As the Chicago Cubs celebrated winning the World Series last month, much of the attention centered on the contributions of young, homegrown players such as National League MVP Kris Bryant and third-place Cy Young finisher Kyle Hendricks. But under the guidance of general manager Theo Epstein, the Cubs also loaded up on external talent — and salaries — in preparation for their title run. Even after adjusting for MLB-wide increases in the cost of a win,1For the purposes of this story, all references to payroll increases will be adjusted by measuring changes relative to differences in how much marginal salary bought a win above replacement (averaging together the two methods of WAR found at FanGraphs and Baseball-Reference) between the seasons in question. Chicago boosted their payroll by 169 percent between 2014 and 2016 — a record two-season increase among World Series winners.2At least since 1985, the first season for which Baseball-Reference.com has team payroll data.It’s a fitting coincidence that the previous mark was set by a team whose roster was basically completed about 20 years earlier. On Dec. 12, 1996, the Florida Marlins signed free-agent outfielder Moises Alou to a five-year, $25 million contract, the final major piece of a team built largely through a wild, two-year spending spree. When they celebrated their own championship 10 months later, Florida would face accusations of having “bought a World Series,” but they also set a model that many teams would follow over the ensuing years, sometimes to great success — as with the Cubs — and often to great failure. Either way, the era of the store-bought championship bid was at hand.As the hot stove burned in late 1996, the Marlins had only been around as a franchise for four seasons, and they’d never finished a season with a record above .500. But they were also trending upward: Florida had improved its winning percentage in each season the team existed, including an 80-82 record in 1996, which saw them post a 31-24 record (a 91-win pace) in August and September.Marlins owner Wayne Huizenga, already a ’90s business icon for his ownership stake in video-rental titan Blockbuster Entertainment, was also sinking more money in his teal-clad expansion club. Before the 1996 season, Florida GM Dave Dombrowski had upped the team’s payroll by 38 percent,3A raw increase of $6.6M, even as the MLB-wide rate of marginal salary per WAR slightly decreased. from 25th in baseball to 16th, reeling in such big-name free agents as Kevin Brown, Al Leiter and Devon White. The next winter, Dombrowski poured even more cash into Florida’s roster — increasing payroll by another 33 percent — when he signed more free-agent deals with Alou, Alex Fernandez, Bobby Bonilla and Jim Eisenreich.Unsurprisingly, all that aggressive free-agent spending gave new Marlins manager Jim Leyland significantly more talent to work with. If we use Tom Tango’s WARcel projections as a simple way to judge how many wins Florida could have expected to get out of its new acquisitions (as well as the cost of losing the players it jettisoned in the process), the Marlins picked up more than 15 net WAR in talent for 1997 alone through Dombrowski’s wheeling and dealing over the previous two offseasons, another modern-era4Again, going back to the advent of complete team-payroll data in 1985. record for an eventual World Series winner. But having an existing talent base is still preferable to buying one. According to my regression, each extra WAR a team starts with before a two-year stretch of acquiring players is worth about a 60 percent bigger boost to its World Series odds than a WAR of talent snagged from outside the organization over the following two years.9In a way, this makes sense: Existing talent is generally younger, with more potential production (and more cost-controlled years) ahead of them with the club than players poached from other organizations.Plus, the effect of supercharging a roster has lost some of its potency since Florida spent its way to glory. If we split up our data into two sets — one that covers the 1987 through 1997 seasons, and the other for every year since — a one-standard deviation increase in payroll and talent meant a 40 percent bigger bump in World Series odds back then than it would now.Anecdotal evidence bears this idea out, as well. Before the 2016 Cubs made good use of their massive talent influx, the 2004 Boston Red Sox were the last team to win a championship with anything close to the same expensive quick-build strategy as the 1997 Marlins employed. And they were the exception to the rule that most free-spending title bids fall short. For instance, extreme shopping sprees by the Yankees in the early to mid-2000s and, more recently, by the Los Angeles Dodgers yielded a grand total of zero titles, despite the huge financial advantages. (Yes, the Yankees won it all in 2009, but that year they spent about 21 percent less — relative to the MLB average — than they had at their peak in 2005.)Despite fears of a “bought” championship dating back to the advent of free agency (if not earlier), baseball is a tough sport to rig with pure cash. Not only is the postseason incredibly random, but it’s difficult to even predict whether a team will be good enough to get there in the first place. The very best team in baseball only has something like a 15 percent chance of winning it all going into a given season, and although there’s no such thing as having too much talent on a roster, putting that philosophy into practice gets mighty expensive mighty fast. That’s why World Series winners such as the 1997 Marlins — and, now, the 2016 Cubs — are destined to remain historical anomalies, ranking among the very few teams in history who ever saw a complex multiyear building scheme end in anything other than frustration. At the time, no modern team had ever won a championship following such a talent binge. When the defending-champion New York Yankees spent the off-seasons of 1994 and 1995 loading up on players, that meant adding 9 net WAR of talent and 56 percent more payroll5A raw increase of $8.2M over seasons where the cost of WAR was decreasing.. In terms of totally overhauling their roster from the outside, mainly through the power of free-agent shopping, the Marlins were on another level entirely.And it worked out well. Together, Brown, Fernandez, Alou, Bonilla, White, Leiter and Eisenreich combined for about 19 of the Marlins’ 37 total WAR, a big reason — along with 1993 trade pickup Gary Sheffield — why 77 percent of Florida’s wins that season were generated by players who debuted with other franchises. (No World Series winner had gotten less from its homegrown players in the previous 74 years.) In combination with a few homegrown talents such as Livan Hernandez,6Hernandez grew up playing in Cuba, but he defected in 1995 and signed an amateur free-agent deal with the Marlins. Charles Johnson and Edgar Renteria, Florida’s mercenary contingent powered the Marlins to 92 wins and the NL Wild Card, then navigated the team through a difficult string of postseason opponents en route to a seven-game World Series triumph over the Cleveland Indians — the first-ever Series win for a wild-card team.Of course, just as swiftly as it had been built, Florida’s talent machine was disassembled part by part. By spring training of 1999 — just 16 months after Renteria lined a walk-off single up the middle to seal Game 7 — 22 of the 1997 Marlins’ top 26 players by WAR had been purged from the team’s roster, scattered across the majors in baseball’s most infamous “fire sale.”Critics railed against the cynicism of instantly gutting a championship roster, a blight that still taints that Florida team’s legacy. But the Marlins were undeniably trendsetters: In the decade leading up to 1997, only 13 teams had been built in a similar manner, increasing payroll by at least 25 percent and adding at least 5 net WAR of talent over a two-year span. In the two decades since, 36 teams have tried the same tactic — nearly 40 percent more attempts per season.7And, again, this is inflation-adjusted — it measures payroll increases over and beyond changes to the MLB-wide price of a win. With MLB teams’ revenues exploding in the post-strike era, more and more teams began mounting Marlins-style tactical spending strikes.As for whether that strategy has been a reliable path to victory, well, it’s complicated. Based on a logit regression using MLB data since 1987,8With the dependent variable being whether a team won the World Series or not — yep, this is an all-or-nothing analysis, in the spirit of the 1997 Marlins’ build-and-tear-down model — and the independent ones being the amount of WAR a team started out with before its two-year spending spree, and the amount of WAR talent added and percentage change in payroll over the subsequent two years. increasing payroll and/or talent can certainly increase a team’s probability of winning the World Series: If an average team were to add as much to its roster as the 1997 Marlins did, its championship probability would increase sevenfold.