so far this year, and in general his power numbers are higher than in previous years.  I do not know yet if his power increase will hold over the season because HR/FB% usually hovers around league average for most players, and Escobar’s is above average at the moment.  Also, home runs do not count in BABIP since they are technically not in play.

            Another indicator to look at is plate discipline.  Escobar’s BB% and K% are greatly improved over his career values.  While it is April, BB% and K% tend to stabilize more quickly than other stats, which suggests that Escobar has made real improvement in that area.  Other plate discipline indicators to look at are swing% and contact%.  Unfortunately, Escobar’s O-swing%, which measures the rate of swinging at pitches outside the strike zone, is at 39%, which is much higher than his career value of 33%.  Escobar is also making contact on those outside the zone pitches more than in previous years.  In fact, Escobar’s general contact% is higher than in previous years.  Escobar’s increased swing% and increased contact% is interesting considering his increased BB%, which would normally indicate that he is taking more pitches, not swinging more.  I think more time is needed to resolve this contradiction.

Escobar’s elevated offensive performance appears to be due to making more contact than in previous years.  Since Escobar has a fairly solid distribution of batted balls, he will likely hold his BABIP at its current value.  If Escobar starts whiffing on pitches outside the strike zone more, his contact rate could fall and his K% could rise.  If this happens, his offensive production will fall. 

           

            

The Offensive Philosophy of the Royals

            There are two major aspects of winning a baseball game: scoring runs and preventing runs.  There are obviously many different components of each of these aspects, but everything in baseball comes down to one of these two things.  Given that the Royals pitchers are doing their share of run prevention so far, I am going to write about scoring runs.  However, since the season is only 17 games in, I am going to write about the Royals run scoring capabilities since 2002, giving me a much larger sample to analyze such that conclusions may be more meaningful.

            In this analysis, I will be looking at some of the components of scoring runs, but certainly not all.  These are ones that jumped out at me whilst perusing FanGraphs.

            First, for a frame of reference, I looked at runs scored per game.  It’s fairly simple, but of great importance, since scoring more runs means more wins.  From 2002-2013 (remember-only 17 games in 2013), the Royals have scored a below average number of runs per game every season except in 2003 and 2011, which has certainly contributed to the generally abysmal seasons in those years.  In 2003, the Royals scored 5.16 runs/game, which was the highest in those years, while so far this year the Royals have scored only 4.00 runs/game, which is the lowest in those years.

            What is interesting here is that the Royals have had a massive amount of roster turnover from 2002-2013, including managers, GMs, position coaches, and players.  Despite these changes, the Royals offense remained below average.  Is there some organizational philosophy that just won’t go away?  Digging into the data, it seems to be so.

            During those years, the Royals have consistently had an above average GB%, which has led to a consistently above average BABIP.  The Royals have had a fairly consistent above average contact rate, which goes hand-in-hand with the relatively lower K% and BB%.  Like I have said in a previous blog post, I can live with a lower BB% if a lower K% comes along for the ride as well.  Having a higher contact% can help compensate for a lower BB% as well.

            So that’s all good and fine and dandy, but the Royals still score a below average number of runs per game despite those numbers.  Why?  The answer lies partly with pitches outside the strike zone.  Unfortunately, the Royals have generally swung at pitches outside the strike zone more often than average and make contact on pitches outside the strike more often than average.  Swinging at pitches outside the strike zone leads to weaker contact.  So, while the overall contact rate is above average, some of that contact is pointless.  The other part of the answer lies with the Royals’ HR/FB%.  Every single year, the Royals’ HR/FB% was below league average.  Fewer home runs equals fewer runs.

            What happened in 2003 and 2011, then?  In 2003, the Royals had a jump in BABIP and were as close to league average in HR/FB% as they ever were.  In 2011, the Royals had an extremely high BABIP.

            The Royals’ offensive philosophy revolves around making lots of contact at the expense of walks.  Thankfully, the Royals have emphasized ground balls in their contact, which leads to the consistently higher BABIPs, which helps compensate for the lack of walks.  However, in emphasizing ground balls, the Royals have in effect de-emphasized home runs.  Basically, the Royals’ offense relies on run scoring components somewhat out of their control, as the location of defenders and the pitchers themselves have some degree of control over BABIP.  The Royals need to sustain a BABIP much higher than league average in order to have an even average offense.  While the Royals have sustained a higher than league average BABIP, it doesn’t seem to help them score more runs.  A change in philosophy is needed.  The Royals need to emphasize limiting swings on pitches outside the strike zone and emphasize hitting a few more home runs.

What's Wrong with Mike Moustakas?

           As Dayton Moore’s Process matures, the Royals need their young players to develop into stars, or at least above average players, in order to make the playoffs.  Mike Moustakas, one of the most prominent alumni of the Best Farm System in the History of the Universe and Everything, is one of the young players who MUST improve for the Royals to have a shot.  There’s just one problem with that so far.  Over 1,000 major league plate appearances, Mike Moustakas’ offense has been terrible.  I’m going to show you some career numbers to back up that statement while comparing them to the league average numbers for third basemen in 2012.  Keep in mind that league averages fluctuate from year-to-year, so the comparisons might be slightly different if I compared his career numbers to all third basemen in 2011.

            I’m going to focus on things that appear drastically different from the league averages.  First, his triple slash line is .246/.298/.388.  The league average was .262/.323/.415 in 2012.  Moustakas is below average in all three categories.  Naturally, Moustakas’ BB% is below league average (6.4% vs. 7.7%), but he also strikes out less than league average (17.7% vs. 18.9%).  I can live with a lower BB% if it comes with a lower K% and more contact.

            Contact.  Therein lies the problem for Moustakas.  While Moustakas makes contact at an average rate, over his career, Moustakas has a .278 BABIP, which is far below the league average of .301.  Moustakas’ contact goes for hits less than league average, which affects his triple slash line.  The distribution of batted balls is telling.  Of Moose’s batted balls, 47.2% of them are fly balls, and the league average in 2012 was 35.0%.  Fly balls generally go for outs more often than ground balls and line drives.  As expected with a high fly ball rate, Moose’s ground ball rate and line drive rate are below league average.  If Moose hit more home runs with those fly balls, that would help his triple slash line, but only 6.9% of his fly balls go for home runs against 11.1% for the league average.

            The biggest factor I can see in the numbers affecting his offensive performance is this: a 19.4% career IFFB%.  IFFB% represents the percentage of fly balls that go for infield flies, which are just about equal to a strikeout.  The league average is less than half Moose’s rate.  Moose not only hits way more fly balls than league average, but way more of those fly balls are essentially AUTOMATIC outs.

            So, why does Moose hit so many fly balls?  Well, I am not a scout, so I can’t comment on mechanical issues with his swing.  I suspect that his swing has too many similarities to a golf swing, which would lead to more fly balls.  However, I can comment on the location of pitches at which he swings.  Baseball Heat Maps, a website by Jeff Zimmerman, an avid baseball researcher and Royals fan, has these data.  Against left handers, Moustakas generally swings at pitches he should.  Against right handers, however, Moustakas’ most concentrated swinging area is at the very high end of the strike zone, and slightly inside.  If a large proportion of Moustakas’ swings are at pitches high and inside, he is going to make a lot of weak fly ball contact.  In order for Moustakas to improve his offensive output, Moustakas must lay off high pitches against right handers and make the pitcher throw lower in the strike zone.

            This work does not necessarily invalidate previous analysis concerning Moustakas’ potential home run rate this year.  The work contained in this post is looking at past data to explain past performance, while the previous post about home run rate is using past data to predict future data.  If Moustakas hits so many fly balls, some of them are going to go for home runs, since he has power.  The Royals have new hitting coaches this year; I just hope they can spot any mechanical issues, if there are any, and fix them.  It will be up to Moustakas himself to develop better plate discipline.

Evaluating the New-Look Starting Rotation

In 2012, after 10 games, I looked at how the starting pitching was doing using various statistics.  In 2013, after 10 games, I'm going to do the same thing, since that will give an interesting comparison to how the starting rotation has improved since last year.  Keep in mind that the small sample size rule applies here, and one game can change statistics drastically.  Like Mendoza's start last night.

In 2012, after 10 starts, the rotation averaged just over 5 innings per start.  In 2013, after 10 starts, the rotation averaged just under 6 innings per start.  Improvement!

In 2012, the rotation had a 6.49 K/9 and 4.38 BB/9, which was awful.  In 2013, the rotation had an 8.65 K/9 and 1.97 BB/9, which puts the Royals 4th in MLB in both categories, which is absolutely fantastic.  Improvement!

I would like to note here that I do prefer K% and BB% instead of K/9 and BB/9 since starters rarely pitch 9 innings, but since I used K/9 and BB/9 last year, I wanted to stay consistent for comparison's sake.

In 2012, the rotation had a 1.52 WHIP.  In 2013, the rotation had a 1.30 WHIP.  Improvement!  The Royals rank 15th in the MLB in WHIP, but there is very little separating the middle third of MLB in WHIP.

So, by those measures, the rotation has vastly improved from last year.  They pitch longer into games, and they strike more guys out while walking fewer.  Unfortunately, the Royals have been a bit unlucky so far since the staff has a .320 BABIP, which ranks 25th.  To compensate for that unluckiness, the Royals have a 77.2% strand rate, which ranks 8th.  While guys are getting hits, the Royals are not allowing many of those runners to score.  Having a high strikeout rate and low walk rate helps prevent those runners from scoring.

The burning question here, though, is this:  Is this performance sustainable?  It's possible.  I expect the K% rate and BB% rate to regress to the mean, but I think this staff is capable of above-average rates in both categories.  Just not as great as they are now.  Guthrie and Mendoza have never shown high K% rates, but Santana and Shields have in the past.  Davis is the unknown.  Partially compensating for the regression in K% and BB% will be an improvement in BABIP.  I don't think they'll allow a .320 BABIP all year.  Another factor is the LOB% (the strand rate).  I think that will regress a little bit, too, due to a lower K% and a higher BB%.  When Danny Duffy and Felipe Paulino return will be a progress check point.  If it is difficult to find starting pitching spots for those two guys, the Royals will be in a good situation.  If those two easily replace two of the current starters, then the Royals will probably not be in the hunt for the playoffs.

Overall, the starting rotation as a unit has been in the top 5 in MLB in the first 10 games.  I don't think that is sustainable.  A top 10 rotation is possible, but I think a top 15 rotation is the most likely.  After a 2012 rotation that ranked near the bottom 5, I'll take the improvement.