Take My C.J. Wilson
C.J. Wilson was paid $77.5M this off season to come in and be a front line starter for the Angels. With Jered Weaver and Dan Haren already on the team, he wasn’t going to be the only ace of the staff, but he has pitched like one. It is now time to sell high on Wilson’s great 1st half of the season. He has 9 Wins, a 2.33 ERA (6th among qualified starters) and a 1.17 WHIP. These stats put him within the top 20 SP in production for the 2012 season (16th SP with ESPN’s Player Rater). His stats don’t match his perifial numbers so he has a good chance of regressing quite a bit in the second half of the season.
The main reason that Wilson should look to regress is that he is completely out performing his Walk and Strikeout rates. Here are his values before tonight’s start:
K/9: 7.2
K%: 20%
BB/9: 4.0
BB%: 11%
Somehow he is able to maintain a near 2 ERA with a 4.0 BB/9. Here are 5 pitchers with similar K/9, BB/9 and their ERA:
Name: K/9, BB/9, ERA
Anthony Bass: 7.7, 3.5, 4.70
Mike Minor: 7.6, 4.0, 6.20
Randall Delgado: 7.3, 4.4, 4.52
Aaron Harang: 7.0, 4.0, 3.51
Justin Masterson: 7.0, 3.5, 3.92
The lowest ERA from the group is 3.51 which over a point higher than Wilson’s 2.33 ERA.
The average ERA of the 5 values is 4.57 which is fairly close to Wilson’s xFIP (4.10) and SIERA (4.23). I would not be surprised to see Wilson have an ERA twice his current value over the 2nd half of the season based on maintaining his Walk and Strikeout numbers.
The other key to the take out of these numbers is that he is not striking out a ton of batters. Among the qualified starters, he is in the middle of the pack (50th of 101) with 85 strikeouts. His K’s are not at any kind of elite level. His walks look even worse. He has the 8th most walks in the league with 46. His BB/9 is the 13th highest when looking at the same 101 qualified pitchers.
The reason he has been able to keep his earned runs down is that he is 8th among qualified starters with a 0.247 BABIP. The low BABIP has led to career high LOB% of 77.5%. It may seem that his BABIP is likely to increase as the season goes on. Several factors will be keeping it low though. Since becoming a starter, his BABIP has never been as low as it is this season, but he has had good BABIP values in the past (0.266 in 2010 and 0.287 in 2011). Also, he is a ground ball pitcher with 52% GB% so far in 2012. Finally the Angels defense as been solid and have the second highest team UZR total in the league.
A final factor I see working against C.J. is that he has only 27 earned runs counting against his ERA vice the 34 total runs he has allowed. Home teams always seem to give the home defenders errors in order to keep good pitchers’ ERAs as low as possible to make them look better. A possible manipulation seems to exist with Wilson. He has been getting quite a bit of home cooking with 6 of the 7 unearned runs coming at home. The difference between his RA and ERA is 0.60 with the league average difference being around 0.35. He may see an increase in ERA as the number of unearned runs associated with him declines.
The Angels brought in C.J. Wilson this off season to become good, reliable starter. He has been that pitcher and more so far this season. The problem is that several different factors look to derail his season. I would sell high on him now while his value may not be any higher.
Jeff, one of the authors of the fantasy baseball guide,The Process, writes for RotoGraphs, The Hardball Times, Rotowire, Baseball America, and BaseballHQ. He has been nominated for two SABR Analytics Research Award for Contemporary Analysis and won it in 2013 in tandem with Bill Petti. He has won four FSWA Awards including on for his Mining the News series. He's won Tout Wars three times, LABR twice, and got his first NFBC Main Event win in 2021. Follow him on Twitter @jeffwzimmerman.
Jeff Weaver? Angels fans better hope not.
Somehow Jeff always manages to sneak in there for a come back.
Thanks. I am one of the few people biased towards Jeff.
Been reading for 3 years about how CJ Wilson outperforms his peripherals and it’s a good time to sell. I think it’s time to ignore his peripherals, he’s proven that he can sustain the performance of being a top pitcher.
Agreed- Wilson and Johnny Cueto have been repeatedly predicted to “regress” due to less-than-stellar peripherals over the past two years. Maybe someone ought to take a look at WHY they are able to outperform their peripherals so consistently instead of assuming that they operate like the rest of the league. A look at similar historic pitchers could shed some light as well. GO.
Walks and strikeouts are not peripheral numbers… I repeat: NOT PERIPHERAL. They are front and center.
Walks notably up. Strikeouts notably down = be wary. Pretty easy.
Yup K/9 and BB/9 ain’t really peripherals. Smarten up lads. Zimmerman has a legitimate point here.
Noted, I’m a dummy; but the point remains: it would be cool and helpful to see an in-depth analysis of pitchers whose results consistently defy what is expected based on their K/9, BB/9, BABIP, etc.
In Wilson’ case, the interesting question is this: what makes him so effective against right handed hitters?
I can’t easily look and point to one thing, but the splits are very funky. He strikes out lefties like you’d think he would. His rates against righties look like he should be taking on some damage… but he isn’t — tiny homer rate and really low average.
Another year, another “sell high on C.J. Wilson” article.
Blah blah blah stats. Blah blah blah peripherals. Can it Einstein, all I want to know is how he looks in a pair of jeans.
He’s never outperformed his peripherals this extreme a degree. And no matter how you slice it, losing a K/9 and adding a BB/9 is no good.
The best thing about this article is I might be able to buy low on Wilson in a few of my leagues now. Thanks guys.
Hopefully people will take my Scherzer for CJ Wilson offers.
I find it funny how people come on here and comment and whine about authors outlining stats and peripherals when that is the basis of the website. Relax people. And enjoy the quality writing.
When a player is able to sustain a level of performance for a long period of time, you should examine why that is instead of writing it off as luck.
Guys like Cueto, Vogelsong and Wilson will routinely be written off here while guys like Scherzer and Norris are hyped up.
I think people come here for fantasy advice, and most the time it’s sound and the logic behind it is very good, which is why I continue to come here. But I disagree with Zimmerman’s assessment of Wilson, and clearly others do as well.
Long period of time in this case being… two seasons. And a half. And never to a degree like this.
People forget just how many major league pitchers there are. One player doing something statistically aberrant for 2 or 3 years doesn’t necessarily tell us anything about him – it would be shocking if in such a large sample of players there weren’t pitchers outperforming or underperforming ERA estimators for years at a time.
Wake me when Wilson does it for 5 in a row.
@mcbrown: I hope he keeps it up or you could be sleeping for a loooong time.
He won’t maintain the 2.30 ERA or whatever he’s at now, but I don’t see why he won’t still do well. A big reason he’s outperforming his FIP, etc is because of the D he has behind him. I don’t see that D getting any worse in the 2nd half. That will continue to help Wilson stay out of trouble
Hey, this coin came up heads 65 times out of 100. Time to just admit this coin just knows how to outperform.
you realize it’s <.5% that a coin tossed 100 comes up heads 65 times…it can happen, absolutely, but it's a worthwhile question at that point whether we're missing something…like is it a weighted coin?
having said that…i don't think we're at that point yet with a guy like wilson. i'd sooner wager on external factors like defense in texas and anaheim holding down his ERA.
0.5% chance? So 1 out of 200?
How many starting pitchers there are in MLB? Even if we go with only 3 ‘good’ starters per team that gives us 90 pitchers, and then one of those 90 having a 1 out of 200 season isn’t all that unlikely.
@James here’s why that logic is wrong:
let’s assume that the 1/200 number is true (it isn’t…see below). let’s also use your n value of 90 starting pitchers. the expected value of having >1 of these 1/200 performances each season is relatively high, as you correctly conclude. colloquially, random variation will explain a pitcher or two significantly outperforming fip/xfip each year (using that <1/200 number).
however, the *same* pitcher having this very rare statistical variation 3 years in a row suggests that there might be something that fip/xfip are not measuring in relation to that specific pitcher. to put it the parlance of the coin toss, it would be like the same coin giving that same 1/200 outcome three times in a row. do you still feel as confident after 3 go-arounds that you're dealing with a fair coin? you shouldn't.
now, as for cj wilson…the 1/200 number is derived from the coin toss problem. i assume the corresponding number relating to outperforming fip/xfip is much larger because of the relatively small sample sizes involved in the different outcomes that form the basis of those pitching measures. so from a math perspective, the chances of cj wilson's performance being related to random variation are much, much higher than getting a fair coin, tossed 100 times, to give you 65ish heads three times in a row.
all of this is a round about way of saying–there is merit to both sides of this debate and why it's generating a robust discussion. i only responded originally because the original poster (knowingly or unknowingly) couldn't have picked a better number than 65 to stir up some debate.
tons of ground balls, a pitcher friendly park, the highest average velocities on his pitches of his career, outstanding defense behind him, 2 of his most common road parks are also pitcher parks. sounds like a guy you should be going after, not dumping
Well, I’ve heard the argument for why he will regress, what do you haters have to say about him NOT regressing?
You can’t just throw the science out the door just because you want to. How can you vehemently defend Wilson with a BB/9 so high and a BABIP so fortunate?
….and “no,” the Angels D is not so good as to prevent regression. That is folly.
I don’t think anyone here is interested in “throw[ing] science out the door”- this is FanGraphs, after all.
Quite the contrary, in fact: I think many of the “non-regressers” here are arguing that the authors should dig deeper- look at other numbers, other stats, other trends, other circumstances, do MORE analyses to explain why peripherals don’t match results SO CONSISTENTLY with guys like Wilson, Cueto, Vogelsong, etc. instead of passing off the results as luck b/c K/9 and BB/9 and BABIP don’t agree.
Wilson, Cueto, Vogelsong… what about Carlos Zambrano? Is he not in the club anymore after he SHOCKINGLY underperformed his peripherals last year?
It’s funny how good we are at identifying these players with hindsight but not with foresight! It’s almost as if their results were due to… I don’t know… random variation over a large sample of pitchers?
“It’s funny how good we are at identifying these players with hindsight but not with foresight!”
That’s kind of the point, isn’t it? Foresight has had folks predicting Wilson, Cueto, etc. falling off the cliff for years; that’s why they are interesting.
You make a good point (a few posts above) that we’d expect a few outliers in such a large population of pitchers, but I don’t see why that means they should be totally written off as outliers (as you seem to imply) instead of subjected to further analysis.
Yes, that is exactly the point, and you have made it for me. People predict a HUGE number of pitchers who outperform for one or two or five years to regress, and the majority do, and some (as we would expect by chance alone) don’t. In arguments like this one, people forget the people who DID regress and remember only the ones who didn’t. Then with the benefit of hindsight they say people have incorrectly been predicting regression for them so they must be special. It is confirmation bias, pure and simple.
I can’t even take seriously an argument that says Ryan Vogelsong, who was out of baseball for the four years prior to 2011 and has a whopping one and a half seasons of FIP outperformance too his name, has been predicted “for years” to regress and has thus defied the skeptics. Cueto is the only one of this group who has something approaching a track record at this point, and like I said elsewhere, wake me when he does it for five years.
SlimJim:
This is not science – baseball stats and models are not science. No. It’s less than that and all for fun.
Okay, since you asked for the “hater” side, really, you can read many counter arguments up the page, but spelling it out for you:
Wilson is right on his career stats this year, even a little better. All the arguments being made about his walks and strikeouts, FIP, and whatever can be washed away with one or two excellent starts.
Think CJ might be capable of getting on even more of a roll? I do. Nobody’s hitting him. Nobody’s going deep off him. His team is real good behind him. He never has any bad starts. Prognosis: good.
Oh and since this is nominally a fantasy article, I would like to point out that Wilson, whether or not he will”suck so bad going forward that his year end ERA or (insert your stat of choice here) will end up over 3.25,” is one of the few guys that has all the factors going forward that put him in the very top tier of guys that get wins, get them painlessly, and not make a Gavin Floyd all over your team’s stats.
The author is not saying Wilson is a bad pitcher, or that he will be a bad pitcher the rest of the season. The author says that this is likely the peak of Wilson’s value this season, so he would sell.
While you may disagree about the magnitude of regression, I do think it is hard to argue that Wilson will continue pitching like this the rest of the season.
If you want to ride out Wilson’s good season, fair enough. If you want to get the most value of of Wilson an sell, now seems like a pretty good time.
There is a big difference in believing that a player’s peak value is now, and that a player is going to suck so sell him now.
No, every player is either the best in the world or below replacement level. There is no middle ground. This is the basis for all sports discussion.
His actual walk and strikeout rates (%’s, not /9’s) are almost identical to 2010. The evidence is certainly piling on that he can “outperform” his peripherals, but even in 2010 his ERA was 3.35, not 2.30, and his career BABIP is .282 (a reasonable number for a pitcher with positive BABIP control), not .250ish.
A 3.40ish ERA is the best reasonable projection you can give him moving forward if you assume his strikeout and walk ratios stay stable. That’s solid, but with a 7.50ish K/9 (which though weaker as a predictor than straight %, is relevant as a result in IP capped roto) and 1.25ish WHIP, it’s more of a low end #2 fantasy SP or high end #3 than a stud.
FWIW, last year his K% was 22.5% and BB% 8.1%, and that yielded a 2.94 ERA for him (with a .287 BABIP and 8.2% HR/FB, both perfectly nominal in his career context). In 2010 and 2012 those numbers are/were roughly 20% and 11%, and for his career (including his bullpen time) they’re 21% and 10%.
The problem with what he did in 2011, even more than “overperforming” his “peripherals,” was that his SwStrike% and F-Strike% were both quite mediocre. This year they’re even worse, so in terms of true outcomes, it looks like the 2010/2012 Wilson is the “real” Wilson more than the 2011 Wilson. He may outperform his true outcomes, but we still have to regress for the degree to which he’s capable of doing so, and right now he’s well overperforming even that. Even for pitchers that can control BABIP and HR/FB to an extent, it’s still folly to predict those numbers to be any better than .275 and 7%, that’s really the upper (or lower) bounds of a viable prediction for even the best of the best at controlling those marks.
Any reason why you wouldn’t be able to expect 2010 CJ Wilson for the rest of the year? Alot of the surrounding stats are right around where he is now, and he finished with a 3.35 ERA.
Considering his home park went from TEX to LAA, I’d say he could be just a little bit better than that from here on out.
Well even that year his BABIP was .266 and HR/FB was 5.3%, and his career rates are .282 and 8.5%, respectively, the former of which is above average and would account for him “outperforming” his peripherals a bit, just not by three quarters of a run. If he maintains these K and BB rates with a .280 BABIP and even a 7% HR/FB, better than a 3.40 ERA is unlikely, while worse is not.
All season I’ve been watching Wilson’s SIERA to ERA. Right now his SIERA is almost two full runs higher than his ERA. So yesterday I dealt him for Ryan Howard coming off the DL. Now reading this post gives me more peace of mind it was the right move for the second half of the season.
I hope you’re making this up.
that’s the worst trade i’ve ever heard of
My 1B has been a platoon of B LaHair and L Duda. I just lost Stanton’s power for a month. My SP rotation is still C Lee, D Price, J Lester, A Wainwright, and M Buehrle. Under those circumstances I do Wilson for Howard 10 times out of 10. Evidently the haters have never owned Howard for one of his patented second half rampages – 0.996 lifetime OPS post-All Star.
Just got Jay Bruce straight up for Wilson. Thoughts?
I would not be surprised at all to see Wilson regress – he is something like third in the AL in ERA after all.
Comparing him to his career performance is a good start, but he is moving out of a great hitters park where he wilted in the summer and into a cozy pitchers park with a great defense behind him.
It seems like a classic sabrmetric mistake to undervalue the performances of lefty starters who exceed their K/BB rates. I saw Jerod Washburn be underrated for years (lowest ERA by a SP in franchise history I believe – in the steroid era no less).
There is a lot of variance in the numbers so I agree with the articles premise, I’m just not sure you should really expect a significant drop off at this point.
“perifial”? i mean, come on.
I like the article…..and the comments. I can’t help but think he’ll regress quite a bit. I don’t care what anyone says about his perferals….perifls….Oh, who cares how I spell it? If a pitcher is so good as to be able to avoid hits and home runs more than others, why then can’t he just avoid them hitting the ball all together? I know pitching to contact is used. But what about pitching to no contact with runners on or in scoring position? Or pitching to more ground balls with a runner on first? I agree with McBrown above. Unless he does it for a much longer time I’ll just chalk it up to luck. At least more luck than many would think.
Next….Someone mentioned Jeff Weaver above. When do we see the “take my Jared Weaver” article? How long until he becomes Jeff? OR, how long does his luck last?
I will add one last thing. With the two of them out performing there outer stats, is there something in the LAAOA defense and or coolaide? How’s that for a peripheral substitute?
Maybe it’s time to stop citing Wilson’s xFIP as opposed to his FIP. Last time his HR/FB was above league-average was 2008, when he was still a reliever. Dude is probably just really good at inducing weak fly balls. Since 2008, even when he was pitching half his games in Texas his HR/9 never exceeded .64 over a full season.
Haha, I just traded for Wilson and then I saw this article. I hope he doesn’t regress. I traded Teixeira for him with Howard coming back. I also have Youk on my bench. Felt like I needed another SP after previously losing Luebke and Beachy for the season.
i buy “selling high” on wilson, but from the trades i’m reading–Ryan Howard?–people seem to be selling low and forgetting that ZiPS likes him at 3.5 era/fip ROS. i mean, we can say “sell high on Player X”, but if the market isn’t there for him because everyone is thinking the same thing, then it’s not really applicable fantasy advice because the love/hate pendulum has pushed too far one way.
i also don’t buy that babip must or even is likely to regress over the next 100 innings–it might take much longer to get back to expected, which we we see again and again. this is point is doubly applicable to Wilson, who benefits from the Angels defense and the combo of velocity and movement he gets on his pitches (per PitchF/X) that comparables like Harang and Bass don’t have.
I traded CJ for Justin Upton …. keeper league …. I feel pretty good, despite JUP sucking this year.
KFC