Left Field And DH

Dave · December 29, 2009 at 2:24 pm · Filed Under Mariners 

Let’s face it, the weakest part of the Mariners roster right now is the offense. And, because people are generally used to team building through improving weaknesses (even though that’s not always the best idea), a lot of fans want the Mariners to add more offense to the roster. Traditionally, left field and designated hitter are two spots where you expect to get a decent amount of offensive production, so it’s pretty common to see people suggest that the M’s should pursue another hitter that plays one of those two spots, in addition to the first baseman that is clearly needed.

I’d like to suggest that the team is set at both LF/DH with what they have. And they can actually expect some pretty decent production from those two spots, given a four man combo platoon/rotation of Milton Bradley, Bill Hall, Ken Griffey Jr, and Ryan Langerhans (or Michael Saunders, but I’d rather he play every day in Tacoma than play a part-time role in Seattle).

Each position will be given something like 700 plate appearances in 2010, split at approximately 500/200 against RHP/LHP. With the four guys currently penciled into those spots, the breakdown should look something like this.

LF, vs RHP: Langerhans (250), Bradley (150), Hall (100)
LF, vs LHP: Hall (125), Langerhans (50), Bradley (25)

DH, vs RHP: Griffey (250), Bradley (150), Various (100)
DH, vs LHP: Bradley (100), Griffey (50), Various (50)

The regular line-up against lefties is easy – Hall in left and Bradley at DH. Both hit LHPs well historically, while Hall is the better defender, so Bradley gets to rest his body when a southpaw is on the mound. Against righties, it gets a little more complicated – Bradley will play when he can, with Langerhans/Griffey essentially splitting the other spot depending on Milton’s health and whether Wak wants to maximize his defense that day.

Using this kind of job sharing plan, the total PA breakdown would be as follows: Bradley (425), Griffey (300), Langerhans (300), Hall (225), Various (150).

In terms of decision making, three of those spots are fixed. Bradley and Griffey are on the team – that’s already been decided. Because of their respective health problems, you have to build in expected injury time for both, which is why the unnamed Various players are accounted for. That may be Mike Carp filling for a few weeks when Junior is on the DL or Wak using the DH to give his regulars a half-day off, but there will be playing time at DH doled out to guys who don’t begin the year on the roster.

That means the only two potential players you could replace are Langerhans or Hall, and in reality, you probably can’t toss Hall off the roster – he’s due $1.3 million from the club for 2010, and given his ability to backup a lot of positions while also serving as a RH platoon LF, you’re probably not going to be able to find a guy who offers the needed flexibility that this roster requires. So, that just leaves Langerhans.

He’s penciled for 300 plate appearances, and he’s something like a +1 to +1.5 win player over a full season, so you’re expecting about +0.5 to +0.75 wins in production from him in that role. Let’s say you decide you want to replace him with a better hitting outfielder, a guy who is a +2.5 win player over a full season. We’ll call him “Johnny Damon”, just for fun.

Obviously, you’re not going to limit that guy to Langerhans share of the playing time. So, you reduce the number of days that Bradley and Hall get in left, with Bradley shifting more to a DH role and eating away at some of Junior’s playing time. Perhaps the new allocation of resources looks like this:

“Damon” (600), Bradley (400), Griffey (200), Hall (100), Various (100)

You get Damon’s +2.5 wins, but you lose Langerhans +0.5 win and some portion of Hall’s expected production against LHPs. You also lose the ability to play the match-ups depending on who is pitching on each day, and you make Hall the backup center fielder, making the team worse on days when Gutierrez can’t play. All told, you’re punting about +1 win of value, so the real upgrade is about +1.5 wins.

How much do you think the team should pay for that +1.5 win upgrade? Wins are going for about $3.5 million apiece on the market this year, so you can’t expect Damon to sign for less than $7 million per year, and rumors have him asking for more like $10 to $12 million per season. Even if you think the M’s could get Damon (or someone like him – remember, he’s just a placeholder for Better LF Hitter Guy) for $7 or $8 million, the team would be paying about $5 million per actual win added.

There are simply much better ways to spend the remaining money in the budget. The team currently has ~0 expected value from first base, as Carp is probably replacement level-ish. As we talked about a few weeks ago, a guy like Lyle Overbay could add +2 wins in value at first base, and he’d cost at most $7 million, and probably less (assuming Toronto would kick in some cash to get rid of him). The team also has a bunch of near replacement level options for the #5 starter spot, so adding a starting pitcher from the Smoltz/Martinez/Sheets/Wang/Bedard group could easily add +1 to +2 wins in value, while certainly coming at a lower cost than adding an offensive minded left fielder.

The Mariners do not have enough money for an LF, 1B, and an SP. If they get good value, they might be able to afford two quality players at those spots. It seems clear to me that 1B and SP are the cheaper, easier routes to upgrade. The current group can handle LF/DH, and provide good value to the team. The M’s don’t need to add another bat there. There are better ways to upgrade the roster, even if it doesn’t satisfy some people’s constant need for “a power bat”.

Comments

127 Responses to “Left Field And DH”

  1. thehemogoblin on December 30th, 2009 1:49 am

    Games played at SS since 2006:
    Bill Hall: 0

    Jack Hannahan: 2

    I don’t know what you’re talking about… we’re fiiiine… /sarcasm

  2. eponymous coward on December 30th, 2009 1:52 am

    You think? I obviously don’t have the WAR formula memorized, but turn just five of his warning track fly balls into homers and his OPS shoots up by .041 points. That seems pretty significant by itself and depending on where you put him it could be a conservative estimate. Plus that totally disregards that fact that if he played somewhere where it was easier to hit homers, pitchers would probably throw him fewer strikes and he might actually walk thirty times. It seems like all that could add up to at least a quarter win, but like I said, I don’t have the formula memorized. Maybe it’s less than I’m envisioning.

    WAR isn’t park-neutral? How on earth can you use it to compare values of players in different parks if it’s not park-neutral? And shouldn’t the point be to adjust it to be truly park-neutral?

    Jose Lopez doesn’t magically become a better hitter in a different park that doesn’t relentlessly scre RHB like Safeco- his skillset remains the same, the environment the skillset is used in is different, and so the results are different. Judging player value needs to be separated from the environment they play in. WAR doesn’t do that?

  3. eponymous coward on December 30th, 2009 2:02 am

    I don’t know what you’re talking about… we’re fiiiine… /sarcasm

    I’m not exactly sure Hannahan is the answer as Wilson’s backup either, but he’s closer to an answer than Hall is. (Ideally, we’d have a roster that had a proven backup to Wilson instead of a cripple at DH, and someone who looked a lot better than Hall did last year as an OF to platoon with Langerhans in LF and spell Bradley every so often, Bradley would be the DH, and you’d sign a 1B… but that’s not going to happen this year.)

  4. leon0112 on December 30th, 2009 4:14 am

    It seems to me that Saunders will play in Tacoma until Bradley blows up or Griffey gets injured. One of those things will happen at some point in the season. Until that happens Saunders get more AB and playing time in Tacoma.

    It would seem to me that Z has played this well. Bradley can either exceed expectations or Saunders will have a slot to fit into when Bradley blows up.

    Let’s find a 1B and SP.

  5. nepacific on December 30th, 2009 5:54 am

    As for Griffey, don’t underestimate the importance of tickling Ichiro and supporting Bradley’s fragile ego. That could be quite a few wins right there.

  6. hansk on December 30th, 2009 6:40 am

    I think like most people here I love all the moves so far this season with the exeption of the Morrow trade and the Griffey bring back.

    I understand the latter wa

  7. hansk on December 30th, 2009 6:42 am

    I think like most people here I love all the moves so far this season with the exeption of the Morrow trade and the Griffey bring back.

    I understand the latter was a business decision, but it may come at the cost of bringing Branyan back. Which leads me to the question: How many wins did we give up by having Griffey/Hall/Langerhans/etc. filling those 500 or so at bats that would have gone to Branyan?

  8. Paul B on December 30th, 2009 7:11 am

    Well, yes, he is hurt by Safeco Field, but let’s not overstate it. The difference between Lopez in Safeco and a nuetral park is probably something more like 1/10th of a win.

    You think? I obviously don’t have the WAR formula memorized, but turn just five of his warning track fly balls into homers and his OPS shoots up by .041 points. That seems pretty significant by itself and depending on where you put him it could be a conservative estimate.

    Your intuition is correct. The difference is much more than one tenth of a win.

    Lopez:
    Career, home OPS: .689
    Career, road OPS: .740

    That’s a significant difference, but last season it was crazy-off-the-charts different:

    2009, home OPS and sOPS+: .641 65
    2009, road OPS and sOPS+: .883 138

    Last year, at home, Lopez at the plate = Rob Johnson (.615 OPS)

    Last year, on the road, Lopez at the plate = Russell Branyan (.867 OPS)

  9. Paul B on December 30th, 2009 7:16 am

    Judging player value needs to be separated from the environment they play in. WAR doesn’t do that?

    I don’t see any splits for park in the Fangraphs player data. Am I just missing it?

    What I’ve always done is go to baseball reference and look at OPS+ (which is park adjusted) as well as look at OPS for the home road split (which is what I did for Lopez).

    If there is a better way, point me to it and I will use it in the future.

  10. Paul B on December 30th, 2009 8:05 am

    Games played at SS since 2006:
    Bill Hall: 0

    although, career wise, he has played 261 games at short, more than any other position except for 266 games at third.

    I think a bigger concern with Hall is his hitting. His wOBA the last 4 years is .369, .317, .297, .261.

    Not a good pattern. The M’s are really depending on a regression to the mean in 2010, as opposed to the trend representing a decreasing underlying skill set in which case he’ll be out of baseball very soon.

  11. mattlock on December 30th, 2009 8:30 am

    but why hasnt the deal happened yet (either in the Lee Deal or the Morrow deal?)

    This comment was made in relation to Overbay. This is my question, exactly. What does this mean? Does it mean that it has been discussed and they haven’t agreed on anything yet? Does it mean that it just hasn’t happened yet? Or does it mean that it won’t happen?

    Given the Jays desire to get rid of him and the Mariners need for a cheap reliable option at 1st, it seems odd that two deals involving the two teams could have come and gone with hardly a whisper of his name–that is, unless it’s something that just isn’t going to happen.

    Perhaps I’m just over-analyzing things…

  12. Willmore2000 on December 30th, 2009 8:36 am

    Back to Vlad Guerrero for a minute, do his knees prevent him from playing 1st base some of the time, and couldn’t Bradley play 1st base the rest of the time?

    Couldn’t we also then start with Saunders as the regular LF, and only demote him to AAA if he does not perform. Sure, this makes Langerhans an overpaid bench warmer, but to test out what we have in Saunders, I think it’s worth it, if it doesn’t work out, back to the platoon solution.

    So a regular lineup would be something like:

    1B/DH: Guerrero/Bradley
    2nd/3rd: Lopez/Figgins
    SS: Wilson
    LF: Saunders
    CF: Gutierrez
    RF: Ichiro!

    Guerrero, supposedly, is relatively cheap, so it would also leave some cash for a Sheets/Pedro signing for the #4 spot in the rotation.

    Again, this all hinges on Guerrero’s knees, and if he can play even 40-50 games at 1st, but if he can, then I think it could work.

  13. Toddk on December 30th, 2009 9:32 am

    â– Brzeczyszczykiewicza on December 30th, 2009 12:43 am
    Bradley can’t be counted on even if he stays healthy. With his mental problems in addition to his injury history, it’s unlikely he plays the whole season no matter where he plays. So if we get a DH and make Bradley play LF everyday and he eventually gets hurt, it’s not a big deal because we didn’t expect him to make it through the entire season anyway, and he was still worth the $3 million investment, and the lineup still has a good DH so it’s no worse than if we still had Bradley and didn’t sign a DH.

    But the team only has ~$10M left to spend. How do you propose filling 1b+SP+DH on that budget?

    Also, is Brzeczyszczykiewicza pronounced exactly as it’s spelled?

  14. georgmi on December 30th, 2009 9:34 am

    Now Lopez is getting $2.3 mil this year while Langerhans is getting $.5 so if you want to get into value (which I assume is what the guy who said Langerhans is better than Lopez was referring to)

    No, when I said it, I meant that Langerhans was a better baseball player than Lopez. I will admit to some hyperbole in the statement, but I meant it to be read the way it came across.

    I don’t know the details of Langerhans’ 2007, but at first glance I find it hard to believe at first glance that 244 PAs spread across three different teams in a single season is likely to be representative of a player’s actual talent, when all the other seasons have had him significantly better than that.

    What is your academic background in statistics? Humble me.

    I have to say I’m more interested in having a discussion than in humbling anybody, but since you asked so nicely, I have degrees in Economics and Information Systems, and while it has been a few years since I actually designed and ran an experiment to prove an hypothesis, I do still know the difference between a standard deviation and a least-squares regression.

  15. amnizu on December 30th, 2009 9:47 am

    Dave,

    Great analysis. It really brings out the roster flexibility that Z has built into the lineup going into 2010 even with Griffey occupying a bench spot. My only real concern here is what happens if (and probably when) 2 or 3 of these guys get hurt. If you look back at Bradley, Griffey, Hall and for arguments sake Jack Wilson you can pretty much count on 2 or more of these guys being hurt at the same time.

    This is why I tend to lean towards the M’s needing more flexibility out of their 1b option, someone like Luke Scott who you have mentioned in the past that can also cover LF if needed.

    This would allow more flexibility for the front office in who they call up from the minors to cover the vacated roster spot.

  16. rebuilder on December 30th, 2009 9:47 am

    Sign Branyan to a one year deal, Try to pick up Sheets if he is to expensive or signs somewhere else maybe take a look a V. Padilla as a SP option. Padilla does a good job of killing innings and posted a war of 2.0. Dont give out more than one year contracts on either guy for we need some flexability with 2011 payroll in we are to sign Felix.

  17. Prion on December 30th, 2009 11:18 am

    I’d love to see Randy Johnson back in a Mariners uniform. If we get him and Lyle Overbay I’d be pretty happy.

  18. TumwaterMike on December 30th, 2009 12:52 pm

    I like Sheets but I think he’ll go elsewhere on a 2 or 3 year deal. I don’t hink the M’s will take the risk on more then one year. Maybe resign Washburn for 1 year, if he does well trade him at the deadline again. Also Bedard would be an option but probably not until May. Fister could hold down that starter position until then.

  19. nathaniel dawson on December 30th, 2009 3:23 pm

    Well, yes, he is hurt by Safeco Field, but let’s not overstate it. The difference between Lopez in Safeco and a nuetral park is probably something more like 1/10th of a win.

    You think? I obviously don’t have the WAR formula memorized, but turn just five of his warning track fly balls into homers and his OPS shoots up by .041 points. That seems pretty significant by itself and depending on where you put him it could be a conservative estimate. Plus that totally disregards that fact that if he played somewhere where it was easier to hit homers, pitchers would probably throw him fewer strikes and he might actually walk thirty times. It seems like all that could add up to at least a quarter win, but like I said, I don’t have the formula memorized. Maybe it’s less than I’m envisioning.

    On second look, you may have been closer than I thought. Safeco depresses runs scored by about 5%, so last years team that scored 640 runs would have been expected to score about 660 in a nuetral park. With Lopez getting about 10% of the plate appearances, that adds about 2 runs to his offense. Probably a little more because he was better than the average Mariner hitter last year and right handers are hurt more. So maybe as much as 3 or 4 runs better in a neutral park.

    And for a couple people that asked, Fangraphs WAR (if that’s the version you were refering to) does not make any adjustments for park or league. At least not unless and until they switch to wRC+.

  20. eponymous coward on December 30th, 2009 5:01 pm

    And for a couple people that asked, Fangraphs WAR (if that’s the version you were refering to) does not make any adjustments for park or league. At least not unless and until they switch to wRC+.

    So Lopez is more like a 2.5-3.0 WAR player if park effects are removed, instead of a 2.0-2.5 WAR player? That’s a fairly substantial difference in value, and places him pretty solidly as a solid MLB 2B… and he’s STILL younger than Jeff Clement.

  21. mw3 on December 30th, 2009 5:50 pm

    Plus that totally disregards that fact that if he played somewhere where it was easier to hit homers, pitchers would probably throw him fewer strikes and he might actually walk thirty times.

    No he won’t. He has trouble identifying what is and is not a strike, he will never walk a lot. Look up players with simillar numbers in their early career, they usually max at 30-35 walks in their mid to late thirties. If they can stick that long.

  22. nathaniel dawson on December 30th, 2009 6:17 pm

    So Lopez is more like a 2.5-3.0 WAR player if park effects are removed, instead of a 2.0-2.5 WAR player?

    I suppose that would depend on which WAR metric you are using. Rally includes park and league effects, using TotalZone instead of UZR, and has Lopez at 1.7 WAR last year.

  23. DAMellen on December 30th, 2009 7:19 pm

    eponymous coward,
    Here’s my assumption (and for the record it’s totally possible that I’m wrong on this and if that’s the case, I’m totally willing to hear all about what’s right):
    I was under the impression that one of the major components of WAR is wOBA which DOES factor in park effects. However parks do not treat all hitters equally. I definitely think the Safe is a lot more damaging to pull power righties like Jose and Mike Cameron than it is to pull hitting lefties like Raul or Russ or even to less pull centric righties like Edgar and Bret Boone. I don’t think I’m alone on that (I’m not am I? AM I?). For that reason, I assume the park adjustments are imperfect. They can scale all Mariners hitters up by 5% because SafeCo allowed 5% less runs than the average park or whatever, but I had assumed that they do not factor in hitting style. Of course, I could be wrong. If they really are able to determine exactly how much SafeCo hurt Jose, then I’m even more impressed by sabermetric analysts than I was already.

    And then Nathaniel posts something that shows that my assumption was wrong and that I landed on the correct conclusion completely by accident. Eh, what the hell. I’ll post this anyways.

  24. Dave on December 30th, 2009 7:22 pm

    WAR on Fangraphs includes blanket park effects which adjusts everyone the same way – this works for retrospective value (which is what WAR is for) and not as well for true talent level discussions (which is what you guys are talking about in regards to Lopez).

  25. DAMellen on December 30th, 2009 8:12 pm

    Yes! I was right!

  26. nathaniel dawson on December 30th, 2009 11:30 pm

    WAR on Fangraphs includes blanket park effects which adjusts everyone the same way

    So I was wrong about that. I thought I had heard several times that Fangraphs didn’t use park factors.

    It’s easy to assume that Safeco hurts Lopez more than other players – certainly it hurts him more than left-handers – but as far as I know, no one has ever really studied the issue in depth. Without any knowledge to tell us any different, it’s probably best to say that Safeco affects Jose Lopez the same way it affects all other right-handers.

  27. ripperlv on December 31st, 2009 2:48 pm

    Nicely written and analyzed article with some good comments. I learned a few things.

Leave a Reply

You must be logged in to post a comment.