Lots of people suggested deadening the ball when discussing trying to limit three outcome at bats. Maybe that’s what MLB did. Now shouldn’t they think about shrinking the strike zone? Then you have a ball that doesn’t leave the yard as easily and pitchers who have to come into the hitter. Sounds like more balls in play right?
Joe can’t we get some stats by hitter obp vs shift when trying to bunt versus swinging away? I have been one of the ones saying forever let them shift, and whether through natural selection of different hitters making it through the minors, or practice and skill development, you will find more major league hitters who can bunt.
And we need a pitch clock, a batters box clock and the Astro mic.
I was at the Royals/ Ranger game that fateful Saturday night in 1977 when the brawl broke out that you mentioned Brett and Willie Horton being involved in. In my 60+ years of following The Great Game I’ve never seen anything remotely like this fight. All other baseball dust ups look like church socials compared to this one. I’d love to hear George’s story. Unfortunately no video exists that I’m aware of. Thanks for the memory.
Speaking of Miggy, I'm surprised you didn't mention that his 3000th hit is probably the last one we are going to see for quite some time (10+ years if not more since no one is remotely close now).
If you want to take one of your deep dives into a topic you might want to write about the very real possibility that when Miggy retires next year (I'm assuming he does when his contract expires) he will be Baseballs last career .300 hitter. I know it seems crazy but if you look at active players with .300 batting averages only a handful of guys fit that category (5000 plate appearances) and everyone but Miggy is almost certain to fall under .300 as they hit the backside of their career (including Trout). With pitching the way it is today it seems incredibly unlikely we will ever have another .300 career hitter once Miggy retires.
As Joe says in the last paragraph, the game is always adjusting for these tilts towards pitchers and hitters. I wouldn't say never for a .300 hitter because if offense stays this low, I imagine rule changes are coming.
I do think we are done with 300 win pitchers though, the relief pitcher bell just seems like one you can't unring.
The problem is that baseball has been moving away from specialist hitters (the Vince Coleman base stealing speedster, the Carew/Boggs/Gwynn low power high average pure hitter etc) that we used to have lots of in the 70s, 80s, 90's and even early 2000s. There isn't a place now for pure .300 hitters unless they also do a lot of other things (power, good defense etc). I think that dooms .300 hitters because those players are incredibly rare. Combine that with the dominant relief pitching that's depressing hitting and there is no going back.
One of my very favorite baseball TV memories is watching Victor Martinez giving a ration of s**t to shifted infielders one night, and then the slowest man in baseball laid down a perfect bunt single. I was roflmao, you better believe! ( And then he never did it again! )
More strikeouts than hits?? maybe that's a signal to change HR strategy which doesn't seem to be workig too well. defensive shift not all about stats, theres a psychological ipact ,i e mental battle between batter and 95 throwing southpaw.
Why must it be a bunt? How hard is it to slap a grounder to the left side (like hitting one to the right to move a runner to 3b)? With only one infielder there, I'm guessing batters could hit at least .400 when they do that. So unless you're Ted Williams...
There's got to be something they've done to the ball itself. The year-to-year comparison of batted balls shows that the likelihood of a ball hit 95+ mph off the bat is only slightly down this year, but then chances of such a ball leaving the yard are WAY down from last year. It should go up slightly as the weather warms up, but jeez, not that much.
Last year there were about 10.2 baseballs hit 95+ mph off the bat per game. Those events became hits 48.8% of the time (i.e. hitters have a .488 BA when they hit the ball 95+ mph), and were homers 11.96% of the time.
This year, so far, there are slightly fewer such events, only 9.93 per game, a decrease of about 2.6%. The batting average on those is only .451, a 7.6% decrease from last year. Maybe the humidors, maybe the increased shifting, etc. could account for that, but probably not.
But the real kicker is that the chances of those batted balls becoming a homer have dropped to just 8.91%, a decrease of 25.5% from 2021!
Given that hits are coming off the bat at high speeds at a rate almost the same as last year, this can clearly not be blamed on humidors. That would change the exit velocity, I would think.
Which means it has to be the drag coefficient on the ball. Maybe the leather is slightly rougher. Maybe the seams are slightly higher. I don't know. But I bet MLB knows EXACTLY what it did, and it just isn't telling. Joe made a comment once a couple of years ago about how the PGA knows exactly what the golf ball does in every way imaginable, and was incredulous that MLB said they really didn't know. I would posit that MLB does in fact know, but they pretend ignorance so they can screw around with things when they perceive a need to do so and then plead innocence when things backfire on them.
Maybe this is overly hopeful, but if the lack of home runs is due to some intentional change to the baseball, that could be a first step toward bringing back more action to the game. Next step has to be something that increases hits in the field of play. I think if you make it harder for pitchers to overpower hitters - some change to the ball to cut spin rates, fewer pitchers on the roster, pitch clock, whatever - the shift might go away on its own as hitters can more easily beat it.
If not, go ahead and outlaw the shift. It’s not some longstanding tradition anyway.
It's mind boggling that players don't bunt to the weak side of the shift. I thought it must be one of those unwritten codes of baseball-"it's unmanly to bunt against the shift". I guarantee that if Popovitch or Belechick were coaching baseball, they would require their players to learn to put the ball in play to the weak side of the shift, and then all of the other managers would also require it, and that would be the end of the shift. It's ridiculous to see one player on the left side of the diamond, and yet virtually all left handed batters refuse to take advantage of that.
I think that the pitch clock may help the offense some. Pitchers have been quite vocal that they "need" the extended time between pitches to be able to go max effort on every throw. Take away that extended time and maybe the velocity drops by a tick or two while the spin rate comes down as well (especially because of the cumulative effect—with each pitch, they're bit further from fully rested). It doesn't need to be a dramatic change to make a real difference.
If that's the case, one thing you'll see is a bigger drop from the most marginal talents—the middle relievers who don't have the arsenal to start nor the weapons to pitch in high leverage, but are able to get some outs with 98 mph—but not with 96. And if those guys go, then the whole shape of the game alters to something closer to its longtime form.
While for anyone my age (and Joe's too - hooray 1967!) would think that the shift is not aesthetically pleasing, I can't find evidence that it is driving the decline in BA. Batting average on balls in play, and its converse, Defensive Efficiency, have been pretty much rock solid consistent within a narrow range during my time watching baseball. Sometimes it inches up, sometimes down, but always in the same .290 to .310 band. It seems that the shift has kept the DEF from falling well below norms, not caused it to increase significantly.
The MLB BABiP is currently the lowest it's been in a season since 1989. It had been between .293 and .303 every year from 1993 to 2019.
Then it dropped to .292 the last two seasons.
Now it's just .283.
There are also fewer balls in play this year, so far, than there were in 2019, which was already about 10,000 fewer balls in play than we had in 2009. Right now we're on a pace for about 95,350 balls in play, and at a rate of .283, we would see about 8,500(!) fewer hits than we did just three seasons ago. [EDITED: Originally I said 2,500 hits, but my math was wrong on that one.)
The shift (and in my estimation, changes to the ball itself) are having huge consequences on the game, and not in a good way for the fans.
Data based on BABIP and other counting stats from Baseball-Reference.
Comparing April to full seasons is a meaningless exercise. It is like comparing an apple to an apple tree.
The Babip is really not any different than it was last year. Last year in April, the Babip was .283. This year, in an April not over yet (with the better weather part coming) and a shortened spring training, it is .283. Last year the batting average was .232 and OBP was .309 in April. This year with a partial month, (and the short ST, and less games) it is so far .231 and .307. Short, with a definite chance to be higher in 5 days when the month is over. Doubles and triples are up slightly over the full month last year. (1.69 per game vs 1.61) So are singles (5.03 to 4.89)
No, the difference is the home runs, and the reason is simple - the universal humidor. This isn't some grand mystery. It is not because of the endless pitchers. It is not because of the shift, it is simply because there are less home runs (.9 per game this year vs. 1.14 last year in April)
The funny thing is, we are getting what we cried for in 2019. Too many three true outcome PAs. Home runs are out of hand. Strikeouts are out of hand. We need more balls in play.
Now that we have that (the percentage of PAs with balls put in play are the highest in April since 2017) we panic. Baseball changes slowly, and we panic at every single change. K/9 were down last season and this season is less than last year so far after going up for 15 consecutive years. This is in part because baseball is slowly adjusting to the shift. They are slowly starting to give more PAs to action hitters rather than three true outcome pull the ball hard every time guys. (If they ban the shift as they seem to intend to next year, K/9 will go up close to half a K per game. You heard it here first.) Pitches per PA also went down last year after steady yearly gains since 2015. It went from 3.82 to 3.96. Down last year to 3.9, it is again at 3.9.
The humidors will again, eventually result in less PAs given to all or nothing guys and more to less limited hitters, resulting in more action in the game, (more steal attempts also) as well as shorter games. Which is what everyone has asked for.
The runs will eventually balance out. This is my 50th year of watching baseball, and the average runs per game over that time is 8.97. This includes a long steroid era and the launch angle revolution. There are two years in that stretch that there were less runs per game than there have been so far this year (and this year will not end up there. I am sure it will end up over the 8.13 we had in just 2014). 26 of those have ended up between 8.5 and 9.5 runs per year. Anything between those two numbers should be considered within range of a normal season.
I think 9 (or 4.5 per team) is about the sweet spot. I think if we allow baseball to evolve, we will get there or close to it, soon, with more action, shorter games, less K's and less home runs.
And we won't let it happen. Because baseball fans are the panic express. Especially in April. They will ban the shift, they will liven the ball or stop the humidors, and we will be right back where we started. More strikeouts, more home runs, more pitches per PA, less balls in play, and longer games, and fans will be right back to complaining about that, even though they (and the MLB overreacting to their overreactions) caused it to happen.
OK, so I dug a bit deeper. You correctly point out that the April/March BABiP and league BA are similar to last year. I am surprised that those numbers change that much later in the season, but I am willing to stand corrected. I will note however that last year's .283 in Mar/Apr was the lowest such mark in at least 25 years, which is as far back as I bothered to look. (Every Mar/Apr back to 1996 was between .290 and .302 before 2021.) So that is still disconcerting in my mind.
As for your assertion that the balls in play are actually up, I concur, though not to the degree that you state. I have them at 6.71/G this year, compared to 6.5/G last April, though that is still well below any other number in the last 26 seasons, barring 2020.
Unless you are taking SF and SH into account. I don't think they should be, because those are surrendered outs, not attempts at a hit that ended up in play. Those are up slightly from last year - perhaps in an effort to beat the shift? - but are still in a historic nadir overall.
Anyway, I plotted Mar/Apr ball-in-play data going back to 1996 and then graphed it to make it easy to read, and posted it on my old blog.
I used Baseball-Reference data and did the DER calculations myself based on the formula from the MLB site, and then multiplied it by 10 to get the scale similar to the scale of Hits/Game, Homers/Game and Balls-in-Play per game.
In any case, it seems to me that while DER is generally pretty steady around 0.7, it is up notably in the last few years, to 0.712 and now 0.713 in 2021 and 2022 respectively. The difference between that and the 0.695 we had in 2019 is about 2.5%, which, maybe doesn't sound like much, but over the course of ~30,000 events works out to something like 750 fewer hits for the season.
You can see also that Hits/Game (yellow) have been in steady decline, despite a few blips, for ~25 years, and that BiP/G are also generally down in that span, though they are up slightly this year.
So the HR/G are way down form the last couple of years, but the extra BiP/G are not turning into hits because the defensive shifts are eliminating those, too. Fans want more action, and continuously watching grounders to deep second that used to be line drive singles and doubles in the gap are not, in my mind, answering that call.
I'm fine if they want to monkey around with the ball to try to get more action into the game, but teams build their offenses and pitching staffs around what they anticipate their players can do, and that is based largely on what they've done in the past. If you're going to screw with things, then teams will end up building their rosters based on incomplete or misleading data, which seems a little unfair to me.
The idea that defensive shifts are the one thing driving down Babip really has no backing, and looking only at the last several years gives you a skewed sense of perspective.
The average Babip for the last 49 years (DH era) is .292. Last year it was .292. I have no doubt it will end up somewhere close to that this season. There have been 9 seasons it was at or below the level it is this April for a full season, and 10 where it was at the .300 that seems to be your goal.
I don't know where you got your balls in play data. I suspect you are talking about hits. Balls in play are At bats - home runs - strikeouts +sacrifice flies (they take sac flies out of at bats, but they are a ball in play) Last April it was at 22.96 per team per game, this year it is at 23.76 thus far. That is 1.6 (counting both teams) more per game.
Of course the number of balls in play has been pretty steadily going down for decades with the constant rise in Ks and the rise in home runs The nadir for a full season was 2020 when it was 23.13. It went up to 23.67 for the full year last year. Not much, but at least the right direction, and the way it was plummeting, just going in the right direction for two years is a step.
Also comparing HR/game to recent years is rough. Recent years are the aberration, not this tiny April with the hitters cold. Last year was the 24th season since we went to 32 teams. The last 6 years have finished 1st (2019), 2nd (2020) 3rd (2017), 4th (2021) 6th (2016) and 7th (2018) in home runs per game. (2001 is number 5) We had less than 1 per game (per team) in 2010, 2011, 2012, and 2014 (2014 was less for the full season than it has been so far this year) and baseball was fine. Oh, and also less than 1 per game every year prior to 1994 (steroid era begins in earnest) except for 1987 (juiced ball controversy). I feel we will end up close to or over one per game after the summer.
I don't think the shifts are the *one* thing driving BiP down, necessarily, but they're a big part of it. Hitters have no choice but to try harder and harder to swing for the fences, leaving us with very little but the three true outcomes.
My BiP were really *Hits* in play, I now realize. Sorry about that. I have recalculated using the formula you cite and came up with the same stats you had for the last couple years so I know we're on the same page.
Early season Hits in play per game are up slightly this year, but are still WAY below historic levels, and the general trend has been downward since about 2004. They averaged 7.8 HiP/G from 1996 to 2018, which was the last time we saw an increase, at least for the early part of the season. This month's 6.7 extrapolates to something like 4500 fewer hits in play over the course of the season. Even if it's half that bad, that's still like 2000 fewer hits in play compared to the ~20-year average. Getting the BiP/G up by 1.6 doesn't do us much good if only an extra 0.2 of them become hits. That's a .125 batting average on those extra balls in play.
With that said, I think you're comparing a bit of apples and apple trees yourself here. You cannot use the average BABiP for half a century, because BABiP changed significantly and more or less permanently in 1993. Between 1973 (when the leagues made some changes after the severely low-scoring 1972 season) and 1992, the MLB BABiP was between .280 and .287 every year but two. In the 1987 "Rabbit Ball" year, it jumped to .289, and in 1981 it dropped to .279, probably because a bunch of games were missed during the warmer (good hitting) months with the strike. Otherwise it was remarkably consistent.
Then in 1993 it jumped to the unprecedented height of .294 and pretty much stayed at or above that level for most of the next three decades. Excluding shortened seasons (which can have a disproportionate effect on this, apparently, depending on which games get missed) the full season BABiP had been between .293 and .303 every year from 1996 to 2019. I don't think we can count 2020 because it was such an anomaly in so many ways.
So you can't just take an average of two wildly disparate run-scoring environments and say, "See! We're still within a standard deviation!" or whatever. BABiP dropping to ~5-10 points works out to 500-1500 fewer hits over a season, and a lot of those are because of the shift, but I admit that the ball is being messed with, too.
Nitpick: SH maybe, but it has been shown that players are no more likely to hit a fly ball with a runner in third and <2 out. That is, they aren’t intentionally giving themselves up.
I’ll look at the blog post when I’m on a computer rather than my phone.
I've been watching roughly the same amount of time. The first game I remember going to was during the interval when the Yankees were playing at Shea Stadium while their home in the Bronx was being remodeled so sometime in 1974-75. I too agree that ~9 runs per game is the sweet spot. I'd only note that, from my aesthetic POV, how that sweet spot is achieved is probably as relevant as how many runs are scored. Even given similar levels of runs scored per game, the Whitey Herzog Royals offense of line drives and aggressive running was more fun for me than the Earl Weaver Orioles offense of walks and three run homers.
I’m not yet convinced that it is the humidor as opposed to the ball itself. Storage vs construction. If the former, we get need to dial back moisture fraction by fraction to fine tune. If the latter, control doesn’t seem to be as precise. (See eg 1987 and 2019)
1. This year is a tiny sample size of early season (pre warming weather) and without normal spring training.
2. You mention the composition of the balls. That, and not shifting, has changed dramatically this year and last compared to the prior seasons. New balls began to be used last year - unfortunately mixed with old - and are completely in play this season. Shifting has been gradually getting more common for a while going back a decade or so.
3. I'm not sure of your source. (Not criticizing, just not sure.) Using BB-Ref and reversing this for DER simply because it was the first one I found instead of BABIP which, within rounding errors, is simply the converse.
2022 .703
2021 .695
2020 .693
2019 .688
2018 .691
2017 .688
2016 .688
2015 .689
2014 .690
2013 .692
2012 .691
2007 .686
2002 .695
1997 .686
1992 .703
1987 .697
1982 .702
1977 .698
1972 .713
Unless one puts a ton of weight on the first ~15 games of the year with cold weather and an unusual Spring Training, that looks like a flat line to me. Where did shifting start and how does one pick it out from this collection of DER?
I would not call ~10% of the season a tiny sample size. For an individual, sure. But in the aggregate, we're still talking about well over 150,000 at-bats, which is plenty to make in inference like this. I admit that offense always picks up when the weather warms, but not to this extent, I don't think.
I used B-R for my data. I would imagine that even if the offense picks up a bit, we will still end up with the lowest BABIP this season in ~30 years, and considering that last year and the year before were also the lowest since 1992, well, that is a trend. Correlation does not necessarily IMPLY causality, but it can suggest it.
I don't have a feel for DER or know how significant those perturbations are, but we all have a pretty good feel for batting average. When you're talking about ~100,000 events, a 9-point difference in batting average works out to about 9,000 fewer hits. (My math was wrong on my first post, which I have since edited.)
I prefer DER over BABIP for this because it includes reaching on errors. If the shift takes away a hit but the batter reaches because the fielder dropped it, that shouldn’t be counted as a success for the shift.
Even assuming this year’s sample is representative of the whole, there would be 1.5 more batters reaching per 100 balls in play than 2019. That is something like 1 for every 3-4 games, an insignificant amount imho.
Arguing definitions is of course silly. But fwiw I think 10% is a tiny sample not because of the raw number but because it is not drawn randomly from the whole. Instead it is 100% of the April games. Extrapolation from that is dangerous.
Joe you cite multiple hard data pointing to the dearth of hitting and then talk about the shift in subjective even poetic terms.
The shift is killing / has killed the offensive game. The stats from the minors when they enforce the infielders on the infield rule show hard line drives to short left and right fall as hits as was the case in baseball for 100 years, not long outs. Most Players aren’t going to bunt their way on. It just seems like capitulation from the get go.
Also, the ball definitely isn’t carrying as expected. Have seen many balls already that look like no doubters die on the track. It’s not just the April cold.
Like where pitcher, batter, and fielder all look amazed that the ball ended up in a glove and not row 10.
The shift is only as effective as it is due to all of the surrounding issues with pitching, because even with seven players on one side it still makes more strategic and tactical sense for most players to pull a ball for a homer. Start working on the pitchers, and you'll see hitters start working against the shift on their own.
Since the beginning of the game, a line drive hit over the infielders head well I front of the outfielder was a guaranteed hit. Every time. Solid contact meant a hit.
Shift has taken that away and so player has to swing for fence to go over the defense.
Just because you can doesn’t mean it should be allowed.
Players playing the three true outcome game and then the launch angle revolution preceded the rise of the shift. The shift is a reaction to the three true outcome game, pulling everything and swinging for the fences with two strikes and not the other way around.
Before 2009, there had only been one year when K's and home runs accounted for over 20% of PAs. It has been over 20% every year starting in 2009, pretty steadily rising from there, reaching a high of 26.59% in 2019.
When this started, the shift was an oddity at best, something the Rays did every once in a while, and didn't really become used league wide until about 2016.
Joe, it’s bandbox, not sandbox. The term bandbox comes from the small box used to store old fashioned shirt collars. See, even you can learn something new about baseball!
Lots of people suggested deadening the ball when discussing trying to limit three outcome at bats. Maybe that’s what MLB did. Now shouldn’t they think about shrinking the strike zone? Then you have a ball that doesn’t leave the yard as easily and pitchers who have to come into the hitter. Sounds like more balls in play right?
Joe can’t we get some stats by hitter obp vs shift when trying to bunt versus swinging away? I have been one of the ones saying forever let them shift, and whether through natural selection of different hitters making it through the minors, or practice and skill development, you will find more major league hitters who can bunt.
And we need a pitch clock, a batters box clock and the Astro mic.
I was at the Royals/ Ranger game that fateful Saturday night in 1977 when the brawl broke out that you mentioned Brett and Willie Horton being involved in. In my 60+ years of following The Great Game I’ve never seen anything remotely like this fight. All other baseball dust ups look like church socials compared to this one. I’d love to hear George’s story. Unfortunately no video exists that I’m aware of. Thanks for the memory.
Speaking of Miggy, I'm surprised you didn't mention that his 3000th hit is probably the last one we are going to see for quite some time (10+ years if not more since no one is remotely close now).
If you want to take one of your deep dives into a topic you might want to write about the very real possibility that when Miggy retires next year (I'm assuming he does when his contract expires) he will be Baseballs last career .300 hitter. I know it seems crazy but if you look at active players with .300 batting averages only a handful of guys fit that category (5000 plate appearances) and everyone but Miggy is almost certain to fall under .300 as they hit the backside of their career (including Trout). With pitching the way it is today it seems incredibly unlikely we will ever have another .300 career hitter once Miggy retires.
As Joe says in the last paragraph, the game is always adjusting for these tilts towards pitchers and hitters. I wouldn't say never for a .300 hitter because if offense stays this low, I imagine rule changes are coming.
I do think we are done with 300 win pitchers though, the relief pitcher bell just seems like one you can't unring.
I don't think rule changes will help.
The problem is that baseball has been moving away from specialist hitters (the Vince Coleman base stealing speedster, the Carew/Boggs/Gwynn low power high average pure hitter etc) that we used to have lots of in the 70s, 80s, 90's and even early 2000s. There isn't a place now for pure .300 hitters unless they also do a lot of other things (power, good defense etc). I think that dooms .300 hitters because those players are incredibly rare. Combine that with the dominant relief pitching that's depressing hitting and there is no going back.
One of my very favorite baseball TV memories is watching Victor Martinez giving a ration of s**t to shifted infielders one night, and then the slowest man in baseball laid down a perfect bunt single. I was roflmao, you better believe! ( And then he never did it again! )
Something that is masking the decline of offense a bit this season. We have a universal DH.
Good point!
More strikeouts than hits?? maybe that's a signal to change HR strategy which doesn't seem to be workig too well. defensive shift not all about stats, theres a psychological ipact ,i e mental battle between batter and 95 throwing southpaw.
Why must it be a bunt? How hard is it to slap a grounder to the left side (like hitting one to the right to move a runner to 3b)? With only one infielder there, I'm guessing batters could hit at least .400 when they do that. So unless you're Ted Williams...
There's got to be something they've done to the ball itself. The year-to-year comparison of batted balls shows that the likelihood of a ball hit 95+ mph off the bat is only slightly down this year, but then chances of such a ball leaving the yard are WAY down from last year. It should go up slightly as the weather warms up, but jeez, not that much.
Last year there were about 10.2 baseballs hit 95+ mph off the bat per game. Those events became hits 48.8% of the time (i.e. hitters have a .488 BA when they hit the ball 95+ mph), and were homers 11.96% of the time.
This year, so far, there are slightly fewer such events, only 9.93 per game, a decrease of about 2.6%. The batting average on those is only .451, a 7.6% decrease from last year. Maybe the humidors, maybe the increased shifting, etc. could account for that, but probably not.
But the real kicker is that the chances of those batted balls becoming a homer have dropped to just 8.91%, a decrease of 25.5% from 2021!
Given that hits are coming off the bat at high speeds at a rate almost the same as last year, this can clearly not be blamed on humidors. That would change the exit velocity, I would think.
Which means it has to be the drag coefficient on the ball. Maybe the leather is slightly rougher. Maybe the seams are slightly higher. I don't know. But I bet MLB knows EXACTLY what it did, and it just isn't telling. Joe made a comment once a couple of years ago about how the PGA knows exactly what the golf ball does in every way imaginable, and was incredulous that MLB said they really didn't know. I would posit that MLB does in fact know, but they pretend ignorance so they can screw around with things when they perceive a need to do so and then plead innocence when things backfire on them.
https://baseballsavant.mlb.com/statcast_hit_probability
MLB did intentionally deaden the ball - the Athletic snagged the internal memo – but I don't think we know how they deadened it:
https://theathletic.com/2375121/2021/02/08/mlb-changes-baseball-deadening/
Sandbox? Maybe meant bandbox? Never heard of a stadium being called a sandbox.
I had the same thought when reading. Not that I know what a bandbox actually is, other than a word for a hitter-friendly baseball stadium.
I actually went to my Dickson Baseball Dictionary thinking I have never heard of “sandbox” in referring at a stadium. It’s not in there.
Maybe this is overly hopeful, but if the lack of home runs is due to some intentional change to the baseball, that could be a first step toward bringing back more action to the game. Next step has to be something that increases hits in the field of play. I think if you make it harder for pitchers to overpower hitters - some change to the ball to cut spin rates, fewer pitchers on the roster, pitch clock, whatever - the shift might go away on its own as hitters can more easily beat it.
If not, go ahead and outlaw the shift. It’s not some longstanding tradition anyway.
It's mind boggling that players don't bunt to the weak side of the shift. I thought it must be one of those unwritten codes of baseball-"it's unmanly to bunt against the shift". I guarantee that if Popovitch or Belechick were coaching baseball, they would require their players to learn to put the ball in play to the weak side of the shift, and then all of the other managers would also require it, and that would be the end of the shift. It's ridiculous to see one player on the left side of the diamond, and yet virtually all left handed batters refuse to take advantage of that.
I think you ought to read up on what made those coaches smart and effective at their jobs.
I think that the pitch clock may help the offense some. Pitchers have been quite vocal that they "need" the extended time between pitches to be able to go max effort on every throw. Take away that extended time and maybe the velocity drops by a tick or two while the spin rate comes down as well (especially because of the cumulative effect—with each pitch, they're bit further from fully rested). It doesn't need to be a dramatic change to make a real difference.
If that's the case, one thing you'll see is a bigger drop from the most marginal talents—the middle relievers who don't have the arsenal to start nor the weapons to pitch in high leverage, but are able to get some outs with 98 mph—but not with 96. And if those guys go, then the whole shape of the game alters to something closer to its longtime form.
While for anyone my age (and Joe's too - hooray 1967!) would think that the shift is not aesthetically pleasing, I can't find evidence that it is driving the decline in BA. Batting average on balls in play, and its converse, Defensive Efficiency, have been pretty much rock solid consistent within a narrow range during my time watching baseball. Sometimes it inches up, sometimes down, but always in the same .290 to .310 band. It seems that the shift has kept the DEF from falling well below norms, not caused it to increase significantly.
Here's some evidence for you:
The MLB BABiP is currently the lowest it's been in a season since 1989. It had been between .293 and .303 every year from 1993 to 2019.
Then it dropped to .292 the last two seasons.
Now it's just .283.
There are also fewer balls in play this year, so far, than there were in 2019, which was already about 10,000 fewer balls in play than we had in 2009. Right now we're on a pace for about 95,350 balls in play, and at a rate of .283, we would see about 8,500(!) fewer hits than we did just three seasons ago. [EDITED: Originally I said 2,500 hits, but my math was wrong on that one.)
The shift (and in my estimation, changes to the ball itself) are having huge consequences on the game, and not in a good way for the fans.
Data based on BABIP and other counting stats from Baseball-Reference.
Comparing April to full seasons is a meaningless exercise. It is like comparing an apple to an apple tree.
The Babip is really not any different than it was last year. Last year in April, the Babip was .283. This year, in an April not over yet (with the better weather part coming) and a shortened spring training, it is .283. Last year the batting average was .232 and OBP was .309 in April. This year with a partial month, (and the short ST, and less games) it is so far .231 and .307. Short, with a definite chance to be higher in 5 days when the month is over. Doubles and triples are up slightly over the full month last year. (1.69 per game vs 1.61) So are singles (5.03 to 4.89)
No, the difference is the home runs, and the reason is simple - the universal humidor. This isn't some grand mystery. It is not because of the endless pitchers. It is not because of the shift, it is simply because there are less home runs (.9 per game this year vs. 1.14 last year in April)
The funny thing is, we are getting what we cried for in 2019. Too many three true outcome PAs. Home runs are out of hand. Strikeouts are out of hand. We need more balls in play.
Now that we have that (the percentage of PAs with balls put in play are the highest in April since 2017) we panic. Baseball changes slowly, and we panic at every single change. K/9 were down last season and this season is less than last year so far after going up for 15 consecutive years. This is in part because baseball is slowly adjusting to the shift. They are slowly starting to give more PAs to action hitters rather than three true outcome pull the ball hard every time guys. (If they ban the shift as they seem to intend to next year, K/9 will go up close to half a K per game. You heard it here first.) Pitches per PA also went down last year after steady yearly gains since 2015. It went from 3.82 to 3.96. Down last year to 3.9, it is again at 3.9.
The humidors will again, eventually result in less PAs given to all or nothing guys and more to less limited hitters, resulting in more action in the game, (more steal attempts also) as well as shorter games. Which is what everyone has asked for.
The runs will eventually balance out. This is my 50th year of watching baseball, and the average runs per game over that time is 8.97. This includes a long steroid era and the launch angle revolution. There are two years in that stretch that there were less runs per game than there have been so far this year (and this year will not end up there. I am sure it will end up over the 8.13 we had in just 2014). 26 of those have ended up between 8.5 and 9.5 runs per year. Anything between those two numbers should be considered within range of a normal season.
I think 9 (or 4.5 per team) is about the sweet spot. I think if we allow baseball to evolve, we will get there or close to it, soon, with more action, shorter games, less K's and less home runs.
And we won't let it happen. Because baseball fans are the panic express. Especially in April. They will ban the shift, they will liven the ball or stop the humidors, and we will be right back where we started. More strikeouts, more home runs, more pitches per PA, less balls in play, and longer games, and fans will be right back to complaining about that, even though they (and the MLB overreacting to their overreactions) caused it to happen.
OK, so I dug a bit deeper. You correctly point out that the April/March BABiP and league BA are similar to last year. I am surprised that those numbers change that much later in the season, but I am willing to stand corrected. I will note however that last year's .283 in Mar/Apr was the lowest such mark in at least 25 years, which is as far back as I bothered to look. (Every Mar/Apr back to 1996 was between .290 and .302 before 2021.) So that is still disconcerting in my mind.
As for your assertion that the balls in play are actually up, I concur, though not to the degree that you state. I have them at 6.71/G this year, compared to 6.5/G last April, though that is still well below any other number in the last 26 seasons, barring 2020.
Unless you are taking SF and SH into account. I don't think they should be, because those are surrendered outs, not attempts at a hit that ended up in play. Those are up slightly from last year - perhaps in an effort to beat the shift? - but are still in a historic nadir overall.
Anyway, I plotted Mar/Apr ball-in-play data going back to 1996 and then graphed it to make it easy to read, and posted it on my old blog.
http://www.boyofsummer.net/2022/04/blog-post.html
I used Baseball-Reference data and did the DER calculations myself based on the formula from the MLB site, and then multiplied it by 10 to get the scale similar to the scale of Hits/Game, Homers/Game and Balls-in-Play per game.
In any case, it seems to me that while DER is generally pretty steady around 0.7, it is up notably in the last few years, to 0.712 and now 0.713 in 2021 and 2022 respectively. The difference between that and the 0.695 we had in 2019 is about 2.5%, which, maybe doesn't sound like much, but over the course of ~30,000 events works out to something like 750 fewer hits for the season.
You can see also that Hits/Game (yellow) have been in steady decline, despite a few blips, for ~25 years, and that BiP/G are also generally down in that span, though they are up slightly this year.
So the HR/G are way down form the last couple of years, but the extra BiP/G are not turning into hits because the defensive shifts are eliminating those, too. Fans want more action, and continuously watching grounders to deep second that used to be line drive singles and doubles in the gap are not, in my mind, answering that call.
I'm fine if they want to monkey around with the ball to try to get more action into the game, but teams build their offenses and pitching staffs around what they anticipate their players can do, and that is based largely on what they've done in the past. If you're going to screw with things, then teams will end up building their rosters based on incomplete or misleading data, which seems a little unfair to me.
The idea that defensive shifts are the one thing driving down Babip really has no backing, and looking only at the last several years gives you a skewed sense of perspective.
The average Babip for the last 49 years (DH era) is .292. Last year it was .292. I have no doubt it will end up somewhere close to that this season. There have been 9 seasons it was at or below the level it is this April for a full season, and 10 where it was at the .300 that seems to be your goal.
I don't know where you got your balls in play data. I suspect you are talking about hits. Balls in play are At bats - home runs - strikeouts +sacrifice flies (they take sac flies out of at bats, but they are a ball in play) Last April it was at 22.96 per team per game, this year it is at 23.76 thus far. That is 1.6 (counting both teams) more per game.
Of course the number of balls in play has been pretty steadily going down for decades with the constant rise in Ks and the rise in home runs The nadir for a full season was 2020 when it was 23.13. It went up to 23.67 for the full year last year. Not much, but at least the right direction, and the way it was plummeting, just going in the right direction for two years is a step.
Also comparing HR/game to recent years is rough. Recent years are the aberration, not this tiny April with the hitters cold. Last year was the 24th season since we went to 32 teams. The last 6 years have finished 1st (2019), 2nd (2020) 3rd (2017), 4th (2021) 6th (2016) and 7th (2018) in home runs per game. (2001 is number 5) We had less than 1 per game (per team) in 2010, 2011, 2012, and 2014 (2014 was less for the full season than it has been so far this year) and baseball was fine. Oh, and also less than 1 per game every year prior to 1994 (steroid era begins in earnest) except for 1987 (juiced ball controversy). I feel we will end up close to or over one per game after the summer.
I don't think the shifts are the *one* thing driving BiP down, necessarily, but they're a big part of it. Hitters have no choice but to try harder and harder to swing for the fences, leaving us with very little but the three true outcomes.
My BiP were really *Hits* in play, I now realize. Sorry about that. I have recalculated using the formula you cite and came up with the same stats you had for the last couple years so I know we're on the same page.
Early season Hits in play per game are up slightly this year, but are still WAY below historic levels, and the general trend has been downward since about 2004. They averaged 7.8 HiP/G from 1996 to 2018, which was the last time we saw an increase, at least for the early part of the season. This month's 6.7 extrapolates to something like 4500 fewer hits in play over the course of the season. Even if it's half that bad, that's still like 2000 fewer hits in play compared to the ~20-year average. Getting the BiP/G up by 1.6 doesn't do us much good if only an extra 0.2 of them become hits. That's a .125 batting average on those extra balls in play.
With that said, I think you're comparing a bit of apples and apple trees yourself here. You cannot use the average BABiP for half a century, because BABiP changed significantly and more or less permanently in 1993. Between 1973 (when the leagues made some changes after the severely low-scoring 1972 season) and 1992, the MLB BABiP was between .280 and .287 every year but two. In the 1987 "Rabbit Ball" year, it jumped to .289, and in 1981 it dropped to .279, probably because a bunch of games were missed during the warmer (good hitting) months with the strike. Otherwise it was remarkably consistent.
Then in 1993 it jumped to the unprecedented height of .294 and pretty much stayed at or above that level for most of the next three decades. Excluding shortened seasons (which can have a disproportionate effect on this, apparently, depending on which games get missed) the full season BABiP had been between .293 and .303 every year from 1996 to 2019. I don't think we can count 2020 because it was such an anomaly in so many ways.
So you can't just take an average of two wildly disparate run-scoring environments and say, "See! We're still within a standard deviation!" or whatever. BABiP dropping to ~5-10 points works out to 500-1500 fewer hits over a season, and a lot of those are because of the shift, but I admit that the ball is being messed with, too.
Nitpick: SH maybe, but it has been shown that players are no more likely to hit a fly ball with a runner in third and <2 out. That is, they aren’t intentionally giving themselves up.
I’ll look at the blog post when I’m on a computer rather than my phone.
I hit enter too soon. There was more to that. The blog post is just a graph, but it'll show up better on a laptop screen than your phone.
I've been watching roughly the same amount of time. The first game I remember going to was during the interval when the Yankees were playing at Shea Stadium while their home in the Bronx was being remodeled so sometime in 1974-75. I too agree that ~9 runs per game is the sweet spot. I'd only note that, from my aesthetic POV, how that sweet spot is achieved is probably as relevant as how many runs are scored. Even given similar levels of runs scored per game, the Whitey Herzog Royals offense of line drives and aggressive running was more fun for me than the Earl Weaver Orioles offense of walks and three run homers.
I prefer that style as well.
Thanks for the monthly data!
I’m not yet convinced that it is the humidor as opposed to the ball itself. Storage vs construction. If the former, we get need to dial back moisture fraction by fraction to fine tune. If the latter, control doesn’t seem to be as precise. (See eg 1987 and 2019)
1. This year is a tiny sample size of early season (pre warming weather) and without normal spring training.
2. You mention the composition of the balls. That, and not shifting, has changed dramatically this year and last compared to the prior seasons. New balls began to be used last year - unfortunately mixed with old - and are completely in play this season. Shifting has been gradually getting more common for a while going back a decade or so.
3. I'm not sure of your source. (Not criticizing, just not sure.) Using BB-Ref and reversing this for DER simply because it was the first one I found instead of BABIP which, within rounding errors, is simply the converse.
2022 .703
2021 .695
2020 .693
2019 .688
2018 .691
2017 .688
2016 .688
2015 .689
2014 .690
2013 .692
2012 .691
2007 .686
2002 .695
1997 .686
1992 .703
1987 .697
1982 .702
1977 .698
1972 .713
Unless one puts a ton of weight on the first ~15 games of the year with cold weather and an unusual Spring Training, that looks like a flat line to me. Where did shifting start and how does one pick it out from this collection of DER?
I would not call ~10% of the season a tiny sample size. For an individual, sure. But in the aggregate, we're still talking about well over 150,000 at-bats, which is plenty to make in inference like this. I admit that offense always picks up when the weather warms, but not to this extent, I don't think.
I used B-R for my data. I would imagine that even if the offense picks up a bit, we will still end up with the lowest BABIP this season in ~30 years, and considering that last year and the year before were also the lowest since 1992, well, that is a trend. Correlation does not necessarily IMPLY causality, but it can suggest it.
I don't have a feel for DER or know how significant those perturbations are, but we all have a pretty good feel for batting average. When you're talking about ~100,000 events, a 9-point difference in batting average works out to about 9,000 fewer hits. (My math was wrong on my first post, which I have since edited.)
I prefer DER over BABIP for this because it includes reaching on errors. If the shift takes away a hit but the batter reaches because the fielder dropped it, that shouldn’t be counted as a success for the shift.
Even assuming this year’s sample is representative of the whole, there would be 1.5 more batters reaching per 100 balls in play than 2019. That is something like 1 for every 3-4 games, an insignificant amount imho.
Arguing definitions is of course silly. But fwiw I think 10% is a tiny sample not because of the raw number but because it is not drawn randomly from the whole. Instead it is 100% of the April games. Extrapolation from that is dangerous.
Joe you cite multiple hard data pointing to the dearth of hitting and then talk about the shift in subjective even poetic terms.
The shift is killing / has killed the offensive game. The stats from the minors when they enforce the infielders on the infield rule show hard line drives to short left and right fall as hits as was the case in baseball for 100 years, not long outs. Most Players aren’t going to bunt their way on. It just seems like capitulation from the get go.
Also, the ball definitely isn’t carrying as expected. Have seen many balls already that look like no doubters die on the track. It’s not just the April cold.
Like where pitcher, batter, and fielder all look amazed that the ball ended up in a glove and not row 10.
The shift is only as effective as it is due to all of the surrounding issues with pitching, because even with seven players on one side it still makes more strategic and tactical sense for most players to pull a ball for a homer. Start working on the pitchers, and you'll see hitters start working against the shift on their own.
Since the beginning of the game, a line drive hit over the infielders head well I front of the outfielder was a guaranteed hit. Every time. Solid contact meant a hit.
Shift has taken that away and so player has to swing for fence to go over the defense.
Just because you can doesn’t mean it should be allowed.
Goaltending is illegal in basketball.
Offsides is illegal in football.
Players playing the three true outcome game and then the launch angle revolution preceded the rise of the shift. The shift is a reaction to the three true outcome game, pulling everything and swinging for the fences with two strikes and not the other way around.
Before 2009, there had only been one year when K's and home runs accounted for over 20% of PAs. It has been over 20% every year starting in 2009, pretty steadily rising from there, reaching a high of 26.59% in 2019.
When this started, the shift was an oddity at best, something the Rays did every once in a while, and didn't really become used league wide until about 2016.
And I’m that Derek Jeter fan whose book you signed and so reluctantly inscribed.😉
Joe, it’s bandbox, not sandbox. The term bandbox comes from the small box used to store old fashioned shirt collars. See, even you can learn something new about baseball!