Something Has Paused the Home-Run Spike
Is it even necessary to go over the background? I will quickly go over the background. Back in 2014, there was mounting concern that baseball no longer featured enough offense. Pitching had taken over, and we saw run-scoring plummet. Strikeouts, as you know, have only continued to go up over time, but that’s been countered by a sudden spike in home runs. There were 723 more home runs in 2015 than there were in 2014. There were 701 more home runs in 2016 than there were in 2015. There were 495 more home runs in 2017 than there were in 2016. Compared to 2014, the number of home runs hit in 2017 was up by a staggering 46%. That’s what baseball became — a showcase for the three true outcomes. Strikeouts and homers were in.
The surge in power made some people happy. It made other people sad. The remaining people were indifferent observers of a changing game. No matter your opinion, we were all left to wonder: now what? Would the power surge continue? Would the ball continue to fly? Would more and more batters continue to swing up, enthusiastically trading contact for dingers? It’s easy to observe a trend that’s already in the books. It’s more difficult to know where it’s going. Most of the time, anyway.
And in this case, I’ve given it away with the headline. What’s become of the home-run spike in 2018? There’s been, to be sure, no shortage of dingers. But the home-run count isn’t going up. It’s actually taken a step backward. Whether it’s signal or whether it’s noise, the home-run spike isn’t spiking.
Here’s the easiest way of looking at things. It doesn’t make sense to compare partial-season data against full-season data, so in the following table, you’ll see league-wide numbers for the last three seasons, through the early part of June:
| Through This Date | HR/Batted% | HR/FB% | GB% | BB% | K% | wOBA |
|---|---|---|---|---|---|---|
| 2016 | 4.2% | 12.5% | 45.2% | 8.3% | 21.1% | 0.315 |
| 2017 | 4.7% | 13.4% | 44.5% | 8.7% | 21.5% | 0.317 |
| 2018 | 4.4% | 12.4% | 43.3% | 8.6% | 22.4% | 0.314 |
Strikeouts have gone up almost a percentage point. That’s unsurprising — the strikeout trend continues unabated. Walks have hardly budged. Offense overall is slightly down, as reflected by wOBA. We have seen a slight shift toward hitting more fly balls. And yet, consider those fly balls. The rate of homers per fly ball has come back down a full percentage point. The rate of homers per batted ball has also, of course, come back down somewhat. It’s not like the league is back to its 2014 level. But it’s also not particularly close to its recent 2017 level, standing closer to 2016 instead. Perhaps, in time, when it comes to the home-run surge, we’ll see 2017 as the apex. It’s too early to arrive at such a sweeping conclusion, but it’s something to pay attention to.
For the sake of showing more information, I’ve broken a few of these numbers down. Using our splits-leaderboard feature, it’s possible to look at league-wide numbers on a week-to-week basis. Here’s how the last three seasons have begun, in terms of league grounder rate:
You can see things shifting downward. Whenever you reduce sample sizes, you increase the relative influence of noise, so on the week-to-week level, numbers bounce around. Still, we can say that more balls are being hit in the air. This is a trend with a variety of causes. And it’s hard to blame hitters, considering how fly balls have been rewarded.
Now here’s how the three years have begun, in terms of homers per fly ball:
And, very similar to that, here’s the plot of homers per batted ball:
Earlier this season, there were a lot of cancellations, and there was a lot of cold, inclement weather. That seemed to me like a reasonable explanation for why the early numbers were where they were, and as you can see, by week five or so, the home runs reappeared. For a few weeks, you can see the 2017 and 2018 lines mirror one another. But then there’s another observable separation. Over the past month, home runs have slipped well behind the 2017 pace. They’re even a little behind the 2016 pace. The home run is very much still alive in the game today, but something has reeled the trend back in.
As it happens, I’m not the first person to get to this. You probably heard about MLB’s scientific committee that investigated the properties of the baseball itself. That committee included one Jim Albert, and Albert has already written on the early numbers in the 2018 season. Here’s an analysis post from May 28, and here’s another from June 4. Albert makes use of some of the available Statcast information, and of particular note is that launch angles and exit velocities are up. If anything, we’d expect there to be *more* home runs, since hitters seem to be hitting the ball a little harder and a little higher. Clearly, that’s not what’s happening. And in the second post linked, Albert folds in the potential influence of lower temperatures. Temperature explains a part of this, but not nearly the whole.
Here’s how Albert sums up his first post:
There is clearly a decrease in home run hitting in 2018 and balls hit with specific launch angles and launch speeds are less likely to be home runs. As I noticed, 2018 hitters are actually hitting with higher launch angles and higher launch speeds, but the balls appear to have more drag. Of course, the big question is why this is happening, and this motivates further exploration.
And here’s a takeaway from his second:
By using a new model using three variables (including temperature), I find that the 2018 home run still is about 300 behind what one would expect based on the 2017 history. So temperature explains some, but not all of the drop in the 2018 home runs.
Home runs are down. That much we know. They’re not down by a dramatic amount, and it’s possible, as a fan, you hadn’t even noticed. It’s not like hitting for power is in any way endangered. Still, consider what the scientific committee found. In their collective judgment, home runs took off because the baseballs themselves experienced lesser drag. They didn’t have a definite answer for why that would be. Now, the early numbers suggest something has changed again. It seems as if the balls must be encountering more drag, relative to 2017, which only makes things all the more confusing. Plenty more analysis will have to be done, and I doubt I’ll be the one doing it. For now, know the baseball itself just continues to fascinate, with minor changes leading to broad repercussions.
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.
Home runs may be down slightly, but strikeouts are still way up, on a continuing trajectory to who knows where. That, and not pace of play, is the game’s most pressing and serious problem.
For years I’ve been arguing in favor of a 10/10 rule, where teams can have only 10 players on the roster eligible to pitch — with an exception made if the score gets lopsided (10 runs up or 10 runs down), in which case they can send someone else to the mound.
This simple rule change would force all pitchers, starters and relievers, to face more batters — which means conserving their energy, managing their arm strength, potentially leading to more balls in play and more action for the fans.
Do you like this idea? If not, let me know what’s wrong with it.
What happens after game four of a five-game stretch when your starter only lasts three innings and all five of your relievers went over the last two days?
Your rule is begging for guys to get hurt.
A) If all five of your relievers have been used in the past two days, you might want to bring up a fresh pitcher from the minors before the next game.
B) Injury to pitchers can result from throwing too hard as much as from throwing too often. The point here is that pitchers will have to learn to pace themselves again instead of going all out all the time.
I agree with what qacm said about using minor leaguers, and also with the general idea that a strengthened focus on pacing and managing arms would make these difficult situations less frequent. Also, such a system would bring value to the “innings eater” type, which, in my opinion, is a skill worth rewarding. Is 10/10 the perfect number? Idk. But I think a rule in the same spirit is worthy of discussion.
Sigh.
Look on the bright side: at least you didn’t have to face the Ohtanis.
It wouldn’t do anything. When you account for the starting pitchers on the roster who no-way-no-how-never are going to pitch in a day’s game anyway you’re left with today’s starter, tomorrow’s starter, and 7-8 relievers. what does that even accomplish?
Unless you’re talking about a 25-man roster restriction. i.e. 15 batters and 10 pitchers by rule. And in that case, the MLBPA would never agree to it, because that’s going to put an awful lot of voting members out of work.
We have no data to prove this would reduce strikeouts. We have no data to know how this would impact injuries. There has never been the need from a rule book perspective to define a roster position. There might not be any need to change the rules. There might be other ways to address the strikeout issue. I don’t like the “we have to do something why not this” approach either because the law of unintended consequences always raises its ugly head.
We voted this guy down into oblivion for posing an idea? Even when he asked for feedback? C’mon guys…
Do I think it’s the best idea? Certainly not, and I agree with Gavin’s injury concerns. But qacm brought up a reasonable point and was looking to discuss with others about it. There’s nothing wrong with that, and it deserves more than a simple thumbs down on his comment.
The downvotes are the feedback.
can’t it be mostly the weather as april was very cold? april HR/FB was 11.9% vs 12.8 last year. May it was 13% in 2018 vs 12.8 last year. So we might be almost at the same pace again but are still chasing the lost ground of april.
I would be surprised if the rest of the year HR pace was significantly lower than last year.
maybe there is a small change in the ball but I think overall the pace will be about the same.
The 12.8% is from May of 2016. In 2017, HR/FB in May was 13.6%. So I think even if you just look at May, the data corroborates the table that Jeff provided, showing that 2018 is trending closer to 2016 levels
Temperature is addressed in the article.
IFFB% is 10.5% this year after being 9.6, 9.5, 9.7, 9.6 over the last four years. Might explain some of the higher launch angle and exit velocity without homers.
more high fastballs maybe? high pitches do induce more pop ups than low pitches, but I haven’t seen stats whether high pitches are up.
I agree someone should look at IFFB in more detail. The difference holds for the “first half” of the past few seasons. I’m curious about how this relates to launch angle data too. Average launch angle indicates hitters have been closer to the “sweet spot” in 2018, but that data point doesn’t consider distribution. Maybe the result is due to excessive launch angle events?
Yeah, it seems to me that rather than just looking at the overall launch angle, we ought to look at the percentage of balls that are hit within a particular range of angles that are likely to produce home runs. Two balls hit 100mph at 30° have the same average velocity and launch angle as one ball hit 30° down into the ground and another ball hit 90° straight up, but the first two will probably produce 2 home runs and the second two will probably produce 0 home runs.
More light hitters realizing launch angle matters and trying to hit it higher?
Xoomwaffle,
Interesting point, do you know the the exit velocities of IFFB’s?
Blame the humidor?
Seems a simple enough thing to check. HR/FB in ARI this year to last, and HR/FB everywhere else this year to last?
Raw numbers was something like -0.8 homers per game. from an article done a couple weeks back. That would make it about 28 homers for 32 home games. That would be about 10% of the 300 homers we’re missing if the 300 homers didn’t already take that into account.
By my quick math, there have been about 1.5 fewer HR per 100 PA. There have been 0.85 more IFFB per 100 PA. The change in HR due to the humidor is somewhat less than 28 per 32 (although it probably has the greater impact)
There has been bout 2000 PAs (1033 home PAs for Ari) in Ari this season for both teams. 2000*1.5/100 = 30 homers less.
Baseball needs the home run right now. 2017 saw the most runs scored since 2009, when the K% starting increasing in earnest. 2009 was also the last year where the league hit for an average above .260… after hitting above that each of the previous nine years.
I read but I didn’t understand why it needs the HR?
I think he means that HRs are about all we have now if you want to score runs.
This is fine by me. Homers are fun, but mostly because they’re rare. Move the outfield fences back while you’re at it. Get more balls in play. More triples=better baseball.
More balls in play = more defensive highlights, baserunning highlights, more action in general. I’m not against huge dongs, bat flips, or pitchers getting ks, but I like a balanced game, too.
“I’m not against huge dongs”
Lana, can we please bring phrasing back?
Homers are fun because they’re homers, not just because they’re relatively rare.
Yea, exactly.
Actually let’s change the ball to a tennis ball and the bat to an aluminum bat so I can get out there and hit some 500 footers. Everyone wants to see that right?
Homers are fun because they’re supposed to be impressive. With a juiced ball they are less fucking impressive. Period.
How many years till they lower the mound? Obviously something has to be done to lessen strikeouts. I love watching all the oldtimer announcers talking about just putting the ball in play (instead of striking out) like it is easy to do.
seems to be completely backwards to be complaining about homeruns when it’s the single most exciting thing in your sport.
old farts who are all about the smallball are irrelevant and declining. making your game conform to their aesthetics is bad aesthetics as well as bad business.
My favourite thing about baseball is fielding. I love defensive highlights way more than homeruns, and my point of view of aesthetics in baseball is as valid as yours, so no, it’s not backwards to complain about homeruns if what one values about baseball is how a shortstop fields an impossible grounder, how a first baseman scoops a horrendous throw, and other things related to that. I understand homeruns are very valuable and represents wins, but to me they aren’t particularly exciting to watch in most cases, so I’m happy they are a bit less frequent than last year.
Has there been any investigation/discussion in regards to the seams on the balls? Higher seams would cause more drag on batted balls and we have seen some pitches this year that have had eye popping movement, again a potential effect of higher seams.
Let’s go back to using hand-wound balls, pitching underhand, no gloves, and use the same ball the whole game. Now that’s base ball! And no spiritous or vinous liquors may be sold on the Grounds!
Could also be implementations in stadiums as well that reflect a drop. There’s so many variables to this that I personally stopped caring about it. Pitches could be lower in zone, stadium dimension changes, weather patterns, rain outs, different hitters in favorable parks, etc etc. In 2016 there were 14 stadiums with a +1 Park Factor), 2017 had 15, 2018 only has 13. The biggest thing I took from this is a changing of a roster for a team that affects the power hitters of a lineup.
for example : Oakland in 2016 was 28th with .829 PF (big move that year, Yonder Alonso and Billy Butler, one finding success the other DFA’d) : 2017 they jumped to 6th with 1.103 (Alonso allstar, Ryon Healy, Matt Joyce, Chapman): 2018 They are dead last with .712 (No Alonso, Healy, Joyce) with the only true masher left on that team being the real Khrush Davis.
The Orioles are also another, where they fluctuated with Chris Davis, Trumbo, Cruz (the one year he was there), Machado. I’d bank money on roster makeup and outlier years (Chris Taylor comes to mind) and being in a favorable position such as parks, or a spot in the lineup, or in general just bad scouting more-so than just drag on a baseball.
Weather entails a lot more than temperature (humidity, wind, pressure, player subs due to rainouts). Also, simple things like park schedule imbalances could be at work.
Its not reasonable to expect the ball to change every year. They probably backed off a bit on the balls juice or gave it more drag, or maybe its just weather and the humidor in Chase. HR are down more in the NL than AL. It could also be that K’s are getting to the point its impacting power.
Whatever it is, this is still a historic HR era in MLB that is tainting stats at levels not seen since the steroid era (which was also another juiced ball era but without the K’s)
I wonder how much of this is a consequence of the flyball era. People throw around advice like “change your launch angle” as if it’s something easy to do. I’m sure LOTS of players are trying to make those adjustments but it turns out adjusting a swing is really, really hard. Some players are adversely impacted and see their offense decline (while racking up K’s).
I suspect the numbers will bounce around for several more years as players experiment with their swing. Some will find success and see a big improvement. Others will fail and end up reverting back to their previous approach. Eventually, we will largely settle into stability as people have either tried it and stick with it or have tried it and given up.
If there’s more drag in 2018 wouldn’t more 2018 line drives go for homers vs. 2017 line drives? Wouldn’t the inverse be true for 2018/2017 fly balls?
In another universe… increased velocity by large % of everybody’s staff is bringing more swing and miss, not so much called strikes which appears constant. Weight of bats must be reduced to help overcome velocity concerns. How? Shudder… composite bats on horizon?
Been seeing a lot more pitches up in the zone. Especially the high heat. Been fun to watch as a fan of pitching.
Spin rate??
I would like to see this factor explicitly addressed. I may be flat out wrong, but my impression is that back-spin has a non-trivial effect on fly ball distance. It doesn’t strike me as obvious what, if any, relationship we might expect between increased launch angles and spin. It seems highly likely that several physical factors and pitcher-batter process interactions affect both, and probably not independently.
Is there good work out there looking at things like how much “control” pitchers can exert over batted ball spin, specifically back spin? What’s the relationship between pitch location, pitch trajectory, and pitch spin with batted ball spin? If there are predictable influences of pitch selection and execution on back spin of batted balls, I would fully expect front offices to attempt to leverage that with suggestions for their pitching staff.
There are several things worth mentioning here IMO.
1) MLB’s scientific community should credit Nate Silver for doing all of the work.
2) As mentioned, this could be noise. On that note, you are comparing the cold third of the season to an entire season – that might work itself out.
3) Most importantly, none of this means that anything has changed. All it means is that hitters have been less successful. When the ball changed, hitters made some changes to capitalize on the new ball and it worked. I am not sure that anything wouldn’t have worked – hitting is just easier now. Let’s pretend that increased launch angle is objectively a good thing – at some point you are overdoing it, that’s for sure! I also always argue that trying to hit high drives is not sustainable – its the same thing everyone has always been trying to do – the difference is that when you miss they now go out of the park ore often. The reason kids and players at every other level don’t go fly-ball heavy is that it is a bad idea – it works great when it works, but when it doesn’t you look like Cody Bellinger. A mis-swing on an attempted flyball has very little chance of working out well. Its not sustainable which is a problem… I wouldn’t be surprised if league rates go up or are consistent with last year as more players figure out how to cope with the juiced ball era – but flyballs aren’t the inevitable course of the game – they were more of a reaction to a new environment.
Lastly, I have some real questions about EV. Pitch velocity varies quite a bit and it seems like they change it every few years. We don’t seem to have those conversations about EV. it seems like almost everyone is generating more EV this year, which makes me question its accuracy. Remember last year when they changed the way pitch velo was calculated? Perhaps it is all part of a all-or-nothing approach, which will surely lead to decreased outcomes sooner than later….or the data isn’t accurate, which I have no doubt we will say in the future.
How can you have this discussion without touching on how the rising flyball rate affects the LINE DRIVE rate?
Using B-R’s trajectory classifications:
— This will be the 3rd straight year with both a rise in flyballs and a drop in line drives.
— Line drives are the best class of batted balls, by far. This year’s OPS splits are 1.557 for line drives, .835 for flyballs, .509 for grounders. A flat 80% of flyballs this year are outs, compared to 37% of line drives.
The pursuit of flyballs is also linked with a rise in strikeouts, both anecdotally and logically.
So there was bound to be a tipping point past which the extra flyballs do more harm than good, even if all conditions remained fixed.
Comparing 2018 projections against last year’s totals, with each class of result as a percentage of all PAs, we find net changes of:
-0.6% in line drives
+0.7% in flyballs
-0.8% in grounders
+0.8% in strikeouts
That boils down to trading some liners for flies, and some grounders for whiffs. Other things being equal, that’s an obvious net loss of offense.
The success of the flyball approach in the past 3 years — the steady rise in scoring despite a drop in line drives — lived on a steady increase in the value of each flyball. Such a trend should be expected to peter out.