Stark complains that historians just come up with good-sounding arguments that fit their agenda and do not check to see if their arguments have any bearing in reality. He is especially harsh to biblical historians. Accordingly, Stark’s goal with the book is to create testable hypotheses, and then test them against concrete data to see if the hypothesis has any validity. He takes the 31 cities in the Roman Empire with more than 30,000 people, along with whether they are in the East or the West, are a port or inland city, and whether they had a church by 100 AD, 180 AD, or neither, and applies statistics to answer the hypothesis.

Stark found support for the following hypotheses:

• The growth of Christianity in the Empire is consistent with slow growth of 3.4% per year, via network growth, rather than mass conversions. (Growth of Christian names in contracts in Roman Egypt, and Christian grave inscriptions in Rome, both fit the curve very well.)
• Cities closer to Jerusalem, and port cities, (both more quickly traveled to) were more likely to have a church. 71% of cities within a 1000 mi of Jerusalem had a church by 100 AD and 100% had one by 180 AD. By 100 AD, only 7% of the cities farther than 1000 mi had a church.
• Hellenic cities got a congregation sooner than Roman cities, since Greek religion was more compatible with Christianity: many Greek gods were mortals who acquired divinity, and Christian theology was rooted in reason.
• Larger cities were more likely to have Christian congregations (more chance of finding someone similar) since it was easier to get a nucleus of people, but not for Cybel and Isis because paganism did not inspire missionary activity.
• Port cities and Hellenic cities were more likely to have temples to Cybel and to Isis: the former because a port made a city more accessible, and the latter because the eastern religions emphasizing emotionalism (celebration, joy, passion) and individualism (paganism was a community religion) were more similar to Greek religion.
• Cities with a Jewish Diaspora community were more likely to have a Christian congregation: Stark asserts that despite Paul being the missionary to the Gentiles, the actual biblical record shows that he got most of his converts from Jews and God-fearers. There must have been continued conversions from less-observant Jews over the centuries, since writers talk about conflict between the two groups.
• Cities with a temple to Cybel or Isis were more likely to have a church, which Stark attributes to the fact that Christianity is also an eastern religion (emotionalism and focus on individual rather than community), so it is less of a leap, while offering more to the adherent.
• Gnosticism was more prevalent in port cities and larger cities (ease of travel and enough people to nucleate), but a Jewish Diaspora was not a significant factor.
• Gibbon’s assertion that Christian rulers after Constantine tried to stamp out paganism was largely false. Emperors appointed large numbers of pagans (or people of unknown religion, which were presumably pagans not advertising the fact, since Christianity was the approved religion), roughly in a roughly 50/50 ratio until at least 455.

Thus, the story of the growth of Christianity can be seen as follows. Paganism was oriented around the community: there were rituals known to work, and it was essential for the community to do them to be in good standing with the gods. The gods, however, were somewhat capricious and of dubious character, moreover, there were lots of them, so people were not motivated to missionary work. The eastern religions of Cybel and Isis offered more for the individual, including an emotional experience, and were closer to monotheism, which inspired more fervor on the behalf of the adherent. These spread through the eastern cities that were easy to travel to. Most people who convert to a different religion are weakly tied to the original religion, and tend to convert when their social circle becomes predominately of the new religion. Converts also are much more likely to a religion with fairly similar beliefs. Christianity was a better monotheism than Cybel and Isis, so was able to take advantage of the popularity of Cybele and Isis. Christianity also offered a way to be part of God’s family without slavish observance of the Jewish Law, so it appealed to God-fearers (Gentiles who worshiped God but did not want the burden of the Law) and to Diaspora Jewish who were not observant. Christianity also spread through the easily accessible port cities and the larger cities, and then continued to slowly grow, partly because of consistent conversions by Diaspora Jews.

Stark gives good summaries of paganism, Cybel, Isis, and various Gnostic sects (asserting that there is not a generalizable “Gnostic” theology). He also is very consistent in creating formal hypotheses and then testing them. However, he is too quick to say “the hypothesis is confirmed!” and entirely too confident. The data is small (31 cities) and pretty noisy. One or two inland cities having a church earlier, for instance, would change some hypotheses from strongly confirmed to merely supported. Overall to book feels like someone who has a conservative Christian axe to grind who has discovered statistics. Sure, you can turn the crank and get some numbers, and it is even helpful to do so—historians should attempt to verify if their ideas are supported by the actual facts we know. But the problem is, we just do not have enough facts for most hypotheses to be confidently confirmed statistically. Sure, the numbers might reach the threshold, but how big are the error bars?