As you’re probably aware, more home runs were hit during the 2017 season than in any prior year. A lot more, actually, as the 2017 total of 6105 round-trippers was a whopping 415 more than the previous record season in 2000.
More after the jump on the season of the long ball.
As shown below, the seventeen year gap between record home run seasons is the second longest of the modern era, so it should not be a surprise that a new record has been set.
What is more surprising is the suddenness of the change to the higher long ball totals, coming on the heels of the longest down trend of the modern era, as shown below.
Looking at year-over-year changes in home run totals, this is the third straight year of increases, something that’s happened only twice before in the post-war period (the other two times, in 1985-87 and 1998-2000, also culminated in new record seasons).
Looking at the 3-year trend, the more than 40% jump since 2014 in the largest of the post-war period.
Nearly every hitter now has at least a little pop, as show below for home runs by batting order position. The “heart” of the batting order in positions 3-6 now account for less than 60% of home runs, while the top and bottom each account for more than 20%.
In terms of numbers of players at different home runs levels, those hitting ten or more home runs has expanded from almost none in the dead ball era to nearly 8 per team today, while those compiling 200+ PA and hitting no home runs at all have virtually disappeared (last year marked the first time with fewer than 5 such major league players for three straight seasons).
Looking at the same data on a line chart shows the 1-9 home run group at its lowest level since World War II while the 10+ and 20+ groups are at record highs, and the 30+ group nearly so.
Looking at in game home run trends, here are a couple that caught my eye. The first line on the chart below shows percent of home runs hit with a batter platoon advantage, increasing steadily through the early 1980s, then declining until the mid-1990s, and mostly steady since then. The rise can be attributed to the growth in switch hitters and to greater use of pinch-hitters (the latter peaking from 1958-62 at 2.96% of home runs, and with no 5-year period since 1983-87 above 2.5%). The decline phase from the early 1980s through the mid-1990s coincides with modern reliever usage and managers increasingly using relief pitchers to counter batter platoon advantages. The steady state since the mid-1990s reflects further refinements in reliever usage with 7th, 8th and 9th inning specialists, regardless of platoon advantage or disadvantage; the most recent period has also seen larger bullpens and shorter benches, giving managers fewer pinch-hitting options.
The second trend in the above chart is the increase in home runs with 0 outs, reflecting both a greater number of players with home run clout and a change in batting approach from trying to get on base to trying to do some damage. Illustrative of this change in batting approach is the percent of PA with zero out and the bases empty (occurring at the beginning of each half inning and following home runs with zero outs), with the 5 highest marks since 1950 all occurring in the past 5 seasons.
Home runs by inning are shown in the chart below. Relief specialization by inning has indeed reduced the frequency of late inning home runs. Of course, a decrease in late inning home runs means an increase somewhere else, in this case in the early innings, a trend also indirectly related to greater dependence on relief pitching; the dotted line showing a marked and steady decline in short starts indicates that managers today are far more likely to stick with an ineffective starter in the early innings, mainly because of the need to get a least a few innings from that starter to avoid overtaxing the bullpen.
The charts below show color coded team home run totals by season. The pre-expansion period shows a general rise in home run totals through the period for all teams, with the Yankees in the vanguard of that trend in the AL, and the Giants and the Phillies in the NL. Some of those season home run totals in the early years are especially low, led by the White Sox who failed to reach triple digits over a full decade (99 total home runs from 1902-11).
The influence of ballparks on home run production has always been a significant factor and remains so today. Sudden changes in team results often coincide with moves to new ballparks.
Looking at home run rankings, again color coded by team and season, provides the results below. The longest runs leading the majors in home runs are the 7 seasons (1949-55) by the Dodgers and 6 years (1936-41) by the Yankees. Expansion era dominance is less pronounced, given more teams and greater player movement between them. Thus, the longest expansion era run leading the pack is only three years by the Braves (1965-67), Tigers (1990-92), Mariners (1997-99) and Rangers (2001-03).
Just a brief opening observation:
It took the big leagues through 1985, if my count is correct, to produce 17 teams who slugged 200+ HRs.
2017 produced 17 in one season.
Your count is correct. Nice observation.
Of those first 17 seasons of 200+ HR, only one team did so in consecutive seasons. Hats off to you if you knew it was the ’63-’64 Twins.
As with many of Doug’s elaborate statistical posts, this one is easy to admire but not easy to comment on: the data is too rich to absorb in detail over a short period and without a set of governing questions. Many of these posts don’t generate the long comment strings that the work should deserve, and I have a suggestion (chiefly to Doug and Andy) about what might be done.
This sort of post is general resource research that it would be helpful to be able to access with ease again and again in the context of later discussions, where consulting the data would help us comment more interestingly and accurately. To allow this, my idea is that HHS add a new tab for “Research,” initially populating it with appropriate past posts, such as this, perhaps retitled in a drop down menu (this one, for example, something like “MLB HR/yr”). This would not preclude discussion upon initial posting, as nsb has initiated here, but the most valuable discussion (and acknowledgment of work like Doug’s) might actually be the uses made of these data presentations at later times.
As things now stand, I’m afraid I’ve lost track of many of Doug’s best research posts that I had intended to study in greater depth than my initial survey, and the same will otherwise probably happen with this one.
Thanks for the suggestion, epm.
I will go back and categorize older posts, and try to create a search utility you can use to find them. Will let you know when it’s ready.
Another observation: the 6105 HRs of 2017 is an increase of 7.24% over the 5693 of 2000, but the 40104 Ks of 2017 is 27.7% more than the 31396 of 2000.
Some more observations:
The 2017 HR leaders included three playing their first full season, Gallo, Judge, Bellinger. Four of the other top eleven bested their previous high seasonal output in double digits: Stanton by 22, Smoak by 18, Moustaka by 16, and Morrison by 15. Martinez, playing only 119 games, hit 7 more than his previous high. Khris Davis bested his 2016 mark by one, 43 over 42, but his previous high was 27. Davis, by the way, batted .247 the third year in a row. Can that be some kind of record?
It ties a record shared by the following (100 PA min. in all years):
Ricky Gutierrrez, .261 from 1997-1999
Tim Foli, .252 from 1982-1984
Alex Kampouris, .249 from 1936-1938
Mike Stanley, .249 from 1990-1992
Mookie Wilson, .276 from 1983-1985
Judge and Bellinger set modern era rookie home run records in their respective leagues. Last time that happened in the same season was in 1925 with Lou Gehrig (20) and Mandy Brooks (14), though Brooks only tied the NL record set by Butch Henline in 1922.
Going back before the modern era, Buck Freeman set the NL rookie record with 25 home runs in 1899 that was not eclipsed until Wally Berger swatted 38 in 1930, that record matched by Frank Robinson (1956) and surpassed only by Bellinger.
Khris Davis is the first AL player with 40 HR, 100 RBI and less than 3 WAR in consecutive seasons. Ryan Howard and Adam Dunn have done it in the NL, with Dunn under 3 WAR for both seasons (2007-08) combined.
On something of a tangent, I have to wonder what the net effect of everyone swinging for the fences has on scoring.
There was a brief discussion on the last post about whether it would make sense to have your best hitters batting leadoff partly because a) it would get them more plate appearances over the course of a season but also b) it would mean having more runners on base at the top of the order. Of course, the obvious problem with that approach would be that you would have worse hitters coming up to drive them in.
It does seem like the Nellie Fox/Richie Ashburn types are a dying breed.
I remember Whitey Herzog writing about Willie Wilson and how Lee May (I think it was) taught him to swing harder, which led to much harder hit outs and a 50 point drop in his batting average.
I can help but wonder if someone like Billy Hamilton wouldn’t benefit enormously from either chopping down on the ball and beating out infield dribblers or using a much heavier bat that would force him to control his swing more and lead to little flares just over the infield.
Dee Gordon is one of the very few guys I can think of who uses an approach even remotely like this but his plate discipline is still pretty lousy and he still manages to strike out 90 times a season.
You would think that some manager would realize that getting on-base an extra 40 or more times a year- especially someone with some speed- might be more productive in the long run than maybe 5 more home runs and 100 more strikeouts.
Am I overlooking something here or does what seems to be going on now not make much sense?
You are overlooking nothing. Very little about modern society, baseball or non-baseball, makes sense.
Hmmm. Follow the money.
Ah, yes.
The old “Home run hitters drive Cadillacs and singles hitters drive Chevy’s” theory.
Of course nowadays EVERYONE drives HumVees and Lexus’s (or is it Lexi?).
And does anyone remember who it was that got Matty Alou to change his approach?
Was it Harry the Hat?
It was absolutely Harry the Hat Walker who turned Mateo into a real contributor in lieu of prior status as a fourth outfielder…..Stargell, Clemente, Clendenon, Bailey, Burgess and Pags – even Gene Alley; they all hit better than the league norm for their positions
It probably would be good to bear in mind that we are mostly retrograde oldtimers, longing for the days of lost youth (Voomo will have to be honorary in that category, since he is, from my standpoint, still stuck in the midst of his younger days). I’m with Hartvig and Voomo on this, but in a decade or two, I think we’ll look quaint, and with better reason than we’d like to acknowledge.
Baseball people are now entranced with what modern technology like statcast has revealed — they’re in a search for guys who can hit “barrels” and turn them into a .650 bip (“barrels in play”), and, for now, that seems to entail lots of Ks as well. But embracing technology and advanced stats isn’t something HighHeatStats partisans should look at askance, and while I think we’re right to value multi-tool players, fielding is very strong, the stolen base is still an important facet of game strategy (though not as it was forty years ago, which may be an appropriate recalibration), and hitters and coaches may just have to work through some years of learning how to optimize barrels without maximizing Ks as well.
The most exciting plays in baseball, the triple and inside-the-parker, were largely lost before any of us were even born. Baseball continued to have variety and interest enough to capture us all, but imagine how the old-timers kvetching online back in the Jazz Age burnt up the fiber-optic cables with their fury (which, I suppose, is why none of their posts has been preserved).
These are good points, epm, and I’d stand with you. I’d like to make one counterpoint, and a couple of other observations:
1. Counterpoint: Maybe you were only talking about the posters in this particular thread, but if you were talking about HHSers overall, some of us are not “retrograde oldtimers.” I’m only 31 (and have been commenting here since it was on the baseball-reference blog, I would guess in about 2007 or 2008… which was a REALLY long time ago). I’m getting older (as my waistline, wrinkles, and general lack of energy when it comes to caring for my two-year-old would tell). Nonetheless, point taken.
2. I think you make exceedingly good points about the overall quality of the game. The particular type of hitter we’ve lost is someone that you CAN still find – but Dee Gordon, Juan Pierre, and Ichiro seem to be the last of that dying breed. I suspect what is needed is Moneyball: constant re-evaluations of what is inefficient in the market, and loading up with that to the point of winning. We saw the Royals and Mariners do (and try to do, respectively) this with defense. We’ve seen teams try it with relief pitching, to varying degrees of success. The MAIN problem with a “Moneyball-style” approach at this juncture is, to me, obvious: teams only look for the inefficiencies in Free Agency, rather than in drafting and player development. One day, someone smart will come along and draft against the market, figuring out how to do that and make a winning team for pennies on the dollar. Then you’ll see people trying to copy the results, rather than the process, which is what gets us to this place, anyway. Teams will ALWAYS do what works. We just need someone to prove that the aforementioned style can work. When they do, other teams will follow.
3. It’s easy to lament the strikeout-heavy game played today, and to miss all the balls in play. So let’s not rule out something that COULD (maybe SHOULD) happen. That is, some sort of rule change. Pundits speak as if like baseball rule changes are impossible. But the NBA added the 3-point line 30 years ago; baseball added the DH 45 years ago. The NFL redefines fundamental and basic words like “catch,” “fumble,” “interference,” etc., on what sometimes seems like a weekly basis. If baseball gets much more out of hand, MLB will do something (whether it’s simple, like increasing the thickness of bat handles, or more drastic, like moving back fences, changing the composition of the ball, etc., remains to be seen).
Well, Doom, I hope that the potential trends you describe in your second point ease current tendencies enough that the more drastic steps of your third point don’t need to be taken. I guess that in baseball I prefer to let the unregulated market prevail. There are exceptions — I was glad to see the mound lowered in 1969 — but piling on with the DH four years later only confirmed for me the moral superiority of the NL, which I’d been taught as a religious belief two decades before.
On your point 1, I’ll hide behind my weasel-word, “mostly.” You’re not mistakable for an oldtimer (and neither is Voomo). I do worry that HHS’s contributor base has been aging as it has been shrinking (not to mention the disappearance of all gender variety). The more different perspectives the better. At least we seem to have a balanced commentariat when it comes to feelings about the Yankees.
World-series winning teams finishing last in MLB in home runs.
2012 Giants
1982 Cardinals
1965 Dodgers
1924 Senators
1906 White Sox
Any others?
I got the same 5 teams that you got.
For a while I was convinced that the surge in homers was because of the juiced ball, and that may very well be the root cause of this, or part of the root cause, along with the focus on launch angle, the desire to not hit into the shift, and so on. But I wonder if part of this could also be attributable to the effect that Stephen Jay Gould noticed over thirty years ago, that there’s a decline in variation of batters over time. Perhaps those with no power have simply been completely squeezed out of the league?
AlbaNate:
I think you’re on to something. Scouts at one time looked for ballplayers. Now they look for ballplaying athletes with particular skill sets, and, to a degree, size. In fact, the very organized way the game is run at every level from T-Ball up winnows not the grain from the chaff so much as those who conform to the various position models and those who don’t. Hard to see a Nellie Fox getting a look now. Ed Lopat would be running an accounting firm.
I’m not sure whether we have an HHS tradition of pausing to note the deaths of outstanding umpires, but reading about Doug Harvey today — a HoF umpire whose active years stretched back to my high school days — I began thinking about how great an impact individual umpires have upon games, and how little we actually understand what that impact is. (By “we” here I guess I mean the general fan; it may be that teams collect or contract for detailed data on individual umps.) Almost everything I know about umpires is based on narratives; the few stats that might be cited in an account of an umpire’s career — seasons/games umpired; years of postseason selection; years as crew chief — are uninformative when it comes to the specifics of an umpire’s unique impact upon games.
But we now possess tools that can change that. We see location on every pitch and that can be cross-matched to umpire calls systematically. (I have to believe teams already have this information in their possession, but I’ve never encountered anyplace where it can be generally accessed. It’s easy to imagine that official distribution would be opposed by the umpires’ union.) There is replayable video of every call on the field in every game, and although it may be that there are cases where no camera angle clarifies the accuracy of a call, the vast majority would be clear, whether there was a protest or not, and could be tabulated.
A few weeks ago, I was complaining about the complexity of fielding stats and the way that B-R incorporated in fielding calculations proprietary data from a private corporation called Baseball Info Solutions (BIS), data that we could not see, where the background of the data was basically a “black box” for us (or maybe just for me). Since that time I’ve learned a lot more about BIS and, in fact, have been poring over the first three or four large volumes of data compilations and methodological analyses (titled The Fielding Bible) that clarify a lot about what BIS does. I haven’t gotten the current one, Vol. IV, because it would cost me $25 instead of the $1.99 I paid for the three outdated volumes, but the first three volumes have confirmed me as a prophet: I predicted that if I had access to BIS data I wouldn’t know what to do with it, and I was absolutely right!
But, basically, using computerized tools for video analysis, BIS was anticipating StatCast-like data about fifteen years ago, and the result is that it’s able to give painfully detailed characterizations of each fielder — this right fielder has a great arm and uses it well to get baserunner kills, but is terrible at choosing efficient paths to fly balls and misses many catches others would make; that second baseman goes well to his right but not his left, pivots well on DPs, and has poor arm speed . . . etc., etc., quantifying the results each season both in terms of plays made and plays not made.
I think the same type of data should be compiled for umps (and if MLB can’t do it via StatCast for labor reasons, BIS can do it as a private company accessing broadcast materials). Every pitch and play is influenced by the individual skills and proclivities of the umps on the field — how consistent the home plate ump’s strike zone is, and how it may be generally displaced from a objective one; how consistently base umps maintain optimal positioning for calls and what percentage of their calls are disconfirmed by video data; disputed call rates; ejection rates and circumstances; and, of course, the new and easily tabulated figures on disputed calls confirmed/reversed (and more controversial characteristics, like out-call patterns on a DP second-base pivot). Looking at a description of the latest Fielding Bible, which mentions analysis of pitch framing in terms of distributed credit among pitcher, catcher, and umpire, I wonder whether BIS may not have already begun this sort of thing.
In my various rants against the way “chance” or “luck” is used as an explanatory fiction in advanced analyses, I’ve talked about ball-in-play outcomes being outcomes of three dimensions of skilled agency, exercised by pitchers, batters, and fielders, with virtually no residue left for a non-agential factor like “luck.” But from the players’ point of view, the ump’s call is luck — generally predictable, but nevertheless, like true luck or chance (e.g., pebbles and wind gusts), outside the agency of the players themselves, and occasionally, like the pebbles that did in Lindstrom and Kubek, adding a “random” determinative element to a play. We have the tools to change that. Once an ump’s skills and individual style are understood and a part of each game’s background framework, then play outcomes can be analyzed as a four-way combination of skills and goal-directed agency, the umpire’s being added to the pitcher’s, batter’s, and fielder’s.
If there is a spectrum of umpire quality, it matters and it affects game outcomes. We should be able to track it, understand it, and quantify it. After all, in every game, the umpires are on the field and part of game action more than any single player. They matter.
So here’s to Doug Harvey: he was a good one, even if we only know this through anecdotes, general testimony, and the fact of his many contract renewals and postseason game appointments.
E p m,
Going slightly tangential here, but after 50+ years of watching MLB, I couldn’t name the 5 best umpires over that period. However, I believe everyone could name the 5 worst. That belief, right or wrong, might speak volumes about the anonymous nature of good umpiring.
Harvey, Weyer, Pelakoudas-they were as much a part of the NL (and my youth) as Mays and Aaron
Hi, is this thing on? Haha. If you don’t remember me, I used to be a fairly regular commenter around these parts years ago. Before real life took some turns and I couldn’t stop by as often as I wished, until eventually I was never stopping by HHS at all. I somehow stumbled back on this site tonight and figured I’d check out what’s going on. Glad to see you’re still keeping HHS kicking, Doug!
Anyway, this is really interesting. Although I was aware of it by now of course, I somehow didn’t even notice the stark increase in HR production this past year during the season, astonishing rookies aside. Do y’all think this is the culmination of trends or a sudden spike that will fade and be forgotten?
Nice to see a once-familiar screen name returning! Doug’s the man: no Doug, no HHS.
My thought on your question is that it may be like El Niño years: a pattern of ups and downs, but steadily rising and spiking when conditions are periodically ripe. Now that StatCast has taught teams how ball-in-play outcomes are affected by bat speed and launch angle combinations, I expect coaching staffs are going to learn how to use high-tech monitoring to train young guys with those payoffs specifically in mind, and the proportion of players effectively aiming for line drives that could reach the seats is likely to rise in the long term towards some equilibrium point higher than the ones we’ve seen in the past.
I don’t know whether anyone will spot this comment, but it concerns our upcoming CoG rounds, and I think the issue I’d like to discuss is one we may want to consider before Doug opens the first of what I assume will be four rounds, given today’s Hall announcement.
I should note at the outset that I’m not sure I remember the rules of the CoG process clearly, so this may be a wasted post: my understanding is that we will be tasked with choosing 4 new inductees, selected from 11 holdover candidates (Luis Tiant, Dick Allen, Kevin Brown, Dave Winfield, Bill Dahlen, Manny Ramirez, Richie Ashburn, Graig Nettles, Bobby Wallace, and Andy Pettitte), perhaps enlarged by a player named through a Redemption Round process (who at some time in the past had little support), plus qualifying players born in 1973.
I think we may be encountering a problem that was not relevant to birtlecom’s original concept: the fact that the high number of open slots for the CoG — which, by birtlecom’s rules must be filled — may exceed the number of eligible candidates that we would have wished to elect under the conditions of previous rounds. To some degree, this was an issue in 2017 as well. In 2018, the problem is that, if I have the rules right, the enlargement of the pool provided by the 1973 group has to supply fresh candidates for four rounds, rather than the one or two rounds that was the general rule throughout the original process.
There are certainly potentially deserving candidates in the 1973 pool, including a couple who are still active (perhaps a problematic issue in itself). But that pool does not seem to be an abnormally rich one, and I feel it’s likely we’ll wind up looking to the holdover candidates to supply most of the electees. This may have the pleasant effect of bringing into the Circle some long-time favorites of a core group of supporters, but my concern is that given the small size of the pool relative to the number we must select, we may wind up forced by the rules to lower the bar for the CoG.
Obviously, that in itself isn’t a really good outcome, but what concerns me more is that we may be forced to allocate membership to multiple borderline candidates this year, while potentially having few slots available in a future year when the Hall may admit only one or two players, and the birth year for CoG consideration is one rich in excellent candidates.
So I wonder whether before we launch the next CoG round, we may not want to consider: (1) whether to hold a Redemption Round, and (2) whether to modify the election process in some way (perhaps by providing an option to vote for “None of the above; retain this slot for future election use” after, say, two rounds — this sort of option might also allow us to postpone voting on active players, if we’d rather not do that).