Saturday, April 24, 2004
The game has been over for five minutes and the Mariners dugout is empty, save for Freddy Garcia. Freddy is still sitting in the dugout, in the same place he sat durig the ninth inning, with an empty Gatorade cup in his hand, the same cup he had in his hands during the ninth inning. He's visibly disgusted.
How Do You Measure Pain?
Rowdy at Honest Wagner still wants to claim "Worst pinchitter ever" title for Abraham Nuñez, on the basis that Nuñez's three-year line as a PH is worse than Mabry's three-year line.
It's just a matter of how you measure pain. Nuñez is like the continuing drip of a Chinese water torture, whereas Mabry is like a kick in the gonads.
But when you consider the rest of the Mariners bench last year and this year, I feel as if someone kicked me in the gonads, even while I was strapped down with the water dripping on my forehead.
You Think You've Got It Bad?
Rowdy at Honest Wagner is frustrated with Abraham Nuñez's contributions as a pinchitter:
Nunez was 11 for 104 as a pinch-hitter from 2000 to 2003. The full numbers are .106 / .174 / .154. That's an OPS of 328. Bob Smizik wasn't kidding when he called Nunez 'amazingly inept' as a pinch-hitter. I knew he was bad, but I didn't realize he was that bad until I looked it up.Nuñez's numbers as a pinchitter are truly atrocious numbers. But, Rowdy, we Mariner fans can do you better. See, we got to see John Mabry come to the plate 24 times last year as a pinchitter, posting a line of .053/.217/.053/.270.
Is Abraham Nunez the worst pinch-hitter ever? I'm serious. I think he could be.
Mariner fans should be concerned, though. Nuñez is a toolsy no-hit middle infielder, just the kind of player that the Mariners love to pick up. He'll probably be part of a Kendall deal, should that happen.
Friday, April 23, 2004
- the oil to lube the Mariner chassis?
As the Mariners continue to scuffle, my feeling that a trade for Kendall is imminent becomes stronger. The Mariners are clearly ready to get rid of Davis, and Davis didn't help his cause with the first wild pitch tonight. (That one went underneath his glove and between his legs.)
Bavasi might look at a trade as a way to shake up the club and strengthen the catcher position at the same time. If he includes a pitcher from the 25-man roster in the deal, the team also creates a roster space. When Soriano comes back from the DL, a roster move will need to be made - could that be the impetus?
Added comments: Just to clarify, I am not in favor of trading for Kendall. While it would improve the team near term, I think the money could be better spent elsewhere on the team, particularly in the outfield. Longer term, Kendall's contract is almost certain to weigh down the team in 2006 and 2007.
In other words, it's not the optimum use of money now, and it's a bad use of money long-term.
Reaching an agreement on salary is the big issue. My foreboding is that if the Mariners continue to scuffle, Bavasi will decide that something needs to be done. The Mariners will then increase the amount of Kendall's salary they assume or add prospects to the deal.
Quinton McCracken or Conor Jackson
Last December I posted about Conor Jackson, a stud player in the Diamondbacks organization that David Cameron mentioned to me at the USS Mariner pizza feed last December. Conor Jackson, essentially, is the player the Mariners gave up so we could have Quinton McCracken on the roster. That's because when the Mariners signed Greg Colbrunn last year they forfeited a first round draft pick, the 19th overall, to the Arizona Diamondbacks. Since the Mariners now traded Colbrunn for Quinton McCracken, the Mariners effectively traded that draft pick for McCracken
The Diamondbacks used that draft pick to select Conor Jackson, out of the University of California. Playing 1B for Yakima in the Northwest League last year (his first year of proball), in 257 ABs Jackson hit .319, with a .410 OBP and .533 SLG.
This year Jackson is playing for the Diamonbacks high-A affiliate in Lancaster. Here is his line in Lancaster so far this young season:
G AB H TB 2B 3B HR BB IB AVG OBP SLG
14 51 19 32 4 0 3 13 0 .373 .492 .627
Are Some Fly Ball Pitchers Also Good Double Play Pitchers? - Part 2
In a previous post, I presented data on double play rates achieved by various pitchers, adjusting the data for each pitcher's ground ball/fly ball tendencies. The analysis used data on the 142 pitchers who logged at least 100 innings pitched during the 2003 season. Details of this review are in my first post, Are Some Fly Ball Pitchers Also Good Double Play Pitchers? - Part 1. That post also has a table showing the data for all 142 pitchers.
In this post, I'll discuss some of my thoughts on these data. Feel free to leave your own thoughts using the comments link at the bottom of this post.
I think the first question that arises with the data is verifying that the variations are real and not just ordinary variability. In some ways, the data set seems much like data on Batting Average on Balls in Play (BABIP). There are both good and bad pitchers distributed throughout the data set, just as we see with BABIP data. Because we would expect that the best pitchers would be best able to adjust their pitching style to get double plays when needed, our intuition tells us that the data should show some segregation of good and bad pitchers. That doesn't mean we should dismiss the results, though, because we do evaluations of this type just so we can challenge our assumptions.
To address this question, similar analyses should be conducted for different years, looking to see if the same pitchers are consistently better or worse than expected in enabling their infielders to turn double plays. If this is a skill that some pitchers have, we should see the skill manifested consistently from year to year. Conversely, if pitchers rankings consistently change greatly from year to year, that would suggest that the statistic has a lot of ordinary variability. I obtained the raw data for this review from Baseball Prospectus Statistics, which only had data for the 2003 year.
The table below shows the net extra DPs recorded by each team pitchers in the data set, as compared with those pitchers expected fly ball adjusted DP rate.
In 2003 Seattle led baseball with 40 more DPs turned than expected. Seattle for outpaced the next closest teams, the Mets and the Marlins, each of whom recorded 25 double plays above expected. Of the five Seattle pitchers in the 142 pitcher data set, four of them ranked in the top 20 for NetDP. Piñiero had the lowest NetDP, and he still ranked 50th out of the 142 pitchers.
Other teams in addition to the Mariners had pitching staffs that consistently ranked near one end of the NetDP continuum or the other. The Mets, for example, had three of their four pitchers in the top 25, and their fourth was ranked 58. The Marlins had their top four pitchers ranked between 3 and 33, with their fifith, Beckett, at 106. At the other end, the Cubs had five pitchers on the list, with rankings of 88, 104, 127, 129, and 134. These consistencies suggest that the variations in NetDP are not random, because if NetDP were random, we wouldn't expect to see such consistencies among so many teams.
The observation that there is consistency among teams, however, points to a second question to consider about the data, specifically, the extent to which the data reflect factors other than pitching? If there is consistency among teams, that consideration suggests that Net DP is a team defense statistic, not a pitching statistic. Additional work with zone factors might help answer that question.
At the moment, I think that NetDP may be more a team statistic than a pitching statistic, simply because of its consistency within teams. If this is the case, the loss of defense in the Mariners infield this year could compound with the weakened outfield defense. In my review of the Mariners Infield two months ago, I made the following summation:
The Mariners are punting defense in the infield, hoping that the added offense will offset the defensive losses. This is not necessarily a bad strategy. Given the fly ball tendencies of the Mariners pitching staff, it makes more sense to sacrifice defense in the infield than the outfield. … If Olerud can regain some of his lost power and Boone doesn't slide, there may be enough infield offensive improvement to offset the defensive losses. If, however, Olerud continues to decline and Boone's year is similar to his 2002 season, the Mariners infield will be noticeably worse offensively as well as defensively in 2004.Were I writing that review now, I would be more harsh in my assessment of the changes in the Mariners infield defense. While the losses in outfield defense are significant, the Mariners pitchers may have also have thrived in 2003 because of the ability of the infield to turn double plays when needed.
I hadn't looked at the effects of adding Aurilia this closely until I prepared this analysis. Before doing this review I was relatively indifferent towards brining in Aurilia. After looking at this information I now conclude that spending the extra money on Aurilia was probably wasted. Given that the modest increase in offense that Aurilia will provide is largely offset by the loss of defense, the added salary paid Aurilia not appear justified. There is a very good possibility that in 2004 the Mariners will pay more and get less at shortstop than they would have obtained had they kept Guillen and just added a serviceable backup infielder.
Thursday, April 22, 2004
Are Some Fly Ball Pitchers Also Good Double Play Pitchers? - Part 1
In a previous thread, I presented some data on the relationship between various pitchers' double play rates (DPR) and ground ball to fly ball ratios (GB/FB). DPR is the number of double plays turned behind a pitcher, divided by the number of double play opportunities. GB/FB is the ratio between ground balls and fly balls hit by batters facing a pitcher. Foul balls are counted only if they are caught for outs.
To summarize, because most double plays involve ground balls, DPRs for ground ball pitchers are larger than DPRs for fly ball pitchers. The data I presented confirmed this, but the correlation between DPR and GB/FB was not as strong as I expected. I also showed that in 2003 Ryan Franklin's DPR was surprisingly high, particularly given his extreme fly ball tendencies. I hypothesized that perhaps Franklin was effectively adjusting his pitching approach to produce more ground balls in double play opportunities.
To look at these effects in more detail, I adjusted each pitcher's "expected" double play rate to reflect his GB/FB tendencies. For example, if a pitcher typically generates twice as many ground balls as the average pitcher, that pitcher's expected DPR should be twice as high as the average DPR. If his DPR is less than twice the average, then he is of less than average effectiveness in converting double play opportunities. Conversely if a pitcher generates half as many ground balls as the average pitcher, that pitcher's expected DPR ratio should be one-half the average DPR.
Note that if an extreme fly ball pitcher is able to adjust his pitching style to induce more ground balls in double play situations, this modified DPR measure will pick up that pitcher's adjustment. Similarly, if an extreme ground ball pitcher loses some of his ground ball effectiveness in double play situations (for example, he loses some bite off his pitchers when he uses a slide step), the adjusted DPR will identify that as well.
As before, I limited my analysis to the 142 pitchers who logged at least 100 innings pitched in 2003. This focuses the review on pitchers who provided a substantial portion of their team's workload. It also reduces the number of pitchers to a manageable number.
For each of the 142 pitchers I used the following data:
- Innings Pitched (IP)
- number of double play opportunities (DPOpps)
- number of double plays made (DP)
- GB/FB ratio
I then created an expected DPR for each pitcher by scaling the group average DPR by the pitcher's individual GB/FB ratio. For example, if a pitcher had GB/FB ratio of 0.75 his adjusted DPR would be:
Adjusted DPR = 0.136*(0.75/1.21)The expected number of double plays (DPExp) for each pitcher then becomes the Adjusted DPR times the total number of double play opportunities, DPOpps. I then compared the number of actual double plays, DP, with DPExp to see if the pitcher generated more or fewer double plays than expected, given his relative GB/FB tendencies. Continuing our example from the previous paragraph, if the pitcher had 150 DPOpps, his DPExp would be:
DPExp = 150*0.084By subtracting DP from DPExp, we can now determine the DPs the pitcher contributed above or below DPExp. Let's call that NetDP, and let's assume that our hypothetical pitcher had actually had 18 double plays turned behind him. Then:
NetDP = 18-13Note that 18 double plays is what the average pitcher would be expected to produce (0.121*150 = 18). So rather than concluding this pitcher was of average effectiveness in generating double plays, we would conclude that he was actually more effective, given his overall tendencies.
Below is the table. I'll present some of my thoughts on the data in a separate post.
|Seo Jae Weong||NYN||188.3||0.75||129||13||10.9||2.1||1.7%||58|
I just completed a long-delayed update to the site template. The major changes are:
- updating and expanding the linksto other blogs in the right sidebar.
- Adding comment and trackback features for each post
Worst Songs Ever
I can't believe Sports and B's aren't all over this one: Worst Songs Ever.
My contribution, I'm not ashamed to say, was "Afternoon Delight" by the Starland Vocal Band. To me for a song to be utterly and unredeemably awful, it needs to sink below mere schlockiness. Many of the other songs, while bad, are easily forgettable and pass from consciousness easily and painlessly. Some songs, though, continue to torment you by looping continuously through your head for hours or days. AD is in that latter category. I've despised it from the first moment I heard it, and yet I'll remember it until I die. I'm half-expecting my kids to play it at my funeral as a cruel joke to those few who will mourn my passing. (I'll be dead, so I'll finally be free of it.) For readers young enough to have never heard AD, be thankful, be thankful.
Note to David: while holding down the Alt key, type in 0202 on the numeric keypad (for upper case) or 0234 for lower case. And do it in Blogger, because if you do it in an editor and paste it into Blogger, Blogger might give you some weird characters when it publishes. (You know how easily Blogger gets confused.)
Nice suggestion from Brian Rust, as a comment at Mariners Musings:
Couldn't you just rename the right-field restaurant "Eating Raul"?
Midnight Mariner Ramblings
Some comments from tonight's game:
- Long and late work hours lead to naps on the sofa during the middle innings.
- Upon awaking I noticed that Hudson was having trouble going low and away on left-handed hitters. This was apparent during Hansen's at bat. The first pitch, which Hansen lined foul to the left, was about waist high, but the target was low. On the second pitch, Melhuse set up away again, and Hudson hit the target. On Hanson's homerun, Melhuse again set up low and away, and Hudson was on the inner half about waist high.
Same thing happened on Spiezio's single in the 8th.
- Sending Hansen to bat for Davis was a good move by Melvin, and I would have said so even had Hansen not produced. I have railed before about how little utility a left-handed pinch hitter adds to the current roster, but Melvin did find the right time to use Hansen.
- So far Jarvis is not the Carrara Evil Dead sequel I feared. I no longer cringe when he starts warming; I now save my cringes for Myers and Hasegawa.
- As I've mentioned before, it's still early to be drawing any but the most tentative conclusions about particular players. With that caveat, though, I think one safe conclusion is that Hasegawa will perform in 2004 pretty much as Hasegawa has performed for his entire career except for one five month stretch in 2003. In Hasegawa's case, the small sample size is his 2003 performance through early September. His performance since then just correlates with the rest of his career.
- While Arthur Rhodes certainly looks better than he did in the last half of 2003, he does not look like the Arthur Rhodes of 2001 and 2002. Nevertheless, barring injuries, I expect that my mid-season it will be clear that the Mariners should have kept Rhodes and let Hasegawa walk.
- Melvin obviously has doubts about McCracken's offensive abilities, or he wouldn't have had Cabrera on deck to hit for McCracken at the end of the game. McCracken has only 6 PA's so far this season, further indicating that Melvin is not expecting McCracken to provide offense. Of course, considering how Melvin didn't use Colbrunn last year even when he was available, we shouldn't infer too much from the McCracken's sparing plate appearances.
- Hudson induced the Mariners to hit five double play groundballs tonight. That started me wondering about the efficiencies of individual pitchers and of team pitching staffs in obtaining double plays.
Knowing that Hudson is an extreme ground ball pitcher, I first looked at data on ground ball to fly ball (GB/FB) ratios. I used data for all 142 pitchers who logged at least 100 innings pitched in 2003. In 2003 Hudson had a GB/FB ratio of 2.04. That was the tenth highest GB/FB ratio among those 142 pitchers with at least 100 innings pitched in 2003. Here is the complete list of the top 20:
GB/FBFor comparison, the Mariner pitcher with the highest GB/FB ratio on the list of 142 pitchers was Joel Piñiero at 1.13, good for a ranking of 68. The Mariner pitcher with the lowest GB/FB ratio (with at least 100 innings pitched) was Ryan Franklin at 0.63, ranking him 137th out of the 142 pitchers who logged at least 100 innings. The pitchers with lower GB/FB ratios than Franklin were:
GB/FBTo convey how extreme the Mariners fly ball tendencies have been, here are the team GB/FB ratios for the 2003 season:
GB/FBAs indicated, 18 of the 30 teams in MLB had higher staff GB/FB ratios than Joel Piñiero, the Mariner starting pitcher with the highest GB/FB ratio. Seattle had the lowest GB/FB ratio of any staff in baseball. As I and others have pointed out many times, this means that the Mariners in 2003 were the team most dependent on outfield defense. Accordingly, weakening the outfield defense will have a bigger impact on the Mariners in 2004 than it would on any other team in baseball.
Next I considered the efficiency of the top 20 ground ball pitchers in converting double play opportunities. In the table below, "DP rate" is the ratio at which a pitcher achieves a double play as a fraction of the total double play opportunities, i.e., situations with less than two outs and first base occupied. The "DP Rank" is where that pitchers DP conversion rate ranked among the 142 pitchers, with 1 being the pitcher who had the highest DP rate out of the 142 pitchers. "DP Rank %" is the pitchers DP rate ranked as a percentage of the 142 pitchers, i.e., a 10% ranking means the pitcher's DP rate was in the upper 10% of DP rates.
GB/FB DP DP DPThose data are pretty interesting. While extreme ground ball pitchers convert a higher percentage of double play opportunities than their fly ball counterparts, the correlation is not as strong as I thought it would be.
Name Ratio Rate Rank Rank %
--------------- ----- ---- ---- ------
Lowe_Derek 3.26 0.16 34 24%
Brown_Kevin 3.09 0.17 28 22%
Webb_Brandon 3.06 0.12 90 64%
Westbrook_Jake 2.74 0.22 3 2%
Day_Zach 2.60 0.17 18 13%
Halladay_Roy 2.34 0.19 8 5%
Brower_Jim 2.24 0.22 3 2%
Jimenez_Jose 2.20 0.16 31 22%
Cook_Aaron 2.16 0.13 70 53%
Hudson_Tim 2.04 0.16 31 23%
Zambrano_Carlos 2.02 0.17 16 12%
Mulder_Mark 1.92 0.19 9 6%
Perez_Odalis 1.80 0.19 9 6%
Estes_Shawn 1.80 0.16 29 22%
Clement_Matt 1.79 0.13 61 47%
Torres_Salomon 1.78 0.16 29 22%
Batista_Miguel 1.78 0.14 55 43%
Maddux_Greg 1.68 0.16 29 23%
Suppan_Jeff 1.60 0.15 40 32%
Pettitte_Andy 1.60 0.12 70 57%
Now for the real shocker. Guess which Mariner starting pitcher had the highest DP rate in 2003??
None other than Ryan Franklin, with a DP rate of 0.17, good for a #31 ranking out of the 142 pitchers. That's right - Franklin, one of the most extreme fly ball pitchers in baseball, ranked in the top 25% of these 142 pitchers in inducing double play groundballs in 2003. Given his fly ball tendencies, I think that's pretty amazing. I don't know whether he was just lucky in 2003, or if he actually is one of the best pitchers in adjusting his pitching style to match the situation.
Another surprising result: the Mariners starter who was least efficient in obtaining double plays was Jamie Moyer, DP rate = 0.12, ranking 91st in our list of 142 pitchers.
Here is the complete table for Mariner starting pitchers:
GB/FB DP DP DPNote: All of the data used in this review are from Baseball Prospectus Statistics. The BP Gb/FB ratios differ from the GB/FB ratios at many other sites, such as ESPN.com and MLB.com. I believe this is because BP data are for all batted balls, whereas STATS calculates GB/FB rations only using balls that are converted to outs.
Name Ratio Rate Rank Rank %
--------------- ----- ---- ---- ------
Franklin_Ryan 0.63 0.17 31 22%
Meche_Gil 0.83 0.15 52 37%
Garcia_Freddy 0.94 0.15 56 39%
Pineiro_Joel 1.13 0.14 59 42%
Moyer_Jamie 0.80 0.12 91 64%
Tuesday, April 20, 2004
Be Careful What You Wish For
Amid the Mariners initial sputtering this season, we've had various bloggers expressing hope that the team's lack of success would cause management to realize the error of their ways, fire Bavasi, and bring in more progressive managment. Being the eternally optimistic blogger that I am, I must respectfully disagree about the outcome of such a turn of events.
Let's assume for a moment that the team owners concluded that hiring Bavasi was a dreadful mistake that needed to be corrected. As the owners ponder Bavasi's replacement, they review the Mariners performance, particularly observing that the years since 1995 have been the most successful in team history. They also note that the Mariners have won more than 90 wins every season during the Gillick years, capped by the 116-win season of 2001.
Given that background, which of these scenarios do you think is most likely?
- Owners conclude that the team performance in the Gillick years is an aberration, amd they need to find someone to change the team legacy. Accordingly, they fill the GM spot with someone who will rework the team's entire operating approach.
- Owners conclude that the team's overall plan is sound, but they unfortunately picked the wrong person to carry on the team legacy. Accordingly, they fill the position with another person whom they feel is best able to continue the team's Old School legacies and traditions.
Frankly, the earliest I could ever see Bavasi being replaced is two years from now, and I doubt that it would happen even that quickly. Even then I believe that the team would stay close to the "old school" principles they believe to be the foundations of the Mariners successes since 1995.
Riding the Pine This Week
My work schedule this week will prevent anything but cursory blogging.
With that, a couple of quick comments.
- The optimist in me is hoping that I will be proven wrong in my comments about Ibañez, Jarvis, Spiezio, etc. But just as the first ten games did not define the Mariners season, neither do the last four games. It's still far too early in the season to make any but the most general conclusions. A few things are clear (team defense, particularly in the outfield, is worse), but whether team is better or worse this year is far from clear. We also cannot draw any conclusions yet about any specific players.
- Sabernomics is a nice blog, worth checking out if you like a combination of sabermetric numeric work and baseball business/economic stuff.
Sunday, April 18, 2004
Mark Kotsay's Wife!
From Chris caldwell of At Least The Red Sox Have 1918:
Whoever the lead announcer of the Oakland-Anaheim game is, he just told America, in so many words, how much of a MILF he thinks Mark Kotsay's wife is.For those three or four people who might be wondering what Chris is talking about, click here for more information.
A Clutch Hitting Double Clutch
Nate over at Hope springs eternal... weighs in on clutch hitting:
Does it seem to anyone else that the Mariners are stranding a lot of runners? Well, they're not, even though it seemed that way to me. The Mariners are hitting quite well in so-called 'clutch' situations. As a team overall, the Mariners have a .318 OBP. With runners on, it jumps to .352, and with runners in scoring posn., it goes up to .404. If you don't believe 'clutch hitting' is a skill, this is yet another ominous sign, since they've actually been lucky to score as many runs as they have (M's are 26th in the majors in OPS, but 20th in Runs scored). If you do believe in 'clutch hitting,' then this isn't an area likey to improve.To assess whether the Mariners are doing well or poorly with men-on-base ("MOB") or runners-in-scoring-position ("RISP"), we need to make the comparison a bit differently. In general, batting statistics go up in MOB and RISP situations. That's because pitchers pitch differently and teams defense differently with MOB and RSP. For example, pitchers use slide steps, they're more conscious of not walking a hitter, they're distracted by the runner, or they try to induce groundballs or popups. Teams may move fielders in or back, and when infielders hold runners to the bag, that opens up holes in the infield. So we shouldn't compare the Mariners performance with MOB and RISP with the team average OBP; rather we should look at whether the Mariners increase in hitting in those situations is larger or smaller than the average expected increases in those situations.
With the help of ESPN stats, I extracted the following information for all American League teams in 2003:
DifferenceApplying these percentages to the Mariners 2004 performance cited by Nate yields:
Situation OBP from Overall
========= ==== ============
Overall .333 -
MOB .346 +3.64%
RISP .352 +5.49%
OBPI agree with Nate's overall conclusion that the Mariners have actually been relatively efficient in getting "clutch" hits. The difference just isn't quite as big as he indicated.
Situation Expected Actual
========= ======== ======
Overall .318 -
MOB .330 .352
RISP .336 .404