(Click image to enlarge)
We all know that run differential is the driver in terms of team record, but teams can generate run differential in different ways. Teams can be great at scoring runs or preventing ones, doing both, or, oddly enough, doing neither. So would you rather have an elite run scoring or run preventing team? The answer, at least over the past 10 years of data, seems to be you want an elite run preventing team.
The graphic above shows the probability that a team playing from 2001-2010 would end the season in the top 8 in league run differential based on the various combinations of runs scored and allowed. I define elite as finishing in the top 25th percentile for runs scored (>=805.5 runs) or allowed (<=698.75 runs) based on all teams over the past decade. Numbers in parentheses represent the number of times that category occurred and the number of times that category finished in the top 8.
(Why the top 8? Because, generally, this captures the four best teams from each league and the eight playoff teams in a given season. Teams that finished in the top 8 made the playoffs 78% of the time.)
52% of teams that finished with a top 8 run differential were elite in at least one category. Teams that were elite run preventers had a higher probability of finishing in the top 8 than elite run producers (64% vs. 57%). That gap gets a tad bit larger if we restrict the comparison to teams that were elite run producers and non-elite run preventers (and vice-versa).
Teams that were only elite in terms of run prevention had a 56% chance of finishing with a top 8 run differential compared to elite only run producers (48%). And this makes sense, particularly if you look at the general relationship between runs for and against and run differential. Run prevention (r = -.72) has a stronger relationship to run differential than run production (r = .67) over the course of the decade.
So while the past decade was undoubtedly an offensive era, when teams managed to put together elite pitching and defensive clubs they had a greater chance of finishing with an elite run differential than teams that simply built elite slugging teams.
Think about 2010. Many prognosticators argued that the Giants were lucky to make it past the Phillies and win the World Series against the Rangers. Upon closer examination, this makes no sense.
The Giants scored only 697 runs all season, but they only allowed 583--for those counting at home, that's a run differential of 114. That differential was good for 4th in all of baseball and an expected record of 95-67. The Phillies, by contrast, scored 772 runs and allowed 640. They were expected to win 96--only one more than the Giants. As for the Rangers, they scored 787 runs and allowed 687, good for the eighth best run differential in the majors. They were expected to win 92 games, four fewer than the Giants. All three teams had what could be considered elite pitching staffs, but the Giants had the best of the three.
As we've seen, run prevention can provide more of an edge to teams than run scoring. Add that to the fact that elite run preventing teams had the same probability of making the playoffs during the decade as run scoring teams (63%) and run preventing teams had a higher probability of winning the World Series (7% to 4%), and the Giants don't look quite so lucky--they just look good.