The game of baseball is centered on home: pitchers throw the ball over home plate, batters hit home runs, and fans root for the home team. Apparently, baseball’s preoccupation with home is no accident. According to a recent study by Denver psychologist Howard Markham, the average divorce rate in cities that have major league baseball teams is 28 percent lower than in cities that lack a major league franchise. Markham says this is because baseball inspires “communication, fun, and friendship,” and though his study does not mention it, baseball does have a rather curious record of promoting matrimony.

Over the years, baseball promoters have carried numerous “Will You Marry Me?” proposals on their electronic scoreboards. Some have hosted pregame weddings in which couples walk “down the aisle” (from the pitching mound to home plate) under a canopy of baseball bats held by players lined up on each side. And in what has to rank among the all-time tackiest promotions, the Atlanta Braves once held a “Wedlock and Headlock Night,” which featured a mass wedding ceremony before the game and a professional wrestling match afterwards.

Although Markham’s research is not intended to be taken seriously, it does raise two intriguing questions: What regional differences exist in divorce rates? And what factors explain these differences? Answering these questions should offer some clues about how to reduce the high divorce rate in the United States today.

Perhaps the easiest way to think of regional differences in divorce is to think not of baseball but of another summer pastime: bathing suits. It is strange but true that the divorce rate of any place in the continental United States can be predicted by knowing the number of days out of the year women there wear swim suits. The more warm weather days suitable for bathing attire, the higher the divorce rate. To illustrate, draw a line across the midsection of the continental United States (along the northern border of North Carolina, Tennessee, Arkansas, Oklahoma, New Mexico, and Arizona, and continuing through southern Nevada and central California). More than 60 percent of the total United States population lives above what might be called “the tan line.” Yet less than half of all divorces occur here, and each of the ten states with the lowest divorce rates is above the tan line.

Ironically, Teddy Kennedy’s home state of Massachusetts is number one in marital stability. Donald Trump’s New York is tied for second (and would have fared slightly better if Donald had not abandoned his first wife for a former winner of the Miss Resaca Beach, Georgia, bathing suit contest). Conversely, six of the seven states with the highest divorce rates are found below the tan line. And four of these six—Alabama, Arkansas, Oklahoma, and Tennessee—are located right in the heart of the Bible belt.

The irony here, of course, is that the Bible is hardly neutral on the subject of divorce. The Old Testament prophet Malachi reports that God “hates divorce”; in the New Testament, Jesus condemns divorce “for any reason except sexual immorality.” Given such teachings, one would expect a geographic region known for its devotion to the Bible to have lower rates of divorce.

The truth is, Scriptural teachings do appear to make a difference among the devout. According to data compiled by the National Opinion Research Center, the prevalence of divorce among weekly churchgoers in the United States is 17 percent. This is less than half that for people who claim “no religion” (37 percent) and well below the marital dissolution rate for people who attend religious services less than once a week (32 percent).

While frequency of church attendance is a reliable predictor of marital stability, it has little effect on state-by-state differences in divorce because the size of the weekly churchgoing population does not vary significantly by geographic region (except in the West, where attendance is lower and divorce higher). Religious affiliation, however, does have an effect on state-by-state differences. Catholics, Jews, and Mormons are less likely to divorce than Protestants. (Two in ten Catholics have been part of a failed marriage, while three in ten Protestants have experienced divorce.) Thus, one of the reasons for the relatively low divorce rates in the Northeast and Midwest is that these regions have a higher concentration of Catholics than other geographic areas.

According to author Michael McManus (a Protestant), the lower rate of divorce among Catholics reflects the fact that their parishes take the responsibility of marital support and counseling more seriously than Protestant churches. In his book Marriage Savers, McManus argues that “America’s churches are part of the divorce problem.” He notes that 73 percent of all weddings take place in a church, yet most ministers do not require couples to go through premarital counseling. Rather than being “blessing machines,” McManus believes churches need to concentrate on offering premarital counseling and marriage enrichment programs like those sponsored by Marriage Encounter.

While the impressive track record of such programs underscores the value of religious teaching in reducing divorce, University of Texas sociologist Norval Glenn believes part of the reason for the correlation between lower divorce and high church attendance is that people involved in a local church are socially rooted. That is, they are integrated into a local network of cross-generational social relationships which offers them support and expects them to adopt certain norms of behavior and attitude (like the expectation that their marriage will not end in divorce).

Glenn believes couples enmeshed in such networks are far less likely to divorce than couples who are more socially isolated. Indeed, his research shows a strong correlation between residential movement and divorce—a connection he believes is due not just to the absence of social integration among transplanted married couples, but also to the fact that martial unions joined in transient geographic areas are often more unstable.

“Where there is a great deal of residential movement, persons on the ‘marriage market’ probably will not have as much information about the characteristics and background of one another as they would if most of them had lived in the same locality for a long time,” Glenn notes. “A high level of residential movement,” he adds, “is likely to increase the probability that persons with dissimilar backgrounds will marry.”

Frequent mobility also increases the chances that a married individual will meet someone who seems more desirable than his or her current spouse, Glenn says. And it increases the likelihood that an individual in a troubled marriage will seek advice from someone who does not know the other spouse and cannot offer the kind of balanced perspective that is often needed to help resolve marital problems. Thus, one of the major reasons for high divorce rates in the Sunbelt is that this area of the country has experienced greater population growth and residential turnover than more stable areas in the Northeast and Midwest.

To be sure, there are other influences at work. For example, divorce rates are affected by the stringency of divorce law. Clearly, the liberalization of state divorce laws during the 1960’s and 1970’s helped fuel the dramatic surge in divorce that took place throughout the country during this period.

Policy considerations, however, have only a minor effect on regional differences in divorce because divorce laws do not vary considerably from state to state. The one exception is Nevada, which has the nation’s most liberal divorce statute and (not coincidentally) a dramatically higher rate of divorce than any other state in the union. (A divorce can be obtained so quickly and easily in Nevada that some people establish temporary residence in the state for the expressed purpose of getting a divorce.)

In addition, regional differences in divorce are influenced by several demographic and cultural factors. Specifically, people in the Sunbelt tend to be slightly younger than people in the Northeast and Midwest (which has a modest effect on divorce rates since most divorces occur during the first seven to ten years of marriage). Southerners also tend to marry at a slightly younger age than Northeasterners (which is relevant because people who marry young are more likely to divorce). And Northeasterners are slightly less likely to marry than people in other regions (which reduces the size of the population that could get a divorce).

In many ways, it is too bad that high divorce rates cannot simply be attributed to the absence of professional baseball. For it would be far easier to expand the number of major league baseball teams than to adopt the sort of daunting reforms that would be needed to make a meaningful dent in the divorce rate.

But if we are serious about reducing divorce in this country—and we should be—then we should seek to overturn government policies that permit unilateral no-fault divorce. We should call for churches to encourage greater participation in premarital counseling, marriage enrichment, and community-building programs. We should implore businesses to reduce the volume of forced geographic transfers and job-related travel. And we should work diligently to change cultural attitudes about divorce, helping Americans see the long-term damage divorce causes.

Obviously, curbing divorce in America will not be easy. But a nation that still honors Lou Gehrig’s heroic perseverance and still cherishes historic Wrigley Field ought to be able to regain its commitment to the time-honored ideal of marital permanence. We ought to be able to renew our vows “to love and to cherish, for better, for worse, for richer, for poorer, in sickness and in health, ’til death us do part.” We ought to be able to find our way home.