Archive for June, 2010

Revisiting a Blown Call from the 2009 Playoffs

By now, just about everyone knows about how umpire Jim Joyce blew a call during Armando Galarraga’s start against the Cleveland Indians, which cost him the 21st perfect game in major league history. Instantly after Joyce’s error was discovered, fans were calling for Joyce to be fired. However, he certainly isn’t the first umpire fans have wanted removed from the game, and he certainly won’t be the last.

For those that remember, Tim McClelland had his own controversy during the 2009 ALCS between the New York Yankees and the Los Angeles Angels. Long story short, during Game 4, Nick Swisher was at third base when Johnny Damon lined out to center fielder Torii Hunter. Swisher tagged up and presumably scored, until the Angels appealed that Swisher left third too early. McClelland agreed with the Angels and called Swisher out. Here is the video of the play. (Only the first 45 seconds is necessary to watch.)

From personal experience, it seems like we blame an umpire for a bad call without ever attempting to understand why the bad call was made. From the video, it’s clear that McClelland wasn’t directly watching Swisher when Torii Hunter caught Johnny Damon’s fly ball, but no one, not even FOX announcers Tim McCarver or Joe Buck, even bothered to explain why he wasn’t looking at Swisher. Back in October, I found a possibility as to why McClelland blew the call, and I will walk you through my reasoning here.

1. McClelland’s Positioning

Note: I have worked two summers as a baseball umpire for a middle school league. I understand that middle school children and professional athletes are very different, but I still had to learn positioning for certain plays, similar to the MLB umpires. During tag-up plays, I’ve been taught to line up the runner and fielder such that I can see both when the catch is made and when the runner’s foot leaves the base. Basically, this is exactly what you’d expect.

From the video I posted above, if you time when Damon made contact to when Hunter made the catch, it’s about 3.75 seconds. Now, I assumed that McClelland hesitated before making an attempt to get into a position to accurately watch Swisher’s tag because he had to decide on an appropriate reaction to the batted ball (similar to a fielder deciding what he has to do to field a batted ball). This is understandable because a person cannot be expected to know where or if a batted ball is going to be caught the very instant it is hit. Therefore, I assumed that it took McClelland about half a second to decide where the ball was going to land and if it had a possibility of being caught, so he should have had about 3.25 seconds (3.75 seconds of ball flight – 0.5 seconds of hesitation) to react.  However, it’s tough to estimate this time because we don’t know exactly what he was thinking when Damon first made contact.


As for his starting position, it is virtually impossible to know where McClelland started because there isn’t any video I found that showed where he was at the beginning of the play. However, because of the rarity of a pickoff attempt at 3rd base and he was still in motion when Swisher left 3rd, I decided that he wasn’t close to the base when the play started. So, another assumption I made was that he was probably a distance from 3rd base that roughly mirrored the positioning of the 1st base umpire when Damon put the pitch into play.

The fact that McClelland was moving when Swisher left 3rd base poses a question: Why wasn’t he in position yet?

A. He hesitated longer than the 0.5 seconds I assumed and he didn’t move fast enough to compensate.

B. His hesitation was close to my 0.5 second estimate, but he was slow in moving to his position.

Look at the picture above again. The center of the oval where I think McClelland started to where he ended up at the time of Swisher leaving 3rd is roughly 1/4 the distance of the basepaths, or about 22.5 feet. Now, from my quick Google research, I found that the average walking speed is about 3 MPH and the average jogging speed is about 6 MPH. Accounting for McClelland’s age, I decided that his jogging speed was about 4.5 MPH. Doing the math, he should have taken about…

(4.5 mi/hr) x (1 hr / 3600 sec) x (5280 ft / 1 mi) = 6.6 ft/sec

22.5 ft / (6.6 ft / sec) = 3.41 sec

…to move from where he started to where he ended, which is very close to the 3.25 seconds I estimated earlier. So, I’m willing to bet that the correct answer was closer to B than A; he hesitated an appropriate amount of time, but didn’t move fast enough to get into position.

McClelland certainly could have gotten into a better position like I mentioned above, but only if he moved at a faster speed. However, from all of us watching many games, I’m sure we can all agree that umpires are not the fleetest of foot and rarely, if ever, even appear to move at a fast jog. I bet a jog for McClelland is probably a slow jog for the average person. Therefore, I think our answer evolves from B into a C that I didn’t even consider listing:

C. He did nothing wrong. The play just happened too fast, so he was in his best possible position.

2. McClelland’s Vision

No, I’m not recommending that he needed glasses (even after wrongly calling Cano safe at 3rd in the same game). Tim McCarver emphatically stated how McClelland wasn’t directly looking at Swisher, which caused his error in judgment. But what I couldn’t believe was that McCarver, nor Joe Buck, nor anyone else not even related to the FOX broadcast made any mention of peripheral vision in relation to this play. For this analysis, I found that this organization states that normal peripheral vision is about 180 degrees. Examine the following two pictures:


As you can understand, a person’s peripheral vision should decrease as he/she gets older, so I accounted for this by showing McClelland’s as being less than 180 degrees in the picture on the left. I know that I subjectively picked where the two lines are, but they are not intended to be exact nor did I even know what McClelland’s vision was like (neither should you), so I estimated that he saw at least part of Swisher before he left 3rd. Now, I’ve already shown that it was probable that the play in real time occurred too fast for McClelland to get into a good position to make an accurate call, so that was probably why he didn’t line up Swisher with Hunter when the catch was made. The picture on the right shows Swisher leaning forward in anticipation of leaving 3rd base. Once he started leaning, I think McClelland assumed that Swisher was off the base, and thus thought that Swisher had left the base before the catch was made. If the time of ball flight had been longer, McClelland could have gotten into a better position, and he most likely wouldn’t have wrongfully called Swisher out.

In review, I found that the time between when Johnny Damon made contact with Scott Kazmir’s pitch to when Torii Hunter caught Damon’s fly ball occurred too fast for Tim McClelland to properly move into position to line up Nick Swisher at 3rd base with Hunter in center field. With the probability that McClelland had declining peripheral vision, just like many people his age, he saw Swisher lean forward out of the corner of his eye, and thus thought that Swisher left the base much earlier than he actually did. Together, I feel that McClelland did the best that he could in making the correct call, but the play simply happened too fast for him.

Even if you feel that I made too many assumptions here, my main point of this article was for you to learn to understand why an umpire made a certain call before jumping to conclusions that he was out to get a certain player or team, or that he’s incapable of being a good umpire in MLB. You don’t need to go into as much analysis as I did here, but you can at least watch some replays on TV and see if they hint at why an umpire made a mistake. Umpires are not “out to get” particular teams or players. I want you to believe that these guys really are trying the best they can.

This article was originally posted on Off The Mark in October 2009. Portions of the article were rewritten for cohesiveness and relevance to the present.

The Year of the Pitcher? A Holistic Analysis

Thus far, the Year of our Sport 139 — or the Year of our Lord 2000 and 10 — curious whispers have grown to sly murmurs, and in unity they portend: the Year of the Pitcher. Already, fans have indulged in the sight of two perfect games and a third de facto perfect game. By contrast, this time last season, we fans were discussing Albert Pujols’ interminable Power and Joe Mauer’s surprising Pop, all the while swirling the snifter of Slugging; but this year, tales of a Resurgent Carlos Silva and the dazzling Kid Stephen Strasburg have seized our headlines. This, the media assures us, is the Year of the Pitcher.

But is it? Perhaps it is the year of the Pat Burrell — the aging slugger — or perhaps the year of the missing needle? One thing is certain: Our teams are scoring fewer runs. The 2010 MLB’s runs per game has reached pre-Clinton lows:

Of course, taken with a tankard of greater perspective, this recent descent does not appear too outrageous. It in fact puts us closer in line with historical performances:

Read the rest of this entry »

Larry Rothschild and Strikeouts

The legend of Dave Duncan is well known.  Pitcher A stinks and is released by his current team.  The Cardinals sign Pitcher A and he miraculously becomes a good pitcher.  Dave Duncan’s effect on these pitchers can be seen in the groundball rates.  Another NL Central pitching coach has a similar effect on pitchers’ strikeout rates.  Every year from 2001-2008, Cubs’ pitchers led all teams in strikeouts.  In 2009 they finished tied for second.  The northside pitching staff has seen plenty of turnover throughout those years, but the one thing that hasn’t changed is Larry Rothschild.  He has been the Cub pitching coach since 2002.

So does Rothschild really have an impact on his pitchers’ strikeouts.   To find out I compiled a list of all pitchers the Cubs acquired from outside the organization between 2002 and 2010.  According to the Sabremetric Library, K/PA becomes reliable after 150 batters faced.  After limiting my list to only pitchers who faced 150 batters as a Cub, I found their K% before they joined the team and during their time with Rothschild.  Here is the list.

Pitcher          K% Before    K% After    Difference
Matt Clement       17.60%       23.20%      5.60%
Antonio Alfonseca  14.70%       17.90%      3.20%
Shawn Estes        17.90%       14.70%     -3.20%
Mike Remlinger     22.30%       24.80%      2.50%
Greg Maddux        17.20%       15.50%     -1.70%
Glendon Rusch      16.70%       17.20%      0.50%
Latroy Hawkins     14.70%       19.90%      5.20%
Kent Mercker       15.60%       22.90%      7.30%
Ryan Dempster      18.00%       21.10%      3.10%
Jerome Williams    14.70%       12.30%     -2.40%
Bob Howry          20.30%       21.00%      0.70%
Scott Eyre         17.00%       23.60%      6.60%
Ted Lilly          19.60%       20.60%      1.00%
Jason Marquis      13.90%       12.60%     -1.30%
Neal Cotts         19.90%       22.70%      2.80%
Rich Harden        23.30%       29.10%      5.80%
Aaron Heilman      20.40%       20.80%      0.40%
Kevin Gregg        20.80%       23.80%      3.00%
Tom Gorzelanny     14.80%       23.70%      8.90%
John Grabow        20.50%       15.30%     -5.20%
Carlos Silva        9.80%       17.30%      7.50%

Only 5 out of the twenty-one pitchers in the list saw their K% decrease under Rothschild.  I assume Maddux didn’t learn anything he didn’t already know.  Clement, Gorzelanny, and Silva are the big ones.  Clement’s three best years were with the Cubs, and Gorzelanny and Silva were borderline major leaguers when they arrived in Chicago.  Even though he had Kerry Wood, Mark Prior, and Carlos Zambrano during the strikeout streak, we need to give Larry Rothschild credit for his influence on the high strikeout totals.

Strasburg’s Debut vs. 29 Other Clubs

What if Stephen Strasburg had debuted against a team that can actually hit? In what should strike anyone as a ridiculous criticism, there have been a few people to point out that Strasburg didn’t exactly face a “real” major league line up in his record breaking major league debut. (A friend of mine joked that he made his 7th AAA start on Tuesday.) Yes–record breaking. I’ll add to Jack Moore’s point: Since 1920, only there have been only 67 games in which a  pitcher has struck out 14 or more batters and walked none. Strasburg’s outing was noteworthy because his is the only one in which that happened with 24 or fewer batters faced. Given that only 66 other pitchers have ever done something like what Strasburg did in his debut, the level of the team he faced seems like a pretty trivial point. This was dominance like we rarely ever see.

Nevertheless, what if Strasburg had faced a “real” line up in his debut? One of the beauties of sabermetrics is that we get to have this argument with math. If, against a real line up, you think he would have looked ordinary and I think he would have looked pretty amazing, we can set aside arbitrary opinions, lay out some points of agreement, and use our calculators to answer the question. Well, that’s a stretch. But at least we can get a sense of the significance that the Pittsburgh lineup made.

Strasburg was obviously on last night. Did he bring his best stuff? Maybe. What we saw last night wasn’t his true talent level. Nobody is that good consistently. Let’s call the talent level Strasburg brought to his debut his instantaneous talent level. That instantaneous talent level faced was in the run scoring environment that the Pittsburg Pirates create. The comination of the two was a .194 wOBA.  Sabermetrics gives us a tool for calculating match-ups known as  a log5 calculation. If we assume that a .298 wOBA is the Buc’s real talent level, we can isolate Strasburg’s instantanous talent level and give the most rational possible answer to the question “what if he’d faced a real line up?”

I’ll cut to the chase and save you some algebra: his instantaneous wOBA-against was .218.

Going back to our log5 calculations, that means the Yankees, if they brought their MLB leading .361 wOBA to face Strasburg last night would have wOBA’d .248. That’s something like the Astros without Lance Berkman or Hunter Pence.

Here’s the same calculation for every other team in the league.

Yankees		0.243
Red Sox		0.240
Reds		0.234
Brewers		0.227
Tigers		0.226
Twins		0.225
Blue Jays	0.223
Rays		0.223
Braves		0.222
D-backs		0.220
Rangers		0.219
Cardinals	0.219
Phillies	0.219
Rockies		0.219
Royals		0.217
Nationals	0.216
Dodgers		0.216
Cubs		0.215
Marlins		0.213
Mets		0.213
Angels		0.212
Giants		0.212
White Sox	0.210
Athletics	0.210
Padres		0.202
Indians		0.202
Orioles		0.196
Mariners	0.195
Pirates		0.194
Astros		0.182

Anyway, we’ll never know what St. Stephen would have looked like against one of 29 other clubs on June 8th, 2010. That’s not the point. The point is that we witnessed one of the great pitching performances in the history of baseball. It was dominance. This post sheds a little light on what dominance means.

The Nationals’ Unique Fanbase

Tom Verducci, in a recent article on Bryce Harper, mentions that the Nationals averaged only about 12,000 households viewing each home game last season.

It occurs to me that the Nationals may be the only team in the country where the “fan base” is more likely to go to a game than watch it on TV.  After all, the Nationals to a certain extent positioned their new stadium as a prime location for D.C. power players to have business meetings and discuss the future of our Great Nation.  Of course, the Nationals have not been putting a great product on the field of late, which will diminish any team’s fan base.  But the Nationals’ current path, of increasing respectability borne on the back of several marquee names (Strausberg, Zimmerman, Zimmermann, and now Harper), is precisely the sort of attention-grabbing roster construction that would make an afternoon ballpark business meeting trendy.  Perhaps more than any other city, the Nationals have access to a unique demographic, one with money to spend but questionable rooting interest in the team.

To investigate this, I found stadium attendance and TV ratings from the 2009 season.  The bigger the ratio of game attendance to TV households, the larger the percentage of assumed fan base attends games:

The Nationals were the only team who averaged more fans in the seats than households tuning into the game (the shocking part of this is that their TV ratings were up 67% over 2008).  The Yankees and the Red Sox, as expected, were at the bottom.  The Marlins and Rays both had two different cable networks (FS Florida and SunSports) showing their games, which increased their household viewing numbers.  The Braves’ large number is due to their TBS days and two cable networks (FS South and SportSouth).  The source I used did not have television numbers for the Blue Jays.

There are a couple of factors likely working against the Nationals here.  One is that they are a recently-transplanted franchise which has not had the opportunity to build deep roots in its new city.  The team that arrived in Washington in 2005, and the stadium in which they first played, did them no favors.  Nevertheless, no other team comes even close to the Nationals’ ratio.

I’m no economics major, but these numbers seem to suggest that certain teams are pricing their tickets appropriately.  The Athletics and the Nationals’ average 2009 ticket prices, $24.31 and $30.63 respectively, resulted in the highest ratios of game attendance versus TV audience.  In those cities, it seems, ticket prices are encouraging fans to watch games in person.

There is another interesting aspect to this data.  Much has been written recently about how a new stadium no longer “saves” a team.  Baltimore and Cleveland’s new stadiums in the 1990s ushered in many years of big crowds and increased revenue.  Writers have pointed to Pittsburgh and indeed Washington as examples of how the novelty of a new stadium is wearing off faster these days.

Yet look at Baltimore and Cleveland.  They rank 4th and 9th respectively in ratio of game attendance to television audience.  Their beautiful ballparks are still saving them from an even more precipitous decline in fan base interest.


tv numbers:

attendance numbers:

A version of this article first appeared on my blog.

A Proposal for Replay in 30 Seconds or Less

You probably already know why I’m writing this post (but if you’re reading this in 2014, Armondo Gallaraga was robbed of a perfect game in the 9th with 2 outs on 6/2/2010.)

Replay gets discussed a lot these days, and there are those in favor of it and those against it. The main reason for it is that replay is more accurate than an umpire, which has been demonstrated over and over again. There are two reasons against it. One is the tradition of umpires, the other is that it will slow down the game. As far as I can tell, it’s futile to argue with people when they love the tradition enough: if you’re sufficiently committed to tradition, no other value will persuade you to give up your stance. I don’t share that love of tradition, but I won’t enter a futile argument here either.

The lost time due to replay is different. We can count seconds and we can try to balance lost time with increased accuracy. Moreover, as technology improves the amount of time lost rewinding tapes and whatever else they had to do in the NFL in the 1980s goes away. So, theoretically, 100% of time used for review is actually spent making a decision. How much time is worthwhile? We’d have to have a discussion about that, but I’m going to throw out 30 seconds. If we could have 30 second replay, it would be worth it. A controversial call on the field typically takes more than thirty seconds anyway, because umpires huddle (but never change the ruling) and managers come out on the field and argue the call (which never has any effect except to get the manager removed from the game.)

Still, it takes a long to time review the play from every angle to come up with the best judgment that the video evidence supports. You just couldn’t do all the work necessary in 30 seconds, so it looks like we’ll have to settle for a longer review time or sacrifice the accuracy we desire.

That is a mistake. The reason replay takes so long is that we think that the goal is to produce the best judgement that the video evidence supports, using time the way we should in a courtroom, where no minute is more valuable than the freedom of the innocent-but-accused. The reviewer must check the play from all angles. He must double check it. He must confer with other reviewers for their opinion, and then come to a consensus. It’s an inefficient system, which in criminal trials is fair and good, but it’s not good for entertainment.

Replay doesn’t have to be a courtroom. Give five reviewers access to all the available video. Give each 30 seconds to decide what the call should have been. Then the vote. They don’t talk about it, they just vote their own best judgment. No changing the vote once cast. The majority rules the day. Suppose for a moment that there is a .75 probability that each of them makes the right call. Then the probability that the majority is correct is .896. (Binomial distribution probability.) Such reviewers would botch the call (as a group) just 1 in 10 times. Furthermore, in the preponderance of replay cases, video evidence is completely clear cut and it takes less than 30 to make a determination that’s right with a probability of 1.

This 30 second replay system would eliminate the vast majority of all erroneous calls in baseball. It wouldn’t be a fail safe system. It would require that we abandon our standard of having evidence that fully justifies our conclusions to all so that no one could come to a better conclusion on the basis of the evidence. But we shouldn’t let perfection be the enemy of the good. And it’s a good thing to preserve perfection.

Jeter, Jeter, Numbers Beater

About a month ago, I wrote an unintentionally controversial article about some puzzling patterns in Derek Jeter’s early season numbers.

There were two main contradictions within his statistics: that he was on pace for the best power numbers of his career while posting his highest ground ball rate ever, and that he was posting a career-low strikeout rate while swinging at a dramatically larger proportion of pitches thrown outside of the strike zone.

My critics claimed that I was reading too much into the stats too early in the season. Under normal circumstances I would have agreed, but it was more than the numbers themselves that were puzzling. When a person hits more home runs on fewer fly balls and makes better contact with worse pitch selectiveness, the results contradict the logic, no matter how small the sample size.

Four weeks later, I think it’s appropriate to revisit the situation and see how things are shaping up.

Overall, the discrepancies have become less dramatic, but the contradictory trends are still in place.

As I predicted, his power numbers have come back down to earth. He’s now on pace for 16 homers (down from 26 at last writing) and 98 RBI (down from 130). Neither would be a career high, but both would be above his norm.

But Jeter’s unprecedented groundball tendencies haven’t abated. Over two-thirds of balls off his bat (67 percent) have been on the ground—by far the highest such figure in the American League. While that’s a slight decline from the 71 percent mark he posted last month, it’s by far the highest of his career and a full 10 points above what he posted from 2002-09.

Meanwhile, his 16 percent HR/FB rate is the highest it’s been since 2005. Coincidentally, the 2005 season was the only other time in his career that his groundball rate hit 60 percent. So basically, the more he puts the ball on the ground, the more likely it is that each fly ball he hits will clear the fences. I’m not sure if that’s really a contradiction, but it’s certainly an odd correlation.

One thing is clear: this isn’t a common trend. This year, Jeter is the only player in the AL who has both a groundball rate over 50 percent and a HR/FB rate over nine percent.

But the more dramatic (and interesting) statistical oddity stems from the collapse of Jeter’s plate discipline.

Over his career, Jeter has been one of the most selective hitters in baseball, hacking at less than 20 percent of balls out of the strike zone. This year, that number has ballooned to 31.3 percent. Simply put, he’s chasing bad pitches. That’s not an insult or a criticism—that’s an indisputable, objective fact.

The sample size isn’t too small to start drawing conclusions. Jeter has seen 288 pitches outside the zone and swung at 90 of them.

As one might expect, this trigger-happy approach has had a negative effect on his walk rate, which, at five percent, is a career low. It’s less than half of the walk rate he posted last year.

Similarly, you’d expect his strikeout rate to shoot up into the stratosphere, right?


While Jeter’s 14 percent whiff rate is a sizable increase from last month’s nine percent figure, it’s still inexplicably lower than it ought to be, given Jeter’s history and his newfound aggressiveness.  How is that possible?

My first thought upon revisiting these numbers was that, in addition to swinging at more pitches off the plate, Jeter was starting to be less discriminatory with pitches thrown in the zone. That made sense, and I was embarrassed that I hadn’t thought of it a month ago.

But it turns out that’s not right either—in fact, it’s actually the opposite. This year, Jeter has chased a career-low 69 percent of balls in the zone, compared to 74 percent for his career. Simply put, Jeter is swinging at more bad pitches and fewer good ones.

I plugged in the numbers and found that, while 80 percent of the pitches he’s swung at since 2002 were good, just 69 percent of balls he’s chased in 2010 would have been called strikes.

And yet, Jeter’s 86 percent contact rate is the best of his career.

This just doesn’t make sense.

Ichiro, Grounded

Ichiro Suzuki is a singular talent. He is fast, seeming to fly all over the outfield. He steals bases with great success — 45 while being caught only twice in 2006. He has a legendary cannon for a right arm. And he hits like a machine. He racked up over 2,000 hits in his first nine major league seasons, and was the second fastest to that milestone (to Al Simmons, who reached it in 12 fewer games). He does not hit many home runs or a particularly impressive number of doubles or triples. He piles up his hits by putting the ball on the ground and running to first base. Joe Posnanski recently explored Ichiro’s hitting prowess, coming to the conclusion that he is one of the few truly unique players in history.

What if Ichiro were slow? What type of player would he be if he were not able to knock out so many ground ball hits? For one possible answer to this question, I compared the rate at which Ichiro reached base on ground balls to the American League average. Baseball Reference lists batting average splits by hit trajectory, and I compiled the league numbers from 2003 through 2009. The league statistics for the first two years of Ichiro’s career, 2001 and 2002, differ significantly from the next seven, possibly due to different methods of categorizing ground balls. I omitted those two years, as well as this season’s small sample. During the seven seasons from 2003-2009, American League players consistently reached base at a .240-.245 average on grounders. Ichiro beat the average in every year, with a high of .368 on ground balls in 2007.

To get an idea of what type of player Ichiro would look like without all those extra hits, I normalized his ground ball hit rate to the league average for each season and recalculated his batting totals. This adjustment cost him 42 hits in 2007 and 44 in 2004, and no less than 14 in any of the seven seasons. For simplicity, I removed only singles from his batting line. I then searched for a player with career statistics similar to Ichiro’s adjusted totals and came up with an interesting candidate. The following table displays Ichiro’s statistics over the past seven years (again, discarding 2001 and 2002) without his extra ground ball hits, side by side with Curt Flood’s career averages.

Table: Curt Flood versus Adjusted Ichiro
Statistics Per 600 Plate Appearances
Adj. Ichiro Curt Flood
Doubles 19 23
Triples 6 4
Home Runs 8 7
Walks 37 38
Strikeouts 56 53
Average .291 .293
OBP .338 .342
Slugging .391 .389
OPS .729 .732
BABIP .313 .314

This adjusted version of Ichiro is a near statistical clone of Flood, who played center field in the 1960s for the Cardinals. Flood was also a great fielder, winning gold gloves in each of his last seven full seasons. He is most famous, however, for starting a chain reaction that led to the free agency system when he refused a trade to the Phillies after the 1969 season. It marked the effective end of his career at just 31, as he played only 13 games in a comeback attempt in 1971. Despite now being inextricably linked with baseball labor history and not often mentioned for his playing ability, Flood was actually a surprisingly valuable hitter. He played in the offensively-challenged 1960s and was significantly above average with the bat for most of his career. Ichiro’s adjusted stats, in today’s era of power and on-base percentage, would be far less impressive.

Of course, this is all merely a thought experiment. Ichiro probably does have some control over where he hits the ball, as his New York Times profile explains. FanGraphs’s Jack Moore also explored Ichiro’s propensity to hit grounders to the opposite side of the infield at an abnormal rate. Thus, simply removing some of Ichiro’s hits does not show us the player he would actually be if he were not so fast. If he could not get on base so often on this type of batted ball, it is likely he would adjust his approach and that his numbers would reach a different equilibrium. What removing all of the extra singles does show us is just how much value those ground ball hits supply. Without them, Ichiro’s numbers look fairly pedestrian, especially in the era in which he has played. With them, Ichiro is Ichiro, and he will go down as one of the greatest and most dynamic hitters in baseball history.

This article originally ran at Ball Your Base.

Is Scott Rolen a Hall of Famer?

Note: Article was originally submitted to Bleacher Report on March 22nd, 2010, before the beginning of the 2010 season. For a link to the piece, visit Joe Regan’s bleacher report article.

As anyone who has read examples of my past writing can attest, I tend to focus a lot of my historical analysis pieces on the Hall of Fame. Today, I will divert myself from that path a little bit to argue my Hall of Fame case for a great in our generation whose contributions have been highly underrated: Scott Rolen.

A lot of words come to mind when discussing Rolen: grinder, a “veteran presence,” scrappy, gritty. I hear these a lot. One thing I do not hear a lot, for whatever reason, is “great.”

A quick look at his hitting numbers, for example, do not scream “great.” In 14 seasons, and 7,382 PAs, Rolen currently sports a line of .284/.370/.498, with 283 HR, good for a 124 OPS+/128 wRC+ . His HR total is good for 146th all time, behind players like Garret Anderson and Miguel Tejada, and he has never led his league in any statistical category, from the “old school” categories of BA, RBI, and hits, to the more analytical categories of OPS, OPS+, and others. Sounds like a classic “good, not great” player, correct?

I disagree completely. His 128 wRC+ puts him in the same category rate-wise (albeit sans-full decline stage) as Paul Molitor and Tony Perez. Molitor, if you recall correctly, spent a good chunk of his career as a DH, while Perez spent over two-thirds of his career at first base. Rolen, on the other hand, has played every inning of his career at third base.

Positional differences, obviously, is not enough. While fangraphs approximates an average third baseman to be worth 1.5 WAR per 600 PA more valuable than an average defensive first baseman with the same batting numbers, this does not address the fact that Perez had a longer career. Heck, Perez is not even the issue here. The issue is Rolen.

Everyone’s favorite new “quick reference” defensive statistic is UZR. Also, most everyone (myself included ) recognizes Adrian Beltre to be a fantastic defensive 3B. UZR reflects this, rating Beltre at +104.5 UZR at 3B since 2002, and a +13.9 UZR/150.

Scott Rolen? 102.1 UZR, 15.5 UZR/150.

I think most people would acknowledge that Rolen can flash leather. I doubt too many would think he is as good as Adrian Beltre.

Total Zone (which can be found on the player pages of Sean Smith’s website ) is not quite as gung-ho over Rolen as UZR, but at +93 since 2002 (and +141 overall), it’s close. The aforementioned Beltre is rated at +79 since 2002, and +96 overall. While I am not prepared to state that Rolen is a better defensive 3B than Beltre, any system that recognizes Beltre’s elite abilities, and then also states Rolen shares said abilities, is a perfectly credible system to me.

So what we are left with is a good hitting, great fielding player at a position that is in the middle of the defensive spectrum, and we are tasked to determine his place in history. Once again, Sean Smith provides a great point of reference for us, which his top 500 positional listing. At 94, we find our subject, Scott Rolen. Behind him are, well, a lot of all-time greats.

One could argue that Rolen still needs to post good seasons to make it to The Hall. Fine. According to Rolen’s fangraphs page , his CHONE projection rates him to be a +3.0 WAR player in 2010 (which is solidly above average). Even if Rolen breaks down rapidly (to the tune of +3.0, +2.0, and +1.0 seasons), that would push him up on Sean Smith’s rankings to Brooks Robinson status.

Maybe he did not peak well enough? Peaks are important, but I would argue that Rolen had many outstanding seasons. According to his page, Rolen did not post a season of sub-4.0 WAR baseball from 1997-2004. 4.0 is usually considered the level of an all-star player. At age 34 in 2009, he posted a +4.8 (according to Sean Smith’s system), yet another terrific season.

To further emphasize “excellence,” one can use a “junk statistic” called WAE, or “Wins Above Excellence,” calculated by subtracting four from every individual season’s WAR total, and defaulting to zero if the number is less than four.

Using Sean Smith’s WAR totals, Rolen’s WAE clocks in at 19.9. The aforementioned Brooks Robinson? A WAE of 15.3. Looks to me that Rolen passes the “excellence” test.

I have already written that the Hall of Fame should measure players more about the way they help their team win, rather than the hype they generate. Fact of the matter is, Rolen is a great baseball player, and I hope when his name hits the ballot (perhaps in 2019? 2020?) that the BBWAA evaluates his career correctly.

Believe the Hype: Looking Back at the 2005 Draft

With the 2010 MLB Draft just around the corner, and excitement in the air, many writers have been cautioning optimistic fans.  Throughout the 1990’s, very few first rounders contributed anything meaningful in the big leagues. However, I believe times have changed, and improved scouting and statistical analysis have lead to better draft choices across the board.  To see this theory in action, we have to look no further than the stacked first round of the 2005 MLB Draft.  Only five years after the fact, 18 of the 30 draft choices have posted a positive WAR in the big leagues.  All statistics courtesy of FanGraphs and

1. Justin Upton:  5.7 WAR

Drafted out of high school, Upton has already made a big impact at the Major League Level, posting a .388 wOBA for the Diamondbacks in 2009.  The team has locked him up, and he appears to be a superstar in the making.

2. Alex Gordon: 4.3 WAR

After strong showings in 2007 and 2008, Gordon has struggled in recent years, and the Royals have sent him down to AAA to learn how to play left field, where he is currently mashing minor league pitching (1.207 OPS).

Read the rest of this entry »