The Royals Are Basically the Best Contact Team Ever

A funny thing about doing a job like this is that you’re expected to write all the time. Which means you’re expected to write early in seasons, and you’re expected to write analytically, and when you write analytically early in seasons, you can end up looking like a moron. Earlier this season, for example, I remember writing reasonably positive things about Carlos Peguero. And early last year, over at JABO, I wrote about how the Rockies weren’t striking out anymore. You can guess what happened to Peguero. And you can guess what happened with the Rockies. The Rockies resumed striking out. They lost lots of games.

So, this isn’t anything new, but early information can mislead. We’re always just trying to separate what’s real from what’s fake, and we tend to be too impatient about that. But every so often, you can spot something legitimate. Several months ago, also at JABO, I observed that the Royals were barely ever striking out. They were on a potentially historic pace, and it was definitely something to stay aware of. To be honest, I at some point stopped being aware of it, but then I noticed again. The season’s about over, now. The Royals are thinking about the playoffs. They’ll be taking with them one hell of a group of contact hitters.

In the first half this year, the Royals, as a team, posted the lowest strikeout rate in baseball. Then they got to the All-Star break, and since that point, the Royals have again posted the lowest strikeout rate in baseball. This isn’t a huge shock, because the Royals didn’t strike out very much a year ago, so this definitely didn’t come out of nowhere. Still, it’s about the magnitude of what we can see.

There’s a graph right below this. To put AL and NL teams on a similar scale, I looked at team strikeout rates by non-pitchers. Then, for the whole group of teams, I calculated the average and the standard deviation, so then I could assign for each team a z-score, reflecting relative strikeout rate. In case you’re not familiar with z-scores, a rating of 1.0 would mean that team’s stat was one standard deviation higher than the average. Same idea goes for negative numbers. Now, here are the relative team strikeout rates:

2015-relative-strikeout-rates

The Cubs have the highest relative team strikeout rate, north of average by more than two standard deviations. At the other end, we see the Braves, striking out far less than the average team. But then the Royals are practically there on their own. The Braves’ team z-score is -1.43. The Royals’ team z-score is -2.56. Put another way: the difference between the Royals and the next-closest team, in terms of not striking out, is greater than a full standard deviation. The Royals are an outlier, constantly putting balls in play when so many other teams have been willing to come to terms with whiffs.

The style is more or less team-wide. At this point, 13 Royals this year have batted at least 100 times. Only one of them has a strikeout rate that’s higher than the average. That one is star outfielder Alex Gordon, and his strikeout rate is higher than average by less than two points. The Royals haven’t made contact with swings any more than, say, the A’s, but they’ve put almost a third of all strikes in play. They lead baseball, by almost two percentage points. They like to swing, and they like to hit the ball fair.

I got curious about that giant z-score, so I dug into history. I went all the way back to 1950, and for every team, I calculated a z-score for that season. In this table, you’ll find the lowest relative team strikeout rates, excluding pitchers to keep the leagues even. I know there was baseball before 1950, but the further back you go, the less recognizable the game. So this’ll suffice.

10 Lowest Relative Strikeout Rates
Season Team K% Z-Score
2015 Royals 15.5% -2.56
2002 Angels 12.6% -2.44
1982 Indians 9.8% -2.36
1969 Braves 9.2% -2.31
1962 White Sox 9.6% -2.29
2004 Giants 12.5% -2.29
1975 Cardinals 9.0% -2.28
1961 White Sox 8.3% -2.24
1994 White Sox 12.5% -2.23
1986 Red Sox 11.3% -2.20

The Royals don’t have the lowest raw strikeout rate ever, but doing that would be almost impossible, given the changes in eras. Strikeouts, as you know, have gone up, so once you take that into account, then this year’s Royals really emerge. Relative to the rest of baseball, no team has struck out less often than this year’s Royals, beneath the average by more than two and a half standard deviations. Behind them you get the 2002 Angels, and it’s somewhat encouraging to know that team won the World Series. If you’re curious, last year’s Royals are four slots away from the table, with a z-score of -2.10. Those Royals seldom struck out. These Royals kicked that up a notch. Or down. I don’t know.

Something else to observe about the Royals — though their bullpen has gotten so much attention, last year their hitters led baseball in Clutch score. And this year, again, their hitters lead baseball in Clutch score, by a big margin over the Twins. Clutch hitting has been a big part of the Royals’ success, and it’s only natural to wonder if there’s a relationship between clutch team batting performance and contact ability. Here’s a plot of batting Clutch against those strikeout-rate z-scores, going back to 1974:

team-clutch-strikeouts

Would’ve been nice to find something, but nothing has been found, here. If you just look at the line, then you see a very slight relationship, but the r-squared number is pathetically small. There’s not much to indicate a link. Certainly nothing that would explain the magnitude of the Royals’ recent offensive clutch-ness. Contact hitting has worked for them, but it’s not a recipe, necessarily, for other teams to follow.

And make no mistake, there are several ways to put together a quality offense. This year the Cubs have the highest relative strikeout rate, and the Royals have the lowest, but Cubs non-pitchers have a 105 wRC+, against the Royals’ 101. The Royals very rarely strike out, but they’re also last in baseball in walks drawn, and the power is right around average. You know that the Royals aren’t an offensive juggernaut. This is more of a description of style.

But it stands to reason it’s the style preferred by Dayton Moore. For better or worse, this is what we have, and the team he’s built stands out from the pack. With the season about over, the 2015 Royals are on the verge of posting the lowest relative team strikeout rate since at least 1950. That’s not the kind of thing that happens by accident, and the Royals aren’t successful accidentally.



Print This Post



Jeff made Lookout Landing a thing, but he does not still write there about the Mariners. He does write here, sometimes about the Mariners, but usually not.


Sort by:   newest | oldest | most voted
Alex
Guest
Alex
7 months 29 days ago

Jeff — thanks for the interesting article.

One thing I’m curious about (and haven’t the ability to check) is the correlation between strikeout rate and power. More particularly, which teams historically have had the best combination of low strikeout rate and high [home run rate / slugging percentage / iso. / etc.]?

I ask because on the table above, one team that stands out — other than the Royals, of course — is the Blue Jays, who have a K% well below average yet who lead the league in home runs. Perhaps this is not surprising, as (my recollection is that) the Jays’ sluggers (Bautista and EE) have tended to have lower than expected K rates. The rest of the league leaders in homers are mostly clustered towards the other end of the spectrum (the Yankees are slightly below [above?] average in K%). It is one thing to avoid strikeouts at the expense of power. It seems to be another entirely to avoid strikeouts while also hitting home runs.

I add the caveat that perhaps I’m entirely wrong about all of the foregoing, in which case I apologize for wasting everyone’s time.

StroShow
Member
7 months 29 days ago

I’d be more interested in knowing how the Jays offense compares to the league offense on a per season basis. They’re so far out in front that it’s bound to be a pretty interesting comparison, maybe more for the modern era than the Ruth times.

Basically, is the Jays offense this season really noteworthy, or just “almost” noteworthy?

Jimmy
Guest
Jimmy
7 months 29 days ago

Jays have a team wRc+ of 115 which puts them in top 20 for all team seasons since 1950

Jetsy Extrano
Guest
Jetsy Extrano
7 months 29 days ago

Jeff, you like scatterplots with labeled outliers. We all like scatterplots with labeled outliers. Can you just have Fangraphs add a page that picks two random stats and scatterplots them for player-seasons? Reload for two new random stats scatterplotted.

Dustin
Guest
Dustin
7 months 29 days ago

I’m blamin all those games vs the Twins.

Dustin
Guest
Dustin
7 months 29 days ago

Blaming

Dusting
Guest
Dusting
7 months 29 days ago

Blamin

Dust
Guest
Dust
7 months 29 days ago

Blam

dbminn
Guest
dbminn
7 months 29 days ago

Thanks for the review, Jeff. I have a related comment and a question:

Comment – I don’t think clutch is the best parameter in this case. The Royals are not only clutch, they have a high WPA. Since the clutch index is relative to WPA, there would seem to be some underestimation in your analysis.

Question – Two outs, bases loaded or runners on 1B/2B, ninth inning. Game tied. Which lineup would you rather have hitting – Astros or Royals? How about 8th or 9th inning?

Maybe a simpler analysis, looking at the probability of BIP, singles, doubles, triples and homers, would give some insight into how many times each team would take the lead.

Intuition would say the Royals, with a higher base hit probability, would win more games than the Astros, who rely on the 3-run homer. Especially considering the superiority of the Royals bullpen during the season. Of course, intuition doesn’t replace analysis…

Again, great topic.

psualum
Guest
psualum
7 months 29 days ago

You can look at this in 2 ways..
Royals maybe be better with a man on 1st and 2nd, two outs and game tied because of
1) Propensity to put ball in play
2) Superior bullpen can hold 1 or 2 run lead

But if your 1st and 2nd, 2 outs, and down 1 or 2 runs, you would much rather have someone from the middle of the Astros lineup provide a chance at a HR then have to rely on a Royal Rally to comeback and win.

dbminn
Guest
dbminn
7 months 29 days ago

I agree that could be true, psualum. I chose the Royals and Astros because their wOBA and WRC+ are essentially the same. Their records in one-run games are very different (23-16 Royals v. 19-29 Astros).

Non-contextual stats would tell you that runs are runs. But in close games late, that may not be true. I don’t know for sure. I think I’d rather have someone who gets base hits 10% more frequently than someone who hits a homer one more time per hundred but strikes out at a greater rate (7% difference).

Brian Snyder
Guest
Brian Snyder
7 months 24 days ago

We can play this game all day

Tie game, 2 outs, Nobody on.

Royals or Astros?

siggian
Guest
siggian
7 months 29 days ago

My answer for #2 would depend on where that game is being played. In KC, I want the Royals team. In Houston, I am more tempted by the Astros team.

The Royals seem to be well constructed for their ball park. Putting the ball in play instead of going for the homer makes sense for them.

Roger
Guest
Roger
7 months 29 days ago

Sure, the game has changed, and the Royals of course strike out more than a dead ball era team, but it’d be pretty interesting historically to see how they stack up.

Billy
Guest
Billy
7 months 29 days ago

Did you read more than the headline? That’s exactly what this article is about.

twm
Guest
twm
7 months 29 days ago

Roger seems to use historically here to as “relative to all teams”, rather than the author’s more narrow use as “1950-present”. Mr Sullivan ends his research at 1950 because the games becomes “less recognizable” as we move backwards through time; Roger, I think, is interested in knowing the comparison with dead ball era teams regardless of these difficulties.

I know Internet trains us to believe otherwise, but snark truly is more effective after you have done the work to understand another person’s opinion/perspective.

The Humber Games
Guest
The Humber Games
7 months 29 days ago

To get a true strikeout rate comp between AL and NL don’t you have to do more than just throw out pitchers? Because whoever would replace them in the lineup would presumably have a non-zero strikeout rate – would it move the needle on your graph at all if you attributed the AL average strikeout rate for a #9 hitter as part of the NL team rate?

TKDC
Guest
TKDC
7 months 29 days ago

I doubt it changes much, but I think if you’re trying to be super even, you’d just cut out DH plate appearances.

Jon
Guest
Jon
7 months 29 days ago

The math here strikes me as a little fishy, since a Z-score of -2.56 isn’t that much of an outlier (1 in 200 observations falls in each of the tails). So it is hard to believe, if the central limit theorem applies, that 1622 teams over 66 years didn’t produce a bigger outlier than this. Not to mention that regression is a bitch and is likely to have an impact on the Royal’s K% even with only 10 or so games left in the season.

I wonder what the tail at the other end looks like. It’d actually be easier to explain a short tail at that end since the extreme outlier high K% teams are likely to have lost a lot of games and benched their worst whiffers eventually. But I don’t know why the teams headed to an outlier season of low K% would purposefully avoid it. Ideas?

Hurtlocker
Guest
Hurtlocker
7 months 29 days ago

“I know there was baseball before 1950, but the further back you go, the less recognizable the game.” I don’t understand this statement?? Why exactly is the game so different?? Because we count more things?

Hurtlocker
Guest
Hurtlocker
7 months 29 days ago

Maybe your article should be “the best contact team ever, after 1950 of course” since the first 60-70 years don’t count?

Concerned Reader John
Guest
Concerned Reader John
7 months 29 days ago

Equipment. Training. Diets. Mound height. Population size. The inclusion of minorities and foreign players. Size of the league. Travel. Pitching rotations (number of pitchers/innings/relievers). Defensive shifts. Snarky internet commenters generating equally snarky internet responses.

Hurtlocker
Guest
Hurtlocker
7 months 28 days ago

Yes indeed. So because of what you list the game before 1950 was likely much harder. Bad lights, few batters eyes, traveling by train for 12 hours. Did you ever hit .400 in little league?? Not so easy I bet, Ted Williams did it against adult athletes that were some of the best of that time. It still counts.

Me
Guest
Me
7 months 17 days ago

I think mound height, exclusion of minorities, and population/league sizes would offset any difficulty train travels would add.

Bad lights would mean (probably) more games played during the day.

There was an article written here a while back discussing how pitchers of that era could have been capable of pitching so many innings because they putting much less effort into the game in order to preserve energy. This and the “3 times through the order” (since pitching a complete game ensures that you will see that pitcher more) theory could help explain elevated averages.

Phantom Stranger
Guest
Phantom Stranger
7 months 29 days ago

I agree it’s a disturbing trend in SABR-friendly baseball press. In the past couple of years most stat searches have started arbitrarily cutting off years before a certain date, usually 1950 or 1961. At the very minimum I would like to see the field expanded to the live ball era.

Baseball is a game built on history and arbitrarily cutting off the sample is a very limiting paradigm change.

TKDC
Guest
TKDC
7 months 29 days ago

I agree. I’d say more importantly any cutoffs should have a rational basis for why the cutoff exists. Without a specific reason, it seems going back to 1919, the end of the dead ball era, is the safest and best way to go.

And the rational basis could be as simple as “this was hard to do, so why do extra work?” but it shouldn’t be just assumed that it is okay to pick some random year to pretend is the beginning of baseball time, especially as you say when you are proclaiming the “best ever.”

twm
Guest
twm
7 months 29 days ago

“basically the best contact team ever” reads as a smirk, to me. Or maybe a wink. It certainly betrays an awareness that “best ever” is always contested, especially when the field constituting “ever” is (somewhat) arbitrarily limited to exclude roughly half the available years of data.

TKDC
Guest
TKDC
7 months 29 days ago

I missed that, but I think you’re right.

Billy
Guest
Billy
7 months 29 days ago

Well, it was different in a number of ways pre-1950: only white players were allowed on teams, mounds as high as 15″ instead of today’s 10″ standard, and there were 16 teams instead of 30.

Ken Burns
Guest
Ken Burns
7 months 29 days ago

Hey Bro, do you even know me?

Jonrox
Guest
Jonrox
7 months 29 days ago

The Royals are also dead last in walk rate (while the Cubs are second best). Interestingly enough, the highest rate teams in both k-rate and bb-rate have rates approximately 50% higher than the lowest in that category (~24% vs. ~16%, and ~9% vs. ~6%)

Baron Samedi
Member
Baron Samedi
7 months 29 days ago

Wow. Contact. *Extended masturbatory hand gesture*

anonymous
Guest
anonymous
7 months 29 days ago

“Baron Samedi says:
September 23, 2015 at 5:44 pm
You write this as if everyone wasn’t already aware that the Royals are trash; always are, always will be.” on the other royals post

did Hosmer fuck your wife?

Ken
Guest
Ken
7 months 28 days ago

What a Hosmer wrecker!

Baron Samedi
Member
Baron Samedi
7 months 28 days ago

Aw. Royals fans so sensitive.

Damaso
Member
Damaso
7 months 29 days ago

interesting article.

i left wondering whether this was all due to the GM collecting great contact hitters, or if we were seeing significant increase in contact rates from imported players?

triple_r
Member
Member
7 months 29 days ago

The Royals play half their games in a ballpark that depresses strikeouts by 6%. Did you park-adjust these z-scores?

triple_r
Member
Member
7 months 29 days ago

Come to think of it, I may have some park-adjusted z-scores on my desktop. I’ll check for those once I get back.

Harold of the Rocks
Guest
Harold of the Rocks
7 months 29 days ago

How does a park depress strikeouts? Isn’t that just a product of the team that plays there?

Does this mean that Wrigey Field and Minute Maid Park do the opposite? Or it it just that half the ABs in said ballparks are taken by a team that strikes out a lot?

LeeTro
Guest
LeeTro
7 months 29 days ago

That’s not how park factors work. (Home Batting Ks + Home Pitching Ks)/(Road Batting Ks + Road Pitching Ks) is the equation. The offense Ks 14.4% at home and 16.8% on the road, while the pitching staff Ks 18.5% of batters at home and 19% on the road. Using just this year, there looks to be a decent impact on Ks. This can come from better batter’s backdrop, Colorado has the thin air that impacts break, and even foul territory could impact that, since there are more foulouts instead of balls in the stands. I’m not sure what KC has that is so helpful, but it seems to be there.

Twm
Guest
Twm
7 months 29 days ago

LeeTro: not certain about all of it, but KC has a lot of foul territory, and as you mention, I know that plays a part in suppressing strikeouts.

Captain Tenneal
Member
Captain Tenneal
7 months 28 days ago

The foul ground would actually help explain why such a high percentage of their contacted balls are fair (53% compared to 50% league average). It’s actually harder to hit a foul ball in KC because foul outs are more common, and those simply look like a ball in play in the box score

Rational Fan
Guest
Rational Fan
7 months 29 days ago

That ’94 White Sox team was remarkable. Stupid strike f’ed everything up.

Their power bats nearly all walked more than they K’ed.

Thomas had 109 walks and 61 k’s in 113 games.
Franco 62 BB’s 75k’s
Ventura 61 BB’s 69k’s
Raines and Cora had more BB’s than K’s

For a team who had a team OPS of 810 – unbelievable – their K rates are even that more impressive.

Bob Hamelin
Guest
Bob Hamelin
7 months 29 days ago

Tell me about it…my team was kicking A that year…

Snowman
Guest
Snowman
7 months 29 days ago

Meanwhile the Braves have the second best contact rate this year, and the last time I looked at the team leaderboards they had the fewest runs scored in baseball.

Eric
Guest
Eric
7 months 29 days ago

this is exactly why I created the stats of HEWCO and CCR. Because the rate stats of batting average, on base percentage, slugging and OPS all consider contact outs as a negative outcome. Contact outs LOWER all those aforementioned rate stats, and are not shown in the numerator of those stats as a positive. Contact outs can move runners up on the basepaths and get your team an RBI, so I don’t quite get how it could ever be a negative proposition to make contact.

LeeTro
Member
Member
LeeTro
7 months 29 days ago

Contact outs can also result in double plays and non-force out fielder’s choices, which are worse than strikeouts. B-R’s calculations have that figured in, and it has the non-K outs as worse. K’s are about 0.01 runs better than the average out, while non-K’s are about -0.005 runs worse. Either way, it’s no more than one run a year.

LeeTro
Guest
LeeTro
7 months 28 days ago

Running the numbers for this year, using RE24, K’s have actually been about 0.0012 runs worse than non-K’s, -0.259 to -0.2578. Either way, it basically has no effect.

wpDiscuz