Lopez, His Hot Streak, and His Future

Dave · August 27, 2009 at 7:49 am · Filed Under Mariners 

You’ve probably noticed that Jose Lopez has been on fire for about three months now. After starting the season with a miserable slump, he was hitting .215/.259/.306 on May 26th. 191 terrible plate appearances that took up nearly 1/3 of his season, and he had accumulated all of three home runs and as many double plays as walks (10). He had played at a below replacement level performance – again. We’d seen this from Lopez before, as he has a history of going into long funks where he’s just an awful player. He’s rebounded before, though, and the hope was that he would rebound again.

Has he ever. Since May 26th, he’s hitting .306/.325/.559, a performance as valuable as it is weird. He’s racked up 27 doubles and 16 home runs in the last three months, but he’s also drawn a whopping five unintentional walks, which is why the OBP is still on the low end despite the fact that he’s been stinging the baseball. For a right-handed pull power hitter to slug .560 over a prolonged stretch of time while playing half his games at Safeco Field is borderline amazing. For a second baseman, it’s incredible.

Whenever you see significant performance changes like this, there’s always going to be people who want to suggest that the most recent results are “real”, and what happened early in the season is something that has been adjusted away. You also see this with people who feel like Russ Branyan has been “figured out”. However, the truth is that both the hot streak and the slump are real, and both inform us about the players abilities. It is the performance in total that is most predictive of future results (when weighted with past history and what we know about how players age, of course), not broken down fractions of that season.

Jose Lopez is not a .570 OPS guy or an .880 OPS guy. He’s neither as bad as he was in April and May or as good as he’s been in June, July, and August. So, the question for the Mariners, then, is the totality of the player one they should want to build around?

After all, Lopez doesn’t turn 26 until after the season ends, and he’s under contract for just $2.3 million next season and a team option for 2011 at $4.5 million before he becomes a free agent. Given his production, he’s a net asset (value of production over cost) to the M’s in the range of $7 million next year and probably something similar the year after. If the M’s paid fair market wages for Lopez’s next two seasons, we think that they’d offer him something like $20 million for 2010 and 2011 – they’ll actually pay him about 1/3 of that.

Given the team’s holes elsewhere on the roster, a lot of people will feel that the team should just count their blessings, be thankful that they have a league average second baseman under contract for a fraction of his value, and use the savings to pursue upgrades elsewhere. There is some validity to this claim – a team can win with Jose Lopez at second base, especially when he’s not making much money, so replacing him shouldn’t be a priority.

However, Lopez in Seattle also represents something of an inefficiency for baseball as a whole. As we’ve talked about, his power is strictly to left field, where Safeco is toughest on hitters. It’s no coincidence that Lopez is hitting .231/.249/.387 at home and .310/.345/.536 on the road. The M’s ballpark is built to suppress hitters with his skill set, which is why the new administration keeps acquiring left-handed players who fit the park better. In addition to the park configurations, the M’s are also likely to field a pitch to contact staff next year, and defense is not Lopez’s strong suit.

Lopez is less valuable to the Mariners than he would be to a team like the White Sox, who play in a park that allows fly balls to carry and have a rotation with several strikeout arms for him to play behind. Just due to the configurations of the stadium and the types of pitchers they’ll have, there’s a real value difference in how Lopez would help the White Sox versus how well he would help the Mariners. In a perfectly efficient market, Lopez would end up on a team like Chicago, with the Mariners then replacing him with a talent whose skills are better suited to their specific needs.

MLB, of course, is not a perfectly efficient market. Adrian Beltre is a terrible fit for Safeco Field too, but the M’s rightly pursued him as a free agent and received more value than they paid to acquire him. Just because it’s not an ideal situation to have a right-handed pull power hitter on the team doesn’t mean that the team should ignore those types of players in total. After all, you can’t have a line-up with nine left-handed hitters. You have to have some guys who can swing the bat from both sides, so if the M’s ship off every player whose value is somewhat muted by Safeco Field, they’ll end up with a mis-configured line-up that is easily shut down by any generic southpaw.

So, where does that leave us? Lopez is a net asset to the M’s, but slightly more of one to other clubs. He doesn’t have the skillset that the organization prefers, but he’s also a productive, young, low-cost player on a team that is trying to add exactly that. He’s prone to wild fluctuations in performance, but he’s also talented enough to become more than what he is right now.

Some people look at Lopez and see Carlos Guillen – a useful player that is viewed as a disappointment due to his lofty prospect status, but one that the M’s gave up on too soon, only to see him break out in Detroit. No one wants to repeat that mistake, certainly, and I can’t imagine that the M’s will be giving Lopez away this winter. But if Kenny Williams (or another GM with a park made for Lopez’s 380 foot power) calls and wants to talk about adding Lopez over the winter, I think the M’s have to not only be willing to listen, but be aggressive in realizing that this off-season is probably going to be the high point of his trade value.

He’s two years from free agency, and is not the kind of player the M’s are going to want to commit long term, big dollars too. As his contract gets closer to expiration, his value willl decrease, and the M’s will likely never be able to get more in exchange for Lopez than they will be able to now. His production level isn’t so high as to be irreplaceable, so if the right deal is presented, the M’s should be willing to pull the trigger on a trade that uses Lopez to help patch another area of the roster.

Lopez is this winter’s J.J. Putz. He has value to the M’s, but likely more to other teams, and Jack should be willing to leverage that inefficiency to make the team better overall.

Comments

110 Responses to “Lopez, His Hot Streak, and His Future”

  1. The Ancient Mariner on August 28th, 2009 8:20 am

    Chris, when it comes to understanding that “an offensive run created and a defensive run saved are equal in value,” that’s not heavy math. Heck, I can understand it, and I’m a humanities geek. In general — luck enters into it, and so results vary year to year, but in general — how many games teams win and lose bears a consistent relationship to the ratio between how many they score and how many they give up; in general, the most important thing is whether you outscore your opponents or are outscored, and by how much.

    Take a team that scores 800 and gives up 700. Let’s say they make some changes, and the next year, they score 700 and give up 600. Are you going to expect them to win fewer games? You shouldn’t.

  2. Dave on August 28th, 2009 8:57 am

    Yeah, it’s not quantum physics. Broken down, it’s just basically this fact: The difference between RS and RA drives wins and losses, not the quantity of either.

    RS – RA = wins. The amount of runs scored or allowed is not that important.

  3. justme on August 28th, 2009 2:28 pm

    I feel like I just got done reading a Dear John letter. Thanks for breaking it to me gently Dave.

  4. snapper on August 29th, 2009 9:47 am

    On the defense/offense trade-off I agree there is not a diminishing value of runs saved if you measure them ex post, I think there are a couple of factors that make the “all defense” strategy sub-optimal.

    1) The range of defensive value is much narrower than offensive (+/- 20 R vs. avg. for D compared to -30 to +70 vs. average for O). Once you get a pretty good defensive player, say +5, it’s extremely hard to improve on that. +10 defensive players are rare, and given the greater variance of defensive stats (only ~1/3 the sample size per season), it may take 3-5 years for the impact of that upgrade to be noticeable.

    Add that to the greater uncertainty of defensive stats, and it is much harder to upgrade most positions by 10 Rs on defense than offense, and actually be sure you get that upgrade.

    2) I think a weak offense with multiple “black holes” really improves the performance of the opposing pitching, beyond the linear individual effects. There is a cumulative effect to a good lineup in seeing more pitches per at bat and per inning that leads to pitcher fatigue.

    A really bad lineup will let the opposing starter pitch more IP and be more effective. You won’t get to feast on the weak underbelly of bullpens very often.

    Finally, your good hitters will be pitched around more often, and if they are not highly disciplined, will likely expand their strike zones and become less effective hitters.

    Given all this I think it is very possible that, given the Mariners current makeup, if you compare adding 50 “theoretical runs” of production (measured in a neutral context) either on defense or offense the offensive “theoretical runs” would translate into more actual runs.

    The 50 “theoretical offensive runs” may well elicit improved offensive production from existing players. The 50 “theoretical defensive runs” may well prove to be more like 20, given variability and measurement uncertainty.

    All else being equal I’ll take a +30/0 (O/D) player over a +15/+15 player every day of the week and twice on Sundays.

  5. Dave on August 29th, 2009 10:49 am

    I think there are a couple of factors that make the “all defense” strategy sub-optimal.

    Just to clarify, I don’t think anyone is advocating for an “all defense” strategy. We’ve repeatedly praised the Branyan acquisition as exactly the type of move the organization should have made at first base, and he’s a plus bat/average glove guy. We want value, regardless of what form it comes in.

    The range of defensive value is much narrower than offensive (+/- 20 R vs. avg. for D compared to -30 to +70 vs. average for O).

    That’s really overstating things. The -30 to +70 is all hitters at all positions compared against each other. You can’t replace Adam Everett with Albert Pujols. You have to remember that the -20 to +20 is relative to position. If you just lumped all fielders together and did a top to bottom difference, you’d find a massive gap between premium defensive shortstops and DH types.

    The gap in offensive value between players at similar positions isn’t that much larger than the gap in defensive value.

    Add that to the greater uncertainty of defensive stats, and it is much harder to upgrade most positions by 10 Rs on defense than offense, and actually be sure you get that upgrade.A really bad lineup will let the opposing starter pitch more IP and be more effective. You won’t get to feast on the weak underbelly of bullpens very often.

    This is a pretty commonly believed myth among the statistical crowd. Relief pitchers perform better than starters, nearly across the board. There is no real benefit from having a relief pitcher enter the game earlier – due to the nature of specialty relief work and the ease of pitching one inning at a time, that reliever is more likely to get your hitters out than the starter is in nearly every case.

    This isn’t true with the elite starters, but you’re not likely to chase them with patient hitters anyway – they throw strikes.

    Given all this I think it is very possible that, given the Mariners current makeup, if you compare adding 50 “theoretical runs” of production (measured in a neutral context) either on defense or offense the offensive “theoretical runs” would translate into more actual runs.

    You’re assuming that we don’t know how to regress our defensive projections to the mean very well. We already build in the uncertainty of defensive statistics into the projections, so no, if you have two equal projections of +50 run upgrades, one offense and one defense, you are not more likely to get the runs from the offense.

    All else being equal I’ll take a +30/0 (O/D) player over a +15/+15 player every day of the week and twice on Sundays.

    You’ll hit your payroll ceiling very quickly, and you won’t have enough money to win the division unless you have a $150 million payroll.

    There’s a reason the smart teams are buying defense right now. It’s the better buy.

  6. snapper on August 29th, 2009 11:16 am

    The gap in offensive value between players at similar positions isn’t that much larger than the gap in defensive value.

    OK, but it is larger.

    This is a pretty commonly believed myth among the statistical crowd. Relief pitchers perform better than starters, nearly across the board. There is no real benefit from having a relief pitcher enter the game earlier – due to the nature of specialty relief work and the ease of pitching one inning at a time, that reliever is more likely to get your hitters out than the starter is in nearly every case.

    You see no benefit in getting the SP out of the game in the 6th? That’s odd to me, since most people (stas guys included) value SPO who can go deep. Very few teams have 4 good RP. Much less 4 RP who can pitch each game in a 3 game series.

    I completely disagree with you on this point. Make the other teams SP throw 100 pitches by the 5th, and you will win a ton of games feasting on the 6th and 7th best RPs.

    You’re assuming that we don’t know how to regress our defensive projections to the mean very well. We already build in the uncertainty of defensive statistics into the projections, so no, if you have two equal projections of +50 run upgrades, one offense and one defense, you are not more likely to get the runs from the offense.

    You can’t estimate defense as well as offense. Full stop. No one can. You have a much smaller sample size of relevant data.

    You’ll hit your payroll ceiling very quickly, and you won’t have enough money to win the division unless you have a $150 million payroll.

    I said, all else being equal. Of course if the defensive player costs half as much per WAR he’s a better buy.

    Also, a perennially contending Mariner team can almost certainly support a $150M payroll, or close to it.

  7. Dave on August 29th, 2009 11:35 am

    OK, but it is larger.

    Not significantly so.

    You see no benefit in getting the SP out of the game in the 6th?

    This has nothing to do with “seeing”. This is just a factual basis of expected performance. Relievers pitch better than starters. This isn’t arguable.

    There’s also the sub-point that the difference in P/PA on the team level is marginal over the course of a game. We’re talking ~10 pitches from the hackiest team to the most patient team. It’s just not a meaningful aspect of decision making.

    and you will win a ton of games feasting on the 6th and 7th best RPs.

    No, you won’t – those guys only pitch low leverage innings when the game is already out of hand one way or another.

    You can’t estimate defense as well as offense. Full stop. No one can. You have a much smaller sample size of relevant data.

    And we account for this in the regression portion of the projections, which is significantly more aggressive for defense than offense. Don’t be a jerk about this. I’m not an idiot.

    I said, all else being equal. Of course if the defensive player costs half as much per WAR he’s a better buy.

    Of course, you ignored cost entirely when stating that you’d rather have a +30/+0 guy than a +15/+15 guy, when we know that the the +15/+15 guy is going cost significantly less given the current way things are valued.

    Also, a perennially contending Mariner team can almost certainly support a $150M payroll, or close to it.

    Unfortunately, we don’t live in that fantasy world where the M’s spend $150 million on their roster. Making decisions on which type of players the M’s should go after based on a hypothetically possible payroll doesn’t do anyone much good.

  8. snapper on August 29th, 2009 5:51 pm

    Relievers pitch better than starters. This isn’t arguable.

    When you make absolute statements like this, you make me question your knowledge of the game. In any given game or inning this may be true, but the amount of these superior innings a bullpen possesses is strictly limited.

    A typical team has 2 or 3 decent RP who can give you maybe 70 IP each. If a team is forced to constantly get 4 IP from its bullpen, you’re going to expose a lot of RP who are not better than the starters.

    No team wants to get to their middle relievers, there’s a reason for this. If you absolutist position was correct, every manager would be itching to take his SP out after 5.

    And we account for this in the regression portion of the projections, which is significantly more aggressive for defense than offense. Don’t be a jerk about this. I’m not an idiot.

    I’m not being a jerk. I do this stuff for a living. If you have 1/3 as much data your projections will be worse, unless D is much more stable than O. Your extra regression to the mean introduces more uncertainty into your estimate. The error band is wider. +10 offense is much less likely to be average than +10 defense is.

  9. Dave on August 29th, 2009 8:05 pm

    When you make absolute statements like this, you make me question your knowledge of the game.

    Do you question mathmaticians when they make absolute statements like 2+2=4? Relief pitchers get more outs than starting pitchers. This is just a statement of fact.

    A typical team has 2 or 3 decent RP who can give you maybe 70 IP each. If a team is forced to constantly get 4 IP from its bullpen, you’re going to expose a lot of RP who are not better than the starters.

    Sorry, but that’s just not true. You should learn more about relief pitcher performance. You’re believing myths.

  10. snapper on August 30th, 2009 10:17 am

    Do you question mathmaticians when they make absolute statements like 2+2=4? Relief pitchers get more outs than starting pitchers. This is just a statement of fact.

    Sorry, but that’s just not true. You should learn more about relief pitcher performance. You’re believing myths.

    You are taking a fact out of context and building a poor argument out of it.

    RP put up better ERAs than SPs for a number of reasons involving their usage. 1) They have the platoon advantage more often ,2) They are protected from long outings when they are pitching poorly, 3) they don’t have to pace themselves, 4) they don’t have to pitch tired 5) The ones who perform poorly quickly lose IP, or end up in the minors, etc.

    However, they are almost to a man worse pitchers than SPs. That’s why they’re in the pen. Many mediocre to poor starters become quite good relievers when switched. Even Mariano Rivera was a failure as a SP.

    The fewer IP your SP give you, the more the myriad advantages RPs have disappear. If a team embraced a policy of pulling every SP after 4 IP, I’m nearly 100% sure the RPs would have a higher ERA in short order.

    Let me put it this way, would you be happy if the Mariners adopted a policy that SP’s would always be pulled after 5 IP, earlier if they are struggling at all?

    Because, your facts, taken in isolation, would suggest that’s a great idea.

Leave a Reply

You must be logged in to post a comment.