The Royals Are A Sabermetrics Team

Jarrod Dyson has helped to form baseball's best defensive outfield. (via Arturo Pardavila III)

Jarrod Dyson has helped to form baseball’s best defensive outfield. (via Arturo Pardavila III)

For two straight seasons, the Kansas City Royals have defied just about everyone’s expectations and played deep into October. The Royals are few people’s idea of a smart, progressive and well-run organization, but that didn’t stop them from getting back to the World Series and taking home the crown this season. After, of course, they finished with the American League’s best record.

Despite all this, many in the sabermetric community still view Kansas City with an air of ironic amusement and skepticism, as if the team’s success is more smoke and mirrors than the result of some well-executed plan. Royals Devil Magic — or rather, #RoyalsDevilMagic — has been a popular phrase on Twitter this month. While many fans are quick to place analytically inclined clubs like the Rays or A’s on a pedestal, the Royals are rarely held in such regard, even as the wins keep piling up. Their run to the World Series in the 2014 postseason happened through a good deal of luck, sure, but they were among the best squads in baseball this season and backed that up in the playoffs.

Numerous reasons exist for why Kansas City has developed such a reputation. The polarizing Wil MyersJames Shields trade (which, funny enough, worked out better for the Royals) spawned not only a bevy of hot takes, but also a tendency to dismiss the organization as incapable of going toe-to-toe with baseball’s shrewder clubs.

Manager Ned Yost, too, has come under fire for his affinity for the sacrifice bunt and an often-archaic handling of his pitching staff. But Yost isn’t the only manager who makes head-scratching decisions (Mike Matheny continues to display his own brand of tactical futility each October), and there’s evidence he’s adapted well enough from his previous mistakes, like when he turned to Wade Davis for a six-out save to close out Game Four on Saturday night.

Yet beyond these criticisms, the Royals have massively outperformed their projected win total the past two years. For writers and analysts to disparage a team’s moves and decisions is one thing, but when the computers all predicted the Royals would finish close to .500 (or below it), the notion their success is the product of luck becomes easier to accept.

Indeed, nearly every projection system, from Steamer to ZIPS to PECOTA, forecast Kansas City to miss out on the playoffs in 2015. That didn’t happen, of course, and while the club has benefited from good fortune at times, its achievements are the result of more tangible reasons than mere luck.

A closer look at how the Royals are run and the manner in which they’ve built their current roster reveals an organization that is smarter and more progressive than it’s given credit for. In fact, if Kansas City had a better reputation within the sabermetrics community, the Royals would be receiving far more praise from analysts and statheads alike for their play this season.

The Royals have risen to the top through a shrewd roster-building vision that has exploited what other clubs value most on the open market and optimizes the playing environment in their home ballpark.

To begin with, the idea the Royals are some old-school organization that eschews advanced stats is plain wrong. Kansas City has a bigger and better analytics department than many realize; the team just rarely divulges much information about that work to the public.

The club has no fewer than four full-time employees who work in analytics, and that doesn’t include any interns or consultants. Mike Groopman, who has spent eight years with the franchise and previously worked at Baseball Prospectus, was promoted to the role of director of baseball operations/analytics back in January. In fact, he just became the first BP alum to be part of a World Series-winning team.

Groopman oversees Kansas City’s analytics department, which also includes John Williams, Daniel Mack and Guy Stevens, whose academic backgrounds rival any front office analyst in the game. Williams attended Yale and has a graduate degree in atmospheric science from MIT. Mack has a Ph.D from Vanderbilt where his studies focused on machine learning, and Stevens co-authored a paper that appeared in the “Journal of Quantitative Analysis in Sports” while he attended and pitched at Pomona. Mack and Williams spoke at the sabermetric conference Saber Seminar this summer.

A Hardball Times Update
Goodbye for now.

These are the types of talented people other organizations are praised for hiring, but the Royals continue to be pigeonholed as a franchise that still clings to outdated ideals and avoids advanced stats altogether. For this reason, the perception remains that they’ve lucked into this winning more than anything else.

Even more impressive than the front-office brainpower has been the way in which Kansas City has applied progressive approaches and taken advantage of undervalued assets to gain an edge on the field. They may not tout themselves as an analytically driven team, but the Royals have excelled with strategies any careful reader of Moneyball should recognize and applaud.

Kansas City’s recent success begins with its defense. That fact won’t come as a surprise to anyone. What some observers might not realize, though, is just how great the Royals are in the field and how this has benefited them in other areas as well.

Back in July, Dan Szymborski wrote about how the 2015 Royals were on pace to be one of the greatest defensive teams in major league history. And indeed, they outpaced just about every other club in baseball with their gloves this season. Kansas City ranked first overall by a wide margin in defensive value and team UZR. The Royals finished second in Defensive Runs Saved behind only the Diamondbacks.

More importantly, this strong performance in the field has resulted from a clear front-office strategy. Kansas City has specifically targeted and developed great outfield defenders in Alex Gordon, Lorenzo Cain, Jarrod Dyson, Paulo Orlando and others. The Royals have paired a fast, athletic outfield with their spacious ballpark, and in turn, been unafraid to acquire fly-ball pitchers who might not be great fits elsewhere but perform well in Kauffman Stadium with the Royals’ stellar defense behind them.

Shields, Chris Young, Jason Vargas, Edinson Volquez and Jeremy Guthrie all have found success in Kansas City the past two years, in part because the team knew each hurler’s weakness would be eased by pitching in an environment suited to his skill set. That each of these pitchers came with his own perceived faults also enabled the Royals to obtain them cheaply, except in the case of Shields. Though even there, the player the Royals traded was an outfielder in Myers who doesn’t necessarily fit that great-defense mold.

Given how undervalued defense remains around baseball, Kansas City deserves credit for committing resources to elite fielders and pitchers who benefit from good glove men in the outfield. In exploiting these market inefficiencies, the team’s front office has constructed a squad that excels at run prevention without ever having to dive into free agency for a high-priced starter.

The Royals haven’t just won with defense, of course. At the plate, they’ve built a lineup filled with high-contact hitters in an era when power has become pricey and highly sought after. Moreover, in a time when hurlers have dominated and strikeouts have soared, Kansas City has assembled an offense that is less susceptible to the effects of elite velocity and pitching’s widespread supremacy.

As Ben Lindbergh wrote recently at Grantland, evidence exists that high-contact hitters perform better against power pitchers, a notion that has been borne out in the playoffs for two years in a row. (The Giants, too, have succeeded with a similar offensive makeup.) Although burly sluggers often can do more damage, they’re less consistent at the plate, and they’re vulnerable when facing hurlers who excel at getting strikeouts. Watching the Mets’ power arms churn through a formidable Cubs lineup demonstrated this concept to a “T.”

Of course, Kansas City proved to be a much tougher foe in the World Series, and that isn’t the result of some small-sample-size fluke. Since the start of 2013, the Royals have posted the lowest strikeout percentage in baseball by nearly two percentage points, the same-sized gap separating the clubs ranked second (Oakland) and 14th (Boston) on the list. The Royals are a clear outlier here, and that their lineup also led the league in contact rate is no accident.

At the end of the day, one still might prefer an offense full of mashers. But the Royals don’t have the cash to spend big money on those kinds of hitters on the open market nor could they trade for them without emptying their farm system. Instead, they’ve built an above-average lineup far more cheaply in a manner any smart front office would love to emulate.

Kansas City also has been on the cutting edge in building one of baseball’s best and deepest bullpens. Relievers like Davis, Greg Holland, Kelvin Herrera, Ryan Madson and Luke Hochevar have given the team an almost unmatched advantage in the late innings. And, as we’ve seen the past two postseasons, that advantage only grows in the playoffs. Other organizations, most notably the Yankees, have sought to copy the Royals’ bullpen blueprint in recent years.

All these stellar arms have helped take pressure off the starting rotation and enabled the Royals to consistently depend on relievers for one- and two-inning spurts rather than the diminished performances of tiring starters in the middle innings. Indeed, for an organization that supposedly eschews progressive thinking, Kansas City has done a great job circumventing the times-through-the- order penalty with a deep bullpen that carries much of the workload.

The club’s relievers led all AL squads in innings pitched this season and finished with the league’s lowest bullpen ERA at 2.72. The Royals’ phenomenal bullpen has allowed Yost to frequently spell his fatigued starters in a way many analytically inclined fans have long been calling for.

Moreover, in Davis, Hochevar, Brandon Finnegan and Will Smith, they’ve also done a great job converting mediocre starters into valuable relievers over the past few seasons. This pattern isn’t a coincidence and has aided their ability to remain among baseball’s best teams at run prevention.

From this perspective, the reasons Kansas City has so outperformed projection systems the last couple years are much easier to discern. The Royals excel defensively, which is the one area of the game we still have legitimate issues measuring and forecasting accurately in the public realm. Yet just because fielding value is harder to quantify than offensive production doesn’t mean the Royals have been unable to gain an edge through valuing defense more than other franchises.

Similarly, one of the major shortcomings of projection systems is how little weight they place on reliever impact. When you hold an advantage in nearly every high-leverage situation, that’s going to end up being a huge benefit to your wins total. That the Royals have excelled in such circumstances lately (and likely more than any other club) helps explain just how they’ve been able to outstrip forecasts of their true talent level.

Perhaps most importantly, when Kansas City called its shot, the team got it right. History is littered with teams who pushed their chips into the middle of the table and came up empty. Larry Anderson to the Red Sox, Carlos Beltran to the Giants, Jon Lester, Jeff Samardzija and Jason Hammel to the A’s, are but a few of the trades that immediately spring to mind as examples where the acquiring team was left unfulfilled. But this year, the Royals landed two big fish in Johnny Cueto and Ben Zobrist, and it paid off in a World Series trophy. Both players were integral to the World Series triumph over the New York Mets. For a team with little margin for error financially, it was imperative the Royals wait for the right time to call their shot, and they did.

Ultimately, Kansas City is still not viewed as a sabermetric team due to a dated reputation that no longer accurately reflects the organization’s capabilities. Instead, the Royals should be regarded as one of the smartest organizations in baseball — a franchise, much like the Pirates or Astros, that has developed a clear plan and carried it out to great success. No, the Royals haven’t excelled solely because they possess some hyper-advanced analytics department filled with mad scientists churning out data and formulas that are well ahead of everyone else. But what they’ve done probably would draw far more accolades from the sabermetrics crowd if a team like the Rays or Cubs had succeeded with similar strategies.

The Royals, for their part, are probably fine with being misjudged and underestimated. They have one of baseball’s best front offices and a collection of hard-nosed players any fan would love to root for. Luck is the last reason for their run to the World Series crown.

References & Resources


Alex Skillin has written for SB Nation, Beyond The Box Score, The Classical, Sox Prospects, Fire Brand of the American League, and Celtics Blog, among others. Read all of his writing on his website, and follow him on Twitter @AlexSkillin.
76 Comments
Oldest
Newest Most Voted
Inline Feedbacks
View all comments
Blue
8 years ago

“Similarly, one of the major shortcomings of projection systems is how little weight they place on reliever impact. When you hold an advantage in nearly every high-leverage situation, that’s going to end up being a huge benefit to your wins total. ”

Using reliever chaining to measure WAR is stupid and lazy. The Royals show exactly why this is the case.

Al
8 years ago

“The Royals are a poorly run joke of an organization!” — Sabermetricians, all year long 2015

“The Royals won the World Series by practicing sabermetrics!” — Sabermetricians, November 2, 2015

Carl
8 years ago
Reply to  Al

Yeah, that’s exactly how this reads.

John C
8 years ago
Reply to  Al

Only those who dogmatically stuck to the script that there’s only one way to succeed in baseball. The Royals built this team perfectly to exploit the characteristics of Kauffman Stadium and to take advantage of a market inefficiency. Speed, defense and contact hitting have all been deprecated ever since sabermetrics came into the mainstream. But that’s a misunderstanding of the science. The Royals are dead last in walks drawn, but they’re seventh in OBP. If you get on base, it doesn’t matter if it’s a walk or a hit. The Royals hit their way on, and they do it well enough. And once they do, their speed and baserunning is extraordinarily good.

They can’t afford to throw a ton of money at pitchers, so they took another tack. Kauffman is a terrible home run park. So you pick up pitchers who throw a lot of fly balls on the cheap, knowing the park will help them out, and your fast outfielders will help them even more. And you back them up with power arms out of the bullpen. Relievers are cheap, relatively speaking. The Royals have them out the wazoo.

There’s more to sabermetrics than walks, three-run homers, and strikeouts. But that’s all systems like Steamer and PECOTA are looking for.

tz
8 years ago
Reply to  John C

John C – great comment to a great article.

Actually, it’s funny about walks and hits. If you just go by OBP then a walk is as good as a single. If you use wOBA, singles are worth a bit more because wOBA was built from run expectancies, which account for the baserunner advancement you get from hits but not necessarily from walks. So you could argue that part of the “sabermetrics doesn’t get it” issue is which generation of sabermetrics are being used. Heck, for a lot of folks OPS is about as far as they would dare to go.

Of course, like any field, there’s surely room for sabermetrics to better explain how a team like the Royals outperforms expectations. While luck is always involved, simply attributing outperformance to luck is just plain lazy.

john
8 years ago
Reply to  tz

There’s no hard and fast rule. If there’s no one on base, a walk is as good as a single. If there’s someone in scoring position, a single is much better than a walk. It’s fine for a team to have a couple of walk-oriented hitters to put in the leadoff and #2 positions, where they’re going to come up a lot with no one in scoring position, but if the entire team is walk-oriented, they’re going to have serious problems because at some point someone is going to have to drive them in. And it isn’t a solution to have low-average hitters with a lot of home runs hitting behind the high-OBP guys, because although it’s true that they’ll hit some multi-run homers, they’ll also strand a huge number of runners in scoring position, and that will hurt the team more in the long run (if you’ve calculated WPA yourself a lot you’ll know this). On the other hand, you can cover up for a team with not-very-good OBP and a lousy BA by hitting a lot of HRs, even if this means a lot of solo HRs, because the outs are generally also with the bases empty–this is how the Astros managed to have the 3rd best Off rating even though they were #21 in AVG, #16 in OBP, and extremely unlucky in sequencing (#29 in Clutch). So the original Moneyball HR+walks-for-everybody model is seriously misguided.

Pat
8 years ago
Reply to  john

I wouldn’t say a walk is exactly “just as good as a single” even with no one on base, because a single can allow the opportunity for the defense to additionally make an error when fielding the ball in a way that a walk generally can’t. It’s not the most common thing to have happen, but it does happen enough that I’d say it should be a consideration

The 1980's: the LAST great decade of MLB...
8 years ago
Reply to  John C

Is here again. Who better to exploit and undermine the three true outcomes philosophy than the Royals, since they last won a championship 30 years ago, in 1985.

If interested, I compared 1980-85 to 2010-15 and here is the results:

Average Team Contact Outs & Errors Gained/Lost
Year 1980’s 2010’s Diff
1980v2010 3422 3046 376
1981v2011 3438 3064 374
1982v2012 3394 2983 411
1983v2013 3354 2996 358
1984v2014 3333 2974 359
1985v2015 3328 2947 381
w/ 81-11 20269 18010 2259
w/o 81-11 16831 14946 1885

Average Team Differential: Hitting Components Gained/Lost
Year Hits 2B 3B HR XBH 1B CO + E
1980v2010 49 -39 13 -35 -61 110 377
1981v2011 -5 -47 8 -48 -87 82 374
1982v2012 46 -32 6 -35 -61 107 411
1983v2013 37 -25 14 -28 -40 77 358
1984v2014 51 -32 10 -14 -37 88 359
1985v2015 11 -28 6 -25 -47 58 381
w/ 81-11 189 -204 57 -186 -333 522 2260
w/o 81-11 194 -157 48 -137 -246 440 1886

Year SB CS SB% Year SB CS SB%
1980 127 62 67.2% 2010 99 38 72.3%
1981 118 64 64.8% 2011 109 42 72.2%
1982 122 62 66.3% 2012 108 38 74.0%
1983 128 62 67.4% 2013 90 34 72.6%
1984 117 58 66.9% 2014 92 35 72.4%
1985 119 55 68.4% 2015 84 35 70.6%
w/ 81 731 363 66.8% w/11 582 222 72.4%
w/o 81 613 299 67.2% w/o 11 473 180 72.4%

Year PA SO EXTRA PA W CO + E H SUM HEWCO Runs RBI
1980v2010 14 -366 381 -45 377 49 381 -16 -25
1981v2011 -22 -381 359 -10 374 -5 359 -46 -53
1982v2012 56 -398 454 -2 411 46 454 -4 -12
1983v2013 14 -388 403 8 357 37 403 24 14
1984v2014 44 -383 426 16 359 51 426 31 20
1985v2015 45 -385 429 37 382 11 429 12 4
w/ 81-11 151 -2301 2452 3 2260 189 2452 1 -52
w/o 81-11 173 -1920 2093 13 1886 194 2093 47 1

It should be noted that in the 1980’s there was only 26 teams rather than 30 and in 1981 there was a strike causing only 2/3rds of the season to be played.

When comparing the 80’s to the 2010’s if its a positive number in the above comparison tables, that is the 1980’s winning the war on the 2010’s, if negative the 1980’s lost the war against the 2010’s.

Essentially it says that the 1980s kicked the ass out of the three true outcomes philosophy, by choosing more consistent contact, contact outs, singles over HRs, doubles, SLG, walks, and strikeouts. In 1983 – 1985, major league teams won 2.5, 3, and 1 more game PER TEAM AVERAGE respectively than 2013-2015.

The 1980’s had worse defense than the modern era but that is it. Their stolen bases were worse in percentage in, but greater in volume and that extra speed helped the 80’s take extra bases, first to third, second to home, even when NOT stealing a base. You can see evidence of this because there were also more triples in the 1980s than the 2010s. Hitting and Pitching were BOTH better in the 1980s. For the decade of the 80s, there were usually 20 to 45 players that struck out 100+ times a year. In 2010s, its over 100+ players that strike out 100+ times a year. 116 players struck out 100+ times in 2015 alone. In the 1980s there were more players that walked more than struck out.

To this end, I created stats of HEWCO, BSM, and CCR which render average/OBP/SLG/OPS useless. Let me know if you would like to see these. Bottom line is the Royals played like a 1980s team and beat all the TTO teams out there, because they didn’t sacrifice contact and consistency. The beauty of it is everyone can hit a single, not everyone can hit a HR.

DrBGiantsfan
8 years ago

Great article. Someone who writes on this site finally gets it right!

Steve
8 years ago

@Al

Exactly, and the same sentiment in 2014, 13,12 ….

Carl
8 years ago

“The Royals are few people’s idea of a smart, progressive and well-run organization…” What?

Carl
8 years ago

If sabermetrics just means “scouting” and “preparation” now, then “sabermetrics” doesn’t mean anything. Nice job wasting everyone’s time, guys.

Eric
8 years ago

No the Royals are NOT a Sabermetric team, they are a collection of individuals that know how to PLAY TOGETHER AS A TEAM -team baseball like in the 1980’s, the last great decade of MLB!! The Royals won because of HEWCO, BSM, and CCR – the stats I created to replace batting average, OBP, SLG, and OPS, which are outdated and do not account for everything good created by offense. Email me if you want to know more about them, thecrazybaseballcoach@gmail.com. I will post a 4 part series on these stats from my linkedin account too.

Shankbone
8 years ago

Very nice article. The Giants have been down this road as well, with a full time squad of analytic guys quietly doing great work. They also went after high contact rate hitters (“hackers”), taking the hit on OBP (and getting mocked for it by most sabermetric publications/bloggers) built a great bullpen, focused on team chemistry/clutch (which is another round of mocking) and built to the team to fit the park via defensive runs saved. Seeing the Mets get their weaknesses exposed was tough, but pitching and defense are huge in the playoffs. The Royals were absolutely amazing in the late innings, they deserve the Ship! Built for the long haul of the season and the sprint of the playoffs. The anger over the Shields trade was pretty amazing, and all the critics were completely wrong. Dayton Moore is one of the best GMs in the game, Friedman might be the most overrated…

John C
8 years ago
Reply to  Shankbone

I wouldn’t call Friedman overrated. How many times did he get the Rays into the playoffs while being in the same division with two of the wealthiest organizations in baseball?

Not taking anything away from Dayton Moore, but it took him nine years to get KC into the postseason, and he only had to get past Detroit once he got them into contention. Friedman banged heads with the Yankees and Red Sox every year.

Shankbone
8 years ago
Reply to  John C

The number is 4 playoffs, and while that is impressive you have to consider how bloated the Yankees and Red Sox were (are?) during that time period of 2008-13. Then again, both teams did win a title, the Yanks in 09 and the Sox in 13. But a lot of the success came from inheriting a franchise player (Longoria), being able to draft 1st for Price and a killer trade in Zobrist (Aubrey Huff). The Rays have good pitching development. However… The drafts after 2007 and Price have been disasters, Price and Zobrist are gone, the Rays are a mess now, and he bolted for the bloat of the Dodgers and their cable contract where he is spending more money on players not playing than he famously did on the Rays on the cheap.

Moore had to build from scratch (well, Alex Gordon) and drafted HS guys in Hosmer and Moustakas – one pick after Price in the later case. HS guys take longer to develop. Looks like he’s got there now. Just saying he has been constantly underrated (and berated in Saber circles) while Friedman has been absolutely lionized.

John C
8 years ago
Reply to  Shankbone

I think Dayton Moore is, and will, get a lot more credit going forward than he has, while if Friedman fails to win a championship with the Dodgers, his stock will deservedly drop.

Moore always had a vision for the Royals, and I give him credit for that, but he made some atrocious player-personnel decisions at the major-league level when he started out. The one thing he did right was Gil Meche, and even that got messed up when the manager he hired ruined Meche’s arm. KC wasn’t going anywhere until Moore finally nailed it with the Greinke trade. That got him Cain and Escobar, and indirectly helped get him James Shields and Wade Davis (when he sent Odorizzi to Tampa in the controversial blockbuster deal). There’s not enough said about that deal. Moore was basically forced to trade Greinke, and he hit it out of the park. Since then, everything he’s done has turned to gold. And I think part of that is because he’s learned to marry his already formidable scouting acumen with modern analytics.

john
8 years ago

It would be more accurate to say that the Royals SHOULD be a sabermetrics team, meaning that sabremetricians SHOULD recognize that the statistical measures which they have been using have gotten things so wrong with the Royals in the last few years (and with other teams built in a generally similarly way, like the Giants in the last few years and the Angels around 2001-5) that they should be an inspiration to make up better statistics. One obvious target is evaluation of reliever importance. WAR as presently conceived just isn’t working. Wade Davis’ WAR this year was only 2.0. Out of the 141 starting pitchers with at least 100 IP (so basically the top five starters on every team), that puts him even with Mike Pelfrey, #71, right in the middle (and less than Wade Miley (2.6), Bartolo Colon (2.5), Andrew Cashner (2.3), Jesse Chavez (2.1), and Robbie Ray (2.1), just to name a few). Does anyone seriously think that Wade Davis’ value to a team is the same as that of average major league starting pitcher? Who would, for example, have thought it would helped the Royals if they could have traded David to the Mets in exchange for Colon for the Series? On the other hand, Davis’ WPA value was 4.01, #6 of any pitcher, starter or reliever. WAR relies heavily on FIP, and FIP doesn’t consider when a PA takes place, so it’s problematic for relief pitchers, whose appearances vary wildly in terms of leverage, so that it doesn’t make sense to weight them equally (this is obviously much less of a problem for starters). For example, from the point of view of a closer who starts the 9th inning with a 2-run lead, it doesn’t (and shouldn’t) matter whether the first batter hits a home run or a single, and they will generally pitch that way, but the difference between a home run and a single in terms of FIP is enormous. WPA is intended to deal with this problem, by relying heavily on situational context, but the problem with WPA alone is that it doesn’t account for luck in terms of BABIP, which is especially important for relievers because they face so few batters. The best solution to this problem, I think, is to value relief pitchers with a combination of FIP and WPA. For example, WPA could be understood to be the basic evaluator, but whereas home runs, walks, and strikeouts would be calculated as WPA is normally calculated, balls in play would be calculated according to some standardized expectation for what a ball in play will produce (that is, a certain number of outs, double plays, singles, doubles, triples, etc.). This would obviously be pretty complicated, but it would give a much better idea of what relief pitchers are really worth to their team.

Cool Lester Smooth
8 years ago
Reply to  john

I’d just use WPA/LI, which puts him at the same level as Corey Kluber, or RE24, which puts him between Max Scherzer and Sonny Gray.

john
8 years ago

Using just WPA/LI has the same problem as using just WPA, that it doesn’t factor in luck on BABIP because it doesn’t give more weight to BBs, Ks, and HRs. And RE24 doesn’t consider game situation at all. What I’m suggesting would be a lot more work, but it would combine the advantages of WPA with the advantages of FIP/WAR and this wouldn’t neutralize each of their disadvantages. It might be better to do this combined metric with WPA/LI instead of WPA, but my first instinct is to say no because this would decrease the value of pitchers who come in in really important situations, but I’m not sure about this.

tz
8 years ago
Reply to  john

What about RE24 divided by final scoring margin? RE24 does consider a big chunk of game situation, but doesn’t capture the difference in leverage due to score and number of innings left. Dividing by final margin would be a fairly painless way of converting it to a number that captures most of the important parts of leverage.

john
8 years ago
Reply to  tz

The probably isn’t weighing leverage (I’m not saying that the idea you’re suggesting here isn’t good, it just doesn’t address the problem here). The problem is that anything which doesn’t take BBs/Ks/HRs into consideration (like FIP does) is going to have too much luck involved.

tz
8 years ago
Reply to  tz

Gotcha.

The issue is that in order to strip out BABIP luck might understate the value of pitchers with an ability to limit damage on balls in play. I agree that much of the BABIP impact is out of the pitcher’s control (defense and a large part of batted ball location), but then you consider a guy like Mariano Rivera, whose .265 BABIP allowed was more than 30 points lower than the MLB average over his career.

I do understand the impact of bad luck/bad defense being magnified when it comes to reliever WPA. Though probably not as much as poor Jeurys Familia right now.

Eric
8 years ago
Reply to  john

I don’t have as much of a problem with Davis’ position in terms of WAR. It is an cumulative stat. Wade was worth nearly as much as Colon’s 194 innings pitched in only 67 innings.

john
8 years ago
Reply to  Eric

So if you were running the Royals, and the Mets came to you before the Series and offered to swap Colon for Davis just for the Series, you’d have accepted? Seriously?

Eric
8 years ago
Reply to  john

I said that Davis was worth about the same as Colon in only a third of the time. He was three times as effective. Had he pitched the same number of innings and kept everything the same, he would have had around 6 WAR. Hardly saying I want Colon over Davis.

john
8 years ago
Reply to  Eric

Oh, I thought you were taking WAR at face value (as many people do).

Dave T
8 years ago
Reply to  john

Of course not, that’s a silly question. Obviously Colon wasn’t going to pitch 3 times as many innings as Davis in the World Series. A short series with multiple days off is pretty much the maximum possible value for Davis, because he can be used so often for more than 1 inning.

Regular season value makes a lot of sense to me due to the difference in IP’s that Eric mentioned.

Dayton Moore
8 years ago

Saying the Royals won the Shields-Myers trade (should be called the Shields Wade Davis-Myers trade) is like saying the U.S. won World War 2 over Germany. Wasn’t a win, it was a demolition

Marc Schneider
8 years ago
Reply to  Dayton Moore

Well, the Russians had something to do with that too. The better analogy would have been the US over Japan.

CoolWinnebago
8 years ago
Reply to  Dayton Moore

The royals won, but i wouldnt call it a demolition, Odorizzi still exists.

Joe Blow
8 years ago

So, 15 years ago, I was laughed off of Royals message boards because my exact same message didn’t fit with the current SABR thinking. Now it’s cutting edge!

john
8 years ago
Reply to  Joe Blow

It still isn’t cutting edge for sabermetrics because people haven’t thought of statistical measures to replace the old ones and give a more realistic value for the Royals and teams like them. At this stage the leading theoreticians are still only at the stage of recognizing that there’s problem.

Cool Lester Smooth
8 years ago

Has anyone looked at whether Contact% correlates positively with RE24-wRAA?

It seems intuitive that high-contact teams will have above-average numbers of sacrifice flies and grounders that advance or score runners, which are often positives in RE24, but negatives in wRAA.

james wilson
8 years ago

The Moneyball comp for KC is exactly right. You take what is most undervalued at the time. But Oakland’s success changed what was undervalued, and so now will KC’s. Good luck to them on that.

The greatest shield KC had from being identified as a sabermetric organization was their manager, who they have survived two consecutive years. On the other hand this may only highlight how bad most managers are and how rare the Botchy’s of the game are. Managing men and managing game requires two different people in the same head.

john
8 years ago
Reply to  james wilson

No, the greatest problem in being identified as a sabermetric organization was that they didn’t go for home runs, walks, and starting pitching, all of which are overvalued by existing favored sabermetrics statistics. And I will admit that Bochy managed circles around Yost in 2014 but Yost really learned a lot (basically by imitating Bochy) and did a great job in the postseason this year in terms of things like getting his starting pitchers out of the game faster, using his relief pitchers flexibly instead of like a robot (hats off for having the balls to use Herrera for 3 innings yesterday when they were behind by 2 runs), not pinchrunning at every possible opportunity, and not bunting in stupid situations.

Obsessivegiantscompulsive
8 years ago
Reply to  john

Like Bochy, Yost appears to have learned from his disappointing first World Series. I noticed the changes in handling his team too, good for him and congrats to the Royals for winning the championship.

Obsessivegiantscompulsive
8 years ago
Reply to  james wilson

So that explains why Bochy is so good! He has a huge head, which helps him fit two different people in the same head!

Andrew
8 years ago

So we now have 2 years in a row where every major public projection system had the Royals as a .500 team, and 2 years in a row they went to the World Series, winning it this year. That seemed to be something of a trend. It was not a good year (relatively speaking) for projection systems, which Dave Cameron wrote about in more detail in this article (http://www.fangraphs.com/blogs/the-year-baseruns-failed/).

Is there any validity to the idea that maybe the pendulum has swung back a little bit, and that we have teams that have more advanced projection systems, thus causing the discrepancy we’ve seen the last few years? If the projection systems developed by the 4 individuals you listed above and used by the Royals are outpacing PECOTA or ZiPS? After all, wouldn’t it make sense that eventually a team would come along and improve on the projection systems most commonly referenced?

Or possibly, could it be that teams have a better understanding of projections, and are taking bigger risks in certain aspects of it, resulting in larger error bars? If 10-12 years ago teams like the Yankees and Red Sox and A’s had an analytic advantage that they no longer have, are we seeing some teams go out on a limb in some areas of the game where the reward is worth the higher risk?

The only way to find out is to continue evaluating Baseruns, PECOTA and ZiPS. If over the next few years we continue to see pretty wild deviations, I have to think that it is at least partly a reflection of the projection system.

Eric R
8 years ago

While the Shields/Davis trade has worked out well thus far — the Royals got 12.0 rWAR thus far from the players they got back, but had to pay them $37M [and still have team options on Davis for $8M and $10M].

Meanwhile, the Rays have gotten 5.1 rWAR for $1.5M of Jake Odorizzi and still have a pre-arb and 3 arb years left. They move Montgomery for Erasmo Ramirez who they got 2.1 rWAR for $522k and also has four team controlled years left.

They have a 22yo from the deal still in AA and turned Myers [after getting 0.9 rWAR for $1M] plus a couple of existing prospects into Souza [1.0 rWAR for $509k plus has five more years of control] and a couple of prospects .


Maybe the Royals are sitting at 12 rWAR for their haul after spending $37M for 5 player-seasons, but the Rays are currently at 9.1 for $3.5M with a whole lot of years of control for the various players involved.

Cool Lester Smooth
8 years ago
Reply to  Eric R

But you can’t just look at how much money they spent.

You also have to account for how much money have the Royals have made by adding two pieces that proved integral to back to back WS appearances, including a victory.

…even before you account for the fact that fWAR is absolutely horrid for evaluating RPs.

BobDD
8 years ago
Reply to  Eric R

Nice point Eric R. I had oversimplified it as a trade for two years of good production at high pay vs. 6 years of average to good play at below average pay. From that standpoint it is still almost guaranteed to be at least a good trade for the Rays. So far it has worked out as well for KC as anyone could have hoped for, but the Rays are line to at least break even and probably come out ahead. So a trade that is probably a win for both teams. But I still think that because of the 2 yrs vs 6 yrs angle that KC was bucking the odds (the Wade Davis of today was a surprise to them), and a similar trade should not be attempted again.

Cyril Morong
8 years ago

They had a .778 OPS with runners on and .701 with none one. With RISP it was .772. Their pitchers allowed an OPS of just .668 with RISP while it was .700 with none on. That is impressive given that teams usually hit better with RISP.

The Royals pitchers allowed an OPS in Late & Close situations of just .629, far below their overall OPS allowed of .710. That beats the normal differential by quite a bit.

Did they build their team to achieve these splits?

Cool Lester Smooth
8 years ago
Reply to  Cyril Morong

It’s pretty intuitive that high-contact teams will improve with runners on more than K-heavy teams, because baserunners prevent the shift.

Cyril Morong
8 years ago

But baserunners prevent the shift applies to all teams, not just high contact teams.

Cool Lester Smooth
8 years ago
Reply to  Cyril Morong

But high contact teams are more affected by the shift than other teams because, you know…the shift only affects balls in play.

Cyril Morong
8 years ago

But I guess we would have to know how much difference in general the shift makes and how often it is used against the Royals vs. other teams.

john
8 years ago

I don’t know if it’s so much conscious at what the market dictates, even if most teams don’t seem to understand it. Relievers are much cheaper than starters.

john
8 years ago

i don’t think this is necessarily true. High-contact teams tend to pull less and so they are less generally vulnerable to a shirt. This may not logically follow but this is the reality of the situation.

john
8 years ago
Reply to  Cyril Morong

And obviously any team with really strong relief pitching is going to have a lower OPS in Late%Close situations because that’s when you use your really strong relievers. This is clearly not an accident.

Cyril Morong
8 years ago
Reply to  john

But the Royals have a much larger than different differential than average. Normal differential is .052. Royals had .091. They designed it that way? How? By putting better pitchers in the bull pen instead of starting?

Also, how do they pitch so well with RISP? OPS is normally .031 higher compared to none on cases, yet Royals see their OPS allowed fall with RISP. Can you find pitchers who do better with RISP?

john
8 years ago
Reply to  Cyril Morong

Part of this may be luck. But a non-luck factor would be, for example, if the manager intentionally brings in strong relievers with people in scoring position in the 6th and 7th inning rather than saving them for pitching the whole 8th or 9th inning, or often not bringing their closer in in the beginning of the 9th but waiting until the other team gets a runner in scoring position (like for example Collins did last night). Another related non-luck factor would be the manager doing comparatively more lefty-righty switching in RISP situations. In all of these cases, it isn’t that individual pitchers do better in RISP situations but that different pitchers are used in RISP situations.

Cyril Morong
8 years ago
Reply to  john

But I guess we would have to know how much difference in general the shift makes and how often it is used against the Royals vs. other teams.

Cyril Morong
8 years ago
Reply to  john

Okay, good points. I guess we need to know if the Royals bring in relievers more often with RISP than other teams. In general, the Roayls only used relievers slightly more often than the league average.

For all of the AL, 34.59% of PAs were with relievers. For the Royals it was 35.88%. That works out to .5 more PAs per game going to relievers.

The Royals starters allowed an OPS of .755. The relievers had .629. For the whole AL, those numbers were .731 & .706. So a very large difference between the Royals and the league.

Is this a conscious strategy, to have mediocre starters and a great bullpen?

john
8 years ago
Reply to  john

I don’t know if it’s so much conscious at what the market dictates, even if most teams don’t seem to understand it. Relievers are much cheaper than starters.

Obsessivegiantscompulsive
8 years ago
Reply to  Cyril Morong

Hey, Cyril, hope all has been well, just wanted to say I miss your writing at Beyond the Box Score. Have you ever updated your lineup regression? The numbers still seems to work (I run the NL numbers against it every year, and I recall them still being close) but was just curious if you ever looked into it again.

Cyril Morong
8 years ago

Thanks for the nice comment. I have not updated those numbers. Maybe I will sometime. I have a blog http://cybermetric.blogspot.com/

john
8 years ago

Assuming of course that it’s predictable whose going to be good. The Giants were in a weird situation this year because their 8th and 9th inning men (Casilla and Romo) were actually less effective than most of their middle relievers (Strickland, Osich, Lopez, and Kontos), unlike in previous years, so although overall the relief pitching was well above average I would guess that the OPS in Late/Close situations wasn’t so good.

Dominik
8 years ago

After Reading moneyball many think that sabermetrics = Walks and the royals don’t walk. For most people sabermetrics are three true outcome hitters, the Astros and Cubs for example are built around TTO hitters and they are considered very sabermetric.

john
8 years ago
Reply to  Dominik

Well walks are one of the TTOs.

Dominik
8 years ago
Reply to  john

Of course they are and the royals don’t walk or strike out and are mediocre at home runs, so they are an anti TTO team.

john
8 years ago
Reply to  Dominik

I agree. I didn’t realize that that was the point you were making.

Eric
8 years ago
Reply to  Dominik

Personally, I have always looked at Sabermetrics like this: its applying strategic and economic thought to the game of baseball. So to that end, its nothing new. This is exactly what strategic planning and economics have been doing for decades in business, but not in MLB. Its turning logic from every angle to gain an advantage from the status quo. So if you absolutely hate the saying, “well that’s the way we have always done it” then Sabermetrics is for you because it also teaches that you can blaze your own path, or explain your version of how you see the game by utilizing statistics and math. Sometimes that is taking “the same old data” and looking at it in different ways, other times it requires starting from scratch and parsing the data into the granularity that you need and creating fields and formulas that don’t currently exist in order to gain an advantage in the marketplace. Personally, I have done this. I find batting average, OBP, SLG, and OPS, archaic. I also think that Runs AND RBI matter. Most people disdain RBI, but I do not. Why? Because offense is about some players who “set the table” (those that score runs) and some players “who eat” (RBI), and some who have the mindset to do both given the situation. Much of the time, most analysts disdain or place little importance on those counting or “volume” stats and give most credence to rate stats. I say, sometimes the rate stats can ring hollow if there isn’t the appropriate volume (R + RBI) backing it up. So to me, these rate stats do not capture all the positive happenings within baseball for my taste. So I built 3 formulas, HEWCO, BSM, and CCR. In order to find the best offensive players, you need to find those with the best mindset situationally and those that have a full understanding of the value chain in baseball. Enter HEWCO, the value chain is PA = backward K + K + PO + HBP + IBB + BB + fb + GO + FO + LO + PE + ME + 1B + 2B + 3B + HR, which reduces to PA = SO + W + CO + E + H. From left to right is increasing value in terms of Runs and RBI, but there is also an equilibrium point. SO’s have no value offensively, Walks have very little positive value, so having the below average walk rate and below average strikeout rate means that an above average rate of plate appearance end in contact, which are all higher up the value chain of HEWCO. BSM = a bases moved stat that I created. CCR = contact consistency rate stat that I created. The end result? These 3 formulas that I created 100% replace AVE/OBP/SLG/OPS rate stats. KC Royals? #1 in HEWCO for 2015 by 150 points, #2? = the Toronto Blue Jays. #15 or 16 The NY Mets, 600 points below the Royals. The two articles by FG:

http://www.fangraphs.com/blogs/alcides-escobar-and-a-bat-and-a-ball/
http://www.fangraphs.com/fantasy/alcides-escobar-elvis-andrus-and-the-perils-of-babip/

Alcides Escobar had a HEWCO value of 704.346
Elvis Andrus has a HEWCO value of 703.063
There were only 18 players with a HEWCO of 700 or higher according to my formula in all of MLB 2015

I think these formulas decipher a lot and it explains why Alcides “works” at the top of KC’s batting order, even better than the older/standard rate stats that I no longer use. Bottom line is that both rate stats and volume stats MATTER. My HEWCO stat can be expressed as a raw whole number value, or a rate based on per game or PA. Let me know if interested in the stats imtigerwords@yahoo.com

Bob
8 years ago

So, the Roylas are sabermetric, because they hire people who have explored the known weak areas of sabermetrics, such as quantifying fielding and the influence of relievers on starters etc. Are you sure these advanced sabermetricians have not put numbers on the power of team chemistry as well?

David Scott
8 years ago

Perhaps what we should be doing is not analyzing KC by the old sabermetrics but analyzing KC to create new sabermetrics.

john
8 years ago
Reply to  David Scott

That should have been the point of this article.

bmarkham
8 years ago

You keep mentioning the projections, but what about BaseRuns? Sure, the projections can be off on teams and it happens every year. Baseball’s not predictive enough for projections to be more than the best over/under. But let’s not forget that this team was a 84 baseruns win team this year. Are we prepared to say the Royals have achieved the ability to beat baseruns, or that baseruns isn’t capturing something about the Royals?

Great article none the less, and I enjoyed the read. I tip my hat to the Royals, they’ve had a great couple of years, and they bet on defensive value which is something that those of us that love sabermetrics should be happy about. I’m happy for their fans, who have had to endure a lot of tough times between the 80’s and now so this has got to be great for them. But the projections had them as a ~.500 team, and by BaseRuns they should have won 84. Maybe they figured out a way to beat BaseRuns, or maybe BaseRuns isn’t as good as whatever Expected Run calculation the Royals use. But I would have preferred to read more about why this may be the case instead of a constant refrain back to beating projections, which they really didn’t beat by much if you put more stock into BaseRuns then the actual win total.

john
8 years ago
Reply to  bmarkham

That will be the logical next step to this absurdity. Sabermetricians won’t even bother to predict actual wins but rather Baseruns. Then after that they’ll start suggesting that we do away with 9-inning games with scores altogether and just have everyone hit with no one on base and declare the winner to be the team with the most Baseruns.

bmarkham
8 years ago
Reply to  john

I don’t get what you mean here. All the projection are on a team level is calculating Expected Run Differentials, which is what BaseRuns is. There’s no way to predict the noise that will occur when sequencing plate appearance events into runs and runs into wins. The projections had the Royals as a .500 team, and by BaseRuns, they were 3 wins better than that. That’s somewhat interesting I guess, but much more important is that they completely broke BaseRuns to the tune of 11 games. My Cardinals made BaseRuns look silly this year too, but the difference is I have no delusions that the Cardinals were actually a 100 win team.

bmarkham
8 years ago
Reply to  bmarkham

should read “that the Cardinals were a true talent level* 100 win team”

KHAZAD
8 years ago

I enjoyed this article. Last year in July, I wrote an article for a site titled “Dayton Moore is playing Moneyball, we just don’t recognize it”. It was met with mostly pompous scoffing. I think there are alot of people that think walks are sabermetrics are all about, and the Royals don’t walk, so they are anti sabermetrics.

Moneyball was all about finding something undervalued by the rest of the league and exploiting market inefficiencies. Then, it was walks and OBP, which are not at all undervalued anymore. Instead, in the Royals case, it was defense, contact, (and limiting strikeouts along the way. DIPS metrics for pitching rely heavily on the ability to get strikeouts while people were completely ignoring the offensive ability to limit those as a skill.) relief pitching, speed, and (dare I say it?) character. While the starting pitching (the most expensive commodity in baseball) was not stellar, they tailored it to their ballpark and defense.

The league will adjust, and five or ten years from now, market inefficiencies may be completely different than they are now. Some team will find new market inefficiencies and analysts will not understand why. Analysts will be slow to adjust, as they always are, but may be able to better see in the rearview mirror what the Royals accomplished while no one was watching or giving them credit for it. The Royals don’t need their validation right now. Flags Fly Forever.

Frag
8 years ago

Something that should be mentioned was that the Royals had a WPA below 20% SIX times during the 2015 postseason:

http://www.sbnation.com/2015/11/2/9657312/2015-royals-world-series

That’s insane, even if contact-oriented teams with good bullpens are better able to comeback from a deficit relative to average teams. If Correa didn’t misplay that ball, we would be talking about a totally different narrative of winning the World Series.

Jose guillen
8 years ago

Dayton Moore has come a long way in understanding what works in Kauffman stadium, but let’s don’t forget the journey- the Jose Guillen contract, acquiring Yuni Betancourt twice, more recently the Francouer extension. None of these players fit this current genius with a plan narrative. He gave Omar Infante a 4 year deal less than 2 yrs ago, and he extended Jeremy Guthrie to find the cash. Also, let’s remember that the Royals had a lot of top 10 draft picks over Moore’s years and really only Moustakas and Hosmer panned out. Also, giving him credit for what Wade Davis and Alex Gordon are doing now, maybe great development, but their current skill set is not what he inherited with Gordon and traded for in Davis. Give him credit for Cueto and Zobrist, but those deals were only necessary to cover up for previous mistakes: Infante, Guyhrie and Vargas making $30+ million combined. He nailed the mid level signings last offseason, but by his record on mid level signing this is either a huge learning curve or outliers. Not to mention he has only developed one average starter in 10 years. The Royals have seemed to have found a market inefficiency, but one of the tenants of Sabermetrics is to look deeper at causality and don’t always take the results based narrative at face value.

Cyril Morong
8 years ago

I looked at the following 7 pitchers on the Royals. These guys were primarily starters and the vast majority of their innings were from starting. I found their weighted average of OPS allowed with none on and with RISP (weighted by PAs-maybe that is not quite right since SLG is TBs over ABS).

Chris Young
Danny Duffy
Edinson Volquez
Jason Vargas
Jeremy Guthrie
Johnny Cueto
Yordano Ventura

With none on, they allowed an OPS of .731. With RISP it was .691. As I mentioned above, the team staff as a whole did .032 better with RISP. So the starters had an even bigger edge. A comment was made above that mabye that .032 differential existed because they brought in good relievers with RISP and their relievers were much better than their starters.

From 2010-14 for all of MLB, OPS with none on was .702 and with RISP it was .733. So normally it goes up with RISP yet the Royals’ starters managed to have it go way down. For the AL in 2015, OPS with none on was .702 and with RISP it was .746.

Zita Carno
8 years ago

And there’s another element that a lot of people had missed but that I spotted right away: Five o’clock lightning. Way back in the 1930s, when the Yankees were embarking on their first great dynasty, right-fielder-cum-first-baseman Tommy Henrich coined this phrase to describe the team’s way of striking back in the late innings when they were behind in the game. Come the seventh, eighth innings (and remember, there were no lights in those days), this would happen again and again, and the Yankees would cut loose and end up snatching victory from the jaws of defeat. Five o’clock lightning. And in this World Series I saw this again and again—twice in extra innings—and I realized that this was the Royals’ mojo, and it was working for them in incredible fashion. And so they took the Series in five games, just as several had predicted they would do. Splendid defense, an incredible bullpen, great contact hitting, baserunning one would give his or her eyeteeth to have (how about those basestealers like Cain, Escobar!), and all those other elements—and yes, they had that mojo. Five o’clock lightning. This was no fluke. The Royals were, and are, and will continue to be, a team for the record books. Hmmm…do I, perhaps, sense a dynasty in the works?

Cyril Morong
8 years ago

“Five o’clock lightning. This was no fluke.”

So they discovered the secret to clutch hitting in late innings in close games?