More on Bunting

Here's an addendum to my last post on bunting. There were 86 successful sacrifice bunts with two strikes. So bunting with two strikes was successful 34.7% of the time in 2004 (86 out of 248 times when you count the 162 strikeouts).

BJ on Baseball Digest Daily

Here is part one of a nice interview with Bill James over at Baseball Digest Daily.

Two interesting points.

First, James' comments about bunting:

'...the general argument against the bunt seems unpersuasive to me. The essential argument against the bunt is that the number of expected runs scored after a bunt attempt goes down in almost all situations when a bunt is used, and the expectation of scoring one run goes up only in a few situations.

But this argument is unpersuasive, to me, because it assumes that there are two possible outcomes of a bunt: a “successful” bunt, which trades a base for and out, and an “unsuccessful” bunt, which involves an out with no gain. In reality, there are about a dozen fairly common outcomes of a bunt attempt. The most common of those is a foul ball, but others include a base hit, a fielder’s choice/all safe, a pop out, a pop out into a double play, an error on the third baseman, and a hit plus an error on the third baseman, or the second baseman if you’re talking about a drag bunt.'

James is of course correct that the argument against the bunt is generally that the run potential decreases after a bunt. For example, using my Desktop Big League Manager, you can quickly see how this works.

The first table below shows the odds with which the sacrifice would have to be successful given a typical hitter (1999-2002) in order to increase the chances of scoring a single run and to maximize the number of runs in the inning. A value of "No" indicates that even if the sacrifice is successful the odds of scoring a single run or maximizing runs goes down.

Outs  Runners  Score 1  Maximize    
0       1st      No        No
1       1st      No        No
0     1st/2nd   79.9%      No
1     1st/2nd    No        No
0       2nd     93.9%      No
1       2nd      No        No

As you can see the only time it makes sense is when runners are on first and second with nobody out and you need one run to tie or win the game. And even then you need to be pretty certain that the hitter can get the bunt down.

However, with a pitcher at bat the odds change since the chances of a pitcher advancing a hitter with a hit go down.

Outs  Runners  Score 1  Maximize    
0       1st     83.1%     No
1       1st     89.3%     No
0     1st/2nd   35.9%    62.3%
1     1st/2nd    No       No
0       2nd     66.1%    96.3%
1       2nd      No       No

Obviously, with a pitcher who can actually get down the bunt it makes sense to do so not only with runners on first and second and nobody out but also a runner on second and nobody out, and even a runner on first when you need a run.

And James is also correct in saying that these calculations are based on only two of the several possible outcomes when a batter squares around to bunt. His comments made me curious to find out how often some of those other outcomes occur.

I took a look at the 2004 play by play data and found that there were 1,731 successful sacrifice bunts in the majors in 2004, an average of 58 per team. Montreal had the most at 100 while Boston had just 12. I also found that there were 515 bunt grounders recorded when there were runners on base, less than two outs, and where the batter was not credited with a sacrifice. Now, it's difficult to gaguge intent from these play-by-play records since some or even most of these may in fact have been instances of hitters bunting for hits with runners on. That said, here were the results of those 515 bunt grounders.

Negative Outcomes
Grounded into double play: 29
Force out: 177
Fielder's Choice, out recorded: 21
Batter out, no sacrifice: 49

Positive Outcomes
Errors: 12
Singles + One or more errors: 29
Singles: 194
Fielder's Choice + One or more errors: 4
Force out + One or more errors: 2

So we can throw out the last entry in that list since the batter was out and yet the official scorer did not rule a sacrifice. So that leaves us with 466 bunt grounders. However, to this list we can add the 162 times a batter struck out on a two-strike bunt attempt with runners on and safely assume that the vast majority of these were perpetrated by pitchers attempting to bunt with two strikes.

So assuming that all rest were sacrifice attempts (which is unrealistic) the odds of a successful sacrifice (taking into account all positive outcomes for the offense) is:

(1731+194+29+12+4+2)/(1731+194+29+12+4+2+29+177+21+49+162) = 81.8%

If you assume that none of the singles were sacrifice attempts the percentage drops to 80.2%. So we can be pretty confident that around 81% of the time a sacrifice attempt is successful. Given this number, however, you can still see from the tables above that there are only a couple of situations where a sacrifice attempt to warranted (obviously the actual percentage will change when considering double plays and errors but since the number of both are similar it seems like these would cancel each other out for the most part). So based on this I'm still pretty comfortable taking Earl Weaver's side in this argument.

So assuming that all the singles were sac attempts, the odds of the other outcomes are:

Double play: 1.2%
Force out: 7.3%
Force out + Error: 0.2%
Fielder's Choice: 0.9%
Fielder's Choice + Error: 0.2%
Error: 0.5%
Single: 8.0%
Single + Error: 1.2%
Strikeout: 6.7%

One of the interesting things here is that the odds of striking out are almost six times greater than the odds of bunting into a double play, which is ostensibly the reason managers direct their pitchers to try and bunt with two strikes. Incidentally, the team leaders in striking out while bunting were:

CIN 16
LAN 15
MIL 12
PHI 12
ATL 11
HOU 10
SDN 9
COL 9
NYN 8
SLN 7

Second, James' has a nice description of the decreasing relative importance of defense over the years which parallels Stephen Jay Gould's argument about .400 hitting. Essentially, as a system stabilizes variation decreases. Variation in fielding has decreased as the game has developed (due to better equipment, standard positioning and a universal set of styles of play not to mention the increase in homeruns and strikeouts), and so there is a smaller relative difference in defensive play across teams. It's not that defense is any less important - after all if you put me at third base for a major league team you'd quickly see why - but it is the case that major league teams don't need to spend as many resources finding top flight defensive players since the number of runs a great defender saves over and above an average defender is smaller than the number of runs a great offensive players produces over and above a mediocre one. I believe this is the case for offense as well but the magnitude of the difference is smaller.

Sabermetric Stats in Seattle

I heard via the SABR-L list that on Wednesday May 25th, The Seattle Times replaced their traditional league leaders tables with sabermetric versions built by one of their columnists, Jeff Angus whose blog I follow, and Baseball Reference's Sean Forman. This appears to be only the second time sabermetric stats have been used to this degree in a paper, the first apparently being the New York Times from 1984-1987.

It's great to see the general fan being exposed to statistics that track better with actual performance. Hopefully, the first of a trend.

Offsensive Offenses

So why are the Cubs under .500?

One reason is that they cannot score runs - just 183 going into tonight - good for 25th in the bigs. At the same time they've hit 52 homeruns, good for 5th in baseball. They once again lead the world in runs scored as a result of a homerun at 45.8% going into Sunday's game.

Of course the underlying reason is that they can't get on base. Here are the OBPs for those who've had significant playing time.

Lee           .446
Hairston      .369
Burnitz       .337
Ramirez       .317
Perez         .315
Patterson     .309
Dubois        .300
Barrett       .293
Macias        .289
Hollandsworth .275
Blanco        .193

So essentially, only three players have an OBP that approaches or exceeds league average. The other eight players are abysmal - but that's not really the problem. The problem is that of those eight only Todd Hollandsworth can really be expected to raise his level of play. The other seven guys are doing pretty much what they do - flailing away at any pitch that comes within three feet of the dish.

On another note Baseball Musings has a chart of runs created versus actual runs that shows the Cubs are scoring 4.4 runs per game when they would be expected to score 4.7 runs per game given their offensive elements. One of the reasons for the underperformance is that so much of their offensive output has come from one player - Derrek Lee. Lee has hit 12 of those 52 homeruns, scored 30 and driven in 39 of the 183 runs (21.3%). Another might be Dusty Baker's propensity to give away outs. Take your pick.

BTW, for Royals fans Baseball Musing has this to say....

"It's difficult to believe the KC Royals are outperforming their expectation, but they are tied with the Angels and the Braves at .4 per game over their predicted value. The Royals are hitting 20 points lower with runners on than with the bases empty, but the hits they are getting are long hits. So they are doing a good job of drving runners around despite a low BA and OBA in the situation."

The Physics of Pitching

Just read an interesting article on the physics involved in pitching published online at the American Scientist. In particular here are a few of the things I didn't know.

1. The theory that Bernoulli's prinicpal is responsible for the deflection of a ball thrown with spin has not been verified experimentally. I thought I had read in Adair that the Magnus force was caused by the decrease in pressure on one side of the ball due to the spin and seams. The authors of this article say that conservation of momentum is a better explanantion whereby the turbulent air generated by the spin is deflected in the opposite direction of the ball's "break".

2. Wind tunnel experiments have not shown that a four-seam grip causes a fastball to drop less than a two-seam grip. The authors speculate that it is a visual illusion. Their theory is that a major league fastball thrown with a four-seam grip does not provide the visual clues that a fastball thrown with a two-seam grip does. With a two-seam grip, the two seams are visible 20 times a second or so which causes a visual pulse or "flicker" that the batter can perceive. The four-seam grip produces a pulse at a higher frequency that that cannot be perceived.

3. Although it makes sense, I hadn't thought about the fact that a curve ball that breaks vertically is more difficult to hit than one that break horizontally. This is the case since the bat's sweet spot is about 4 inches long and just 1/3 of an inch high

4. Dynamic visual acuity - the ability to perceive information in moving objects - is not correlated with static visual acuity. Therefore a person could have geat eyesight in the view of an optometrist but be lousy at picking up the spin on a breaking ball (that must have been my problem). The authors note that:

"get a feel for the range in dynamic visual acuity, consider that most of us can read the label on a phonograph record turning at 33 revolutions per minute, but this would be about the limit of our capabilities. The great Boston Red Sox hitter, Ted Williams, could read one turning at 78 revolutions per minutes, which is far beyond the dynamic visual acuity of the average person."

5. The four-seam grip on a slider is what causes the red dot to appear in the upper right hand quadrant of a slider thrown by a right-handed pitcher. Using a two-seam grip eliminates the dot and therefore gives less information to the hitter.


Their conclusion is that...

"the pitcher should use a four-seam grip for fastballs and curveballs to remove the perceptual clue of the two red stripes and the flicker. Then, he should use the two-seam grip for the slider, to remove the clue of the red dot. These techniques could make a fearsome pitcher even more difficult to hit."

The Hangover Effect Redux

A few weeks ago I wrote about the "hangover effect" that some say causes the Rockies' hitters to perform poorly on the road. I didn't find any evidence of the effect and went on to postulate that perhaps the Rockies have been the unwitting victims of "altitude bias". Once again, I didn't find much evidence to support that speculation.

Finally, I read an article in the Rockies official magazine that referred to a "pitching hangover" that Greg Maddux and others say affects them after pitching at altitude.

Although the article mentioned that the hangover was relatively short in duration, I wondered if pitching at Coors Field had a persistant effect that could be measured. To test this I looked at the 2003 and 2004 seasons and measured the performance of visiting starting pitchers in their Coors Field starts compared with the first start after. The reasoning being that if Coors Field produced a hangover, perhaps the next start for the pitcher which show it through decreased performance. Here's what I found.

                 IP        H     TB      K   BB  WHIP   K/9 BB/9  H/IP
Start at Coors  874.3   1082   1837   588  356  1.64  6.05  3.66  1.24
Next Start      875.3    889   1412   652  310  1.37  6.70  3.19  1.02
Other Starts  18199.3  18175  29175 13471 5933  1.32  6.66  2.93  1.00

As you can see in the starts at Coors the performance was predictably worse than would be expected as evidenced by increased walks and hits per inning pitched and decreased strikeouts. What's interesting is to compare the second two lines which show the pitcher's next start and then all their other starts excluding these two. Although both hits and walks per inning pitched are higher in the next start, the margin is so small that I doubt that shows any kind of real effect. In fact, in 2003 pitchers actually did better in the start after a Coors Field start than in others.

So I wouldn't put too much credence in the "pitcher's hangover" for starts anyway. Relievers might be a different story, however, since they pitch more frequently...

Gray Pitchers and Position Players

In an attempt to be responsive to my public I've broken the number veterans into pitchers and position players. Here are the results...

Pitchers

1950 7
1951 6
1952 8
1953 13
1954 9
1955 13
1956 13
1957 12
1958 12
1959 11
1960 9
1961 9
1962 9
1963 9
1964 14
1965 13
1966 16
1967 17
1968 9
1969 13
1970 12
1971 14
1972 10
1973 13
1974 15
1975 13
1976 11
1977 14
1978 13
1979 23
1980 25
1981 27
1982 25
1983 25
1984 26
1985 26
1986 24
1987 19
1988 20
1989 23
1990 20
1991 22
1992 26
1993 23
1994 25
1995 20
1996 18
1997 18
1998 18
1999 19
2000 26
2001 26
2002 27
2003 34
2004 33

Non-pitchers

1950 12
1951 13
1952 7
1953 13
1954 13
1955 24
1956 17
1957 19
1958 16
1959 17
1960 20
1961 18
1962 17
1963 16
1964 14
1965 12
1966 14
1967 14
1968 11
1969 14
1970 16
1971 19
1972 20
1973 24
1974 25
1975 26
1976 30
1977 29
1978 27
1979 30
1980 30
1981 29
1982 36
1983 41
1984 44
1985 48
1986 46
1987 48
1988 45
1989 37
1990 32
1991 33
1992 33
1993 25
1994 23
1995 24
1996 27
1997 31
1998 27
1999 27
2000 33
2001 39
2002 37
2003 45
2004 43

And here are the graphs:

Coors Field Fun Facts

Here's a few Coors Field tidbits that I gleaned from the Rockies media guide. I was reminded of these after seeing the Rockies low-scoring 3-2 and 3-1 victories the last two days.

• No Rockies pitcher has thrown two complete game shutouts at Coors. Only five Rockies pitchers have thrown complete game shutouts: Mark Thompson, Roger Bailey, Brian Bohanon, Mike Hampton, and John Thomson. Meanwhile Tom Glavine has done it twice both I 1995 and 1997
• On April 26th Juan Pierre hit his first career homerun at Coors on his 729th at bat. He had been the record holder for most at bats without one. Now that falls to Pedro Astacio at 137 at bats
• The Rockies set a record for most consecutive home games scoring at least one run at 361 from 7/5/99 to 9/17/2003. That's over 100 games longer than the previous streak of 258 held by the Boston Braves 1892-1896
• 10 homeruns have been hit in a game five times, most recently 7/2/2002 vs. the Giants when the Rockies hit 3 and the Giants 7
• The 303 homeruns hit there in 1999 is the Major League record (3.7 per game)
• The longest homerun at Coors is 496 feet by Mike Piazza 9/26/97. Larry Walker is second at 493 on 8/31/97
• There has never been a 1-0 game in 801 contests going into this season. The old record was 635 games for the Philadelphia A's (1882-1891)
• There have only been two 2-0 games, one was one of the Glavine shutouts

Although the last two days have been low scoring the simple park factor calculated as:

PF = ((homeRS + homeRA)/(homeG)) / ((roadRS + roadRA)/(roadG))

has Coors Field up around 1.48 for 2005, second only to Jacobs Field. The Rockies have scored 133 runs at home and just 55 on the road while their opponents have scored 131 at Coors and 89 away. Overall that's 12.57 runs per game at Coors as opposed to 8.47 on the road.

Historically, the park factors as calculated by Baseball Reference (using three year weighted averages have been:

Year BPF/PPF
2004 120/117
2003 112/111
2002 121/119
2001 122/119
2000 131/128
1999 129/126
1998 119/120
1997 123/123
1996 129/129
1995 128/128
1994 116/118
1993 120/122

It's interesting that park factors have been decreasing in recent years. Most people attribute this to the humidor that was installed at Coors Field in 2002 to keep the balls at 70 degrees and 50% humidity. There was an great article by Mike Klis of the Denver Post online today that talks about the humidor. I found the following interesting.

"The Rockies believe the pitching problems at Coors Field could be at least partially corrected if the weight specifications MLB has for its baseballs weren't contradictory to its humidor/ball-shipping specifications. By rule, a baseball can weigh from 5.0 to 5.25 ounces and measure from 9.0 to 9.25 inches in circumference.

This is not an insignificant variance. MLB's own study at the University of Massachusetts-Lowell in 2000 found that a well-struck, 5.25-ounce, 9.25- inch ball could travel up to 49.1 feet less than a 5.0-ounce, 9.0- inch ball.

The current humidor settings, however, cannot sufficiently offset altitude's effects to pump the baseballs up to 5.25 ounces. According to Kevin Kahn, the Rockies' chief of ballpark operations, and humidor engineering head Tony Cowell, the balls put in play at Coors Field weigh about 5.12 ounces, on average. "

According to Cowell, prior to the humidor balls were being put into play with weights as low as 4.6 ounces and 8.5 inches.
Cowell is pictured below.

Although I don't doubt that the humidor has had an impact you'll note that scoring began to decline in 2001, the height of the offensive outburst of recent years. Perhaps that was just a fluke.

The Graying of the Game

Recently on the SABR-L list there was an interesting discussion on the “graying of the game” that was spawned by a recent article on Baseball Prospectus of the same name by Nate Silver. Unfortunately, I'm not a subscriber so I don't know what Silver had to say. On the SABR list however, this was a discussion centering on the perception that the average player at the major league level is getting older. The speculation was that free agency would tend to lengthen careers and so post 1977 we should see an increase in the ratio of older players to younger players in the game. Of course, other factors to consider are wider demographic trends of the baby boom generation and the impact of steroids, modern medicine, and modern training techniques.

To try and see what’s going on here I created a query in the Lahman database showing the number of active players in each season since 1950 whose debut had been at least 15 years earlier. In other words, I wanted to see how many veteran players there were in each season. Because there were different numbers of teams in the leagues over the course of those seasons I divided the number of veterans by the number of teams to come up with a Veteran Index. Here were the results:

Year  #  T   VI
1950 18 16 1.13 
1951 19 16 1.19 
1952 15 16 0.94 
1953 24 16 1.50 
1954 21 16 1.31 
1955 34 16 2.13 
1956 27 16 1.69 
1957 28 16 1.75 
1958 24 16 1.50 
1959 24 16 1.50 
1960 25 16 1.56 
1961 25 20 1.25 
1962 23 20 1.15 
1963 22 20 1.10 
1964 25 20 1.25 
1965 24 20 1.20 
1966 27 20 1.35 
1967 26 20 1.30 
1968 18 20 0.90 
1969 25 24 1.04 
1970 26 24 1.08 
1971 30 24 1.25 
1972 28 24 1.17 
1973 33 24 1.38 
1974 37 24 1.54 
1975 36 24 1.50 
1976 39 24 1.63 
1977 40 26 1.54 
1978 38 26 1.46 
1979 49 26 1.88 
1980 50 26 1.92 
1981 53 26 2.04 
1982 60 26 2.31 
1983 61 26 2.35 
1984 65 26 2.50 
1985 68 26 2.62 
1986 65 26 2.50 
1987 63 26 2.42 
1988 60 26 2.31 
1989 54 26 2.08 
1990 49 26 1.88 
1991 50 26 1.92 
1992 52 26 2.00 
1993 43 28 1.54 
1994 42 28 1.50 
1995 38 28 1.36 
1996 38 28 1.36 
1997 43 28 1.54 
1998 39 30 1.30 
1999 38 30 1.27 
2000 52 30 1.73 
2001 60 30 2.00 
2002 58 30 1.93 
2003 70 30 2.33 
2004 69 30 2.30

I then graphed both the raw service time and the Veteran Index.




As you can see the Veteran Index declined from around 1955 to 1968 and then steadily increased reaching a peak in 1985 before declining again until 1999 and then took off again through the present.

While I don’t have anything but speculation to back this up, my view would be that the increase after 1968 was due to the combination of better training and the advent of free agency. Particularly in the early 1980s teams tended to award long term contracts to veterans which tended to keep them in the game longer. In the early 1990s the trend moved towards shorter contracts and consequently the Veteran Index dropped. The combination of steroids and advanced training techniques may account for the recent increase. It should be kept in mind, however, that the baby boom generation (born around 1951) would have reached their 15th season around 1987 so it could also simply be an artifact of the larger societal demographics. The recent increase couldn’t be accounted for in the same way however since it would include those in the baby bust generation (born between 1967 and 1979) which is considerably smaller than the boomers.

Not High on Miles

There was an article on Rockies second baseman Aaron Miles on MLB.com today. The gist of the article is that Miles has been hitting well since (10-27, 1 HR, 1 BB) he was moved to the second spot in the order behind Clint Barmes.

What I thought was interesting in the article were these comments by Rockies manager Clint Hurdle.

"I think he has the freedom now just to be instinctive at the plate with his hands and his bat," Hurdle said. "We've simplified things for him. He doesn't have to work counts to the degree he was earlier.

"We've told him that in offensive counts we want him looking for a pitch he can hit, where before, 3-1, he was taking balls. We tried. He tried. We felt we had better alternatives, so we made the switch. It proved beneficial for both of those guys involved."

Well, it's hard to believe Miles was trying to work the count when he was hitting leadoff since he walked all of 1 time in 87 plate appearances. It's not really possible for him to be less selective is it?

Even with his hot hitting of late his OBP is still only .315, far below acceptable for a top of the order type hitter and even worse in Coors Field. The telling statistic is that Miles has scored just 19 runs in 123 plate appearances despite batting in front of Todd Helton. Contrast that with Barmes who has scored 32 runs in 151 PAs. I'd much rather see Hurdle experiment with Matt Holliday in the second hole - a good baserunner and developing hitter who could benefit from seeing some good pitches in front of Helton.

While he's pretty solid defensively, I just hope the Rockies don't fall in love with Miles and look at him as some kind of long term solution at second base in the same way the Royals thought too highly of Desi Relaford. What would be particularly disturbing is if they trade a second base prospect or hold someone up in favor of Miles or worse yet, sign Miles to a multi-year contract. His career record, and the fact that he's 28 years old, don't add up to the kind of player who actually helps a ballclub. For example, in the minors his highest on base percentage in a full season was .351 in 2003 at Charlotte (and he had a ton of minor league at bats playing seven full seasons). Last year as a 27 year-old rookie for the Rockies he walked just 29 times and record an on base percentage of .329. This, not to mention his 24 extra base hits and .368 slugging percentage in 551 plate appearances - which add up to a miserable 697 OPS. I just can't understand the thinking behind batting a player second who doesn't get on base.

James on the Royals

Following along the theme of "Royal Pain", Bill James, who lives in Lawrence, wrote an interesting article for the Lawrence Journal World that was published on Friday. There, James lays out the following five problems with the Royals.

"1) The economics of the game are stacked against smaller cities.

2) They have drafted relatively poorly in the last 10 years.

3) In the early- to mid-'90s, the Royals didn't know a ballplayer from a skeet shooter and committed themselves to staving off ruin by bringing in a long series of fading stars. This caused the base of the organization to crumble, which increased the financial pressures on the team, putting them in a position from which they never have recovered.

4) When the Royals have had young players that they could not afford to keep, they have uniformly failed to acquire value in exchange. The worst example of this was last year with the Carlos Beltran trade.

5) The Royals have been unable to identify and acquire the kind of affordable, decent journeymen players who could serve as a tourniquet on the organization. They have done this SOMETIMES, but just not consistently enough. Jose Lima was 8-3 for them in a stop-gap role two years ago. They wouldn't make a decent offer to retain him, let him go, and he was 13-5 for the Dodgers."

I couldn't agree more with all five points although points 4 and 1 are interrelated. Given that the Royals are a small market team that had no hope of re-signing the likes of Carlos Beltran, Johnny Damon and in the future Zack Greinke, teams can afford to offer relatively little in exchange. In the Beltran deal of last June I think Baird did about as well as he could.

The strategy of acquiring affordable journeyman talent that can be used as trade bait around the deadline is one that the a small market team needs to employ with impunity and one which the Royals are not employing regularly. For example, one strategy would be to sign several quality relievers as free agents and then trade as many as possible at the deadline for prospects ala the Jason Grimsley trade of last season that netted Denny Bautista. This strategy would also not break the bank since veteran relievers are not priced at the same level as starting pitchers.

The Ugly Inning

I've scored about 30 or so games for MLB.com's Gameday system starting last season in Kansas City. Tonight in the Diamond Backs 10-4 victory I scored what was for me the most difficult inning thus far with the possible exception of the infield fly fiasco in last year in Kansas City.

Here's the play-by-play for the top of the third inning in which the D'Backs scored three runs.

Diamondbacks 3rd (Diamondbacks 7, Rockies 0) -- S. Green singles up the middle. J. CruzJr. reaches on force attempt, throwing error by T. Helton, S. Green to 2nd. With R. Clayton batting, S. Green advances to 3rd on a balk; J. CruzJr. advances to 2nd on a balk. R. Clayton reaches on a fielder's choice out, J. Wright unassisted, S. Green out at home. C. Snyder singles to left-center field, J. CruzJr. scores; R. Clayton to 3rd. B. Webb reaches on force attempt, throwing error by J. Closser, R. Clayton scores; B. Webb to 2nd; C. Snyder advances to 3rd on throwing error by C. Barmes. C. Counsell hit by pitch. C. Tracy grounds into a force out, A. Miles to C. Barmes, C. Snyder scores; B. Webb to 3rd; C. Counsell out at 2nd. L. Gonzalez walks, C. Tracy to 2nd. T. Glaus strikes out swinging.
(3 Runs, 2 Hits, 3 Errors, 3 LOB)

Although there were only three runs scored, the codes used to enter these plays were difficult to figure out. Luckily, during the game I'm on AOL instant messenger with a support person who "talked" me through them.

• S. Green singles up the middle. (S8/G)
• J. CruzJr. reaches on force attempt, throwing error by T. Helton, S. Green to 2nd. (E3/TH/FO/G.1-2)
• With R. Clayton batting, S. Green advances to 3rd on a balk; J. CruzJr. advances to 2nd on a balk. (BK.2-3;1-2)
• R. Clayton reaches on a fielder's choice out, J. Wright unassisted, S. Green out at home. (FC1/G.3XH(1))
• C. Snyder singles to left-center field, J. CruzJr. scores; R. Clayton to 3rd. (S7/L.2-H(UR);1-3)
• B. Webb reaches on force attempt, throwing error by J. Closser, R. Clayton scores; B. Webb to 2nd; C. Snyder advances to 3rd on throwing error by C. Barmes. C. (E2/TH/FO/BG.3-H(UR)(NR)(TH);1-3(E6/TH);B-2)
• Counsell hit by pitch. (HBP)
• C. Tracy grounds into a force out, A. Miles to C. Barmes, C. Snyder scores; B. Webb to 3rd; C. Counsell out at 2nd. (46(1)/FO/G.3-H(UR);2-3)
• L. Gonzalez walks, C. Tracy to 2nd. (W.1-2)
• T. Glaus strikes out swinging. (K)

And just for good measure, later in the game I had the following play:

S9/L-.2-H;BX2(926362)

which translates to:

L. Gonzalez singles to right-center field, C. Tracy scores; L. Gonzalez out at 2nd, B. Hawpe to J. Closser to C. Barmes to T. Helton to C. Barmes to J. Closser.

Hoping for smoother sailing on Tuesday night.

Royal Pain

For those of you who are suffering with the Royals this season you might be interested in the latest exchange from Rob and Rany. It's clear the frustration is growing as Rob and Rany promote the following ideas:

1. Send Andy Sisco to long relief
2. Send Amby Burgos and Leo Nunez back to the minors
3. Give Matt Diaz an everyday job
4. Fire Jeff Pentland, the hitting coach as well as the interim manager Bob Schaefer
5. Play Ruben Gotay everyday, either in KC or in Omaha
6. Release Mike McDougal, Calvin Pickering, Colt Griffin, Emil Brown, Jose Lima, and Ken Harvey
7. Send Denny Bautista to Omaha
8. Send Justin Huber to Omaha from AA
9. Start shopping Angel Berroa
10. Draft the best available player next month, assuming that to be Alex Gordon from Nebraska
11. Hire a manager who knows what he's doing

For my part I strongly agree with 1,2,5,8, and 10. (1) Sisco only earned runs have come in blown save opportunities and as Earl Weaver said, "the best place for a rookie is in long relief". (2) There is no reason Burgos and Nunez are in the majors other than to use up service time that will cost the Royals in arbitration sooner rather than later. (5) Gotay needs to play everyday somewhere, period. (8) Huber is tearing up AA and should be getting his chance against AAA pitching right now. (10) The only way the Royals are going to get better is to draft well and that means taking the best available player and spending the signing money. I'd rather see them spend it there than on replacement level players.

I moderately agree with 3,4,7,9 and 11. (3) Diaz seems to me to project to be an adequate major leaguer but never a starting corner outfielder. While he's better than Brown, it doesn't really matter which one plays given that neither is a long term solution. (4) Well, there is certainly something to be said for firing based on lack of performance. Pentland and Baird seem to say the right things regarding plate discipline but fail to execute. Pentland should take the fall but I'm worried that the cure could be worse than the disease. (7) Bautista looked good in his first start and has actually pitched well in his three of his five starts. He still walks too many guys (17 in 35.7 innings thus far) and so perhaps a stint in AAA will help him make some adjustments he could apply in the majors since young players are unlikely to make major changes at the big league level. What's most exciting about him though is that he's given up just 5 homeuns in 65.3 major league innings while striking out 42 - both pretty good ratios. (9) I can't imagine that this would be fruitful even if Allard Baird decided to do it. As pointed about by Rob and Rany Berroa hasn't shown that 2003 wasn't a fluke so I doubt another team would take him on for two and half more years at over $4M per. (11) It goes without saying they should hire a good tactical manager - something Tony Pena certainly was not. However, I think it more important that they focus on setting some organizational direction and then hiring a manager that buys into the philosophy and will abide by it. Of course I think the organizational philosophy should be based on the "Be the House" approach that was articulated by Dodger's GM Paul DePodesta here. What I don't want to see is a knee jerk reaction that results in the hiring of the anti-Pena per Luther's comment that the human race is like a drunken man who after falling off his horse on one side climbs back up and falls over the other. On the other side of the horse may be a manager who, ironically, might turn out to be another Tony Muser (think Larry Bowa).

The only item I disagree with is 6. I'm just not sure what good it will do to release those guys. The Royals don't really have any better players to replace them with. Why not just see if perhaps Harvey can get hot and then trade him? Same goes for Lima and even McDougal. On the Pickering front I see he's 3 for 33 in Omaha with 7 walks. So at least we know that it wasn't just major league pitching that is the problem. But even Pickering bad as he's been might start raking at AAA, making him tempting for some team looking for a left-handed bat off the bench. Of course I didn't think Lima should be signed in the first place but are you really going to replace him in the rotation with Jimmy Gobble? After all, D.J. Carrasco is now in the rotation. The only case that can be made is to bring up 30 year old Larry Jensen who is giving up a hit per inning in Omaha and on whom I can find absolutely no information. Not much of a choice.

Any thoughts?

Making SOA Real

Taking a page out of Ron's book I've created another blog (the world needs more blogs). This one will be related to my work at Compassion, particularly the effort of the team I'm on at creating a Service Oriented Architecture (SOA). The titles is "Making SOA Real" and it will focus on the decisions and trials we've gone through in settings standards, implementing models, and finally (we hope) rolling out our first SOA-based deliverables later this year.

The motivation for creating this blog was primarily to share our experiences. I hope to post the kind of information I was looking for when I first started thinking seriously about how an organization would move into SOA based development. Enjoy.

A Sea of Green

On Monday night I took my family to Coors Field for the first time to see a game played at altitude - which the Rockies won 7-6 I might add. As I explained to my nine-year old daughter with her eyes glazed over somewhere around the fourth inning, "lots of runs are scored at Coors Field and one of the hypotheses* given for this is that the fences must be very deep so that there aren’t too many homeruns, but one of the side effects is that more singles and doubles fall in front of the outfielders".

* We home school and so any time you can throw in a word from a recent lesson you do

She nodded as her eyes fixated on the cotton candy vendor and I went back into my contemplation of these and other great thoughts. But it occurred to me that I hadn’t ever actually seen any data that would support this hypothesis so I dug into the play by play data for 2004. I reasoned that Coors Field would produce more singles and doubles classified as fly ball or popup than would other parks. I wrote a quick query and found the following:

       Fly Pop Total
MIN    170   9  179
BOS    125   6  131
CHN    120  10  130
NYN    108  11  119
CHA    104   8  112
BAL     99   7  106
COL     94   6  100
NYA     95   3   98
ANA     87   8   95
KCA     84   4   88
SFN     72  12   84
HOU     79   5   84
PIT     74   6   80
PHI     73   7   80
FLO     71   5   76
DET     66   8   74
SLN     67   4   71
ATL     64   6   70
SEA     55  11   66
LAN     55  11   66
OAK     54   7   61
TBA     55   5   60
TEX     51   4   55
SDN     51   2   53
MIL     47   6   53
CLE     41  10   51
MON     43   6   49
TOR     41   4   45
ARI     34   3   37
CIN     34   1   35

This of course is not really what I was expecting. Could it really be possible that there were 170 fly ball singles in the Metrodome but only 34 in the Great American Ballpark, a ratio of 5 to 1? And why would there be 125 fly ball singles in Fenway Park, a ballpark with a small outfield?

When looked at in total Coors Field did produce the most singles at 1,104, just ahead of Anaheim at 1,097. Of those, 556 were line drives with only the Ballpark at Arlington producing more at 557, and 448 were ground balls with Detroit far outpacing the field at 539.

Putting these questions aside I went ahead and ran a similar query for doubles.

     Fly Pop Total
BOS  186  3  189
CLE  157  1  158
COL  146  2  148
SDN  141  2  143
MIN  140  2  142
PIT  131  1  132
SEA  130  0  130
CHN  127  3  130
NYN  127  2  129
ATL  124  2  126
HOU  119  2  121
ARI  114  0  114
FLO  112  1  113
NYA  106  2  108
TOR  105  1  106
TEX   98  0   98
MON   94  0   94
SFN   92  2   94
BAL   91  2   93
ANA   90  2   92
PHI   83  3   86
KCA   77  3   80
OAK   80  0   80
CHA   80  0   80
CIN   76  0   76
SLN   74  0   74
DET   70  1   71
TBA   66  0   66
LAN   56  3   59
MIL   50  0   50

Now this was really disturbing. While Coors Field ranked third, Boston still had 186 fly ball doubles while Milwaukee had just 50. Again, when looked at in total Coors Field produced the second most doubles at 379 (Fenway saw 395) with 188 line drives and 42 ground balls.

So in summary Coors produced the most singles and second most doubles but this analysis did not support the hypothesis that this is because more balls fall in front of outfielders who have to cover the spacious outfield. My daughter will no doubt be thrilled by my analysis.

In thinking about these results I was reminded of Joe Buck calling last year’s World Series and his annoying habit of describing every ball hit into the outfield as a popup, many of which traveled 350 feet or more. Taking the "Buck Factor" in account, the large differences between parks I think can better be explained by differences in how those who score the games interpret fly balls and popups. From my experience scoring games for MLB.com (from which this data is derived I believe) it is sometimes difficult to decide where the line is between ground ball and line drive, popup and flyball, and flyball and line drive. Generally, my criteria is that if a ball touches the ground before it reaches the outfield grass it is a ground ball. If a ball is hit on an arc greater than around 30 degrees (definitely subject to a lot of interpretation), the ball is a flyball if it travel more than 200 feet or a popup if less. Everything else is a line drive. I also use the L- scoring as in S9/L- (single to right on a softly hit line drive) for flares that fall in over the infield.

Of course if anyone has any better data on this I'd love to see it...

Futility

With tonight's 3-1 loss to the Blue Jays the Royals fall to 8-25. Winning at a .242 clip will buy them 123 losses, easily surpassing last year's record 104 loss season. With that kind of frustration it's not hard to see why Tony Pena resigned after the game - a move many Royals fans have been calling for since the first week of the season.

It's also not hard to see how Zack Greinke is 0-4 with his 3.38 ERA given that the Royals have scored all of 116 runs in those 33 games, just 3.52 per game. One note of encouragement regarding Greinke is that he's only given up three homeruns in 40 innings this season. Long balls were his achilles heel last season. His control remains good (only 8 walks) although his strikeout rate is somewhat concerning at 4.72.

What's most interesting, however, is that Mike Sweeney with his homerun tonight for the Royals only run has driven in 29 of the 116, fully 25%. Driving in a quarter of your team's runs in a full season has never been accomplished. Those who've come closest since 1900 are:

                 RBI   Pct
Nate Colbert 1972 SDN 111 22.7%
Wally Berger 1935 BSN 130 22.6%
Ernie Banks  1959 CHN 143 21.2%
Sammy Sosa   2001 CHN 160 20.6%
Jim Gentile  1961 BAL 141 20.4%
Bill Buckner 1981 CHN  75 20.2%
Bill Nichols 1943 CHN 128 20.2%
Frank Howard 1968 WAS 106 20.2%
Babe Ruth    1919 BOS 114 20.2%
Frank Howard 1970 WAS 126 20.1%

Who Touched Me?

Last weekend my family and I visited a church here in our new home of Colorado Springs. The pastor was preaching about the healing of the women in Mark 5 (also found in Luke 7 and Matthew 19). After talking at length about the human need for touch and drawing the application that we, like the woman, shouldn’t be afraid to reach out to Jesus and that no problem is big enough to separate us from Christ’s love, the pastor made a passing reference to Mark 5:30-31.

“At once Jesus realized that power had gone out from him. He turned around in the crowd and asked, ‘Who touched my clothes?’

‘You see the people crowding against you,’ his disciples answered, ‘and yet you can ask “Who touched me?”

In reference to Jesus’ question the pastor said something to the effect that “Of course, Jesus knew who had touched him, he just wanted the women to come forward and admit it.”

I’ve heard this same notion expressed from the pulpit before and it always intrigues me as to why one would assume that Jesus’ question was not sincere. It seems to me it stems from the belief that since Jesus was God incarnate, he therefore was endowed in his human nature with the attributes of God and so would be omniscient and therefore know who had touched him. In an essay called The World’s Last Night C.S. Lewis writes about the timeless nature of God and how this view tends to…

“conceal an attempt to establish a temporal relation between his timeless life as God and the days, months, and years of his life as Man. And of course there is no such relation. The Incarnation is not an episode in the life of God: the Lamb is slain – and therefore presumably born, grown to maturity, and risen – from all eternity. The taking up into God’s nature of humanity, with all its ignorances and limitations, is not itself a temporal event, though the humanity which is so taken up was, like our own, a thing living and dying in time. And if limitation and therefore ignorance was thus taken up, we ought to expect that the ignorance should at some time be actually displayed. It would be difficult, and, to me, repellent, to suppose that Jesus never asked a genuine question, that is a question to which he did not know the answer. That would make of his humanity something so unlike ours as scarcely to deserve the name. I find it easier to believe that when he asked, ‘Who touched me?’ (Luke 7:45) he really wanted to know.”

To me this view makes a good deal of sense and relates to the main point of Lewis’ essay, the second coming. Taking the view that Jesus never displayed any human qualities such as real curiosity of lack of knowledge forces one to make unnatural interpretations of passages such as Mark 13:30. Here, Jesus says:

“I tell you the truth, this generation will certainly not pass away until all these things have happened.”

Most modern evangelicals will interpret the word “generations” here to mean “race” as noted in the NIV from which the above is quoted. While substituting the word “race” seems to alleviate the apparent difficulty of Jesus not knowing when these things would be fulfilled, Lewis points out that this kind of ploy is unnecessary (not to mention pointless since it makes the verse a tautology) since two verses later Jesus also says:

“No one knows about that day or hour, not even the angels in heaven, nor the Son, but only the Father.”

As Lewis says, “To believe in the Incarnation, to believe that he is God, makes it hard to understand how he could be ignorant; but also makes it certain that, if he said he could be ignorant, then ignorant he could really be.”

I should mention that there is also a view of the second coming known as Preterism that holds that Jesus was correct in his assertion of Mark 13:30 since the events he describes culminated with the fall of Jerusalem to the Romans in 70 C.E.

Scoring Runners From Third

Announcers are always talking about the importance of scoring the runner from third with less than two outs. So how often do major league teams convert this opportunity and are some players more proficient in these situations?

To attempt to answer those questions I looked at play by play data for the 2003 and 2004 seasons. Specifically I looked at all situations where the following were true

1. There was at least a runner on third
2. There were less than two outs
3. The play ended the batter’s plate appearance (to make sure and not count wild pitches, balks, passed balls, and pick off errors)

What I found was that there were 21,073 opportunities during those two years and that 51.7% of the time the runner scored from third.

The leaders for those two seasons (with more than 50 opportunities) were:

                Opp    Runs     Pct
Crede Joe        57     41      72%
Lawton Matt      56     40      71%
Cruz Deivi       54     36      67%
Cabrera Orlando  71     47      66%
Tejada Miguel   115     76      66%
Matsui Hideki    93     61      66%
Anderson Garret  71     46      65%
Anderson Marlon  51     33      65%
Cintron Alex     51     33      65%
Renteria Edgar   97     62      64%
Casey Sean       72     46      64%
Damon Johnny     72     46      64%
Sheffield Gary  119     76      64%
Millar Kevin     70     44      63%
Stewart Shannon  59     37      63%
Loretta Mark     67     42      63%
Conine Jeff      75     47      63%
Baldelli Rocco   83     52      63%
Ramirez Aramis   72     45      63%

This is an interesting list and includes both "RBI men" such as Miguel Tejada, Garret Anderson, Gary Sheffield, and Hideki Matsui as well as some hitters you wouldn’t think of that way at all including Deivi Cruz, Orlando Cabrera, and Alex Cintron.

Those at the bottom of the list were:

              Opp Runs Pct
Crisp Coco     59  25  42%
Wilson Craig   60  25  42%
Beltre Adrian  70  28  40%
Burrell Pat    90  35  39%
Ramirez Manny  99  38  38%
Sexson Richie  68  26  38%
Kotsay Mark    55  21  38%
Burnitz Jeromy 56  21  38%
Bonds Barry    68  25  37%
Drew J.D.      55  20  36%
Sanders Reggie 61  22  36%
Valentin Jose  54  19  35%
Hernandez Jose 57  20  35%
Dunn Adam      69  24  35%
Helms Wes      58  20  34%

What’s interesting about this of course is that players you would normally think of as run producers don’t seem to fair very well including Barry Bonds, J.D. Drew, Manny Ramirez, Adrian Beltre, and Richie Sexson.

There are two reasons for this. First, the players at the bottom are those that tend to draw more walks and probably do so at a greater ratio in situations such as these by being pitched around. Since walks were included in the opportunities the percentages will tend to be lower for those types of players. Re-running the numbers and excluding walks the bottom bunch are:

               Opp   Runs  Pct
Soriano Alfonso 65    31   48%
Gonzalez Alex   63    30   48%
Wigginton Ty    63    30   48%
Wilson Craig    51    24   47%
Cameron Mike    56    26   46%
Castilla Vinny  74    34   46%
Jones Andruw    82    37   45%
Aurilia Rich    51    23   45%
Burrell Pat     76    33   43%
Beltre Adrian   63    27   43%
Dunn Adam       54    23   43%
Sexson Richie   57    24   42%
Sanders Reggie  54    22   41%
Hernandez Jose  50    20   40%
Helms Wes       51    19   37%

While this list still includes a couple players with high walk rates such as Adam Dunn and Richie Sexson, it now makes clear the second reason – these players also have high strikeout rates (Mark Bellhorn doesn’t make the list because he had only 47 opportunities although he scored the runner 28 times for a respectable percentage of 59.6%). By the way, when excluding walks Barry Bonds opportunities drop from 68 to 27 but his percentage skyrockets from 37% to 81%).

Overall, when walks are excluded the total number of opportunities drops to 18,793 and the scoring percentage climbs to 56.3%. Here are the leaders with walks excluded:

               Opp   Runs  Pct
Sheffield Gary  100   76   76%
Crede Joe        55   40   73%
Lawton Matt      51   37   73%
Renteria Edgar   86   62   72%
Helton Todd      76   54   71%
Casey Sean       65   46   71%
Ramirez Aramis   64   45   70%
Tejada Miguel   107   75   70%
Anderson Garret  66   46   70%
Guillen Carlos   62   43   69%
Cruz Deivi       52   36   69%
Huff Aubrey      81   56   69%
Beltran Carlos   60   41   68%
Cabrera Orlando  68   46   68%
Damon Johnny     68   46   68%

The real question, however, is whether the ability to plate a runner from third with less than two outs is an actual skill or whether this is a manifestation of small sample sizes and/or luck. To try and find out I took a look at the 87 players who had at least 25 opportunities in each season (excluding walks). Running a regression on the percentage of runners driven in yielded a correlation coefficient of .321 – not random but not exactly strong either.

Of course, the first thought that comes to mind is that there should be a correlation from year to year since the scoring rate should be tied to batting average. After all, players with higher batting averages will tend to drive in more runners from third with less than two outs. In a previous study, I looked at the variability of batting average and found that the correlation was .337, essentially the same as that for driving in runners from third. So is this totally dependant on batting average?

In order to isolate the effect of batting average I calculated the number of runners driven in over and above what would be expected based on batting average alone. In other words, for each season I multiplied the player’s batting average by the number of opportunities to calculate the number of opportunities that the player was “expected” to cash in. Of course, this isn’t a true measure of the number of runs expected since all players will also drive in runs from third on fly balls and ground balls – which is the ability we’re trying to measure – here I’m simply trying to establish a baseline. I then subtracted this from the actual number they did drive in. The leaders in this new category of Runs Over Expected from Third (ROE3) were:

                Opp  Runs   Pct  ROE3
Sheffield Gary  100    76   76%  45.0
Tejada Miguel   107    75   70%  43.0
Renteria Edgar   86    62   72%  35.2
Delgado Carlos   85    57   67%  32.6
Batista Tony     85    52   61%  31.8
Huff Aubrey      81    56   69%  31.5
Rodriguez Ivan   84    54   64%  27.8
Giles Brian      81    51   63%  27.4
Helton Todd      76    54   71%  27.1
Cabrera Orlando  68    46   68%  26.7
Ramirez Aramis   64    45   70%  26.4
Crede Joe        55    40   73%  26.3
Damon Johnny     68    46   68%  26.3
Casey Sean       65    46   71%  25.9
Conine Jeff      68    45   66%  25.8

You can see that many of these players are the same ones in the leaders by percentage but just rearranged a bit and heavily sorted by opportunities.

Once again those at the bottom are dominated by players who barely make the 50 opportunity threshold over the two years.

                Opp  Runs   Pct  ROE3
Edmonds Jim      54    28   52%  12.3
Rivas Luis       50    25   50%  12.1
Stairs Matt      50    25   50%  11.3
Overbay Lyle     54    27   50%  11.1
Wilson Craig     51    24   47%  10.6
Crisp Coco       51    25   49%  10.5
Bautista Danny   52    25   48%  10.3
Figgins Chone    50    25   50%  10.2
Dunn Adam        54    23   43%   9.6
Aurilia Rich     51    23   45%   9.6
Beltre Adrian    63    27   43%   9.0
Sexson Richie    57    24   42%   8.8
Hernandez Jose   50    20   40%   8.2
Sanders Reggie   54    22   41%   7.3
Helms Wes        51    19   37%   5.7

A more fair comparison would to calculate the rate of course by dividing the ROE3 by opportunities adjusted for the number of runs that were expected to come up with a runs from third rate (R3Pct). To illustrate consider Gary Sheffield who had drove in 76 runners (45 more than expected) based on his batting average in 100 opportunities. His R3Pct would then be calculated as:

Expected Scores = 76-45 = 31
Opportunities beyond expected scores = 100-31=69

R3Pct = 45/69=.652

In other words, Sheffield came through 65.2% of the time when his batting average is taken into account.

The leaders in R3Pct are:

                Opp  Runs   Pct  ROE3 R3Pct
Sheffield Gary  100    76   76%  45.0  0.652
Crede Joe       55     40   73%  26.3  0.637
Lawton Matt     51     37   73%  23.4  0.626
Renteria Edgar  86     62   72%  35.2  0.595
Ramirez Aramis  64     45   70%  26.4  0.581
Cruz Deivi      52     36   69%  21.9  0.578
Casey Sean      65     46   71%  25.9  0.577
Tejada Miguel  107     75   70%  43.0  0.574
Anderson Garret 66     46   70%  25.6  0.562
Guillen Carlos  62     43   69%  24.3  0.561
Huff Aubrey     81     56   69%  31.5  0.558
Beltran Carlos  60     41   68%  23.5  0.552
Helton Todd     76     54   71%  27.1  0.552
Palmeiro Rafael 63     42   67%  25.7  0.550
Cabrera Orlando 68     46   68%  26.7  0.549

Those at the bottom include:

                Opp  Runs   Pct  ROE3  R3Pct
Crisp Coco       51    25   49%  10.5  0.287
Wilson Craig     51    24   47%  10.6  0.281
Bautista Danny   52    25   48%  10.3  0.277
Soriano Alfonso  65    31   48%  12.6  0.270
Burrell Pat      76    33   43%  15.5  0.264
Castilla Vinny   74    34   46%  13.8  0.256
Aurilia Rich     51    23   45%   9.6  0.255
Ramirez Manny    76    37   49%  12.9  0.248
Jones Andruw     82    37   45%  14.7  0.246
Dunn Adam        54    23   43%   9.6  0.237
Hernandez Jose   50    20   40%   8.2  0.214
Sexson Richie    57    24   42%   8.8  0.210
Beltre Adrian    63    27   43%   9.0  0.199
Sanders Reggie   54    22   41%   7.3  0.185
Helms Wes        51    19   37%   5.7  0.151

Poor Wes Helms can’t seem to get out of the bottom no matter what adjustments are made.

So now with R3Pct in place we can re-run the regression based on R3Pct to see if we get a clearer picture of what is going on here. When I did so the result was .340, a slight improvement over using batting average alone but still not a strong correlation.

My conclusion would be that players do have some control over driving in runners from third with less than two outs but that ability is closely tied to other more obvious abilities such as hitting fly balls, hitting ground balls to the right side of the infield, and not striking out. These other abilities are not known to be situation dependant and so it’s doubtful that players have much of an ability to bring the runner in from third over and above what their other abilities give them.

It’s also interesting to look at ROE3 in the aggregate for teams. For 2004 the numbers were:

        Opp  Runs   Pct   ROE3  R3Pct
CHA     263   167   63%   96.6  0.502
PIT     295   176   60%   99.2  0.455
SFN     344   207   60%  114.0  0.454
MIN     298   178   60%   98.8  0.452
TBA     339   201   59%  113.5  0.451
DET     316   189   60%  103.0  0.448
CLE     327   195   60%  104.8  0.443
ANA     366   219   60%  115.6  0.440
BAL     375   223   59%  117.5  0.436
TEX     305   178   58%   97.0  0.433
NYA     346   200   58%  107.2  0.423
ATL     314   180   57%   95.3  0.416
SLN     362   209   58%  108.4  0.415
LAN     296   167   56%   89.6  0.410
SEA     331   186   56%   96.7  0.400
BOS     359   202   56%  100.8  0.391
PHI     326   179   55%   92.1  0.385
KCA     297   161   54%   84.2  0.382
SDN     371   203   55%  101.7  0.377
MIL     299   158   53%   83.9  0.373
CHN     285   154   54%   77.6  0.372
OAK     294   159   54%   79.7  0.371
FLO     304   163   54%   82.8  0.370
TOR     312   166   53%   84.9  0.368
HOU     349   184   53%   90.9  0.355
ARI     295   151   51%   76.5  0.347
MON     279   142   51%   72.6  0.346
COL     289   151   52%   71.7  0.342
CIN     303   150   50%   74.2  0.327
NYN     288   137   48%   65.4  0.302

It’s important to remember when looking at the team aggregates, however, that these do not reflect overall how successful a team was in scoring the runner from third but only the aggregation of all opportunities for a team’s players to do so.

To illustrate the difference consider the situation where Corey Patterson triples to lead off the first inning. Derrek Lee then has a chance to score the runner but instead pops up. Nomar Garciapara then hits a grounder to the right side and Patterson scores. In this scenario the Cubs would be credited with a 50% efficiency since they had two chances and delivered once. However, you could also say that Cubs were 100% successful since they did actually score the single runner Patterson in the inning. Overall, then the odds of scoring a runner from third are greater than these percentages. In taking a look at my Big League Manager you can see that from 1999-2002 teams had the following scoring probabilities.

Outs  Runners    Pct
0 3 86.4
0 23 85.6
0 123 87.2
1 3 66.2
1 23 69.5
1 123 67.0

Hawkins' Effectiveness

Saw this interesting bit on the Cubs web site in the story on Mark Prior's denial of using steroids in response to the admission by his mentor Tom House to doing so while a player.

"The folks at STATS, Inc., did a little extra research on LaTroy Hawkins and pitchers and their effectiveness in the ninth inning. Since 1995, Hawkins is 17-for-30, or 57 percent, with a one-run lead in the ninth, while STATS, Inc., says all pitchers in baseball have a 75.1 percent success ratio.

With a two-run lead, Hawkins has a 91.3 percent ratio, while the rest of baseball is at 89.4 percent. And, STATS, Inc., says Hawkins has an 89.5 percent success ratio with a three-run lead while other pitchers have a 95.5 percent rate."

So there is some credence to the view that Hawkins is somehow not well-suited to the closer role.