Blueprint for Building a World Series Winner

Zachary D. Rymer@zachrymerMLB Lead WriterApril 1, 2013

If winning the World Series was a simple matter of adding Y to X to get Z and then breaking out the champagne, more teams would do it. The formula isn't that simple.

This isn't to say that there's not a formula. There is. It's just not that simple.

The proof is in the last 10 World Series champions. Lay them all down on a table and go to town with a scalpel, and you'll see that they have a a few qualities in common. Dust these qualities off and line them up, and you have a recipe for a World Series champion.

The recipe consists of eight semi-simple steps, and it is as follows.

1. Have Three Starters Who Can Log at Least 180 Innings

Let's start with starters.

Video Play Button
Videos you might like

Asking starting pitchers to achieve a specific win total, ERA, WHIP or whatever else would be an exercise in wasted breath. If they're out there for enough innings, luck will do a number on their results-based numbers. That will always be the case as long as there are variations in defenses and ballparks, which essentially means we're looking at a "forever and ever" scenario.

But asking starters to just be out there for enough innings either way? That's fair. Eating innings is something starters do have some control over, and getting innings out of a starting rotation is just as important as they say it is.

With the bar set at 180 innings, a championship team should feature at least three innings-eaters.

The other stuff is all well and good, but it's the column on the right that stands out the most. It shows that eight of the last 10 World Series champions had starting rotations that ranked in the top 10 in MLB in innings pitched.

In a perfect world, every team would be like the 2005 White Sox. Their rotation didn't feature dominant arms, but it did feature four guys who ate innings like Homer Simpson eats donuts. Mark Buehrle, Freddy Garcia and Jon Garland topped 220 innings, and Jose Contreras topped 200 innings.

Buehrle lasted fewer than six innings only twice. Garcia lasted fewer than six innings three times. Garland lasted fewer than six innings four times. Contreras lasted fewer than six innings eight times, but he did go at least five innings in five of those. That's quite acceptable for a back-end starter.

The results weren't always pretty. Buehrle gave up at least four earned runs 11 times. Garcia gave up at least four earnies 12 times. Garland and Contreras both gave up at least four earnies nine times.

The innings kept coming anyway, and they were still plentiful once the White Sox got deep into the postseason. Between the ALCS and the World Series, White Sox starters pitched at least seven innings in each of the nine games that were played.

No team in the above table stands out as a rule-breaker, but the 2006 Cardinals did (for lack of a better word) slither their way into it.

Tony La Russa is one of the great baseball geniuses of all time, and his mind certainly gave the Cardinals a strategic advantage in their last two World Series victories. But he had himself a whipping boy that he punished time and time again in 2006, and his name was Jason Marquis.

La Russa squeezed 194.1 innings out of Marquis even though he had an ERA over 6.00 and gave up more homers and earned runs than anybody in the National League. He made seven starts that saw him log more than five innings despite the fact he gave up at least five earned runs. He made two starts in which he gave up over 10 earned runs but was left in for five innings anyway.

La Russa got innings out of Marquis, sure, but he did more harm than good.

Fair warning, Cardinals fans: Since the '06 championship team is by far the flukiest of the last 10 champion teams, the phrase "Don't be like the '06 Cardinals" shall be heard often in this space.

Don't take it personally (he said in vain).

2. Have Two Starters Capable of a 7.0 K/9 or Better

One thing about the starting rotations of the last 10 World Series winners is that they had variety. If you're going to win the World Series, it's best not to feature three sinkerballers, two soft-tossing lefties and so on. You have to mix it up.

Strikeout pitchers of any kind, however, are welcome anytime.

A would-be World Series winner should feature at least two strikeout pitchers in its starting rotation. Since the league-average K/9 is usually somewhere in the 6.5-7.0 range (see FanGraphs), we'll define a "strikeout pitcher" to be any pitcher capable of posting a K/9 over 7.0.

With the lone exception being the 2005 White Sox, nine of the last 10 World Series winners featured at least two punchout artists in their rotations.

As for the '05 White Sox, it may be puzzling that their starting rotation was so successful without at least one above-average strikeout pitcher. After all, more strikeouts mean fewer balls in play, and fewer balls in play means fewer things that can go wrong.

The trade-off is that the '05 White Sox rotation finished the season ranked fifth in MLB in strikeout-to-walk ratio. That was thanks to a 2.38 BB/9, the second-best mark in the league. Just as more strikeouts mean fewer things that can go wrong, fewer walks mean fewer things that can go wrong.

But the 2010 Giants are the real stars of the show here.

Tim Lincecum led the National League with a K/9 of 9.8, and Jonathan Sanchez made up for his ghastly 4.5 BB/9 with a 9.5 K/9. He also allowed the fewest hits per nine innings of any NL starter, and he had a lower ERA than both Lincecum and Matt Cain (it boggles the mind).

As for Cain, he had a 7.1 K/9. Madison Bumgarner had a K/9 of an even 7.0 in his 18 starts, and even Barry Zito came close to cracking the 7.0 threshold with a K/9 of 6.8.

Giants starters needed to miss as many bats as possible when they took the mound, as the club's batters were highly unpredictable. The Giants finished 17th in MLB in runs scored, yet they still posted the second-best run differential in the NL thanks to their K-happy starting rotation.

Once again, the 2006 Cardinals offer an example of how not to be. They did have two starters with K/9s of at least 7.0 in 2006, but one of them was Anthony Reyes, and he only made 17 starts.

Among Cardinals starters who made at least 10 starts, nobody besides Reyes and Chris Carpenter had a K/9 higher than even 5.0. It's no wonder their starting staff finished with a 4.79 ERA, not to mention a 4.96 FIP (see FanGraphs).

FIP, if you're unaware, stands for Fielding Independent Pitching. It's a stat that measures what a pitcher's ERA should look like by considering only things that pitchers can control: strikeouts, walks, hit by pitches and home runs.

Cardinals starters finished with a high FIP because their lack of strikeouts left a few too many things to chance. Jason Marquis' status as a whipping boy didn't help either.

So again, don't be like the '06 Cardinals. Leaving an excess of things to chance will result in frustration more often than it will result in triumph, as it did for them.

3. Have a Lefty Specialist Who Can Hold Lefty Hitters to a Sub-.700 OPS

To the pens we go, and we shall start with a discussion with the weirdos who make up 10 percent (actual fact) of the earth's population: left-handers.

Lefty specialists must make up a smaller percentage of big leaguers than even that, but they're sort of like hobbits on a quest to steal gold from a dragon. You never know when one might come in handy.

What's a good figure for a lefty specialist to aim for? Well, some digging on revealed that left-handed batters have tended to post an OPS right around the .700 range against left-handed pitchers within the last 10 years.

Anything under that mark will do for a lefty specialist, and the majority of the last 10 World Series champs have had a lefty specialist capable of doing that.

Nine of the last 10 World Series champs had at least one stud lefty in their bullpens. Some were better than others—J.C. Romero FTW!—but they all got the job done.

That the 2010 Giants and the 2011 Cardinals felt compelled to go out and acquire left-handed specialists speaks highly of their importance, and both Javier Lopez and Marc Rzepczynski panned out quite well for their new employers.

Lopez appeared in 27 games for the Giants in 2010, holding opposing hitters (mainly lefties) to a .368 OPS. The Giants went 18-9 in his appearances, and he gave up one hit and one walk in 5.2 postseason innings with four punchouts.

Rzepczynski appeared in 28 games with the Cardinals in 2011. He barely held hitters to an OPS under .700 at .692, but the Cardinals went 18-10 in his appearances. He was money in the NLCS and the World Series, allowing one earned run and striking out eight in 7.1 innings.

The only team that did it wrong was the 2003 Marlins. Lefties compiled a 1.062 OPS against Michael Tejera and an .831 OPS against Armando Almanza. Neither of them was dominant against lefties to begin with, as lefties had a .736 OPS against Tejera and a .772 OPS against Almanza in 2002.

The Marlins got away with it, but they're the exception to the rule.

4. Righty Setup Man with: K/BB Over 2.0, OPS vs. Lefties Under .740, OPS vs. Righties Under .720

The problem with lefty specialists is that most of them only specialize in getting lefty hitters out. A lefty reliever who can hold his own against lefty and righty hitters is a rare sight indeed.

But right-handed relievers who can hold their own against lefties and righties? They're fairly common.

Right-handed batters have tended to post an OPS in the .720-.730 range against right-handed pitchers over the last 10 years, and left-handed batters haven't done any worse than a .743 OPS against righties in the same span. Any right-handed reliever who can live below those numbers in this day and age is doing a fine job.

Give or take a few points in either direction, the last 10 World Series champs all featured at least one right-handed reliever who has managed to do just that while posting a K/BB ratio of at least 2.0 in the process.

Yeah, I stretched the rules a little to include Tim Spooneybarger, whose 2003 season was shortened by Tommy John surgery. I also bent the rules a little to include Mike Timlin, who had his share of issues with right-handed batters in 2004.

Nonetheless, each of the last 10 World Series winners is represented here, and there are virtually no red flags to discuss after Spooneybarger and Timlin.

Right-handed relievers who can get right-handed hitters out are a dime a dozen, but these guys are special because of their matchup versatility. They can handle an inning with two lefties and a righty due up just as well as they can handle an inning with two righties and a lefty due up.

That means less work for the other guys in the pen if things go according to plan, and a high K/BB makes it easier for these guys to have things go according to plan. Fewer walks means a larger margin for error, and strikeouts are the best way to make high-pressure situations pass without harm.

Right-handed relievers like these also have the potential to blossom into something special. Take a couple of guys in our next point, for example.

5. It Takes About a 2.0 WPA to Be a Championship Closer

One thing all closers have in common is that one loose wire that allows them to shrug off pressure and to go work in high-leverage innings. It's something that really can't be quantified.

Not by the save statistic, anyway. Win Probability Added is better. defines WPA as the "sum of the differences in win expectancies for each play the player is credited with." It favors pitchers who dominate in high-leverage situations, meaning it favors shutdown closers.

What kind of WPA defines a true top-shelf closer? Well, there have been hundreds of 20-save seasons in the last 10 years, but only 107 of them have come paired with a WPA of 2.0 or better. That's an average of roughly 10 or 11 per season, or about a third of the league. Case in point, only 10 of the 27 guys who saved 20 games in 2012 posted at least a 2.0 WPA.

Thus, 2.0 WPA looks like a pretty good bare minimum for a top-shelf closer. Five of the last 10 World Series winners have had closers that easily surpassed this mark. Here they are in chronological order:

Keith Foulke, 2004: 2.8 WPA

Jonathan Papelbon, 2007: 3.8 WPA

Brad Lidge, 2008: 5.3 WPA

Mariano Rivera, 2009: 4.0 WPA

Brian Wilson, 2010: 4.2 WPA

Each of these five pitchers had the highest WPA among the relievers in their respective bullpens. As the season went along, there was never any doubt that they truly deserved to be the ones getting the final three outs.

The five other champions on our radar didn't have it so easy. They were forced to improvise with their closer's role, but they managed to stumble on top-shelf closers in time for their postseason runs.

2003 Marlins

Braden Looper had a 2.32 ERA and 17 saves in his first 43 appearances to go along with a 2.96 WPA. However, he began to struggle right around the time Ugueth Urbina was acquired from the Texas Rangers. Looper compiled a 5.93 ERA over his final 31 appearances, as well as a minus-0.72 WPA.

In the meantime, Urbina compiled a 1.41 ERA and a 1.9 WPA in his 33 appearances as a Marlin. Over a full season's worth of high-leverage situations, Urbina's WPA would have been 4.4 (based on's projections).

If so, he would have been a surer thing than Foulke in 2004, Papelbon in 2007, Rivera in 2009 and Wilson in 2010.

2005 White Sox

Dustin Hermanson did fine as Chicago's closer in 2005, saving 34 games and compiling a 2.04 ERA in 57 appearances. But then his back acted up, forcing Ozzie Guillen to find a new closer.

He tabbed Bobby Jenks, who was an attractive option because of his ability to rack up strikeouts.

By the end of the season, Jenks was statistically the best shutdown reliever Guillen had in his bullpen after Hermanson. Jenks finished the season with a 1.3 WPA in 32 appearances. Average that out over a full season, and you get a 3.2 WPA.

2006 Cardinals

Tony La Russa went with a "proven closer" in Jason Isringhausen in 2006, and he kept going to the well even after it became apparent that Isringhausen was a liability. He finished the season with 10 blown saves, an ERA of 3.55 and a WPA of minus-0.7.

Meanwhile, Adam Wainwright dished out ownage all season. Though many of his 61 appearances came in lower-leverage innings, he still topped Cardinals relievers with a 1.8 WPA.

Wainwright didn't become St. Louis' primary closer until the final days of the season, so we don't know what his WPA would have looked like over a full season's worth of high-leverage situations. That he managed a nearly top-shelf closer WPA as a setup man, however, says something.

2011 Cardinals

Fernando Salas handled the bulk of the save opportunities for the 2011 Cardinals, but he also led the Cardinals with six blown saves. Three other St. Louis relievers blew at least four.

Jason Motte was one of them, but he also held the distinction of being not only the surest bet in St. Louis' bullpen, but the surest bet of its entire pitching staff.

Motte had a 1.2 WPA in 2011, the highest among all Cardinals pitchers. From when he recorded his first save on Aug. 28 to the end of the year, he compiled a 0.49 WPA, which would have been a 2.5 WPA over a full season.

2012 Giants

Brian Wilson was the Giants closer at the start of the 2012 season, but he only made two appearances before he had to go in for Tommy John. After many twists and turns, his job eventually passed to Sergio Romo.

Romo had a 1.41 WPA in his final 20 appearances, which would have been a 5.3 WPA over a full season's worth of high-leverage situations.

To put that in perspective, both Jim Johnson (technically) and Fernando Rodney fell just short of a 5.3 WPA in 2012.

If you're looking for further perspective, consider this: Neftali Feliz had a 0.7 WPA in 2011, and Jose Valverde had a 0.5 WPA in 2012. They, of course, closed games during the regular season for the last two World Series losers, and both had their issues in the postseason.

6. First Two Hitters Should Combine for a .340 OBP and 200 Runs

As much as baseball has changed over the years, lineups still look largely the same today as they did back when men wore top hats out to the old ballgame. Managers still put a pair of table-setters in front of the big boppers, and their job description is simple: Get on, get over, get in.

Table-setters on a championship team should get on at least 34 percent of the time (a .340 OBP, in other words) and come around to score at least 200 times.

Of the last 10 champs, six saw their 1-2 hitters combine for a .340 OBP, and two teams came agonizingly close to joining the party. The table-setters of the 2005 White Sox just missed on both fronts, and the table-setters for the 2006 Cardinals fell a little short of a .340 OBP.

Some teams had easier times finding players to satisfy these requirements than others. The 2003 Marlins had Juan Pierre and Luis Castillo locked into the top two spots of their lineup for pretty much the entire season. Ditto the 2004 Red Sox with Johnny Damon and Mark Bellhorn.

But no team did it quite like the 2009 Yankees. They had Derek Jeter and Damon at the top of their lineup, and they did everything that can possibly be asked of a pair of table-setters. 

Jeter's 2009 season will likely go into the books as the last truly great season of his career. He hit .334 with a .406 OBP and scored 107 runs. As a leadoff man, he hit .336 with a .409 OBP. 

The 2009 season certainly was the last great season of Damon's career. He hit .282 with a .365 OBP and also scored 107 runs. He hit .287 with a .367 OBP as a No. 2 man.

What made Jeter and Damon dangerous was their versatility. They accounted for much of the .485 slugging percentage compiled by the club's 1-2 hitters, and they also did damage on the basepaths.

Jeter and Damon combined for 42 stolen bases (30 from Jeter) and were the Yankees' two best baserunners (FanGraphs) in the eyes of Ultimate Base Running (UBR), a statistic that quantifies a player's ability to make all sorts of productive plays on the bases.

Whereas we've been turning to the 2006 Cardinals for examples on how not to be, here we turn to the 2011 Cardinals. They opened the season with Ryan Theriot and Colby Rasmus as their 1-2 hitters, and that combination was doomed to fail from the start.

Theriot had posted a mere .618 OPS as a leadoff hitter in 2010, and Rasmus was (and still is) a high-strikeout guy whose best work in 2010 had come batting out of the fifth and sixth spots.

Sure enough, Theriot and Rasmus bombed as table-setters. Theriot managed a .324 OBP as a leadoff man, and Rasmu swas jettisoned to Toronto in July. In all, he posted a .318 OBP as a No. 2 man.

Tony La Russa never did settle on a solid pair of table-setters for his lineup in 2011, and that cost his offense a shot at being perhaps the elite offense in baseball. The Cardinals finished fifth in MLB in both team OPS and runs scored, as well as first in the National League in both categories.

If they'd had Jeter and Damon at the top of their lineup, their offense would have been unstoppable. And if it had been, they likely would have clinched a playoff spot before the 11th hour.

7. Load Up Corners with 100 Home Runs and .470 Slugging Percentage

It's no longer an established fact that the guys with all the power are going to play on the corners of the diamond, either at third or first base or in left or right field. Power hitters have invaded positions like shortstop and second base that had once been reserved for 90-pound weaklings.

But more often than not, you still find the power hitters playing close to the foul lines, and adding power at the corners should still be a goal for would-be champions.

Prospective champions should look to get about 100 home runs and a slugging percentage in the neighborhood of .470 from their corner players.

The six teams that got at least 100 home runs from their corner positions managed a slugging percentage near or over .470. Certain circumstances kept two clubs from joining the party.

The 2005 White Sox were kept short in both categories because they played Scott Podsednik, a center fielder by trade, in left field. He went homerless and managed just a .349 slugging percentage.

Center fielder Aaron Rowand, meanwhile, hit 13 homers with a .407 slugging percentage. Switch him and Podsednik around, and the numbers look rather different.

The 2007 Red Sox figured they were going to get 35-40 home runs out of left fielder Manny Ramirez, but he hit only 20 home runs and saw his 2006 slugging percentage experience a 126-point drop.

As for how power on the corners can be done right, this is where we get to tip our hats to the '06 Cardinals.

Albert Pujols had one of the finest seasons of his career in 2006. He hit a career-high 49 home runs and led the NL in slugging at .671. Scott Rolen had a solid year at third base, hitting 22 homers with a .518 slugging percentage. Primary right fielder Juan Encarnacion hit 19 homers with a .443 slugging percentage.

Left field was a little bit of a different situation for the Cardinals in 2006, as it featured a revolving cast of characters. Chris Duncan, however, hit 13 of his 21 homers and slugged .605 as a left fielder.

The '06 Cardinals go to show that a team doesn't necessarily have to get 25 homers and a .470 slugging percentage out of each corner spot. Getting a ton of production from one spot and then solid production in the other three spots works just as well.

And yes, it also helps to have one of the greatest hitters ever on your side. It's a shame there aren't more of them to go around.

Two teams came close to satisfying neither end of the 100/.470 rule: the 2010 and 2012 Giants. Because the Giants are the class of the league right now, this may look like an excuse to shoot the 100/.470 rule out of the sky.

But it's not. The elephant in the room is AT&T Park, which is death on power hitters. Many balls that are homers in other parks may not even make it to the warning track in San Francisco. According to, AT&T Park was the 20th-worst park for home run hitters in 2010 and the worst park for home run hitters in 2012.

Giants general manager Brian Sabean knows to tailor his roster to his ballpark, and he did a particularly masterful job of that in 2012. The Giants finished dead last in the majors in home runs, but they found the gaps better than any team in baseball, leading the league in triples with 57.

Take these two Giants teams out of the equation, and the 100/.470 rule for corner players looks like gospel rather than just a neat idea.

8. Strength Up the Middle Is Roughly 12 Wins Above Replacement

Important: Please note that the following numbers were compiled before and FanGraphs agreed to a unified replacement level. The exact numbers are a little different now, but only a little. Just a few decimal points in either direction so far as I've seen.

They say you need to be strong up the middle in order to win ballgames. If you're new to baseball, just happened to scroll down to this point in the article, that means being stout at catcher, shortstop, second base and center field.

Modern stat-heads have debunked many old-timey baseball proverbs, but not this one. Jay Jaffe of and Baseball Prospectus wrote in a 2011 ESPN Insider piece that it is indeed important to be strong up the middle, and he noted that even Bill James has sounded off on the idea.

To measure strength up the middle, Wins Above Replacement is as good as it gets. Up-the-middle players have a wide range of skills, so why not evaluate them with an all-encompassing stat?

Whether you use's WAR or FanGraphs' WAR— has a helpful comparison chart—true strength of the middle amounts to roughly 12 Wins Above Replacement.

Of these 10 teams, six got at least 12 WAR (either version) from their primary up-the-middle players. FanGraphs WAR put the 2005 White Sox, 2007 Red Sox and 2011 Cardinals about close enough to 12 WAR. 

A couple teams almost found out the hard way that not being strong up the middle can be a death sentence, as the 2004 Red Sox didn't take off until they put Orlando Cabrera's defense at shortstop, and the 2012 Giants didn't take off until Marco Scutaro took over at second base. 

Much of the WAR accumulated here came courtesy of offensive contributions, but defensive contributions must not be overlooked. That eight of these 10 teams posted Defensive WARs (calculated by and based off the Defensive Runs Saved metric from Baseball Info Solutions) of at least 3.0 shows that teams can't get away with putting only offensive-minded players up the middle.

The 2008 Phillies are an excellent model for teams to follow. Second baseman Chase Utley, shortstop Jimmy Rollins and center fielder Shane Victorino were all highly productive all-around players in 2008, particularly Utley.

Utley finished 15th among all hitters in Weighted Runs Above Average—or wRAA, which both and FanGraphs use as a primary batting metric for calculating WAR—and he was Philly's best defensive player with a dWAR of 3.5.

Rollins and Shane Victorino finished right behind Utley on the Phillies in dWAR, and only they finished ahead of Utley in UBR (FanGraphs).

So up the middle, the Phillies had their best offensive player, their three best defensive players and three best baserunners. The only area in which they were lacking was at catcher, as Carlos Ruiz and Chris Coste didn't bring much to the table on either offense or defense. 

It wasn't all bad at catcher, though. According to, Ruiz had a 3.85 catcher's ERA in 2008, which tied for fourth-best among MLB catchers. While Rollins, Utley and Victorino put up insane numbers, Ruiz took very good care of Philly's pitching staff. 

As for how strength up the middle is not done, we go to—you guessed it—the '06 Cardinals. They had a catcher and two second basemen who couldn't hit, a shortstop who could only hit singles and a center fielder whose career was in the early stages of a sharp decline.

Yadier Molina gave the Cardinals strong defense behind the plate, but his wRAA checked in at minus-27.0, and he was as big a liability on the basepaths as the next catcher (FanGraphs).

Both Aaron Miles and Ronnie Belliard were in the red in terms of wRAA as well, and they also provided little value on the basepaths. They were only strong on defense, combining for a 1.0 dWAR.

David Eckstein was also solid on defense with a 0.9 dWAR, but he compiled a minus-8.5 wRAA and only a 2.0 UBR. 

Thanks in part to waning health, Jim Edmonds split time in center field with So Taguchi. In the end, he was only good for a 5.7 wRAA, and he played poorer defense than usual with a minus-0.2 dWAR.

So again, don't be like the '06 Cardinals. Unless you plan on hiring La Russa to be your manager, that is.

Final Thoughts

Winning the World Series is not X + Y = Z. It's more like A + B x C/D + XY/(CB x A) x 3.14 = Z.

Is it a perfect formula? No, it's not. If it was perfect, baseball would be more of a computer program than a game. Everything would work perfectly all the time forever, and that would make the game about as much fun to watch as paint drying on growing grass.

However, baseball does resemble the Matrix in an odd sort of way. There are strange variables and rule-breakers (shakes fist at La Russa) here and there, but by and large, all things flow according to the numbers. In this case, the statistical trends that tie the last 10 World Series winners hide in plain sight. 

Numbers are the lifeblood of the game, and the numbers above are the DNA of a World Series champion.

Special thanks to and FanGraphs for the numbers. 

If you want to talk baseball, hit me up on Twitter.

Follow zachrymer on Twitter