The League’s Continuing March Towards Three Outcome Baseball
Writing about baseball in April can be difficult. Things are happening, and the natural inclination is to want to talk about those things, but for most players, we’re talking about 20 or 30 plate appearances. A bunch of starters have pitched one game. Besides changes in velocity, there’s not much we can say about what has happened so far. The Diamondbacks and Twins are the best two teams in baseball right now, so yeah, it’s early.
But while samples are still tiny for players and teams, things tend to stabilize pretty quickly at the league level. And, not surprisingly, the first week of the season was filled with the two things MLB games are becoming known for; strikeouts and home runs.
Through the first week of the 2017 season, the league strikeout rate is 21.9%, almost a full point higher than the record-high 21.1% mark that the league established last year, breaking the previous year’s record of 20.4%, which broke the record previously set in 2013, which broke 2012’s record, and so on and so forth. Setting the all-time league wide strikeout record is just something baseball does every year now.
The increase in strikeout rates drove offense through the floor. But then, in the second half of 2015, home run rate took off, and the long-ball trend hasn’t slowed down since. Despite playing in mostly cold weather in the early part of the year, the league wide home run rate so far this year is almost a match for the 2016 home run rate. Generally, home run rates increase as the season goes on, with warmer weather helping the ball carry further, and lower-quality pitchers getting called up to replace guys who got hurt earlier in the year. The fact that we’re starting 2017 near the 2016 seasonal home run rate suggests that this trend isn’t going away.
And as Ben Lindbergh wrote before the season started, this is exactly what we should have expected.
Spring strikeout percentage has an extremely strong relationship with regular-season strikeout percentage, boasting a .98 correlation from 2006–16, the years for which spring stats are available via MLB.com. (A correlation of 1 would indicate a perfect positive relationship in which the two stats move in tandem.) And despite some tenuous signs that the called strike zone’s dimensions could be contracting this spring after its years-long expansion stalled last season, the strikeout rate still seems to be rising. This year’s 20.2 percent?—?an increase of 0.8 percentage points relative to last March?—?is the highest spring mark in our 12-season sample.
Over the last 11 campaigns, the regular-season strikeout rate has been 1.4 percentage points higher than the spring rate on average, maybe because the pitchers catch up to the hitters as the season wears on. If we add that typical margin to this spring’s rate, we get an expected regular-season figure of 21.7 percent, which would yield about 1,000 more strikeouts than we saw last season. That result could only strengthen commissioner Rob Manfred’s desire to formally raise the strike zone’s lower bound from beneath the kneecap to the top of the kneecap in 2018. (As that use of anatomy indicates, the strike zone is super precise.)
…
Over the last decade, the relationship between spring and regular-season home run rate on contact has been weaker than that of spring and regular-season strikeout rate, but still strong overall (.73 correlation). Last spring’s rate was the highest on record, and sure enough, it augured a record-setting regular season. As Opening Day approaches, this spring’s rate has reached another new high, which suggests that whatever’s happened with homers since the second half of 2015 hasn’t run its course yet.
Strikeouts have been trending up for a long time; home runs trending up more recently, but in a very noticeable way. Both of those trends continued in spring training, and they continued in the first week of the regular season.
There has been one area where, through week one, things don’t look like the last few years, though, and that’s walk rate. So far this year, the league average BB% is 9.3%, up over the 8.2% BB% MLB put up last year, and way up over the 7.7% mark we saw in 2015. Of course, we’re dealing with 7,000 plate appearances in 2017 versus over 180,000 in each of the last few years, but through the handy splits leaderboard, we can actually compare weekly walk rates over the last few seasons.
Going back to 2014, so we have three full years of data plus the first week of 2017, no week with a full slate of games has seen a higher BB% than the first week of the 2017 season.
| Season | Week | BB% |
|---|---|---|
| 2017 | Apr 3 – Apr 9 | 9.3% |
| 2016 | Aug 29 – Sep 4 | 9.1% |
| 2016 | Apr 25 – May 1 | 8.8% |
| 2016 | May 9 – May 15 | 8.6% |
| 2016 | Apr 18 – Apr 24 | 8.6% |
| 2015 | Sep 14 – Sep 20 | 8.6% |
| 2016 | Apr 4 – Apr 10 | 8.6% |
| 2014 | Apr 14 – Apr 20 | 8.6% |
| 2015 | Sep 7 – Sep 13 | 8.5% |
| 2014 | Apr 28 – May 4 | 8.4% |
Weeks with a league-wide walk rate over 9% haven’t just randomly happened. The last time we saw a weekly walk rate over 9% before last August was May 9 to May 15 of 2011, when the league wide BB% hit 9.2%. In other words, we’d gone over five years without a weekly walk rate over 9% before last August, but now it’s happened twice in the last six weeks of the MLB regular season.
As Jon Rogele noted in his piece on the 2016 strike zone, the strike zone appeared to slightly contract last year after several years of significant growth, and Jeff pointed out that in spring training this year, we saw fewer low called strikes than in prior years. And with an unusual spike in the first-week walk rate, it may be possible that the strike zone is going to shrink again this year, even without the MLBPA’s agreement to change the rulebook definition of where the bottom of the zone begins.
But because the caliber of stuff of Major League pitchers just continues to get better, increasing the walk rate may just offset the rise in strikeouts to some degree. Because right now, it looks like MLB hitters just can’t hit strikes from MLB pitchers with anything close to the same regularity as they did 10 years ago.
In 2008, the first full year of PITCHF/x data, the league rate on contact in the strike zone (Z-Contact%) was 88%. It basically stayed there for four years, until it dropped to 87.3% in 2012, and then it hung out close to that mark for the next four years.
Last year, in-zone contact fell to 86.5%, and after a one-year pause, the league strikeout rate shot up again. The league’s in-zone contact rate through the first week of this year? 85.3%. Because the best pitchers are the only ones who have made two starts thus far, these numbers aren’t an apples-to-apples comparison to full year numbers from prior years, and this will likely go up once the number of innings is more regularly distributed between pitchers, but the early data suggests we’re in for another year well below the established Z-Contact% norms we got used to over the last decade.
Which makes sense, given that every team now has six or seven guys throwing 95 with knockout breaking balls in their bullpen, and that it’s not even that weird that we have starting pitchers sitting 97 or 98 for extended outings these days. It’s just harder to hit pitches that go this fast and move this much, and even if the league is moving to take away the expanded strike zone that pitchers have enjoyed in recent years, offense will only increase if hitters can hit strikes.
In week one, they didn’t, so league wOBA was still just .316 even as the ball continued to fly out of the ballpark.
It’s just one week, and things can obviously change. Perhaps the league walk rate will sink right back to to closer to 8% this week, and this will all look like nothing but small sample size theatre. But given what we already knew about how the game was changing heading into the year, the strikeout and home run trends aren’t surprising, and the league has made it clear that it would like a smaller strike zone, so perhaps the jump in walk rate shouldn’t be surprising either.
But if you were looking for reason to think the move towards three true outcomes might slow down, well, the first week of 2017 offers basically no hope for you. Unless the first week spike in walk rate was a fluke, we may very well be headed for a record low number of balls in play this year.
Dave is the Managing Editor of FanGraphs.
Let’s move the mound back 5 feet.
Better yet. Let’s move it 5 feet to the left!
Maybe we can have three mounds, kind of like the boxes in golf. Start from the far mound, pitcher gets an 0-1 count. From the near it’s 1-0. Options!
And if the batter kicks the pitch over the fence instead of using his bat, it’s 10 bonus runs!
Or just get rid of fences, and put water hazards in play…..
http://www.fangraphs.com/community/53-things-about-a-53-second-finnish-baseball-video/
Eliminate the mound entirely.
How much of the stress on a pitcher’s body comes from the fact that he’s stepping down as he delivers? Put in a circle similar to that in softball, and let the pitcher make any delivery he likes as long as he ends in the circle (which gets rid of weird double hop deliveries).
Your ideas ae intriguing to me and wish to subscribe to your newsletter.
Doug, that’s a great idea. Somebody needs to tell it to that chucklehead currently masquerading as the MLB Commissioner. (The post has been effectively vacant since Bart Giamatti passed on.)
Hard to say offhand the benefit of flat ground over the mound, but the comparison to softball is a tiny bit glib–the underhand motion is just much more natural which why top NCAAW college softball pitchers may still be asked to pitch both ends of a ‘header, and throw far more pitches than their baseball counterparts, and with less risk of injury.
At some point MLB needs to do something. Baseball is getting more and more boring. Combine the lack of balls-in-play, with the ridiculously sluggish pace, and it’s just not a good entertainment product any more.
Baseball needs more skipping. Walks are lazy. Home runs have slowed down to jogs and that’s more like exercise. And ‘strike outs’ sound really negative. The manager is the only person who is still called ‘skipper’ – but when is the last time a manager actually skipped? Where is the joy in Mudville?
Of course – that does not mean baseball should allow gamboling
I’d say the old-school baseball fans / writers / historians probably collectively shuddered reading this post. But then again, the majority of them probably have never made their way over to FanGraphs to begin with.
I’m old school. And so are some of my best friends. And while we’re having trouble understanding all of the changes in the game, we understand one thing: everything evolves.
Typical condescending drivel.
It seems like theres another cycle.
Teams emphasized pitching down and getting balls on the ground.
Hitters changed their swing paths, to hit those low balls in the air.
Pitching to contact turned into home runs, with better swing paths.
Then they started pitching out of the zone, which upped strike outs.
And now it’s also upping walks, because hitters are adjusting.
So now we see the high fastball come back, and have an increase in pop outs.
Is this too simple? Will 2017 be the Year of the 4 outcomes? HR, K, BB, and Pop out?
I think you may be right. I’ve seen so many hitters so far in 2017 try in vain to get under a high fastball. They sure look silly trying that.
there you go!
Three things that could solve this:
1) Lower the seams on the ball. Lower seams = less break = easier to true up and hit the ball = more balls in play = more excitement.
2) Deaden the ball by some significant percentage, whether 5%, 8%, 10%, whatever. Deadened ball = less flight = fewer home runs = more balls in play = more excitement.
3) “Robot umpires” (i.e., home umpire wears earpiece indicating strike zone breach by the pitch). Less subjectivity = truer balls and strikes called = pitchers have to come in to get *actual* strikes = batters put more balls into play = more excitement.
A groan from the home crowd as Nelson Cruz hits another dinger. Yeah, no.
“Deadened ball = less flight = fewer home runs = more balls in play = more excitement.”
So you think having balls that would have been home runs, which now remain in play (most often leading to fly outs) will be more exciting? derp.
So you think watching a player from the visiting team jog around the bases after hitting a homer against your team would be more exciting than watching them have to run hard to try to make doubles or triples while your team tries to throw them out?
derp, indeed.
You really think fans want to see fewer home runs? Really? REALLY???
I really think fans want to see more action! Really! REALLY!!!
I am a fan. I want to see fewer home runs. (And fewer strikeouts).
I may not be a majority, but I don’t think I’m alone.
“derp” please in english.
and many of those deadened balls will fall for hits.
I advocate for a rule change. Ground Rule Triples to replace GRDoubles… doubles are passe and triples are 90 ft closer to home!
“So you think having balls that would have been home runs, which now remain in play (most often leading to fly outs) will be more exciting? derp.”
I suspect players who kept hitting medium fly balls would rediscover the virtues of the line drive.
I got a better idea. Let’s switch to tennis balls for a week. I do not understand how physics works so I have no idea what would happen, but it would be a lot of fun to see what happens when everyone is thrown out of their comfort zone. In fact, we could change the ball to have different properties randomly every week, just to see pitchers and hitters learn what to do with it. I think this would be fun, although I am less confident anyone else will agree.
If the fraction of balls in play really does change significantly, it decreases the opportunities for defense to contribute value. i wonder if we will see fewer defensive stars.
How many times have you said, or heard someone say, some variation of “That was a great game – a bit of everything happened”? It is quite rare to hear someone say, “What a great game, only a couple things happened.”
Baseball, like any sport, is more interesting when a greater variety of things happen. In a way I am reminded of tennis, which has been dominated by high percentage players like Novak Djokovic and Rafael Nadal, who play endlessly long matches in a war of attrition. The style of play during the 90s and before was both more attacking, but also more diverse (we could see Roger Federer as the great transitional player, who bridged “old” and “new” school styles). Part of what happened is that the courts became homogenized, with less variance as in the 90s. That said, we’re (hopefully) starting to see this break up a bit.
So here’s the glass half full view: the game becoming stagnant will force evolution. Some player, or players, will come along that do other things even better than before. Imagine a line-driving hitting version of Tim Raines, who hits .330 with 50 doubles, 15 triples, only 10 HR, and steals 70 bases (Trea Turner?). That sort of player could open up new possibilities. The problem is, it seems players are told to bulk up and loft the ball and hit more HR, so those 50/15/10 extra base hits turn into 30/5/25, which is unfortunate, imo (there’s nothing more exciting than a triple, in my opinion… in baseball, that is).
We sort of saw this with Trout, who went from a .320 hitting, 25-30 HR, 30-50sb guy, to a .290 hitting, 35-40 HR, 15 SB guy. But in a rather unusual move, last year he went back to .315, 30 HR, 30 SB….which is far more interesting a player than the 2014 version. Hopefully he stays in that range.
Hopefully baseball not only evolves, but moves in cycles. That is, it isn’t simply a linear path, but a spiral-like form: returning to old ways, but at a higher “octave.” I’d like to see a bit of the 70s-80s return. That was the most balanced era that I’m aware of, with a bit of everything happening. But it might take a few years to swing back, if at all.
now here is an intelligent thought! thank you.
help me campaign for Ground Rule Triples.
Along these lines, the more that teams selectively steer their rosters more toward power over contact, the more rare the contact commodity becomes. And that’s basically how the cycle tilts back the other direction. In an era dominated by high K rates, guys like Murphy and Pedroia become more prized bc teams will recognize that having a mix of skills is preferable to a team full of swing for the fence types bc pitchers can adjust to a homogeneous lineup more easily than a mixed one.
This is a process though. The offensive adjustment (to power pitching) won’t happen overnite. But till this point, the thinking appears to be that swinging harder is the antidote to higher velocity. And swinging harder by and large results in 3 outcomes. Again, this will evolve with players becoming skilled at making contact with these higher velocities they are seeing. This is just one cycle of the ever-shifting cat and mouse between pitching and hitting.
Until batters catch up, having a tighter and more precise strike zone and lowering the mound are prob the two most sensible tweaks that can be done quickly.
I wish it worked that way, but I don’t see any sign it will. The game will evolve in the ways that help teams win; there’s no reason to assume those will also be the ways that makes the game more interesting.
Tennis is another sport where the players got bigger and stronger, and the equipment amplified that, but the dimensions of the court were unchanged. The natural evolution led to the game being overwhelmingly dominated by big hitters, serve-and-volley finesse players went extinct, and the sport (men’s especially) is less interesting to watch (and is much less popular) than a generation ago.
In football, on the other hand, the NFL/NCAA had a clear idea of the style of play they wanted to cultivate and they have repeatedly enacted rules changes to push the natural evolution the way they wanted.
Had been noticing anecdotally that BBs were up so far (including several situations where runs have been scored via walks), so thank you for writing about it. I was hoping it was that teams were starting to coach selectivity, but it seems like squeezing…
We all know ‘chicks dig the long ball’ (I kid) so don’t look for any changes that will hurt HRs… $$. Until managers start to impress upon pitchers to challenge hitters the walk rates won’t go down either. Part of the solution is in coaching hitters to use all of the field. Umps could also start calling the high strike.
Does anyone else find it ironic that we are finally coming up with some defensive metrics to a point where it seems MLB management is understanding and valuing defense perhaps more than ever before, and at the same time we’re seeing fewer balls in play to the extent that defense is perhaps the least important than ever before? And that was a long sentence.
I wonder if its even theoretically possible for a defender to be as valuable as an Ozzie Smith, Mark Belanger or Brooks Robinson were in their day. Fewer balls in play and a smaller gap between bad defenders and good defenders have made the impact of having a super-elite fielder like that a low lower (probably!).
Is there any particular reason why HRs particularly are increasing? Is this to imply that other hits (singles, doubles, and triples) are not also increasing, and in fact decreasing? I get some hitters like Adam Dunn are indeed built for the all-or-nothing style of hitting, but is there any information how these trends are particularly affecting slap hitters?
The anecdotal evidence is that more hitters are adopting an upper-cut swing to better combat sinking-fastballs down in the zone. The batters are choosing to become “all or nothing” hitters (since a high % of non-HR flyballs are outs) and that’s leading to more HRs.
Mitchel Lichtman, who seems to really understand what he’s talking about, is firmly in the “something about the ball changed” camp: https://twitter.com/mitchellichtman/status/752944657974386688
Changes in pull% and FB%, or overall talent changes in the game are significant but don’t come close to explaining the level of change, according to his research.
Obligatory error bar advocacy:
We don’t know from this that HR are still high — the previous 2009-2013 lower level is inside our 95% confidence interval. The data’s getting there though.
On the HR/game graph, our current error bar for 2017 is +/- 0.15 HR/G. So draw an error bar from 0.97 to 1.27 HR/G.
Calculation:
HR/PA is 2.9% = 208/7049 (may be off by a couple)
The 95% confidence +/- is 0.4%
(By normal approximation: 1.96 sqrt(p 1-p / n)
Batting stats from http://www.baseball-reference.com/leagues/MLB/bat.shtml
totally wasted opportunity for 3000 words about Ryan Schimpf
FWIW, BB% is always higher in April than the rest of the season, every single year since 1973, averaging 0.6% higher. And the Week 1 rate tends to be a hair higher than April overall, 9.1% vs. 9.0% (using teams’ first 7 games).
And while this week’s 9.3% is the highest of the past 3 years, 4 others in that top-10 table were in April, and another was partly April. And it’s just 3.8% of the season.
Looking a bit farther back, the Week 1 rate was 9.5% in 2010, but the year ended at 8.5%. That April’s rate was 9.4%, but it dropped to 8.3% the rest of the way.
There may be ways to infer something from this Week 1 walk rate, but on its face I see nothing.
Fun fact: This past week’s league BABIP was the second lowest weekly BABIP since 2014 among full baseball weeks (min 1000 IP) and the 7th lowest this decade.
Yeah, this is probably an underreported part of the “baseball is getting boring” story. Not only are walks and strikeouts going up, but quality contact is also going down (thanks to that same accursed velo and movement). BABIP will dip across the game.
Actually, BAbip has only gone UP in the past 5 seasons, climbing from .295 in 2011 to .300 last year.
Besides, if you’re talking about “quality contact,” it’s odd to exclude the optimal result, home runs. If you look at BA on contact, the rise is 10 points over the last 5 years.
I agree that this volume of strikeouts makes the game less fun to watch. But it’s easy to see what the batters are trying to do, and that they’re succeeding, on a macro level.
If they want to “solve” this, do what Coors Field did last season and raise or push back some fences. Having a small percentage of HRs turn into doubles and triples can make the game more exciting and eliminating some short distance fences (I’m thinking particularly of areas like down the LF pole in Tampa) will disincentivize the type of extreme pull hitting that many TTO guys employ.
Other than at Coors, it seems like fences recently have only moved in to try and give fans “more excitement.” Maybe teams should embraced the idea that a ball high off the OF wall can be more exciting than a jog around the bases after a 330 footer.
I think we have to wonder if broad homogeneity in Front Office personnel makeup and data-driven decisions is also affecting this trend, and if so, if it is a “cycle” or something here to stay. Home runs are so much more valuable than singles and steals – indeed, with knowledge of pitch counts and the times through the order penalty, walks may well be more valuable than singles – that it is objectively “correct” to prioritize them in players, to my opinion, the detriment of the game’s enjoyability.
Consider the NBA, where the ever-increasing number of 3-pointers isn’t a game-theory driven “cycle,” but a conscious, data-driven exploitation of the rules as written. Barring significant rules changes, I don’t think that’s just a fad that will reverse itself. I think we may be looking at the same thing here.
Interesting article. I think there are a lot of people who don’t want to see a change, as baseball is baseball is baseball (or whatever folksy term they use). But it changes all the time. For the simplest of stats, look at the BA of the 20th best guy from, say, 1926, and compare it to the 20th best guy from last season. That’s fewer active plays. Or how much relievers are used, and how fast they throw. If you like TTO baseball, then I guess you won’t mind, but I don’t, and I don’t think the trend in baseball is a good one.