Technically, Baseball Really Is More Boring
Pace. The word occupying the most of Rob Manfred’s mind is pace. Or maybe it’s action. They could be tied. Manfred wants more action; he wants a quicker pace. In short, he wants more things to happen. Games continue to grow longer and longer, and that means dead time. Strikeouts continue to rise, and that means fewer balls in play. Home runs continue to rise, and that means…I don’t know, reducing the thrill of the average home run. If I had to distill everything down, I’d say that baseball is most concerned with staying interesting. It doesn’t want to be the boring sport, and it doesn’t want to become the more boring sport.
Personally, I’m not bored by baseball. I know you’re not, either, because this place is selective for superfans. Even if you wouldn’t ordinarily think of yourself as being a baseball superfan, you almost certainly qualify. Think about the website you’re reading right now. We all still find pleasure in the game, and I’m not finding any less than I ever have. If anything, my own level of interest deepens by the day.
But, wouldn’t you know it, but you could say baseball really is more boring than it’s been in a while. Not so much in a way that you’d notice. It’s subtle, as many trends are. But I have numbers to make a case. While I don’t know what to make of them, I’d still like to share them, while I have your attention.
I’m a big fan of win expectancy and related metrics, not for analytical purposes, but because I think win expectancy does a good job of capturing how a game or a given situation feels. Fan emotions track pretty strongly with win expectancy over time, and so it has its uses. One of the numbers that comes out of win expectancy is leverage index, which you can read more about here. The shortest possible explanation is, the higher the leverage index, the more tense the situation. If you have two games, and one of them has a bunch of high-leverage spots while the other doesn’t, you’ll come away from the first one feeling different. You’ll be more drained. Or, you’ll be more ecstatic, or more dejected. You’ll be more something.
Win-expectancy statistics exist going back to 1974. That’s a strange-looking starting point, but it does allow us to cover several decades. So why don’t we consider all that time? On a per-plate-appearance basis, here is how the league-average leverage index has moved:
The peak occurs in 1976, with an average leverage index of 1.05. The lowest point belongs to the 2017 season in progress, with an average leverage index of 0.96. It doesn’t seem like very much, but given how little the line moves around, this does pass for dramatic. For the sake of reference, in 1976, about 10% of all games went to extra innings, and about 34% of all games were decided by one run. In 2017, about 8% of all games have gone to extra innings, and about 23% of all games have been decided by one run. That’s not everything, but those are meaningful indicators.
Relatedly, we can look at the percentage of plate appearances that have come in high- or low-leverage situations. I used the splits as determined at Baseball Reference, so I could search all the way back.
On average, over the whole sample, 20.1% of plate appearances have been in high-leverage spots, and 42.4% of plate appearances have been in low-leverage spots. This year currently sits at 18.4% and 44.3%, respectively. The former is the lowest rate in the whole window. The latter is the highest rate in the whole window. Again, these swings look small, and we’re talking about a couple of plate appearances per 100, but any shift here is notable, given how stable these numbers have frequently been.
Now, in any single year, by any measure, you’re going to have a spread. Which teams in this season alone have played with the highest and lowest average leverage? Here’s the answer to that, after combining numbers for hitters and pitchers:
By this measure, the Cardinals and Brewers have played under the most stress. Remember that leverage index doesn’t have anything to do with a team’s place in the standings — it’s worried about the outcome of just the one game. Like, the Phillies are over there toward the left, not because their games have mattered, but because their games have been fairly close. Now, over on the right side, you see the Reds, followed by the White Sox, Rockies, and Twins. Those teams have played with the least average stress. As a matter of fact, since 1974, we’ve got more than 1,200 individual team-seasons. The 2017 Reds have the lowest average leverage out of anyone. The four lowest teams here are all included in the bottom five. There are four teams from 2017, and one team from 2002. A dropping tide lowers all boats.
As a different but related measure, in here you see the team-by-team picture in terms of rate of high-leverage plate appearances:
The general order is the same, with some minor shifts. For hitters and pitchers combined, 20.7% of all plate appearances for the Cardinals and the Phillies have been high leverage. The Indians and Rockies bring up the rear, at 15.8%. That’s a difference of a few plate appearances per game. The White Sox are third-lowest, at 16.3%. We can look at the same giant sample of team-seasons since 1974. Out of the bottom five teams in high-leverage rate, there are currently three teams from this year. By the previous measure, the Reds have played the most boring baseball. By this measure, the Indians have played the most boring baseball. Some boring baseball taking place in Ohio, relatively speaking.
Of course, leverage isn’t everything. And even teams like the Reds and Indians have found themselves in some awfully stressful spots. And then, there are other considerations, like the context of the pennant race, or matters of player development. There are so many different angles to help to keep baseball feeling fresh, and it’s not like anything in here signifies the death of the sport. I don’t even know how to explain what we’re seeing. I’m at a loss, not that I’ve thought all that much about it. The only conclusion is that the league’s average leverage is at its lowest point on record. How you choose to feel or interpret that is entirely personal. But it’s definitely a statistical thing that has happened.
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.
Does leverage index correctly take into account the uptick in HRs/TTO baseball? Is it possible that a game down 2, man on first is higher leverage than it used to be since higher chance of the team tying it on one swing?
I came to the comments to make this same point. With HR/rate so far outside historical norms that created the leverage index (to my knowledge), you have more potential. But then again, I still think a rally is far more exciting anyway which includes lots of traffic on the basepaths, etc. Just waiting for a bomb is not nearly as great feeling
Similarly, is the defensive component of WAR adjusted downward to account for fewer balls in play?
Sometimes the changing game is painted as positive and sometimes it is negative. It is just a narrative. Baseball is still baseball.
Really? I think it’s totally acceptable to ask if the game is moving in a healthy direction.
This has always been. And baseball hasn’t just been baseball. The game has been repeatedly tweaked to keep it fresh.
The late 60s pitcher dominance was boring so changes were made.
2-1
The pace IS bad. Too many stoppages, too many strikeouts.
This is an unprecedented new direction. TTO baseball can be both dull AND long. A 2-1 game once had the benefit of quick pace and brisk flow. Now the strikeouts mean hundreds of pitches by the 6th. More pitching changes, commercials.
So you end with a game that’s both slow and fairly dull.
It’s a problem
So, that’s your narrative. I am OK with your take. I enjoy watching baseball for what that is worth to you. My favorite era of baseball was the steroid era. It is OK to have opinions on things and write stories about them. It sounds like you might want to consider not watching baseball?
What a ridiculous thing to say. I am in the same boat as bpd. I enjoy watching baseball, but I used to enjoy it a lot more when the game had a faster pace and better flow. Now that the game is changing, rather than ask how we can make it more enjoyable again, your suggestion is to just give up and stop watching?
I feel like after looking into this we need to ask: WHY is LI lowering? Is it better bullpens? Less parity? If a team falls behind early, they don’t fight as hard to come back? My intuition tells me it has to do with manager’s pulling their starters earlier to preserve leads, avoiding the 3rd time through the order penalty, when in the past they would let the starter gut it out until the lead had whittled away.
By the way, the Phillies appear on this list because they have blown so many late leads with their crummy bullpen, it has created lots of high LI PAs.
As I was reading, I was thinking it’s almost certainly somehow connected to bullpen usage. Teams don’t come back as often because they’re facing stronger (and fresher) pitchers. So if you envision a run environment where games are pretty much settled after the 7th inning (as an example), you’d expect to see a much lower LI. There’s just fewer chances for high-stress plate appearances.
Maybe it’s just confirmation bias, but I feel like there’s been more blowouts this year than in recent years.
By that metric, the Nats should be in the same vicinity, rather than roughly in the middle.
This looks like noise. The y axis exaggerates this. Not that p-value is the be all end all, but what’s the chance this occurred randomly? Saying that “it’s definitely a statistical thing that has happened” without a proper null hypothesis and a simple check for p-value to at least start a discussion is not cool. That statistical thing may be noise, which happens more often than any of us like to admit, because that’s a boring answer.
it’s a starting point for a better discussion, i agree. there does appear to be a slight trend over time, just by eyeballing the annual variance, but it isn’t dramatic. a proper statistical analysis of the variance would be welcome.
As someone that analyzes a lot of temporal data, this looks like the exact opposite: a very strong long-term temporal signal with some annual noise like due to other factors (probably mostly variation in parity in a given year).
In any case, LI should go down as run scoring increases. With more runs there is more potential for variation in scores throughout the game and more potential for blowouts and lots of low leverage situations, all low scoring games are close by default.
Since pLI only exists going back to 1974, our true population variance is not really known so lets use a t-test. From 1974-2016, we have a population mean of 1.018 pLI. Our 2017 sample mean is 0.96 pLI and a standard deviation of 0.05 between the 30 teams. To calculate the t-statistic we use the following formula:
t = (.96-1.018) / (.05/sqrt(30)) = approx -6.04
p-value = .0000014
The critical value for this one-tailed hypothesis test we have to beat when our alpha = .05(95% confidence interval) and we have 29 degrees of freedom is -1.699. We can see that our t-statistic easily falls in our rejection region so we can reasonably accept the alternative hypothesis that this years pLI is statistically different from years past.
I understood a couple words here.
You’d think the supposed parity in baseball these days would mean more closely-fought games.
OF course, this analysis, as I understand it, is specific only to the outcome of the single game and doesn’t look at the context in which the game is being played. We might be seeing lower intra-game leverage indexes, but, due to league parity and the extra wild cards, the games themselves might have higher leverage indexes, in the context of the standings?
That would make baseball more interesting — at least for people who care about a team, if not for casual viewers.
Surprised to see the Astros not higher in leverage. Though, I’m guessing not having trust in backup starters who struggle third time through the order with a lead is more nerve wrecking than it actually should be.
It seems like league-wide swings in Leverage Index could be attributed to three things: run scoring environment (more scoring=more blowouts=lower avg LI), parity (better-matched teams=closer games=higher LI), and runners on base in late innings. I’m guessing the primary driver of the decrease in LI this year is that there are fewer baserunners in the 8th and 9th, both because relievers are more dominant than ever and because hitters are striking out more, hitting more homers, and hitting fewer singles.
Paradoxically, LI is both agnostic of run scoring environment (LI values assigned to certain situations are static year-over-year, right?) and driven by it. As Roy_Hallayay notes below, the higher likelihood of a 9th-inning home run may not be reflected in the comparative LI stats above, but a 1-run lead with 2 outs and nobody on in the 9th has to feel more tense to a fan in 2017 than it did in, say, 2012, right?
So while we’re on this discussion, I’ve gone back and forth on the idea of a pitch clock and I think I’ve decided on yes, baseball does need a pitch clock if it wants to keep younger people interested and it needs to be implemented quickly, preferably by next year. I’m thinking the clock starts when the pitcher gets the ball from the catcher/ump and has 10 seconds until the delivery with nobody on base and 15 seconds with runners on base. I’ve timed it while watching games and it seems pretty reasonable. Clocks would reset on pickoff attempts only, and only the catcher/infielders are allowed to talk with the pitcher, the pitching coach is not allowed to come out and talk about whatever useless stuff they usually tell the pitcher. I think this would do a lot of good and am curious as to people’s thoughts
I don’t much like the superstition/routines between pitches/bats – but I totally disagree with a clock. Baseball should not be timed. There’s plenty of ways to market to different generations – and plenty of ways to speed up the game without changing the core of the game itself.
It’s the guys with the superstitions/routines who have changed the core of the game. Pitchers used to get the ball back and throw it again. Now they just stand there and stare in forever.
Pitchers didn’t used to either have to rub in so many new balls (this really sounds eewwww) or get yanked so frequently (omg – another ewww).
Used to be only short relievers who seemingly took forever between pitches. Now – we’ve got a game full of them – and starters seem to me to slow down once they start looking over their shoulder.
Pitch clock doesn’t address either of those. And it still destroys the basic notion that baseball is not a timed sport. It is based on innings/outs – NOT time
That’s not a basic notion of baseball at all. Ballgames are getting longer, and we ought to do something about it. That’s a basic notion.
Before night baseball, there was a soft time cap.
You think young fans are turning away from the game because of a lacking pitch clock? I don’t. You aren’t going to win fans with a pitch clock. Kids live in a world where they have all kinds of bite-sized options at their fingertips – baseball can’t compete with that, because that’s not what it is.
If kids were growing up playing baseball, going to games, watching games with dad on the couch, collecting baseball cards or whatever kids used to do when they were growing up then they would be fans. Its just not the world they live in.
If MLB changes enough, it may manage to alienate its real fan-base. I think that is a much bigger threat than the realistic potential to suck in new fans. Band-aids don’t fix problems, they just cover them up temporarily.
As a Dad who can’t get his kids to sit on the couch and watch baseball with him, a lot of the problem is video games and You Tube. My kids prefer entertainment that is action packed and can be condensed into a few minutes. They will walk into a room and watch the last few minutes of an NFL game because it is clear that it will be over shortly and they can get back to their video games. It’s passive, but it’s short like a You Tube video. They don’t have the interest in watching something passively for an extended period of time. They don’t really like movies, either. It’s a generational thing.
Edit: I want to add a thought that just occurred to me. Maybe the problem is the proliferation of first-person video games. My kids love playing sports, but they don’t love watching them. When I was a kid, I identified with my favorite players, and I loved video games, too. However, our video games involved us playing as someone else rather than being first person. It seems to be that kids today prefer to experience life in the first person, which makes watching someone else play sports boring for them.
That is what I was trying to say. Baseball is anything but instant gratification. I don’t know about the first/third person thing – I think the instant gratification piece is pretty logical and hard to dispute. It is just a fact that most popular video games are first-person-shooters, I don’t know that they are choosing them. They sit and watch other people play games on YouTube which is anything but a first-person experience. Kids haven’t changed, but the world around them has. There are so many extremely easy, gratifying experiences are at their fingertips. You and I would have chosen them too…
My oldest kid is barely old enough to sit and watch the game from the couch. So far it is not going great… When I get this problem solved I will send you a PM!
I agree with everything that you said. I think that we might be talking past each other. I agree that we would have chosen them, too. I wasn’t blaming the kids so much as saying that they are at least partially a product of the world that they live in just like everyone else was.
I will always maintain that the camera view of the game is a huge deterrent in capturing the true game. I would estimate 70% of a televised game is the classic view catcher-batter-pitcher. Why not more split screen, first base-pitcher-catcher-batter, overhead view, …. There are so many other things to see. Hardcore fans know there are numerous game within the game going on but all we see is the traditional shot. When I go to a game I constantly pan the field, positioning of inf/of, signs given, runner leads, look into dugouts, … There should so much more to see.
Agreed. So much of baseball is about the little intricacies. 1/100th of a second could mean a double or a whiff, 1/10th of an inch between a lazy fly ball and a bomb, one step to the right by the shortstop between a great diving stop or a seeing-eye single.
When I go to games, I always like to pick seats in the upper decks because it gives me a better view of all the subtle little changes between at bats–not only the big infield shifts, but outfielders taking a few steps to the left for certain batters, players running to backup throws, the chatter between teammates in the field.
So much of the game is missed on TV because it’s either the pitcher’s back or random shots throughout the crowd.
Agreed. Personally, I would love to see a pov shot from the umpire but that could open up another can of worms with arguing calls, etc.. umpire union would probably never have that
They had a camera mounted on the umpire for a few games of the College World Series which I saw. It made me sick personally every time they cut to it; the umpire’s head wouldn’t stop bobbing up and down.
Perhaps saber is itself at fault for the game becoming more boring. Baseball has always been the sport with more info/data than every other. But think about what made that interesting. It was that that created conversations with random strangers at the park. They could second guess the team and invoke history and everything else (call hit-n-run – nono X can’t control that contact and Y’s too slow – man wish Z was still playing this is exactly the situation he thrived in – ump you’re a bum that’s clearly a strike get some glasses).
There is always a risk with information. That it reinforces technocratic attitudes – reliance on a skilled insider elite, a deus ex machina, and passivity from the outside. But this is still a sport and no fan ever has gone to a game to watch the owner/GM manage their WAR/roster$ – or to watch the manager manage the bullpen/lineup. Gone really are the skillsets that many fans do watch for – digging out of a deep hole, scrounging and scrambling, dealing with defeat as well as victory – and above all the fans own dreams of You know I coulda done this better.
With advanced metrics comes some strange stuff, my theory is baseball is adopting the “blow-out” basketball strategy. More recently in basketball when a team is getting blown out it pulls their starters and “key-players” earlier, to rest them for the next game. They basically give up. Even the back-ups aren’t giving it 100%. I think they have done their studies and the very small chance a team is able to “come back” from a “blow-out”(in baseball probably 6-7 runs) as the game grows longer is not worth depleting your entire bullpen, or even worse using the next day’s stater to pick up innings in a long one in the slim chance you do tie it. Just my opinion from watching games this year.
Haven’t baseball teams always done this? Isn’t that more or less the entire point of the “long man” in the bullpen, to save everybody else from having to pitch in blowouts?
My perception without any data to back it up (take that for what it’s worth), is that teams use their bullpens when losing differently than they used to. Yes, they used to have a long man in case of a blowout, but they might use their second or third best pitcher if they were behind in a close game. That makes it more likely that they might come back. Now, the second best pitcher is the eighth inning set-up man, and the third best is the seventh inning set-up man. Teams behind in a close game will often bring in their fourth or fifth best reliever, which means that 2-run deficits often become 5-run deficits, and we never have the high leverage situations arise.
Yes exactly, IMO it just seems that teams are quicker to throw out the “long man.”
I don’t watch basketball, but I find it hard to believe back-ups wouldn’t give 100%. Aren’t they playing for their next contract?
right, like you said you don’t watch basketball. I seen blow-outs turn into alley-oops contests.
They do play hard, in my opinion. But with the salary cap changing the way it has recently, most good teams have very bad backups (Golden State excused). So the backups just suck and probably seem to not be trying.
Is there any correlation between team wRC+ and average LI? Or team bullpen ERA and LI? Or any other team indices that correlate or can predict a low or high stress team?
I strongly disagree with your assertion that this a community of baseball superfans. If anything, this is the primary place where people would argue that the game is boring – there are even stats to prove it! There are baseball fans here, but there are also a bunch of people that are fans of statistical analysis (of whatever) and people who just enjoy a good progressive narrative. Baseball simply produces so much measurable data and so many finite outcomes that is attractive to those with an interest in statistical analysis, not necessarily baseball fans. Most baseball fans are not particularly drawn to that. I am, but I am also weird. I get the impression from the comments and up/down votes that a large chunk of this site’s patrons aren’t a fan of of the game. Just look at how many people are talking about whats wrong with the game or how it needs to change… I wouldn’t call them fans of baseball – they are fans of something else.
I’d add that I spend more time reading about baseball these days than watching it, even though my team is in a pennant race, and Miguel Sano and Jose Berrios are more fun to watch than any Twins players in a long time. I enjoy the meta-experience of baseball, but I’ve grown to find watching the actual games rather tedious, even if I’m doing something else at the same time. I don’t have any real desire to go to games anymore. I don’t know if that’s baseball changing or me getting older or something else, but I do think Fangraphs is serving me in it.
It may be we’re all becoming Billy Beane after all.
Are you sure that you don’t just prefer the summary as opposed to an actual increasing dislike for the on-field game? The FanGraphs experience is an option that didn’t used to be there. I get that this is a damn fine option for many people and I never really thought of it like that before. For me, I am always far more interested in the real game than the after-the-fact analysis – I like to analyze it for myself. Analysis is a way to summarize the actual games and events that happen.
To me, baseball fan means you enjoy watching the game. This is just a made-up label and everyone is entitled to their own take on it. I hate the idea of proclaiming myself to be a higher level fan than anyone else, but that is how I see it. The only thing about the modern game that grinds my gears is the 20 second delay after every stolen base or play at home plate where they decide whether to review the play or not – then the 2-3 minute review process. I don’t think the game-flow has changed much outside of that.
I’ve been around internet Sabermetrics pretty much since it became a thing. The experience has gotten a little better (though Fangraphs peaked in the NotGraphs era) but it’s not really a ton different now than it was when Baseball Primer was the primary forum. We understand baseball a lot better, and maybe lacking the thrill of figuring the big stuff out is part of it.
I think it’s likely either something about me personally – and there are plenty of possibilities there, my life now is not very similar to what it was in 2002 – or about baseball, or both. I’m not highly invested in knowing the answer.
Our changing lives are a huge part of our changing perceptions. I work all day then come home to two toddlers – watching a baseball game sounds fine to me!
I’m a huge fan of baseball, but I need less downtime in the game. I can’t spend 3 1/2 hours watching a game multiple times per week.
There is so much wasted time that could be lessened without truly harming the game: between pitches, between innings, mound visits, reliever warmups, replays, etc.
If we could get the game down to 2 1/2 hours on average (meaning some games would be less), I’d watch a lot more games.
This is going to happen once ratings fall in the next few decades as young people don’t replace older ones. I just worry that it will be too late.
I don’t think there is anything to fear. So what if there are less fans? The only one who will suffer are the people raking in millions of dollars at the fans expense. They won’t quit playing the game, selling tickets or televising the games. If it wasn’t such a cash grab, it would all be better!
Maybe we need a Predicted Pace metric to make it easier to identify games which are likely to go quickly.
Great article, but this “superfan” IS bored by today’s game and pace of play. I love reading about baseball, but can’t bear to watch more than an inning or two of batting-glove adjustments before I move on to other things.
Ditto. I’ve loved baseball for years, been into sabermetrics since the 1984 Baseball Abstract, and I find 2010s baseball deeply tedious.
looking at trends i’m sure that K’s an HR’s aren’t at all helpful with sustaining leverage situations. And that’s what we’ve got right now.
Ive spent most of my life as a pitcher or pitching coach. So, when I watch games Im focusing on (1) catcher framing and (2) pitch sequencing, and things of that nature. Its interesting for me, but for many others watching baseball is like watching golf, nascar, etc.
As a cardinals fan, id bet most of our leverage index is due to our bullpen giving up leads, so thats not a positive viewing experience … nor is watching our offense.
To me, baseball was best in the 80s, when there were many different styles of play, and there was more action.
I love sabermetrics, and while the strategy application of tto is “smart”, its also boring for the masses.
As for kids, I’ll explain like this … my 4th grade son had a buddy spend the night tonight. Theyve done 4 different things in 25 minutes. They both play sports year round (as do my 2 other kids), but they dont watch them.
We go to some cardinal games and low level minor league games in peoria and joliet, and they enjoy those as events. watch the games on tv … no way.
Seriously, who has 3-4 hours a night to watch baseball? Retired dudes, thats who.
hard disagree.
leverage is not a context neutral measure of excitement.
if, for example, teams tend to score in bunches because of increased homer rate, then you’d expect lower leverage but also higher excitement, as fans think a comeback is always around the corner.
plus, homers are cool.
Imagine for a minute that everything in mlb stayed the same, but walk rates jumped to 20%. While perhaps the walks themselves are too terrible exciting, there would be a huge increase in exciting situations in my opinion. I personally think scoring runs is what I tune in for, and this environment is awesome as far as I’m concerned
More HRs means less emphasis in RISP situations because more of the scoring goes on without runners advancing around the bases. Most HRs are solo. So that means that a higher percentage of the scoring is without much leverage. But hey, the 7th game of the ’16 WS was enough leverage for several years.
It’s worth noting the change in leverage of the current style of play. But the rest of this is way off base.
To quantify fan interest is hard enough. To show a link from that to the leverage measures used here is something I’ve not seen attempted.
Even the “leverage” that might affect fan interest is a matter of subjective perception. I think few fans (including me) could even guess their team’s quintile in the average leverage rankings, across a whole season, just from their own experience of games.
And I think those who do find the current style of play more “boring” are far more moved by the loss of variety, rather than the change in leverage.