Why I’m Down on C.J. Wilson
A week and a half ago, I published my American League starting pitcher tiers, which were based on my projections and ranked pitchers in descending order of dollar value. One of the bigger surprises was C.J. Wilson landing in the third tier, ranking 16th among all AL starters. This ranking was behind guys like Brandon Morrow, Colby Lewis and Max Scherzer, who it is likely that the vast majority of fantasy owners would draft much later than Wilson. So why am I down on Wilson? Let me count the reasons.
After being signed by the Angels and getting out of Texas, the immediate assumption would be that Wilson would be even better. After all, we all know that Rangers Ballpark is a hitter’s haven and Angel Stadium favors pitchers. Many also like to point out Wilson’s home/road splits as more evidence that he is going to thrive even more so in Los Angeles. In the last two seasons as a starter, Wilson has posted ERAs of 3.70 and 3.69 at home, but 2.91 and 2.31 away. Surely, he is pitching much better away from Rangers Ballpark, right? Well, not exactly.
As you are probably well aware by now, I mostly ignore ERA and choose to focus on the peripherals and the expected ERA metrics, like SIERA, as that gives us a much better idea of the pitcher’s talent, stripping out luck. Unfortunately, SIERA is not available for home/road splits, but his xFIP at home has been 4.03 and 3.27. What has it been away? Surprisingly, even worse at 4.11 and 3.53. So his skills have actually been worse pitching away from home. Only great fortune in away games has created the illusion that he is really being hurt by his home park.
His BABIPs away have been much lower than at home for one. Now, Rangers Ballpark does inflate hits on balls in play, but Wilson’s differential is much larger than you would expect just simply based on park factors. Ground ball pitchers typically don’t post BABIP marks well below the league average like Wilson has, so it would take a much larger sample size to convince me that Wilson has some sort of hit suppression skill only rarely seen in ground ballers. Similarly, his LOB% is much higher during away games. The BABIP difference alone probably isn’t enough to explain this. Last, his HR/FB ratio in 2010 was nearly identical both home and away, both being at unsustainably low levels. In 2011, his home mark jumped to 11%, which is above league average and makes sense given the home run inflating powers Rangers Ballpark has. But his away mark was again between 5% and 6% and that’s simply not a level any starting pitcher has the skill to sustain.
Overall, in 2011 Wilson posted a 2.94 ERA, but a 3.44 SIERA. So clearly he benefited from some good luck, but a 3.44 ERA would still be excellent for this year if his luck neutralizes. Well yes, but that assumes he posts similar skills this year, which is not something I think he can do. His 8.3 K/9 is likely to fall given his below league average SwStk% and his BB/9 is probably going to rise given his history and worse than league average F-Strike%. Worsening peripherals are going to cause his SIERA to spike and only another year of below league average BABIP and HR/FB ratio marks will prevent his ERA from rising into the high 3.00 range.
My official projection is:
| IP | ERA | WHIP | W | SO | K/9 | BB/9 | BABIP | HR/F |
|---|---|---|---|---|---|---|---|---|
| 215 | 3.71 | 1.30 | 14 | 189 | 7.9 | 3.4 | 0.300 | 10% |
That ERA projection falls between his SIERA marks over the last two seasons as a starter. He is just barely earning positive value from his ERA and his WHIP actually hurts fantasy teams ever so slightly. His value basically comes solely from wins and strikeouts according to what I am projecting.
Mike Podhorzer is the founder of ProjectingX IQ, an advanced fantasy baseball analytics platform that transforms projection data and in-season performance signals into actionable intelligence. He is the 2015 Fantasy Sports Writers Association Baseball Writer of the Year and three-time Tout Wars champion. He is the author of the eBook Projecting X 2.0: How to Forecast Baseball Player Performance, which teaches you how to project players yourself. Follow Mike on X@MikePodhorzer and contact him via email.
Thanks for the clarification! I’ve always thought that something like this was behind Wilson’s stats, but I’ve never really looked into it. Great article!
Does SIERRA strip out “Luck” or just hold certain things we don’t understand even? And since we don’t understand those certain things, it’s possible that those certain things can be controlled by certain pitchers in ways that, again, we don’t understand and, therefore, hold even.
I’m just asking, and dunno much about SIERRA, but economists often overrate their models by forgetting that ceteris paribus is a useful fiction, but in reality nusquam paribus.
I don’t think those two things are necessarily different. A lot of what we consider luck is what we simply have not determined yet there is a skill difference, but we also are quite sure that a .200 BABIP is made up of a whole lout of luck even if that pitcher does have some BABIP suppression skills better than the average pitcher. We only have 2 seasons worth of innings from Wilson, which is simply too small a sample to conclude he has BABIP and HR/FB skills better than the average pitcher. He certainly could, but at this time, without clear evidence other than he’s done it so far, the best thing to do is assume he doesn’t and project him as such.
I don’t think you realize that xFIP is mostly park-neutralized. I.e., Rangers pitchers will almost always have XFIPs lower than their actual ERAs bc xFIP assumes an average HR/flyball rate, which clearly is not an appropriate assumption for the Rangers home park.
In general, people frequently make the mistake of saying a pitcher has out/under-performed his xFIP without realizing that xFIP is not intended to be compared to actual ERA in cases where park HR factors are significant.
Of course I understand how xFIP is computed. However, Wilson’s HR/FB ratio was only 8.2% last year and 5.3% the prior year. Do you really think that drops further even just because he’s leaving a home run haven to a more neutral park?
I am down on Wilson as well but i don’t buy your theory that a pitcher’s skill changes depending on where he pitches. Those home/away xFIPs dont seem significantly different. If anything it just shows how much of a hitters park Arlington really is. He pitches the same no matter where and his ERA is a point lower out of Texas
as.
Ballparks can also affect a pitcher’s skills, as in strikeout and walk rates. For example, last year PETCO Park inflated strikeouts. Wilson’s skills were not negatively affected by Rangers Ballpark, as the ERA difference solely stems from much better marks in the luck metrics. However, those differences are too large to chalk up to just park effects, so a lot of it is simply good fortune that you should expect to neutralize.
Hopefully this isn’t a question that would take an entire article to answer, but how do you know Petco inflated strikeouts? Is that based on how the Padres pitching staff did at home vs away? Or is it based on Padres hitters home and away?
Could we have more causation analysis on this one. It seems rather bold to hang your hat on a correlative framing of ball park factors when it relates to strike outs and walks. This is particularly true when your meta-level criticism of people pro Wilson is that it is located in a false correlative analysis of ball park factors when it relates to his ERA and FIP/SIERA.
Statcorner.com. It gives park factors for many statistics. PECTO increased strikeouts by 7% versus lefties and 10% versus righties last year. http://statcorner.com/team.php?team=SDN&year=2011&leag=N_L
Does that have anything to do with the fact that the Padres were playing there every game? Just a hunch…
No, park factors are not influenced by the players on the team. That’s not how they are calculated. There may be a bit more complexity, but in general, park factors compares how a team performs in its home parks with away games. So it is telling us that the Padres struck out more at home than away.
The best example of this seems to be the speculation that Jared Weaver’s motion is ideally suited to Anaheim– http://www.fangraphs.com/blogs/index.php/jered-weavers-favorite-rockpile/
Or rather, most extreme, not best.
I feel like the operative term is EXTREME because virtually no one else has this type of pitch type and even if they did the unique marriage to Angels Stadium and Jered Weaver make me think that this is an exception that virtually no other pitcher can readily access
Him having a BABIP lower than league average could be a factor of him having a good defense behind him, right? Texas was pretty good, especially the left side of the IF in Beltre/Andrus.
Yeah, that plays a role too as Texas has had excellent defenses in recent years. The Angels UZR/150 was fantastic last year as well, so he actually might not see much of a decline in defensive support. Though if Trumbo gets lots of playing time at 3rd, that’s unlikely to be a good thing for Wilson’s BABIP.
xFIP is not useful to your argument. xFIP takes things like park HR factors out of the equation, but fantasy doesn’t. If he moves to a park that suppress his HR/FB and pitches exactly the same, his xFIP won’t change at all, but his ERA will.
It does, because like I stated in the article, the discrepancy is too large to be chalked up completely to park factors. Luck was definitely involved as well. He’s posted a mid-5% HR/FB ratio away from Texas the last two years. That’s simply not sustainable in any park.
That HR/FB number does a lot more to convince me than any of the xFIP you used in the article. I missed it at the end of the fourth paragraph up there. I’m still not sure why you used xFIP to show the regression you expect because his road HR/FB rate was unsustainable. FIP and HR/FB would seem to be the right tools for that.
You’re forgetting about the half run in ERA he is going to save by pitching to the Angels’ superior catchers, and not the guy so bad he was traded away for Vernon Wells.
[/sarcasm]
Problem?
hey. Jeff Mathis calls a GREAT game.
down on cj but he’ll have a sub 4 era for sure w. 175+ K
he got his contract and ran, just like cargo, matt kemp… and zito.
He got his contract and ran? It’s generally believed he left a lot of money out on the table to sign with his hometown Angels. He’s the opposite of going with the highest bidder.
where did CarGo and Kemp run to, exactly?
Although I appreciate the analysis on this article I have a certain criticism:
I feel like falling back to league average in virtually all of your analysis is problematic. I think it begs the question of what type of pitcher the league average paints and neutralizes various forms of pitchers into one. For example, his SwStr% being low seems indicative of the type of pitcher he is, a pitch to contact pitcher who generates GB for his defense to soak up. Also, if he is trying to deceive and mitigate the effects of batters making contact, that would indicate that his strike out strategy is divergent from fastball pitchers that thrive on F-Strike and SwStr%
But, I mostly ask for better understanding and appreciate any knowledge you can provide! thanks!
I don’t really care why his SwStk% is low, I am concerned that a low mark means his strikeout rate is bound to fall.
Do you ever get the feeling that a pitcher is just going to be awful in his new contract? I agree with your projection as it basically echoes the Steamer projection, but I also have the feeling that Wilson is just going to implode.
With that said, had it been Colby Lewis switching parks, I take it you would’ve bumped him up based on his splits / k-rate / luck indicators?
Mr Podhorzer, nice article. We are having a little discussion about it on reddit. Someone asked a question about what the implications of .300 babip are. I don’t suppose you want to make an account (30 seconds) and answer him? I post links to fangraphs a few times a week so I’m sure it would generate a little interest if you posted occasionally, even do an AMA (you’ll have to look up what that is). Just a thought. It’s got 2,000 subscribers. Thanks! http://www.reddit.com/r/fantasybaseball/comments/rmwt4/why_im_down_on_cj_wilson_fangraphs_fantasy/c476cyk
Thanks for expounding on your Wilson position. I see a heavy-groundball pitcher with good component skills leaving a hitter’s haven resulting in fewer bombs and a better ERA. I’d be really surprised if he was inexplicably worse than his 1st two years as an SP after moving out of TEX, but we’ll see.
I am certainly on board that C.J. will see some regression this year, but he should still be a nice pitcher all told. What I don’t see is how Colby Lewis will magically be more valuable [i] as a fantasy starter[/i].
Lewis is a fly-ball pitcher in a launching pad park. Last year his home ERA was 5.54 and away it was 3.43. He is likely a better pitcher than his ERA would indicate, but the reality is he still pitches at Arlington half the time. He will also have the pleasure of pitching to Pujols this year, so good luck with that.
To respond to questions about Colby Lewis, just a reminder that he posted a 3.72 ERA (and 3.58 SIERA) in 2010 with a 1.19 WHIP. His skills did take a dip last year, but I’m sure a lot of it was caused by the velocity drop. Still, he didn’t pitch marginally worse, his luck just tilted the opposite way. He’s going to have a better WHIP than Wilson and the wins and strikeouts are going to be close. So it comes down to the ERA, and since Lewis already has the WHIP advantage, Wilson needs to win ERA to earn similar fantasy value.
The blind guarantee that Lewis will have a better WHIP than Wilson is odd to me. It’s 1.201 for Lewis and 1.215 for Wilson the last two years. Lewis *might* enjoy a negligible advantage in the category. The difference in those two WHIPs is 3 base runners in 200 innings. That’s not an advantage.
Even if you give Wilson his 2010 WHIP of 1.245, it’s 9 base runners in 200 IP compared to Lewis’s 1.201… that’s nothing. Wilson beat average BABIP by all of 1.3% last year at .287 so even if he regresses, I don’t think anything close to an implosion is on the horizon. Plus he could improve his control again and mitigate any BABIP regression.
BTW, Lewis’s 2010 ERA was aided by a remarkably favorable HR/FB rate in Texas. It regressed (admittedly too much to 16%) in 2011 and we saw what happened. If he doesn’t beat the 10% HR/FB league average at home, he will have problems. Sign me up for Wilson possibly allowing a few more singles over the home run king pitching in Texas a heavy-flyballer any day.
It almost feels like you’re dismissing Wilson entirely as the case against him isn’t terribly strong as I see it presented. You cite Lewis’ ’10 SIERA which is fine, but Wilson’s was better last year. Wilson’s skills are improving while Lewis’ regressed/stagnated save a 0.4 gain in BB rate. I just don’t see any reason to take him over Wilson.
Interesting debate, Mike.
I find this utterly unconvincing. He’s moving to a better ballpark within a division that he has basically dominated two years in a row. And his base skills are supposed to erode because…..? I can’t think of any fantasy format in the world where Colby Lewis will have more value than C.J. Wilson this year.
Mike, I believe you and many others sometimes rely too much on sabremetrics…and believe they can tell a whole story. I think they have their place in baseball, and I use them regularly.
With that said, if you take a closer look into C. J. Wilson’s career, you’ll find that his where he ranks in swinging strike rate doesn’t line up with his K per 9 like it does with the vast majority of other pitchers. Last season was not an outlier. That’s who he is. Look at his 2010. Look at the pitchers who had a swinging strike rate comparable to his..which was under 7 percent. They were the Bronson Arroyo’s of the world. Arroyo had a k rate of 5.05 that year.
What was Wilson’s rate? 7.50. Randy Wolf, Jon Garland, Carl Pavano, and Jake Westbrook all had better Swinging Strike rates than Wilson in 2010. Yet Wilson’s K per 9 was significantly higher.
Also, BABIP fluctuations aren’t just because of luck. Pitchers have varying degrees of ability to get a hitter to hit a ball where he wants him to hit it. Ask anyone whoever saw Greg Maddux pitched.
Anyway, after comparing Wilson’s Swinging Strike rate each year to his Strikeout rate, it’s easy to see that Wilson’s strikeout numbers aren’t a fluke.
Also, in addition…the idea that a guy could be super lucky on the road, and super unlucky at home for 2 straight years is absolutely laughable…for lack of a better word.
The odds of a major league pitcher having extreme “luck” variations based on whether he’s pitching at home or on the road for a period of 2 years are extremely low. Like, I feel I have a better chance of winning Powerball…than the chances that C. J. Wilson has really great luck for 30+ road starts, and really bad luck for 30+ home starts.
To imply such a theory is to be insane.
Agreed, he has two full seasons of defying his sabr peripherals at this point, yet he is still being treated as a typical pitcher. Sometimes its to your advantage to accept that there are outliers, especially in baseball, where players aren’t machines.
You do realize that 2 seasons of home and away numbers means one full season of home numbers and one full season of away numbers, right? One season is the definition of a small sample size. The odds of a pitcher having extreme luck variations is actually very high. In fact, every year we see it by just looking at the full seasons of many pitchers.
Right. But the idea that a pitcher could be lucky just when he’s on the road for the same span of 2 years seems absurd. I mean, can we even reference another case of such luck from another starting pitchers in the last decade…where they were really lucky on the road for 2 seasons, and very unlucky at home during that same span?
I’m sorry, but the odds of someone being only lucky when they go on the road over the span of 2 seasons seems ludicrous…and is definitely not the same as being lucky over the course of a full season.
With that said, my main point is…when you look at his annual Swinging Strike Rates and compare it to his actual Strikeout rate, 2011 is actually par for the course for C. J. Wilson. Striking out over 8 per nine while only generating an 8.3 Swinging Strke rate is pretty normal for him. Basically, he’s spent most of his career freezing batters.
His BABIP for the season wasn’t crazy low, .287 is not that extreme….I think he’s at worst, in line to stay about the same considering the change in home park.
I’ve got another thing which strikes me; during almost his whole career Wilson has been able to post a better ERA than his FIP, xFIP, SIERA, you name it. So it seems there’s a pattern in this. What could be behind this? Some kind of abiltiy to control the outcome of his groundballs?
You have to throw out his relief statistics because relievers post lower BABIP and HR/FB ratios and higher LOB% marks on average than starters. So they will nearly always beat their expected ERA metrics.
2.70 ERA 1.02 WHIP, 30 Ks in 33.1 innings. Have a full season left, but the kid’s bringin it. I’m callin a Halladayesque season, but a slow July. Let’s see how the season plays out.
…and Barry Zito has a 1.76 ERA and 1.11 WHIP. Wilson has a .217 BABIP and his SwStk% is down again. Seems like a great sell high guy to me, especially if you own one of the 15 3rd baseman who was recently injured!
Youre comparing CJ Wilson to Barry Zito… That’s such a stretch especially since one is in his prime and the other exited his like 5 years ago.
Not comparing him to anyone. My point in mentioning Zito is that ERA and WHIP after 1 month of stats are still rather meaningless. To mention them without looking at SIERA, BABIP, etc is giving an incomplete picture of how he has actually pitched.
So…has anyone actually been watching Wilson pitch? With eyeballs. I think he made improvements last year with his control and they have stuck around so far this year. What makes you think his BB/9 is suddenly going to spike up? It seems to me CJ has demonstrated a genuine ability to limit his walks and pitch to GB contact. I don’t see that changing unless he gets hurt or completely slips out of his mechanics. So going to a pitcher friendly home ballpark should only increase his “luck”. I think he’ll post even better numbers this year than last.
Because it has. He’s back to walking 4 per 9 again. Have fun with that Angels fans.
Guess CJ is having the last laugh, huh?