The 2011 Milwaukee Brewers entered the All-Star Break at 49-43, but that record betrayed their performance in June and July. In both months, the Brewers allowed more runs than they scored; that didn’t stop them from posting winning records in both months. At the break, the Brewers’ 405 runs scored and 406 runs allowed suggested a team that might be able to stick around .500, but hardly a team that could be expected to win at an 86-game pace. Fortunately, the Brew Crew destroyed their competition from late-July onward, burying any questions about their legitimacy as NL Central contenders.
No matter how you feel about baseball analytics and statistics, one truth will continue to hit you over and over again: in order to win a significant number of games, baseball teams need to score more runs than they allow. I know, I know, this sounds like so much mathematical determinism by another one of those statheads that couldn’t beat the speed limit with his fastball. If you’re going to hang on to ANYTHING from the development in baseball statistics over the last 40 years, hang on to the idea that the differential between runs scored and runs allowed vastly influences a team’s ability to win games.
Of course, there’s that pesky little thing called distribution; sometimes, even when a team produces a certain differential between their runs scored and runs allowed, they outperform their expected record thanks to the distribution of their runs scored and runs allowed. If you’re starting to have faith that teams could overwhelmingly outproduce poor run differentials to win ballgames, stop right here: of 196 MLB teams that won 85 or more games from 1996-to-present, only 8, count ’em again, 8 allowed more runs than they scored. Most people might stop at that 96%: COMPETITIVE TEAMS MUST SCORE MORE RUNS THAN THEY ALLOW! Yet, those 4% hold useful truths about how teams play baseball.
Unfortunately, I am writing about this because the 2012 Milwaukee Brewers currently sit at 28-34, having scored 268 runs against 290 runs allowed. Interestingly enough, the 2012 Pittsburgh Pirates have scored 198 runs against 223 runs allowed, but they are currently two games out of first place, boasting a 32-29 record. What gives? How is it that the Pirates and Brewers can post similar run differentials with remarkably different W-L records?
THE WINNING LOSERS
Baseball is not a good sport for general rules. Just about any rule you can think of has notable counterparts. Pitching wins championships, right? Well, yes, except for when hitting wins championships (the St. Louis Cardinals in the last 10 years are a great example of the latter rule). That’s just one example; we could go on and on all day. Each time you feel yourself pulled toward a general rule about baseball, hit Baseball-Reference archives and you’ll soon find antidotes to overreaching, grand ideals. Baseball is as stubborn and gritty toward general rules as David Eckstein was running out ground balls; neither will ever fall in line with the logical expectations of their followers.
Now that we have that out of the way, I can blitz your mind with general rules. Those 8 clubs that won 85 (or more) games while allowing more runs than they scored? They had a surprising number of traits in common. A simple hypothesis: since these teams railed against any common sense about scoring more runs than allowed to win lots of ballgames, they achieved their improbable winning records by doing very specific things very well. First and foremost, those 8 teams:
1997 Giants: 784 RS / 793 RA, 90 wins
2007 Diamondbacks: 712 RS / 732 RA, 90 wins
2007 Mariners: 794 RS / 813 RA, 88 wins
2011 Giants: 570 RS / 578 RA, 86 wins
2009 Tigers: 743 RS / 745 RA, 86 wins
2008 Astros: 712 RS / 743 RS, 86 wins
2006 Red Sox: 820 RS / 825 RA, 86 wins
2009 Mariners: 640 RS / 692 RA, 85 wins
That’s a rather motley collection of ballclubs, if I don’t say so myself. Just glancing over that list, you’re going to say, “wow, there are a whole lot of differences between those clubs.” What’s interesting is the surprising number of similar traits they share, which leads me to wonder if there’s a specific recipe for successfully winning games while allowing more runs than scored.
(1) Play in an extreme ballpark.
(2) Win a disproportionate number of 1-run games.
(2a) Play a disproportionate number of 1-run games.
(3) Accumulate losses in blowout games.
(4) Win a significant / an extreme number of home games.
(5) Post a respectable record during negative run differential months.
(1) Extreme ballparks. This rule doesn’t apply to every club here, which is why these are general suggestions and not universal truths. However, 5 of the 8 clubs on that list played their home games in ballparks with park factors significantly outside of neutral batting and pitching effects. The 2007 Diamondbacks boasted Chase Field (107/107 3-year batting/pitching); both Mariners clubs played at SafeCo (96/97 and 94/95 3-year batting/pitching); the 2011 Giants played at AT&T Park (92/92 3-year batting/pitching); and the Red Sox had Fenway (105/104 3-year batting/pitching). Each of those parks significantly suppressed or increased runs scoring in their respective cases.
(As a note, even some of the “average” parks inch toward extremes. Candlestick Park boasted a 98/98 3-year batting/pitching factor, and Minute Maid went 97/98 for 3-year batting/pitching. Comerica is the odd egg, at 101/101 3-year batting/pitching).
This should not be all that shocking. In fact, if you simply look through league performances year-by-year, many times teams that outperform their run differential have a park that allows them to produce some extreme factor. Looking at teams that score or allow runs at least one standard deviation away from the league/park average, I suspect (a) that team underperformed or outperformed their run differential, and (b) that team played in an extreme park. Playing in an extreme park simply provides circumstances for strange runs scored/runs allowed performances with limited effort required on the part of the club.
(Therefore, we can get (4) out of the way: not surprisingly, these extreme ballpark clubs AVERAGED a .597 winning percentage in their home parks during their improbable seasons. What makes that impressive, you ask? These teams averaged 361 RS / 360 RA at home, and five of those clubs won at least 46 home games while allowing more runs than they scored at home. Now that’s home cooking!)
(2, 2a). One run ballgames. If the idea of using runs scored and runs allowed to examine a team’s expected win total makes run differentials extremely important, then the best way to subvert that significance is to distribute wins unevenly through one-run victories. Whether it’s 1-0, 4-3, or 8-7, a one-run game means that a team overcome its shortcomings (not enough runs, pitching imploded, etc.) to win, or that they simply held on to outlast their opponent.
These eight clubs performed extremely well in one-run ballgames. In fact, the 2008 Astros were the worst of the bunch in one-run games, and they still went 21-21. Every other team won at least six more one-run games than they lost.
Furthermore, most of these clubs played an extraordinary number of one-run ballgames. Every team except for the 1997 Giants and 2008 Astros played at least 47 one-run games; half of the clubs played at least 50 one-run games!
The math is rather simple: if a team plays, say, 50-55 one-run games, and they win 28-to-35 of those games, that team has a great foundation to outplay their run differential.
(3) Blowouts. Logically, every baseball team desires close ballgames. In a close game, one bloop hit, one lucky bounce can allow that team to sneak a victory from their opponent. A silly counter-rule might exist: If a team excels at winning close games, the size of their losing margins does not matter.
This is where this group of teams gets a little rough. Oddly enough, these clubs played in nearly as many blowouts as they did one-run ballgames, albeit with much uglier results. Against their sparkling .585 winning percentage in one-run ballgames, these clubs won only .449 of their blowouts. While their average one-run performance over the course of a season was 188.5 RS / 180.3 RA, their average blowout performance was 226 RS / 273 RA.
There you have it: in a collection of approximately 92 games, these clubs went 414.5 RS / 453.3 RA, but escaped with nearly 48 wins on average. The power of blowouts looks daunting, and I gather that most truly good teams will have records in blowouts that are nearly as good as their one-run ballgames (the 2011 Brewers can boast this feat, for instance). If you’re looking to outplay your run differential, though, winning every close game possible and losing by a mile when you lose is a pretty good practice (if counterintuitive).
(5) Post a respectable win total during “losing” months. Not unlike their bizarre home performances or their blowouts or one-run games, this group of eight ballclubs boast unexpected performances during months in which they were outscored. Collecting each month that these teams were outscored, the average performance by these clubs during outscored months was 314 RS / 373 RA. That’s a pretty daunting margin! Yet, their average record during these “losing” months was 37-39.
The 1997 Giants are my favorite example in this category. From May through July, they were consistently outscored, running through a swoon of consecutive months of 401 RS / 448 RA baseball. That’s pretty bad, but they escaped with a 42-42 record during those months. “How?,” you ask! Well, isolating 21 of their losses in those months, you’ll find a set of games in which the team lost 50 RS / 203 RA!!! Basically, the 1997 Giants were WHIPPED in a quarter of their games during that three month swoon, but their 42-21, 351 RS / 245 RA performance during those other May, June, and July games suggests that the team was indeed pretty good.
BUILDING AND REBUILDING: A REPRISE
The similarities of these clubs prove that they are a truly bizarre bunch. Suffice to say, no General Manager would set out to build a club this way; a GM is not going to say, “well, my club has an extreme park, so I’ll build a club that can win every close game and get blown out otherwise, while outplaying run differentials at home and during their worst months.” Rather, I think these clubs indicate that every so often, there is a perfect storm of elements that combine to help a team succeed against all odds.
Unfortunately, here’s the difference between the 2012 Brewers and the 2012 Pirates. While both clubs claim equally bad run differentials, the Pirates are using whatever strengths they have to win whenever they can. For example, 27 of their 61 games were one-run ballgames, and they won 17 of those games; last night’s 7-1 blowout at the hands of the Baltimore Orioles brings them to 3-8 in blowouts. Between those two categories, the Pirates are 20-18, despite scoring 115 runs against 138 runs allowed. At home, in extreme PNC Park? 19-11, despite 86 runs scored against 83 runs allowed.
Miller Park is shifting into a hitters’ park once again, after a few average seasons; the park’s 3-year 104/104 batting/pitching factors are nearly as extreme as PNC’s 3-year 95/96 batting/pitching factors. Yet, the Brewers are 16-17 at home. The Brewers lost their second consecutive one-run game to the Kansas City Royals, bringing their one-run performance to 10-11. The team’s performance in blowouts is moderate, too, forming a 9-11 mark despite 98 RS /114 RA in those games. All told, if a club scores 170 runs while allowing 187 runs over 41 games, a 19-22 record seems about right.
The significance of these elements will play out during the trade deadline. Theoretically, neither the Pirates nor the Brewers should be in a position to buy talent at the deadline. Both teams allowed significantly more runs than scored to this point in the season, suggesting that they will lose more games than they win if they maintain their current performance over 162 games. However, the intangibles will shift everything; the Pirates are milking PNC Park for every ounce of value it is worth, and they are winning just about as many close games as they can. They have that perfect storm of elements working to the extent that their front office should be compelled to buy.
The Brewers, on the other hand, look like strong candidates to sell. However, aside from the fact that the team maintains the significant core from their 96-win, Division Championship club, it seems to me that the simple fact that the Brewers have so many replacement players on their roster makes it unlikely that the Brewers will trade many players away (thereby creating more holes in the roster). Typically serviceable-to-above average MLB players such as Corey Hart, Yovani Gallardo, and Rickie Weeks remain under contract beyond this season, and veterans such as Aramis Ramirez, Francisco Rodriguez, and Randy Wolf either have contracts or current performances (or both) that make any significant trade unlikely.
In the universe of baseball checks and balances, those unknown laws of the diamond are churning in such a way that staying put at the trade deadline gives the Brewers as much as an advantage as their Division counterparts’ potential trades. The Brewers are straight-up underperforming (or bad?) this year, yet their core of players is strong enough to suggest that a correction in performance can lift them to a competitive level.
Last year, the Brewers excelled in many aspects of the game. If they reclaim any of their close game magic and use their remaining 48 home games to their advantage, they can aim for that strange arrangement of baseball elements that allows them to exceed their run differential. In this regard, their sole competition is the Pittsburgh Pirates; for, if there are any baseball gods at all, surely they will not allow two clubs to compete for a division championship while both clubs allow more runs than they score.