Cust Cussing
Last season, Jack Cust was one of the out-of-nowhere stories of the year. He’d received all of four plate appearances in the majors from 2004 to 2006, and had essentially been relegated to Triple-A slugger while bouncing from organization to organization. The Padres sold him to the A’s for cash after Oakland experienced enough injury problems that they felt they needed a warm body, and Cust responded to the opportunity by hitting .256/.408/.504 and becoming a key part of their line-up.
2008 hasn’t gone as well. He’s currently sitting at .161/.373/.242 through 83 plate appearances, one of the stranger batting lines you’ll ever see. He has just seven hits on the season, but he’s drawn 20 walks in 22 games, so he’s posting a respectable on base percentage despite the fact that he’s just not hitting. Generally, you’d look at a .161 batting average and conclude that the guy is just in a slump, and that regression to the mean will make that bounce back in the not too distant future. Jack Cust isn’t your normal hitter, though. He strikes out at unbelievable 41.4% clip, making contact less often than just about any other position player in recent history. Ryan Howard and Adam Dunn combine to cover the top three spots in single season strikeout totals in major league history, and even they didn’t get particularly close to striking out in 41% of their at-bats with career rates of 35% and 32% respectively.
I was going to show you a chart of the K% for those three plotted next to each other, but realized it wouldn’t be very helpful, because Jack Cust’s markings aren’t on it. See, the Fangraphs K% chart tops out at 40%, and Cust is consistently above that mark. He’s literally off the charts.
When you strike out as much as Cust does, you have to be amazing when you make contact. Last year, Cust was, posting a .366 batting average on balls in play and whacking 26 home runs in 124 games. So far, in April, Cust’s batting average on balls in play is down to .242, and only one of his seven hits are home runs. Even when he hits the ball, it’s not going anywhere, and that’s made him a significant liability at the plate during the first month of the season. With the signing of Frank Thomas to be the regular DH, Cust is going to have to play left field to keep his spot in the line-up. Unfortunately for him and the A’s, he’s about as good at that as he is at making contact. He might be the worst defensive player in baseball – if he’s not, he’s close.
Add it all up, and the A’s have a guy who has to torch the ball when he hits it in order to be a valuable player, and when he’s not driving the ball, he’s the least productive regular in the major leagues. Due to his defensive problems, it’s going to be harder and harder for the A’s to justify penciling him into the line-up if he doesn’t start producing better results when he makes contact. The A’s are surprising a lot of people by standing at 16-10 through 26 games, but they’re not going to be able to stick with Cust killing them in the field and at the plate forever.
Print This Post
Jack Cust does have a highly unusual set of statistics, which makes him hard to evaluate. But led by Billy Beane, the Oakland As have evaluated him and (probably) like what they see.
The most important thing about Cust is that he has a high On Base Percentage (OBP), typically in the high 300s, and that doesn’t seem to be in any danger. (Following the precepts of Billy James, the As are “bigger” on OBP than say, batting average or power.) It’s true, howeve, that the quality of Cust’s OBP has declined, with more walks than hits this year. Last year, Cust’s batting average was almost .100 points higher, and his slugging percentage was more than .250 points higher than they are so far in 2008, while the 2007 OBP was only .035 points higher. Cust does strike out a lot, but he also has an unusually high BABIP (Batting Average on Balls in Play) of over .400 over his career.
Maybe Cust is in a slump. Or maybe opposing pitchers have gotten wise to his potential threat and elect to walk him even though it costs at least four pitches to do so. (Oakland is also big on “intangibles” of this sort; hence they initially placed a higher value on the Giambi brothers than any other team in the league.) In any event, the As will find a way to use him “strategically,” by cutting down his strikeout rate, or by putting him in the line-up at times inconvenient for opposing pitchers (and DH-ing him would eliminate defensive issues). His high OBP is a “down payment” for more production of one sort or another. As long as he maintains that, he is in no danger.
If they could cut down his strikeout rate I think they would have done so already. This isn’t something that will go away. I do agree that DH-ing him would cure defensive issues but Frank Thomas is the DH now and he’s a much better option than Cust. His BABIP should improve and I wouldn’t expect him to hit any less than 8-10 HR (unless he’s sent down/released) but this is a major red flag.
Cust had the best HR/FB rate of any hitter with a qualifying number of plate appearances last year. So far this year he’s in the bottom half with a paltry 8.3%. I don’t know that Oakland could have expected Cust to lead the league in HR/FB again this year, but they had to be expecting at least 20%. A figure of 25% which would be below his career rate of 26.4% would bump his slugging percentage up to around .370.
Looking at his BABIP, last year’s .366 was not that big of a fluke because his line drive percentage was 23.2%. Unfortunately for Cust and the A’s about a third of his line drives have turned into ground balls. So his .257 BABIP isn’t that much of a fluke, either. At 16.7%, his eBABIP would be 30 points higher than his current BABIP.
If these two things happened, Cust could get back to being a useful player. Adding the power improvement and getting back to his eBABIP would give him a weird looking .191/.403/.400 line.
The strikeouts and the defensive issues give Cust no margin for error. He has to take his walks and he has to hit for power to be useful. I don’t see how Oakland is going to suddenly fix his extreme strikeout rate and with Frank Thomas and Mike Sweeney on the roster most of his at bats have to come in the outfield.
Cust isn’t alone, though. Adam Dunn’s HR/FB rate is half his career rate. Ryan Braun is down two-thirds from last season.
It seems strange that Cust would suddenly stop hitting for power. The league could have figured him out, but it seems more likely that this is a small sample size issue and that some more of his fly balls will start leaving the yard.
Cust’s career batting average on balls in play is not .400. I’m not sure where you got that from, Tom.
And, while his OBP is nice and all, on base percentage is not a skill – it is a result of a variety of skills (contact, patience, power, speed, etc…). If we observe that Cust lacks the requisite skills to post a high on base percentage, that is more useful for projecting his future production than what he did in 400 at-bats last year.
Isn’t this what everybody always said about Cust? That he was prone to long slumps where he wouldn’t make contact?
The A’s actually watch his approach and analyze his plate appearances. If they thought he didn’t have the skills to hit in the majors, they would have set him up as a platoon guy. I don’t think we should give up on him any more than we should give up on that 40-year-old future hall of famer who can’t play the field and hit 164/315/301 in less than 100 PAs this season.
With all due respect to the A’s, I don’t think we need to defer to their decision making process in every situation. They are a well run organization, but they’re not infallible.
And, really, there’s a pretty distinct difference between Frank Thomas’ skillset and Jack Cust’s skillset. Jack Cust is trying to succeed with a contact rate worse than just about anyone else in major league history. Even Rob Deer was better at putting the bat on the ball than Cust is.
The bar for how good you have to be when you make contact is so high when you strikeout in 40% of your at-bats that it’s almost impossible to succeed for a long period of time with that deficiency. There’s no reason to believe that Cust’s striekout rate is going to improve, and there are legitimate reasons to believe that it’s a flaw that will keep him from being a productive hitter for any real length of time.
While OBP may not be a skill, BB/PA is surely a skill. And it is one that relatively immune to slumps when compaired to BA. We all know Cust can take a walk, as you mentioned, but can he come up with enough hits? Well its been just less than a month, and when Cust has been given a chance to play he’s never posted a BA lower than .228 in the last 3 years (which he did for the Sacromento Rivercats).
But primary bone to pick is this; there is just no way a player with a .373 OBP one month into the season is either the worst regular in the game or derserves to be given up on. I could list the regulars with worse OPS but that would get old fast, since the list is pretty long, but it does include David Ortiz, Andrew Jones, Jose Guillen, Cano (Cano’s .437 OPS is worst in MLB). No one below him has even close the OBP he has, and we all know OBP is more important that SLG. As was mentioned by JC, he’s got to start hitting HR eventually. Last year he had 15.2 AB/HR, this year it is at 62 AB/HR (62AB/1HR). So unless something is seriously wrong with his swing, he’s going to come around (which is certainly possible, but I haven’t read anything about it). Maybe he won’t hit HRs at that rate again, but if he’s hitting them at about 20AB/HR, he would still finish the season with 25 HRs. And 20Hrs, say 25 2B (he had 18 last year in 124 games), and an OBP of .370ish is nothing to thumb your nose at. Particularly when you consider Cust plays in a pretty serious pitchers’ park.
Actually, I’d suggest that BB%, just like K%, is a function of different skills, but is not a skill itself. It’s an important distinction.
Cust draws walks because he is selective at the plate (skill #1) and has enough power to force pitchers to not challenge him on every pitch (skill #2). However, while he controls how selective he is, he doesn’t control what pitchers throw him, and his walk rate depends on pitchers throwing pitches out of the strike zone. If Cust doesn’t start hitting, he’s going to see more strikes, and his walk rate will fall. Now, if his power and contact skills haven’t diminished, he should theoretically be able to hit those strikes and remind pitchers why they were pitching around him to begin with, but I think it’s clear that his current rate of performance isn’t sustainable.
He just won’t keep walking this much if he doesn’t start hitting soon. He can’t post a .380 OBP and a .160 batting average – no one can.
As for the list of worse regulars, you have to factor in defensive abilities. Cust is a huge negative in the field while playing a relatively easy defensive position. Even when guys like Robinson Cano or Tony Pena aren’t hitting, they’re still proving some positive value to their clubs with the glove. Cust only provides value as a hitter, and when he’s not hitting, he’s a big problem.
Dave Cameron has been predicting Cust’s collapse for a year now, so a stopped clock is right once a year. Cust will blast out of this slump like all the ones before, and again Dave will hope this analysis, like much of his, disappears in time never to be brought up again.
In reply to Dave Cameron,
Per ESPN, Jack Cust has had 601 career ABs and a whopping 249 K’s. The difference (601-249)=352 BIPs. He’s had 143 hits. The ratio of 143 Hits /352 BIP is .406, BABIP. As a check, the ratio of 143Hits /601 ABs is exactly his batting average of .238. And 249Ks/601ABs is just is K rate of about 40% (41.4% to be exact). To use round numbers, a .240 BA/(1-0.4)=.400 BABIP (0.4 represents the K, or “noncontact” rate.)
There is the debate between what I call the “causalists,” and the “time series” analysts. The causalists will look at causal factors such as “contact, power, patience speed, etc.” in estimating a player’s ability. A time series person would say that the best way to estimate a player’s ability is to look at his record–unless it defies other known sabermetric relations. How a player gets there is less important than what he does. The Oakland As (and yours truly) are in the time series camp. Their general manager Billy Beane, was a counterexample; a gifted athlete with all the tools who could only hit .215 because of lack of “intangible” qualities. Other players, e.g. Scott Hatteberg, do better than their tools suggest, because of an abundance of such qualities.
Sabermetrics suggests that Cust will pick up power and contact, thereby lifting these two aspects of his line as JC suggests (because the current depressed levels are a “random,” not explainble, result). It’s possible that these metrics will stay down, but then his OBP would have to fall in tandem. As long as Cust maintains his OBP, the As will give him the benefit of the doubt.
The accepted formula for BABIP is (H-HR)/(AB-K-HR+SF). A home run is not a ball in play as most people understand the term. What you’re measuring is closer to batting average on contact, but still not exactly right since you’re omitting sacrifices.
It’s also a pretty significant generalization to classify sabermetrics as “time series analysis”. There’s a lot more to sabermetrics than the theory of past results equaling future projection. The search for truth in baseball is not limited to on base percentage and anti-tools rhetoric.
The sample size we’re working with from this season is very small to be drawing such rapid conclusions. r-squared for K% doesn’t reach .5 until 150 PA, and Cust has only 83 this season. He struck out a ton last season: 32.5% but regression to the mean dictates that he will get less extreme not more extreme. HR/FB rates don’t have r-squared of .5 until 300 PA, so we shouldn’t make much of his present 8% rate. Ryan Braun is smacking just 7% of his FBs for HR, but I think we can all agree that Braun will bring that number up. And given that he and Jack Cust have just one season in the majors (effectively) I’d say we should reason similarly about each. It would be surprising if a guy with Cust’s history of power (in both MLB and the minors) had a sudden, massive drop in power in at the age of 29. 33% HR/FB is extreme, but I think we can expect him to reach 18% or more this season.
Also, “killing them at the plate” is not a good description of a .370 OBP. Putting a man on base 37% of the time is great. There are vast swaths of MLB batters that will never reach that rate for whom an 8% HR/FB rate is upside.
By the way, BABIP never reaches the point at which r-squared is .5. For almost any player, assuming a BABIP of .300 is fair. I think Oakland has good reason to expect Cust to be a positive producer on offense this year. He’s not a solid defender, but his RZR is better than Raul Ibanez, so in that respect, they’re still competing in their division.
First, I’d agree there are several “skills” that determain BB%, but you could surely break those down again as well. Like selectivity at the plate comes from pitch recignition, quickness with the bat, etc. But anyway, the guys got BBs. Even if pitchers TRY to throw him more strikes, many pitchers just aren’t good enough to do so without laying in a few meatballs. Which as you say, should drive Custs BA/SLG up. But this has yet to be seen. I agree with you no one can keep up a line of .160/.370/.220, and that type of player is not very valuable, but if anyone can keep it up for very long, its a guy like Cust (not that I think he will).
As for overall value. I have a hard time believing Andrew Jones with a line of .159/.226/.256 is making up all extra outs he creates with his bat in the field. And for that matter Jose Guillen, who’s an equally bad fielder as Cust, has a line of .177/.202/.323. There are some 30 players behind Cust in OPS, and serval ahead of him that I would argue are less valuable with the bat due to the fact that Cust still isn’t making that many outs (and 1/2 his games are coming from a pitcher friendly park), and you’re really going to argue the possitional adjustment makes them all better. Sure a slick fielding second baseman like Cano maybe (nearly .200 points of OPS is hard to make up, particularly nearly all of it comes from OBP), but there are some DHs and bad fielding corner outfielder on this list too. Garret Anderson is one ahead of him in OPS who, if I had to guess, has had less overall value to his team in the last month. His line .228/.264/.356, and last year he was a -3 in LF in only 85 games. Cust in 60 in the OF in general was -3 last year.
If Jack Cust is not the worst defensive player in MLB it might be the guy the A’s have at 1st base; Daric Barton. How bad? Mike Sweeney, severe knee injury and all, is playing 1st for the A’s. The A’s have four DH’s and unfortunatley two of them have to play in the field.
Cust has fallen into similar slumps before. Playing the selective endpoints game:
July 4, 2007 – Aug 3, 2007: .197/.337/.318, 35% K% in 83 PA
Mar 25, 2008 – Apr 27, 2007: .161/.373/.242, 31 % K% in 83 PA
That’s not to say the slump isn’t troubling. The best way to estimate a player’s current talent is to create a projection incorporating all the latest performance data, including (in Cust’s case) his latest awful showing. Running a continuous Marcel (such as I and studes have done at THT recently) shows the following:
Date AVG OBP SLG GPA
6/3/2007 .258 .356 .469 .277
7/3/2007 .276 .368 .507 .292
8/2/2007 .263 .363 .477 .283
9/1/2007 .268 .381 .490 .294
10/1/2007 .259 .386 .471 .291
3/29/2008 .257 .383 .464 .288
4/28/2008 .248 .382 .442 .282
(Stats are his Marcel projection on each of the respective date)
The slump has definitely changed our position on his true talent AVG and ISO, although the OBP is still there. So, while your comments are generally correct, the magnitude of the slump is only quantified when all the data are taken account and weighted appropriately. My guess is that Cust’s true talent is as a .280-.290 GPA hitter, and that’s what we’ll see for the balance of the year unless there’s something physically wrong with him.
FYI, Marcel (as of today, and knowing nothing about BIP distribution) sees Cust at a .326 BABIP and .374/.666 AVG/SLG on-contact. All three of those numbers are significantly lower than his career averages.
As for defense, I am not sure what Cust’s defensive projection (true skill) is, but to date this year he has been better (by RZR) than Jose Guillen, Emil Brown, Raul Ibanez, Brad Wilkerson, and Garret Anderson. While I don’t know how that will play out in the future, and while Cust is surely one of the worst defensive corner outfielders in the majors, he’s probably not dead last and may not be the worst even in the AL West.
Jose Guillen is a terrible in the field, but is nowhere near the black hole that Cust is.
It’s true that it’s very difficult to succeed with a league leading K-rate; but it’s also true that it’s very difficult to not succeed with a league leading BB-rate combined with a well above averaged BABIP and a high HR-rate for his entire career. He is pretty much guaranteed to have a high OBP despite all those strikeouts. As long as he continues to see a ton of breaking balls, he’ll keep up the walks, and if that does not continue, he’ll hit more HRs.
It’s also worth noting that while he’s certainly one of the worst defensive OFers, he’s not the worst, and he’s basically been fine this year in every game that was not against the Mariners, which is presumably all you’ve seen of him. He does not make horrible gaffes at the rate he did over the weekend….
What’s the evidence that Cust is such a terrible defender? THT’s RZR places him at .855 in RF last season (where he had most of his days in the field); that number is toward the bottom of all RFs, but NOT LAST. He produced 8 outs out of zone in 382 innings, or in 2.1% of innings played. The highest OOZ rating for an RF was Austin Kearns, 4.9%. If Cust got OOZ like Kearns, that would mean an additional three outs every 100 innings, or about once per 4 games. Yes, an extra out every 4 games is a nice thing to add, but it’s not that huge, and many other RFs have OOZ ratings like Cust’s. So can someone please show me the evidence that Cust is as bad in the outfield as everyone here seems to think?
By the way, if the problem is with Zone Ratings, I’d love to hear about it. ZR is the only quantified measure of fielding ability that I know. (Except fielding percentage, and everyone reading this knows how significant a measure that is.)
Also, Daric Barton is a good fielder, perhaps very good. He ranks in the top 1/4 of 1B right now, so I don’t think that claim has any merit at all. This is based on a small sample of BIZ, but he has 10 OOZ outs this season, which is also among the top.
Better defensive metrics than ZR:
Probablistic Model of Range (Baseball Musings)
Plus/Minus (Baseball Info Solutions)
UZR (Mitchel Lichtman)
Fan’s Scouting Report (Tango)
All of the play by play metrics hate Cust’s defense, and the A’s fans who filled out Tango’s report rated him a 15 out of 100. RZR is nice because it’s so available but it’s not in the same class as any of the metrics above. When it disagrees with +/-, UZR, PMR, and pro scouting reports (there’s a reason Cust has been DH’ing for years) the best assumption is that RZR is wrong.
“All of the play by play metrics hate Cust’s defense, and the A’s fans who filled out Tango’s report rated him a 15 out of 100.”
I found Cust’s PMR, which definitely hates him. Can you point me to his UZR and +/-? Thanks.
Hey Dave, if you’re still reading back this far, what do you think of Dan Fox’s Simple Fielding Runs as a measure of defensive ability? If I remember correctly, it fared pretty well, intuitively, in comparison to Plus/Minus.
+/- is available via subscription to Bill James Online. MGL and Tango periodically release UZR data over at The Book blog.
I like almost everything Dan Fox does. SFR is another solid tool to use when you don’t have play-by-play data to look at. When you do have that specific PBP data, though, I think the best idea is to look at that, and that’s what PMR/UZR/Plus-Minus offer.