Category Archives: statistics

[game 101] Perfection

In his last start, against the Tampa Bay Rays, Mark Buehrle was perfect. And perfection is something so rare and beautiful that all we can do when we’re near it is try to seal it tight to preserve it, and then ask and hope and plead that we can have it again.

Tonight, starting against the Twins, Buehrle complied with our wishes for five more innings, running his streak to 45 straight batters retired, extending back to the start before the perfect game.

I expected quite a different game; I expected Buehrle to show either fatigue or lack of focus. I even thought the Twins had been especially lucky to catch him after his feat. But he was both commanding and loose, easy, and fluid. His quick smiles over to his infielders and his consummate ease on the mound showed me the Twins were in trouble.

A perfect game is so rare I can toss out the entire, small treasury of stats on the subject in a few sentences. There have been 18, over 129 years. Two in this decade, the last before Buehrle’s five years ago, thrown by Randy Johnson, at 40 the oldest pitcher to do so. We once went 34 years without one, from 1922 to 1956, and the drought ended in the most spectacular perfecto, Don Larsen’s World Series gem of all gems.

There are a few hall of famers on the very short list—Cy Young, Sandy Koufax—but a surprising number of near-nobodies—Len Barker, Mike Witt. And that fact gives a clue about how and why a perfect game comes to be.


You will need, first of all, a willing opponent, a team to pick on. The Twins have offered themselves up to such service twice, as have the Dodgers. David Cone was able to feast on the Expos, and Kenny Rogers had the Angels of 1994. In pitching a perfecto against the Rays, Buehrle conquered a mighty team at a mighty time.

You will also need a few great fielding plays. Dewayne Wise provided one for Buehrle by robbing Gabe Kapler of a home run. The fielders, if they are typical of our human species, will grow tenser as the innings mount up, knowing that they may be called upon to save the gorgeous artifact the pitcher is creating. And while they wait for this life-and-death duty, they are often very idle in the field while the pitcher deals out strikes and coaxes easy two-hoppers.

Finally, you will need luck. There is no other word for it. In the long journey of collecting 27 outs, way too much can happen. A pebble can happen; a bunt can happen (poor sportsmanship, most would say); a dropped popup can happen. The perfect game requires the stars to align.

So, tonight, as Buehrle notches out after out, I pause to ask myself, What is it I am watching here? No pitcher has thrown two. And back-to-back-ism is just out of the question—that would only be occurring in one of the twelve extra dimensions conceived of in string theory, right?

But the innings roll on. Bear in mind, Buehrle has a no-hitter to his credit as well as his perfecto. And he has gotten just about every Minnesota batter to chop a frail grounder to his alert and able infielders. Through six innings, all Beuhrle has done is get the ball, throw the ball, and retire the batter. He throws strikes and he works at a brisk and apparently happy pace up there.

In the sixth, he clicks off two more outs, and faces Alexi Casilla. Not a formidable hitter under any conditions, but against Buehrle tonight you give Casilla the odds of a bug near a moving windshield.

Buehrle gets two strikes on him, no balls, and you are ready for him to notch his fourth strikeout and end the inning. But Casilla watches a ball, then lays off another outside pitch. He fouls the next pitch off, and then is curious and patient enough to work the count full.

Buehrle throws a pitch inside and we have a walk. It’s truly inside, but it’s close enough to demonstrate the tissue-thin border between baseball success and failure. End of perfection.

As these things often do, the crack widens fast. Denard Span hits a clean and solid and no-doubt-about-it single to center—end of the no-hitter. With two on, Joe Mauer hits an authoritative double that scores Casilla—end of the shutout, and the game is tied.

Buehrle completes the inning by getting Justin Morneau to ground out with the same trademark gentle chopper he’s induced all night. And the White Sox bring him back up for the seventh. This isn’t a pretty inning, for Buehrle will dispense a hit by pitch, three singles, and three runs. After getting another run off Octovio Dotel, the Twins lead 5-1, and Buehrle has no chance of a win after five magnificent innings and two lousy ones.

Scott Baker spent the entire night much in Buehrle’s shadow. As he’s been doing lately, Baker grinds through his outs, throwing a lot of pitches and engendering a lot of foul balls. He gets the outs, mind you, but he needs a lot of pitches to do it and is always on the brink of a sudden homer.

Baker pitches six complete innings and needs 115 pitches to do it. His only blemish is a solo homer from Jermaine Dye in the sixth. Like Buehrle, a long night of excellent pitching doesn’t get on his won-loss record, for the win goes to reliever Jose Mijares, who presides during the Twin’s outburst in the seventh. Joe Nathan gets the save and the Twins are now tied with the White Sox at two games behind Detroit.

The quest for the postseason is advanced by the win, and the rivalry with the Sox is tilted our way by the two wins in the series. But the scoreless game that played out over the first five innings, and even the single runs that each team scored in the sixth, were much more interesting than the mere won-loss result. Baker fought every batter tooth and nail, while Buehrle tossed bait over the side that the Twins quickly nibbled. Baker was trudging and careful; Buehrle relaxed and cool. They both got similar results, but pitched so differently.

Minnesota fans consider the Chicago rivalry their most bitter. It was just last season that we came down to a single game playoff against the Sox and lost the chance to enter the postseason. So when Buehrle got genuine cheers and a standing ovation when he left in the seventh, it was a moment that reminded me of the honorable nature of baseball. He tipped his cap, and I tip mine.


[game 63] The Sacrifice

The Twins won today, playing a day game at Chicago to start their weekend series against the Cubs. I salute day baseball, but I also can’t usually watch it. The box score tells me Kevin Slowey has gotten himself to 9-2, Joe Nathan has gotten a save, and Joe Mauer and Jason Kubel both homered. Yay, Twins.

But I’ve decided I’ve got to mix up this blog a bit. My initial thesis was that something specific and unique occurs in every game; that it’s possible to write an insightful little essay that helps polish the definition of baseball with the raw material of each game. It’s proving harder than I thought.

Lacking much in the way of reader interaction, I began to swing the blog toward something it appeared my husband enjoyed: chronicles of game outcomes, written with as much pep and wit as I could muster. I stopped ruminating on managerial strategy or pitching statistics, not least because the entries I produced on those subjects needed more research and polish than I could given them in a single night. But in abandoning the theoretical investigation of how baseball is played in favor of a fan’s delight in the specific exploits of Mauer and Morneau, I didn’t have much new to say each night.

I’ve been brooding on this, and looking for solutions. What I’m going to try for the next week is writing about different teams. I will, of all things, cut the Twins loose for a while, though I’ll monitor their box scores closely.

Tonight I watched the Tigers visit the Pirates in Pittsburgh. The game had instant appeal—the Twins are about to face the Pirates next week, and the Tigers constitute our primary foe in the division. And, fun sports parallel fact: the Pittsburgh Penguins simultaneously played the Red Wings in the Stanley Cup finals in Detroit. Major upset there—the Penguins won, and I would like to call into question the planetary alignments that have visited so many woes upon Detroit while allowing so little respite via their sports teams. You will recall that Michigan State couldn’t overcome North Carolina in March Madness; now the Red Wings. Dang!

But only baseball really concerns us here. I just can’t root against the Tigers, despite the fact that somehow that Twins would need to climb over them to win the division.

Rick Porcello, the rookie righthander for the Tigers, is having a solid season. He cruised to a 9-0 victory against the Twins in early May, and I essentially discounted his skills in favor of seeing that game as very hot Tigers versus stone cold Twins. But I see tonight he has promising pitching skills. Porcello has had some tribulations, but since two especially bad outings in April, he’s been working his ERA back down. It’s currently 3.71, and he’s 7-4.

Tonight Porcello had something of a dream experience with interleague play. He won a close game 3-1, pitched 7 innings (after generally not lasting past 6), and had two RBI singles to do as much as he could to win the game by himself. A sweet night for the 20-year-old.

But my main point of focus was on two instances in which runs weren’t scored. In the third, with the Tigers up 1-0, the Pirates led off with two consecutive singles. Pitcher Ian Snell then sacrificed to advance the runners. It’s the typical choice for any pitcher in the NL playbook, and it got the Pirates to the top of the order with one out and men on second and third.

It didn’t work, though. Andrew McCutchen grounded into the worst kind of fielder’s choice in which a runner is out at home. The next batter grounded out to end the inning.

There’s a reason there are four bases and three outs. You can’t keep trading outs for advancing runners.

In the top of the seventh, it’s the Tigers’ turn to try a sacrifice. In this case, Placido Polanco led off the inning with a single. Batting second, Don Kelly was asked to sacrifice Polanco to second. Now, I happen to know nothing about Don Kelly. Perhaps he’s an especially weak hitter and should be used as an NL pitcher is at the plate—asked to sacrifice in order to avoid grounding into the double play. But he hits second in the batting order and I think a bit more should be expected.

The game at this point is 2-1 Tigers, and I know intellectually and emotionally that the sacrifice is the wrong play. Pete Palmer and other baseball mavens have been slicing and dicing this one up for the last 20 years, and the central truth that emerges statistically is that trading outs for bases does not increase a teams chance of scoring. Generally, it isn’t even neutral—it makes it less likely a run will score.

As this inning plays out, we tally up another failure of the sacrifice strategy: mighty Miguel Cabrera, for whom Kelly might have been making way, grounds out and Polanco hikes over to third. I won’t credit Cabrera with a sacrifice because the hit was toward the shortstop and could have cost them Polanco. With the man on third and two outs, Magglio Ordonez must hit successfully or the Tigers lose the opportunity. Yep, it’s a routine grounder.

The moral of the story is that outs are more valuable than bases. Giving one or two hitters a chance to bomb sac flies or place well-engineered bunts is only deferring the moment when a player has to get a true, honest hit. Yes, I’ve seen runners add in a steal or a wild pitch to make that four-base journey without using up all the outs as sacrifices, but as the Win Expectancy calculations from Baseball Prospectus show, rare are the times when playing for one run via a sacrifice succeeds.

The tabulations have all been done, but managers still signal the sacrifice. There are some nuances that the statisticians haven’t accommodated, chief among them the fact that certain hitters are more likely to be successful at accomplishing the sac play than any other. The National League gives us an example with every pitcher, but the Twins have a ready-made sacrifice machine in Carlos Gomez.

So sometimes the play is not a technique for advancing the runner so much as it is a way to avoid letting a hitter bollix up an inning that has a scoring opportunity. It’s also worth noting that anything which moves a runner from first to second greatly diminishes the likelihood of a double play, giving the next hitter a bit better opportunity as well.

There’s justification for the sacrifice, but every time I see one tried, I ask if losing the out is truly the better situational decision. The best justification I can give is that scoring in baseball is so difficult that anything must be tried, but tonight no runs were attributable to sacrifices, and several men were left on base when the sac was tried.

So, the perfect design for frustration: four bases, three outs.

[Game 23] Second Inning

If one team actually played the game perfectly (which would mean the other team couldn’t, quite), baseball would be dull. Tonight, in the first of a weekend series against the Royals, the Twins produce a dull second inning and eight other exciting ones as they hoist themselves over .500.

Kevin Slowey starts for the Twins. In his last outing against the Indians, he was masterful and he has assembled a pretty 3-0 record, though the journey there has required some nail biting.

But the big high hope in the Metrodome is the long-awaited return of Joe Mauer. Mauer’s back woes kept him from spring training, let alone the regular season, so he arrives tonight after a relatively brief limbering up in the minors.

When Mauer comes up to bat for the first time in this month-old 2009 season, the happy Twins fans flash welcome back signs all over the Metrodome. My theory is that Mauer is the secret sauce in our lineup. He gets on base, Morneau et alia drive him in. The singles hitters lower in the order get on base, he drives them in. The hitting flow hinges on Mauer.

My theory has to wait tonight, because Mauer decides to demonstrate his value a tad more directly. Sidney Ponson disposes of the first two Twins hitters easily, then faces Mauer. His first swing of the year is a homer to left. It doesn’t, it absolutely doesn’t, get any better than that.

Mauer is a marquee player but not for his power—he hit all of nine homers last season. He is a pure average guy, with a batting eye that spares him from strikeouts and allows him to outfox the average pitcher. Tonight it’s as if he wants to say he’s glad to be home in the most convincing way possible.

We go to the second, the Twins ahead 1-0. According to Baseball Prospectus, the fewest runs are scored in the second inning. It’s your best bet for getting to and from the concession without missing a score. The only reason not to take the extra time to get the nachos along with the beer is that the second is generally over very quickly.

Now, why should this be so? Even if the first inning included a hit or two, we’re in the heart of the batting order or at least safely above the very bottom. Both the record books and observation show that starting pitchers often stumble in the first, searching for a groove. Do they invariably suddenly find it by the second? So much so as to become Terminator-like?

If the reason lies in the batting order, it may expose the hard-to-prove interdependence of a hitting sequence. It’s not unusual for the cleanup man to lead off the second, or his protector, the number five hitter. Both are supposed to be capable of power; both should be RBI hounds. With no one on base ahead of them, it’s like the light goes off. They are hitting machines who power down when the circumstances don’t remind them of why the hits are needed.

If the reason lies in the pitcher’s intensity, it suggests that the first inning is some kind of practice swing. The average game in the middle of the average season shouldn’t inspire too much in the way of jitters, but perhaps pitchers really need a little ritual of first-inning stress. I don’t have access to the type of data that would frame the question, but it would be interesting to know if a low-scoring second follows a messy first, or if the second is just the Zen-trance pause on the way through the game.

If they needed to audition, the second inning is what most pitchers would put on their highlight reel. Here is what Sindey Ponson does with his: throws 10 pitches, gets a strikeout and two groundouts. Wait—Slowey tops him. Six pitches with a strikeout.

Slowey’s second is positively gorgeous. He faces the Royals’ toughest bats and mows them down. Jose Guillen, the cleanup hitter, pops out on the first pitch, swinging so greedily and foolishly that he has to walk back to the dugout in cloud of gloom.

Billy Butler, the burly DH the Royals have elevated to first baseman, takes a ball. It seems to be a little experiment on Slowey’s part; he’s toying with Butler. The next pitch Butler grounds harmlessly over to shortstop Nick Punto, who makes the throw with time to spare. Basic baseball.

Slowey has used three pitches so far when Mike Jacobs, batting sixth as the DH, steps in. A humming fastball grabs the right edge of the plate as the lefty Jacobs absorbs the news: control pitcher. Mauer sets up a little more inside and the ball rides up a little higher and tighter. Strike two and all Jacobs can do is stare at these things. I assume we’ll go to the changeup, but Mauer and Slowey keep it simpler still. It’s another fastball running just a bit more inside. Jacobs was paralyzed the entire at-bat. You would think Slowey might have a hypnotist act in the offseason.

It was about the most perfect second inning a pitcher can pitch, particularly coming right after the pure high of Mauer’s home run. Beautiful but, strictly speaking, boring. Between them, Slowey and Ponson toss 16 pitches and collect their outs so briskly they look lazy up there. For the game to be great, struggle must be part of it.

And this game has some struggle for both teams. The Royals are hardly out of it with a one-run deficit, so they erase it in the third. David DeJesus singles to drive in two runs and maybe Mauer’s feel-good return will be marred.

In the fourth, we get the M and M boys in their classic formation: Mauer doubles to lead off and Morneau drives him in. And the rest of the lineup gets into the act, pounding on Ponson to the tune of four runs to put the Twins up 5-2.

We end up with a typical Slowey outing: shimmering control, no walks, a handful of strikeouts, and too many hits too close together. Slowey generally looks terrific tonight, but he once again has a brief Mr Hittable funk, coughing up four hits in the fifth that turn into three runs to let the Royals tie the game.

He completes only five innings, gives up a total of eight hits and five runs. It’s enough to bring him to 4-0 after we get another does of the Mauer-Morneau reunion. In the fifth, Mauer walks and Morneau cranks his homer to put the Twins up 5-7. The bullpen keeps the Royals 1-2-3-you’re-out quiet for the rest of the game. Joe Nathan gets the save, Mauer the happy headline

Slowey has perfect second and a sloppy fifth. They’re just innings, innings that show the highs and lows within a win.

[Game 22] April

Running to stand still actually can work in baseball, but it isn’t too impressive in itself. The Twins finish their month of April at .500, which happens to be good enough to put them a half game out of first in the AL Central. Getting into the playoffs is really a matter of not losing too much. And so the Twins squeak on.

There are 12 teams in the majors with worse records, 2 with identical ones, and 15 with better ones. We’re so average we’re driving a Camry and listening to Coldplay. Suddenly I feel so beige.

It’s fair to say that we’re positioned favorably in the division because the division is, in a word, lousy. The Indians had a feel-good come from behind win against the Red Sox last night, but they’ve had little to cheer about this month, with an 8-14 record.

The Royals, White Sox, and Tigers all stand over the Twins at a less-than-lofty 11-10. Guys! Just play the missing game and we’ll all be even again.

But the Royals do have some things to be very happy about this April. Zack Greinke secured another win tonight, against the Blue Jays, making him 5-0. His scoreless streak ended, but it got to an otherworldly 43 innings. He has a near zero ERA and is at or near the AL top in strikeouts. Lots of good news for the Royals, and yet the Royals are only microscopically better than the Twins. It’s a 5-man rotation, and, um, Sidney Ponson is part of it.

Then come the White Sox and Tigers. Fun, tiny fact: the Tigers are the only team with a winning record within the division, and that’s a sterling 3-2. Everyone else is .500 except the Indians suffering at 4-5. Clearly, nothing much has been decided about the fortunes of the AL Central.

The Twins have completed only 8.3% of their games in the division, so April has been inconsequential divisionally. Instead, let’s look at the month series by series. There is a bit of an uptick if you wade all the way through.

We played seven series, split one, won three, and lost three. Are the team colors going to have to change to beige and tan? The hopeful note is a little trendline I could draw, starting in the middle of the month. We sweep the Angels, fall apart in that single day series against Boston, then go 2-1 against both the Indians and Rays. Not sure this truly qualifies as momentum, but in a world of .500 baseball I’m poking under every rock for a little energy.

In the division, we lose the White Sox series, 1-2, and make up for it by taking the one versus the Indians 2-1. The absence of especially bad news isn’t the presence of good news, but we have to build what we can on this bedrock of mediocrity.

Tonight’s game was an especially easy one. We played well, on offense and defense. The Rays played miserably, on offense and defense. Nick Blackburn went seven solid innings. He gave up two runs, and it would have been only one if he’d left a bit before that 100th pitch. Pitching for the Rays, Scott Kazmir had an uncharacteristically poor night. The strike zone eluded him, and the Twins were ready, notching nine hits and eight runs.

It began with a four-run first inning, in which the Rays gave more than the Twins took. The four hits were all singles, Jason Kubel’s a particularly bloopy one. Morneau and Morales walked as we sent nine men up. Kazmir tossed two wild pitches to allow two of the runs to score. Ugly, untidy baseball.

In the fourth, hapless Akiri Iwamura, the Rays second baseman, demonstrated the way to get two errors on one play. He bobbled a grounder to allow Kubel to reach first, then tried to make the throw to first, only to sail the ball well past Carlos Pena. Justin Morneau snuck home on the throwing error.

It wasn’t all bad baseball by the wretched Rays. Denard Span hit a triple, and the sight of him racing round the bases and sliding feet first, then popping right up, was a joy to behold. Brendan Harris used his spot at second in the batting order to smack three hits and score two runs. There were RBIs aplenty: one each for Harris, Morneau, Kubel, Cuddyer, and Young.

Blackburn pitched well throughout. He gave up a run in the third, but ended that inning with a strikeout, and kept the Rays deathly quiet every other inning. Entering the seventh with a pitch count in the 80s, Blackburn finally ran into some trouble. The hits were innocuous enough, but Blackburn wasn’t hitting his spots. Only one run scored. Gabe Gross pecked out the RBI, taking advantage of two prior singles by Ben Zobrist and Dioneer Navarro. It was never a meltdown for Blackburn and he completed the inning, but it wasn’t a flourishing finale either.

The Twins had the game well in hand after the first inning, and won it 8-3. April’s over, and in one sense it’s as if we’ve gained no ground—you’re .500 on day one of the season, too. But we’ve won by not losing, and any minute now there could be a little streak of glory. 

Game 18: Almost

Tonight, a few things only almost happen. But the things that do happen give us a clearer portrait of a pitcher.

Kevin Slowey started tonight against the Indians. He enters the game with a pleasant little 2-0 record, but his 5.89 ERA and 1.64 WHIP suggest problems. In his three prior starts, he’s pitched 6, 5.1, and 7 innings. He studiously avoids walks, allowing at most one a game, and collects an average of 4 strikeouts per start. These numbers are good to excellent, but then we turn to hits. Slowey has coughed up 28 of them. And they’ve turned into runs: 5 in two of the starts and 2 in his most recent contest against the Angels.

These tiny little nibbits of data aren’t sturdy enough to predict or define much. If you have only these stats to stare at, you can conjure up two pitchers. One of them is pitching rather well but letting a lot of balls get hit into play. Luck and fielding deficiencies render them hits; the hits get clustered; runs score. The second pitcher is not in control of each at-bat. He’s not locating the ball; he’s falling behind in the count; he’s tipping his pitches; he’s leaving the ball up. The hitters are feasting; the hits get clustered; runs score.

I like baseball stats, and I like poking around in them, but I also realize that the stats we develop and the analysis we bring to them are actually largely in the service of . . . not watching the games themselves. We want the stats to distill the abilities of the players so we can know what they’re capable of in the future, and so we can compare them, all without the tedium of watching their every move. But sometimes you have to watch the game to find the truth.

I’ve seen two of Slowey’s three previous starts. He is pitcher type one, not type two. He’s had some shaky at-bats, but he is largely in command of the big fundamentals—a plan for attacking hitters, getting first pitch strikes, running out an arsenal of pitches. He doesn’t have a pronounced tendency toward either ground out or fly out, but he does have to suffer through pitching to contact, and he can’t rip through a game piling up Ks.

Now, wait just one minute, the stats partisans will say. The greater value of statistics is that they prevent us from generalization and the bias toward seeing what we want to see. I can claim to watch Kevin Slowey, but I’m watching with my partisan Twins blinders on, and I’m claiming that seeing two games is more insightful than studying the stats for three.

Fair enough. Neither vision is perfect. But tonight’s game clarifies Slowey’s status both under observation and statistically. He holds the Indians scoreless for eight innings, strikes out 7, and get a win. Then there is the matter of the almost.

Slowey accumulates hits, as we have to acknowledge he always will, but tonight he distributes them so fastidiously they can’t do any harm. He doles out 5 of them, each buried in a separate inning. The Indians never produce anything remotely resembling a threat. They trudge up to the plate, inning after inning, to collect an isolated hit sandwiched between Ks and groundouts.

The Twins are having the opposite experience. They munch up and spit out Carl Pavano and secure a 5-0 lead. They snare another two runs off Masa Kobayashi. The beefy bats of the left-handed hitters were going fine, but so, at last, were the righties. Joe Crede, for example, gets a homer, and Delmon Young is 3 for 4. Jason Kubel seems to be swinging from a Barcalounger he’s so relaxed at the plate. He hits two homers and a double, just missing another 4-hit night.

Slowey has thrown 106 pitches through 8 innings.  He hasn’t allowed a walk, and hasn’t had a single stressful inning. With a 7-run lead, he has quite a bit to work with if Ron Gardenhire wants to give him a chance for the complete game.

Every fan of baseball accomplishments wants to see him trot out for the ninth. Every fan of the Twins, if he can remain pure in the quest for victories alone, wants Slowey to enjoy some time on the bench. Passing 100 pitches almost always spells trouble, and if it doesn’t manifest itself tonight it may show up in Slowey’s next two starts.

The diligent students of pitch counts have an important qualifier, though. Fatigue is a serious problem, and leads to injury and mechanical problems. But the 100-pitch threshold does not define fatigue. Throwing a baseball isn’t the problem. Throwing a baseball when already tired is the problem. Tonight Slowey has accumulated a full game’s pitch count but he has never had a tough inning. Does that mean his arm is just fine?

We find out in the ninth, when Slowey gives up three straight singles. None of the hits were especially mighty, and they all needed a little good fortune in the positioning of the fielders, but none of them were bloops either. My head and heart are at odds. I am rooting for Slowey to get the complete game shutout; I am wincing because he could do himself some damage up there.

Gardy fetches him and brings in Luis Ayala. There are none out. Ayala’s job is not just to end the game but, if at all possible, preserve that shutout so none of Slowey’s baserunners come home to cloud his ERA. Ayala gets a strikeout with the bases loaded, the best possible start at untangling this mess. But the next hitter singles to center and one run leaks in. Ayala finishes it off with a double play.

So, almost. Almost a complete game, and almost a shutout, but not quite. We don’t get the special accomplishment, but we do get a pitching gem. It polishes up real good, even if it doesn’t fit the standard setting.


Measuring Pitchers

Glen Perkins made his third start today, hoping for his first win. His two prior outings were both good enough to result in wins, but baseball outcomes don’t go hand in hand with pitching skill.

Against the Mariners, Perkins ran into Jarrod Washburn, who happened to post a nearly identical pitching line but for one small difference: Perkins allowed one run, Washburn none. Against the Blue Jays, Perkins gave up 3 runs and once again made it through eight innings. Unfortunately, that third run, coughed up in the eighth, tied the game.

Perkins has a tiny ERA and a great WHIP, but he’s 0-1. The stories behind his losses remind me why baseball analysts work so hard to create meaningful statistics. Win-loss records don’t define a pitcher’s work but a team’s. ERA isn’t a pure assessment of a pitcher’s worth, since it rests on the assumption that all runs are created equal and ignores game situations. And ERA is cumulative, which lets one bad outing poison a dozen others.

The folks at Baseball Prospectus and that radical theorist who came up with BABIP (batting average on balls in play) have all struggled to create measuring systems that are independent of both the pitcher’s own team and the opposing team.

I was reading about these systems recently, and as impressive and elegant as they are, I also had to ask if this mania for isolating individual performance is really a noble quest for insightful measurement or a problematic desire to take a team sport and pretty much put it in a particle accelerator so we can finally find the quarks and muons. Maybe we will learn something about pitching if we can strip away defensive support, but it’s starting to sound a lot like looking for nucleons in the atoms comprising the curveball.

Still, I like these statistic forays because they invite me to look at the game more completely, and break down the forces that are at work. Setting up appropriate boundaries between what the pitcher does and does not control is the best way to put the right material under the microscope. 

I want to be careful, though, when the analytic system is trying so hard to eliminate the game situation, teammates, ballpark effects, and opponents’ skill that the individual performance we do measure doesn’t occur in a real baseball universe. It is a team sport, and the stats we really need are ones that help us see how all the parts come together to make a win or a loss, a hit or an out.

Perkins first two starts show the poverty of both W-L and ERA. His 0-1 record doesn’t express the high quality of his pitching, while his ERA papers over the serious sin of allowing that tying run. What we really need to compare Perkins accurately with other pitchers is a stat that blends his value to the team (in producing wins and pitching well when the stakes are highest) and his individual skills (unsullied by booted double plays or opposing pitchers who happen to be a little luckier or a little better on any given day).

Today Perkins finally gets his win, and he once again completes 8 innings. Now he’s pitched 24 innings, given up a total of 4 runs, 16 hits, and 4 walks. He’s 1-1, with a closer-grade ERA of 1.50. His performance this season should rank very high.

Going this deep into the game three times running is especially impressive to me. Today he needed only 84 pitches to get through 8, and 58 of them were strikes.

Perkins did just about everything necessary to create conditions in which his team could win. He held off the opposition, pitched briskly, kept runners off base, scattered the few hits and run he allowed, and required no bullpen support except for the closer’s save in the ninth. There are plenty of wins allotted to pitchers that really stem from the offense’s heroics. But Perkins did earn his win—the Twins scored only 3 runs and Perkins muzzled the Angels all game long.

So the stat I want is not the misleading won-loss, in which the offense of both teams has too prominent a part. And I don’t want the rather milquetoast Quality Start, which doesn’t express excellence as much as muddling along. I want a stat that shows what the pitcher contributes to the team, not something that tracks his work as if the journey of the ball to the plate was the only action for which he’s accountable.

The Twins have swept the Angels and pulled themselves up to .500. A 7-7 record isn’t especially handsome, but in the sluggish AL Central that’s enough to leave us but one game out of first. The White Sox, Tigers, and Royals are all tied at the top, while Cleveland is an unhealthy 4-9.

Watching the 2009 season through the fortunes of a single team makes my vision a bit blurry, but I must also note that the Twins sit at .500 after playing four series, the first three of which were against the teams that lead the East, Central, and Western divisions. Now, that’s a potentially misleading observation, because those teams lead after collecting their wins against the Twins as much as by beating other worthies. But at least we’ve been tested against teams playing good baseball.

The Twins have looked a good deal more lively since Friday’s comeback miracle. They get their first day off tomorrow, then face the Red Sox, who are stuck in the same gear with a 6-6 record. The games are in Boston so we’ll be leaving our home dome for the furious fans at Fenway.  

Why 162?

Since I remain committed to providing the results of Twins games, I will announce that they won today, 9-2 over the Angels. I cannot share this information as a witness, though. I missed the game, a little problem that is going affect this blog project many times this season.

Missing one game couldn’t be important. In fact, watching all 162 might actually be a little dangerously obsessive. One game means very little. Then again, why does it take 162 to settle anything?

The analytical tools that we use to measure players, like on-base percentage, on-base plus slugging, and walks and hits per inning pitched, are all poised to answer questions about winning and losing games. Well, that’s axiomatic: the only thing we’re interested in is winning games. But with 162 of them to play, does it matter which ones are won?

Let’s take this from the perspective of the Twins. The point of winning games is to win the Central Division and then get into post season play and win the World Series.

No one claims the 162-game regular season determines who the best team is, though arguably it could. We could rank all the teams by total wins and not bother with the World Series. Here’s where that great sabermetric phrase “You could look it up” comes into play, because the World Series is not consistently won by the team with the most regular season victories. In my memory, I don’t think it’s happened in the last decade.

The goal is the glory of the World Series, and the price of admission is the post season, and the route to the post season is winning the division. The wild card is the backdoor route, a path that is generally blocked in the American League by the big duels in the AL East. It’s not out of the question, but let’s assume that winning the Central division is the only way in.

Major League baseball now uses a schedule that emphasizes divisional play. The Twins will face the Tigers, Indians, White Sox, and Royals 18 times each. We’ll play each of the AL West teams 10 times. Then there are 32 games against AL East teams, from a scant 5 against the Orioles to a top-heavy 8 versus the Blue Jays. Interleague play gives us another 18 contests, all based in the NL Central. There will be 6 meetings with the Brewers (in pursuit of a local rivalry) and 3 each against the Cubs, Cards, Astros, and Pirates.

There are eleven teams the Twins never play: the entire NL East and NL West, plus the Reds. The unbalanced schedule means that every team will have some reasons to gripe or grin. The Twins don’t have to face the current juggernaut of the Marlins, but they also miss the chance to feast on the Nationals, or get a little hitting lift in Coors Field. You could dissect the schedule all week to look for little edges or obstacles, but a certain arbitrary grace must be accepted.

Back to the 162 games. Clearly, the 72 games in the Central itself are the most important. Every win there includes a corresponding loss by the only competitors who count. Meanwhile, teams in every other division are free to win as often as they like, as long as they lose to us.

The 13 games played so far have the Twins at a semi-miserable 6-7. In the division they’re 1-2, also lousy. But when April is over, they will have played only 6 games against Central opponents. Their best possible record, if they sweep the Indians in the week ahead, would be 4-2. Their worst would be 1-5. But the key fact is that they will only have played 8% of their Central contests. May will be a more decisive month: 15% of the divisional games will be played then.

Wins are still wins, and this little schedule discussion isn’t offered as a way to feel a bit better about the lackluster start of the season. The Twins won’t win the Central by only beating Central teams, just as they won’t lose it by repeating their weak show against the White Sox. Any team can nab the division crown by winning predominantly outside the division as long as the rest of the division loses.

As a method for determining the best team, baseball’s structure of an immensely long regular season followed by post season berths determined with an emphasis on divisional contests is far from perfect. It isn’t the staggering disappointment of college football’s BCS, but it isn’t tennis scoring either.

A typical complete baseball game gives each hitter 3 or 4 chances to do something that he will generally do only one-third of the time. To win the game, several of the hitters will have to manifest their average in a small mathematical sample. The starting pitchers, meanwhile, must prove their ERA or WHIP while facing around 25 batters, another small sample.

Ultimately, the result of a single game is a distortion of all the players’ skills. Last night, Jason Kubel hit for the cycle. Tonight, weirdly enough, he collected four hits again. That’s very, very strange—he’s not going to continue doing this. In another testament to the distortion of the individual game, the Indians scored 14 runs in the second inning, and 22 overall against the Yankees. Individual games can’t show which team is best, though they may give a decent hint as to who is best on a given day.

The 3- or 4-game series is a much more telling unit of play. That the Twins lost their first three series, to the Mariners, White Sox, and Blue Jays, is the clearest statement about how they’re doing as a team and in the standings. They’re set to win their first series against the Angels, and tomorrow we’ll find out if it’s a sweep.