From J.D. Martinez to Josh Donaldson, hitters throughout the big leagues have been honing a new approach at the plate, hunting for big flies and eschewing worm burners. It’s a change rooted in the latest metrics, which say balls hit in the air tend to be more valuable than grounders — particularly since the home run surge of 2015 started turning a higher percentage of fly balls into home runs than ever. So, over the last two years, batters have adjusted their swings accordingly, sending ever more balls skyward.

The resulting trend toward fly balls has significantly improved a handful of hitters, helping them achieve far better results than when they slapped more grounders. Some observers have even suggested it could be contributing to the surge in home runs. But a closer look at the data shows that, while there is a sweeping transformation underway, it seems to be hurting as many players as it is helping.

A batter can hit more fly balls by changing the angle of his swing. Instead of the slight downward plane recommended by many instructors, more of today’s batters are adopting uppercut swings that drive the ball into the air. And across the league, the effect is palpable.

Over the past three seasons, the ratio of ground balls to fly balls in MLB has dropped from 1.34 grounders per fly in 2015 to 1.25 this year. For individual players, the changes are even more significant. FanGraphs’ Jeff Sullivan documented a historic number of players who have dropped their ground-ball percentage by 5 percent or more since 2015.

Some of those players have benefited greatly from these swing changes. Oakland Athletics first baseman Yonder Alonso nearly halved the number of grounders he’s hitting so far this year, and he also boasts a personal-best 178 weighted Runs Created plus, one of the best marks in the league. There are similar anecdotes for Martinez, Donaldson, Nationals All-Star second baseman Daniel Murphy and others.

So there is definitely a fly-ball revolution underway in baseball. But that revolution is not without its discontents. Cincinnati Reds first baseman Joey Votto recently disparaged the trend towards fly-ball hitting in an interview with The Cincinnati Enquirer. “I see it with a lot of guys. Everyone tells the good stories, but there’s a lot of s—ty stories of guys who are wasting their time trying things,” Votto said, as quoted in the Enquirer.

Votto is right; being a more productive hitter really isn’t as simple as “elevate to celebrate.” Over the last three years, just as many hitters have suffered by increasing their fly-ball rate as have benefited. Here’s a chart showing each hitter’s change in fly-ball rate from the previous year, in comparison with his change in weighted On-Base Average (wOBA).

Among players who increased their fly-ball rate, it was almost exactly a toss-up as to whether their wOBA would get better or worse. Similarly, players who decreased their fly-ball rate had about a 50/50 split of improving and worsening wOBAs. Overall, the correlation between a batter’s changing fly ball rate and his subsequent change in production is nonexistent. That same lack of correlation holds if you use the more advanced metrics (such as launch angle) tracked by MLB’s StatCast system.

Although there are some fly-ball success stories, plenty of hitters have swung up only to see their wOBA dive down. For every Yonder Alonso there is a 2016 Kiké Hernandez, who spiked his fly-ball rate by 11.7 percentage points, only to watch his wOBA drop by 89 points. Or maybe you’d prefer Cubs outfielder Jason Heyward, the owner of a dreadful wOBA 21 percent worse than the league average in 2016. The Cubs are at the vanguard of the fly-ball revolution, reportedly championing the phrase “there’s no slug on the ground.” Heyward seems to have listened, because he increased his fly-ball rate by almost 10 percentage points after joining the Cubs. But in contrast with the success stories of players such as Alonso and Martinez, the change has had a disastrous effect on Heyward.

So adopting an uppercut swing won’t necessarily make a player great. But it will probably make them hit more home runs. (When players up their rate of fly balls, the consequence is usually more dingers.) The increasing rate of fly balls leaguewide seems to explain some of the explosion in home runs from 2015 to 2016 (although that still leaves the mid-year 2015 increase in home runs a mystery, even setting aside the speculation around — and puzzling evidence for and against — ball juicing).

Home runs are great! But the problem is that fly balls also come with other, less desirable consequences. For example, players who hit more fly balls into the outfield also hit more pop-ups on the infield, which are about as valuable as striking out. Given his aforementioned criticism of fly balls, maybe it’s no coincidence that Joey Votto is also one of baseball’s best at avoiding infield pop-ups — he probably knows the two are related.

Moreover, the conscious effort to adapt an unnatural swing plane could harm a player’s natural hitting motion. Heyward had been a productive hitter earlier in his career with similar fly-ball rates as last season, but his swing mechanics were notably confused a year ago, which resulted in an obvious weak spot against low pitches.

In an interview with CSN Chicago, Cubs hitting coach John Mallee described the work he was doing to improve Heyward for the 2017 season. “He’s trying to mirror the swing that he had then…. It’s not actually making a change; it’s just getting him to who he was,” Mallee said. Bucking the revolution, Heyward has hit significantly fewer fly balls this season, and his production has improved, as well (although he’s still underperforming expectations).

Stories such as Heyward’s show that the fly-ball revolution is not for every hitter. Notably, many of the players who have transformed the most by adopting uppercut swings were underperforming before. Alonso was a below-average hitter last season; Donaldson was a former high draft pick who struggled for years to come into his own. Tinkering with their swing planes might have been the secret to unlocking their full potential. But for players with established mechanics like Heyward, adopting a new philosophy is a riskier proposition. All told, it’s tough to predict whether more fly balls are the missing ingredient for a hitter, or just a harmful distraction.

The Humane Society and Safer Chemicals, Healthy Families seem like two groups that could, conceivably, coexist in adjoining booths at your local Earth Day festival. But these organizations have instead ended up on opposite sides in a debate over how the Environmental Protection Agency should go about regulating thousands of potentially harmful chemicals that Americans come into contact with every day.

This disagreement between the people who would like to help you adopt a puppy and the people who would like to help you avoid hormone-altering deodorant — and conversely, a spot of agreement between the Humane Society and trade associations for the petrochemical industry — goes beyond the details of a specific EPA proposal. Instead, experts say, it reflects a long-standing (and maybe impossible-to-solve) difference of opinion about risk, data collection and the role that animal sacrifice plays in keeping humans safe.

From early 20th century factory workers poisoned by radioactive paint to growing concerns that flame-retardant chemicals used on sofas and pillows could be linked to birth defects and cancer, the history of American industry is rife with cases of workers and consumers learning that a product they thought was safe actually wasn’t. In some instances, the effects of a toxic exposure play out in quick and obvious ways — those factory workers who painted watch dials with glow-in-the-dark paint made from radium began to sicken and die in gruesome ways after just a couple of years on the job. But some health impacts — an increase in lifetime risk of cancer, say — are harder to spot. And the links between those effects and exposure to a specific amount of a specific chemical are harder to prove.

That difficulty is compounded by the fact that, much of the time, we don’t even know what chemicals we’re being exposed to, at what amount, or what risks those chemicals are associated with. The EPA has had the authority to regulate the chemicals used in household products since 1976, but it had no control over chemicals that were already being manufactured and sold before that year: a collection that includes 62,000 substances. Moreover, the EPA isn’t even always aware of what all’s in the stuff you buy. The trade publication Chemical and Engineering News recently wrote about a study in which EPA researchers ground up 100 consumer products, including cereal and baby toys, and found 3,800 chemicals — only 200 of which were chemicals the EPA had expected to turn up.

Last summer, Congress took the first step toward increasing public knowledge about, and regulatory oversight of, toxic chemicals, passing the bipartisan Frank R. Lautenberg Chemical Safety for the 21st Century Act. Among other things, the act made it mandatory for the EPA to evaluate all the chemicals used in consumer products and to apply a risk-based safety standard, rather than one that has to balance a chemical’s risks against its industrial usefulness. It’s such a popular change that even the proposed 2018 EPA budget, which would reduce the agency’s funding by 31 percent and eliminate 50 programs and subprograms, provides increased funding to implement the Lautenberg Act. But putting the plan into action means filling in the holes in our knowledge, which means doing a lot of toxicology research — and this is where groups that might otherwise join hands and sing “Kumbaya” around the peace pole start to disagree.

The EPA published its proposed plan for implementing the new act Jan. 17 in the Federal Register. In public comments submitted to the agency, Safer Chemicals, Healthy Families, an advocacy network that works in education and policy for consumer environmental health, supported what the EPA wants to do. And the Humane Society and other animal-rights organizations didn’t. And that’s because of the tools used by the field of toxicology. This research, aimed at understanding the effects of chemical substances on the human body and where to draw the line between a safe amount and a dangerous exposure, relies on data collected from laboratory experiments. Some of those experiments are “in vitro” — they’re done in a test tube (or, more accurately, in one of dozens of tubes packed together like a honeycomb) using human or animal cells. Others, though, are “in vivo,” conducted in the bodies of living creatures such as rats or zebra fish. Representatives from both the Humane Society and Safer Chemicals, Healthy Families said they want to see the Lautenberg Act implemented. But the Humane Society is concerned that the EPA’s specific plan will increase the number of animals who die on our behalf.

In particular, the Humane Society is worried about an extra step that’s been added to the process. As legislated, the Lautenberg Act was a two-step system meant to help the EPA move quickly through tens of thousands of substances: First, prioritize chemicals based on potential for risk, then evaluate the safety of those designated “high priority.” And the EPA’s proposed plan retains this framework. Once a chemical enters the prioritization phase, the EPA has nine to 12 months to figure out whether it should be low priority or high priority. Low-priority chemicals are effectively set aside, as far as the evaluation process is concerned, and the industry can keep using them as before. Meanwhile, high-priority chemicals enter the evaluation phase, where the EPA has three years to figure out what risk the substance poses to human health and how it should be regulated. But the EPA has added a step: The pre-prioritization phase, a time for figuring out what data exists for a given chemical and whether the agency has enough information to truly know how the chemical should be prioritized. And, unlike the other steps, there’s no ticking clock for pre-prioritization. It could last a month. It could last years. “It creates this new sort of black box where they’ll ask for all the info they think they’ll need for the entire risk-assessment process,” said Kate Willett, director of regulatory toxicology, risk assessment and alternatives for the Humane Society. Willett thinks there’s a lack of transparency here that could result in companies feeling pressured to conduct new animal research for chemicals that might end up classified as low priority anyway, a scenario that, she said, would make those animals’ deaths a waste.

The EPA did not respond to a request for comment on this added phase, but Holly Davies, senior toxicologist in the Washington State Department of Ecology and a member of the EPA’s scientific advisory board on chemical safety testing, said a pre-prioritization step made sense to her because it helps the agency ensure that it doesn’t end up with more chemicals in the prioritization and evaluation phases than it can process in the allotted time frame. She pointed out that there’s no minimum amount of data that a company has to provide as part of pre-prioritization. And, she said, even if a company chose to do more tests at that phase, the information would be valuable no matter what priority the chemical was later assigned. “For me, to do a test in animals and find there is no effect, it’s not wasted,” she said. “Because now you know there is no effect. If testing puts something in the low-priority category, then that’s useful.”

Useful? Or a waste? What’s really being debated here isn’t just the presence or absence of a this phase in the EPA’s assessment of chemicals. It’s something more fundamental: If we want to keep people safe, is it necessary to experiment on animals?

That is a far-reaching question without easy answers. “It’s been going on for years,” said Andy Igrejas, director of Safer Chemicals, Healthy Families. He believes animal testing is a crucial part of protecting humans.

But concerns about animal testing — what animals it’s being done on and how well it works — aren’t confined to animal-rights groups. In 2015, the National Institutes of Health stopped funding research on chimpanzees, the animal most closely related to humans, after a National Academies report found that the research being done on them was unnecessary and unethical. Other research has found considerable problems associated with translating the results of animal studies to humans, both because of differences in biology and because of research problems created by the humans themselves. A 2009 paper examining why animal models weren’t terribly predictive of human responses to chemical exposure noted that animal studies were significantly less likely than human studies to be systematically reviewed, a process in which scientists compile and analyze multiple published studies on the same topic to see if they are pointing toward a common outcome. The same paper found that it was more common for animal studies not to follow basic scientific methodological rules, like blinding to prevent the researchers from knowing which animals have been exposed and which haven’t.

That convergence of practical, ethical and biological problems has created a movement within toxicology to get more research done, faster, and with less reliance on animals. The National Toxicology Program, a branch of the National Institutes of Health, even has a lab that’s dedicated to developing and testing alternative toxicological methods. Primarily, that means using new high-speed testing systems that allow researchers to run the same experiment quickly on hundreds of cells at once and then using mathematical modeling to take the results of many kinds of these studies — research on human cells, research on purified human hormone receptors, research on fish embryos — and combine them into a single prediction of a chemical’s impact on a full-scale human.

Davies still thinks traditional animal research is important, though. Almost every method we have of studying the human biological impacts of chemical exposure is flawed, she said. Nothing, not even a human, is a perfect proxy for what will happen to humans, broadly, when they come into contact with a given chemical. Instead, she said, animal testing is part of a suite of testing strategies that, taken together, can help us understand potential risks. “It’s building up a picture of what we know,” Davies told me. “You have data that’s in vitro, data in non-mammals, data in mice. And they all contribute to the story in a way that makes sense.”

From Willett’s perspective, groups primarily focused on animal rights and groups primarily focused on environmental health agree on more of this than they disagree on. But faced with imperfect methodology, deeply important questions for human health and big philosophical and ethical implications, different people are likely to come to diffferent conclusions about what to prioritize and how to proceed. That’s part of what the EPA is considering now, as it reviews public comments and puts together a finalized plan for the future of chemical exposure research.

And those differences will continue to have big implications for the process of turning science into policy, at the EPA and elsewhere. “If you aren’t increasing the information base, then you aren’t solving one of the main problems,” Igrejas said about the EPA’s toxicology testing. Willett, meanwhile, questioned whether that was true. “From our standpoint, based on experience over the past 30 years or so, simply adding more information to the mix doesn’t necessarily make you make better decisions.”

You’re reading Significant Digits, a daily digest of the numbers tucked inside the news.

10 percent

Ford Motor Co. plans to cut 10 percent of its global workforce, with a focus on salaried employees. The company will cut $3 billion this year. [The Wall Street Journal]

78 percent

An NBC-Wall Street Journal poll found that 78 percent of Americans would prefer that an investigation into ties between Russia and President Trump’s 2016 presidential campaign be led by an independent prosecutor or commission. [CNN]

101 years, 38 days

Age of Bryson William Verdun Hayes, who on Sunday became the oldest person to skydive. [The Guardian]

10,000 boxes

General Mills is setting up a sweepstakes where the prize is one of 10,000 boxes of marshmallow-only Lucky Charms cereal. When you think about it, this is the closest we’re gonna get to what’s essentially an extremely shabby version of “Charlie and the Chocolate Factory,” and based on the already ridiculous sugar content of standard Charms, it will almost certainly result in about as many human casualties. Seriously, you can just buy the marshmallows in bulk on the internet if you’re cool with your UPS guy losing all remaining respect for you. [StarTribune]

$7 million

Organizers of the notorious Fyre Festival, a failed attempt at a Bahamas music festival promoted by “influencers,” borrowed up to $7 million in a mad last-minute attempt to make good on the festival’s lofty promises and avoid becoming an international laughing stock, new documents show. It did not go great. [Bloomberg]

$9.7 billion

Twenty-nine states threaten suspension of a person’s driver’s license as a way to make them pay traffic fines. Californians owed their state $9.7 billion in traffic debt last year, and it turns out that people who can’t afford to pay are not magically transformed into people who can afford to pay the minute you take away their license to drive. [The Atlantic]

If you see a significant digit in the wild, send it to @WaltHickey.

There have been lots of questions, especially among liberals, about when congressional Republicans might turn on President Trump, particularly in the wake of his controversial firing of FBI Director James Comey and the reports late Monday that he compromised classified information. The assumption behind these questions is that at a certain point, something so outrageous will be revealed about Trump that the resulting scandal will … er … trump party loyalty.

But, at least historically speaking, even the biggest scandals don’t wash away partisanship.

We went back and looked at key congressional votes during three relatively recent periods in which a president was accused of wrongdoing: Watergate (Richard Nixon), Iran-contra (Ronald Reagan) and the Monica Lewinsky scandal (Bill Clinton). Two trends stick out. First, partisanship still matters. And in a big way. Second, when defections do come, they’re more likely to come from the centrist wing of a party.

Even as Nixon aides resigned and the Watergate controversy grew around the president in 1973, many congressional Republicans were arguing that the investigations of the president were overly aggressive. Two future GOP presidents, George H.W. Bush (then chairman of the Republican National Committee) and Reagan (then governor of California), called Nixon and assured him that he could get through the scandal. Things escalated in October 1973 when Nixon ordered the firing of the special prosecutor investigating his administration, leading both the attorney general and deputy attorney general to resign, in what is now known as the “Saturday Night Massacre.”

But Nixon retained party loyalists in Congress even after that dramatic move to stop those investigating him. The House Judiciary Committee held a series of votes about recommending Nixon’s impeachment in July 1974. All 21 committee Democrats, and six committee Republicans, voted for the first article of impeachment, which essentially accused Nixon of obstructing the investigation of the Watergate break-in. The other 11 Republicans voted against that article. There were three articles of impeachment against Nixon. Nineteen Democrats voted for all three articles of impeachment. Just one Republican did. A majority of the Republicans on the committee, 10 of the 17, voted against all three articles.

Nixon resigned before the full House voted on impeachment, partly because Republican officials finally began abandoning him. But Nixon didn’t step down until Aug. 8, 1974, more than two years after the Watergate break-in.

The 1998 House votes impeaching Clinton — who was essentially accused of lying and trying to block the investigation into his his affair with White House intern Lewinsky — were also split almost entirely along partisan lines, with the Republican majority ensuring that two of the four articles of impeachment were adopted. Clinton was impeached on a perjury charge in a 228-206 vote; 223 Republicans and five Democrats voted “yes,” while 200 Democrats, five Republicans and one independent voted against it. On the other charge to pass, obstruction of justice, 216 Republicans and five Democrats voted “yes,” while 199 Democrats, 12 Republicans and one independent voted “no.”

In the Senate, in turn, party loyalty saved Clinton from being removed from office. Because removal requires a two-thirds majority and Democrats controlled 45 of the 100 Senate seats, 12 Democrats (along with all 55 Republicans) would have had to vote against Clinton for him to be removed from office. But all 45 Democrats voted “no” on both articles.

Finally, during Iran-contra, Democrats controlled the House but did not push to impeach the president. Special congressional committees in the House and Senate were created to investigate the Reagan administration’s sale of weapons to Iran and the sending of money to the “contras,” a group fighting the governing regime in Nicaragua. There was a final, joint report that blasted the administration for violating a number of laws and suggested that Reagan — whose administration said he was unaware of the actions of his aides — should have known what was going on. The two committees, combined, had 15 Democrats and 11 Republicans. All 15 Democrats but just three of the Republicans signed on to the official report, while eight of the 11 Republicans wrote a minority report dissenting from the majority’s findings.

These were three of the biggest scandals in modern American history, and party loyalty stayed strong almost through the end of each.

The second trend is that there is an ideological component to party loyalty during scandal. In the simplest sense, more moderate members (in comparison to Congress overall) are more likely to split from a president of their party. Let’s look at the same votes and decisions as above for Watergate and Iran-contra, as well as the vote on the initial impeachment investigation of Clinton’s conduct. (To make a more apples-to-apples comparison, since Nixon and Reagan were never formally impeached.) According to DW-Nominate scores, which measure the ideology of each member of Congress based on roll call votes from previous congresses, the few Republicans who broke with Nixon and Reagan were more liberal than the rest of Republicans in Congress then. Similarly, the 31 Democrats who joined Republicans to support the initial impeachment investigation of Clinton’s conduct were generally more conservative than the 175 Democrats in Congress who opposed that GOP proposal.

These votes and decisions make for a messy comparison. They were on different actions and came at different points in each scandal’s development. They also had different stakes, which can affect how strongly partisan lines hold. For example, in the Nixon era, the vote to give the House Judiciary Committee subpoena power as part of its investigation was adopted by a 410-4 margin.

Still, the idea that at some point party loyalty goes out the window and politicians become nonpartisan statesmen doesn’t hold up historically.

What does this mean for today?

The historical data suggests that, first, Republicans overall are likely to stand with Trump amid the controversy. And that has largely been the case. According to a New York Times tally, 139 Democrats and independents are calling for a special counsel to investigate the connections between the Trump campaign and Russia — which Comey’s FBI was investigating at the time of his firing. No Republican in Congress supports a Department of Justice-appointed special counsel focused solely on the 2016 election and Trump. (This is likely because Republicans fear a special counsel is hard to control and could expand and lengthen his investigation, like Kenneth Starr did as independent counsel probing Clinton.)

Only six of the combined 290 Republicans in the House and Senate back the next most aggressive option, some kind of special committee in Congress (along the lines of the ones created amid Watergate and Iran-contra) or an independent outside group (such as the commission of largely retired officials that investigated the 9/11 attacks). Eighty-five Democrats have called for a special committee or investigation. So essentially almost all Democrats support some kind of ramp-up of the investigation, compared with about 2 percent of congressional Republicans.

History also suggests that those who are breaking with Trump, or will do so in the future, are more liberal than other Republicans. That has been true so far as well, with some caveats. In addition to the six who have backed some kind of special investigation, 42 Republicans have said that they had concerns about the Comey firing. These 48 GOP members (15 senators and 33 House members) were more likely to be moderate and somewhat more anti-establishment (meaning they vote against the party leadership on some measures) than their colleagues, according to their DW-Nominate scores. The group of 33 House members is disproportionately associated with the center-right Tuesday Group and includes few from the deeply conservative House Freedom Caucus.

Of course, in one sense, we don’t have to guess which Republicans are more likely to break from the president: There was already a big test of loyalty to Trump, less than a year ago. After he won the GOP presidential nomination, a small bloc of senators and House members refused to endorse him, and a disproportionate number came from the left wing of the party.

That was in the summer of 2016. The biggest scandal of the Trump campaign came in the fall, with the release of the “Access Hollywood” tape on Oct. 8. The number of anti-Trump Republicans expanded slightly after its release, but the vast majority of Republicans in Congress still supported Trump on Election Day, despite his unpopularity and the series of controversies about his conduct, most notably bragging about sexual misconduct on tape.

Our data is capturing only formal votes, and it could miss other decisions made by members. Perhaps the most important example is Howard Baker, who was the top Republican member of the Senate Watergate committee, which aggressively investigated Nixon and generally worked in a bipartisan way. Baker and the other two Republicans on the committee joined with its Democrats to obtain tapes of the president’s Oval Office conversations. (At the same time, even Baker felt loyalty to his party, as his aides at times discussed the committee’s processes with the Nixon White House.)

Who might be the Baker of today? John McCain, of Arizona, is the only Republican in the Senate currently calling for a special committee to investigate the president. Lindsey Graham, of South Carolina, is urging Trump not to pick a GOP politician to replace Comey as the FBI director. Arizona’s Jeff Flake and Nebraska’s Ben Sasse, both during the campaign and last week amid the Comey firing, have questioned Trump’s decision-making and judgment.

Overall, though, the lesson from history is simple. In the wake of Comey’s firing, Democrats are encouraging Republicans to “put country over party.” In making this call, they are essentially asking congressional Republicans to do something politicians historically have not: push for punishments and the potential removal of a president from their own party.

Subscribe: Apple Podcasts |
ESPN App |
Download |
RSS |
New to podcasts?

How big of a scandal is President Trump’s firing of FBI Director James Comey? The FiveThirtyEight Politics podcast crew, senior writer Perry Bacon Jr. and Brandon Rottinghaus, a professor of political science at the University of Houston who studies political scandals, try to figure that out. The team also discusses how the recent resignation of Census Bureau Director John Thompson could affect preparations for the 2020 census.

You can listen to the episode by clicking the “play” button above or by downloading it in iTunes, the ESPN App or your favorite podcast platform. If you are new to podcasts, learn how to listen.

The FiveThirtyEight Politics podcast publishes Monday evenings, with occasional special episodes throughout the week. Help new listeners discover the show by leaving us a rating and review on iTunes. Have a comment, question or suggestion for “good polling vs. bad polling”? Get in touch by email, on Twitter or in the comments.