AL Cy Young Crowdsourcing Results

On Friday, we ran a poll asking who you’d vote for the AL Cy Young Award. With 5,500 votes in, the votes were decisive, to say the least.

Felix Hernandez captured over 60 percent of the votes, and no one else was even close. Francisco Liriano finished second with just under 11 percent, while CC Sabathia came in third at a bit over 9 percent. Cliff Lee was fourth at 7 percent, while David Price rounded out the top five at just over 4 percent. Clay Buchholz, Trevor Cahill, C.J. Wilson, Jered Weaver, and Jon Lester were essentially non-factors – we probably could have left them off the ballot without seeing any real change in the outcome.

There are a few interesting thing about the results, to me. I was surprised by the measure with which Felix dominated the voting, honestly. As often as the FanGraphs crowd is referred to as “stat zombies”, the perception is that our authors and readers just look at the numbers and make no adjustments beyond what the leaderboard says, yet the guy who dominated the voting doesn’t lead the league in any of the “FanGraphs Stats” – WAR, FIP, xFIP, or WPA.

Cliff Lee is first in WAR, yet finished a distant fourth, as voters presumably held his recent struggles and higher ERA against him. Francisco Liriano is first in both FIP and xFIP, but only 1 reader out of 10 gave him their vote, likely because of his lower innings total and his higher ERA. And perhaps most surprisingly, CC Sabathia finished third, even though he’s not appreciably better than the also-rans in anything other than Win-Loss record.

Sabathia ranks 15th in FIP, 11th in xFIP, and 10th in WAR, yet he did significantly better than guys like Jon Lester and Jered Weaver, who have him beat in most of the “advanced metrics”. Let’s compare Lester and Sabathia, for instance, as both play in the AL East and are the aces of large market, nationally prominent clubs.

BB/9: Sabathia (2.84) over Lester (3.36)
K/9: Lester (9.69) over Sabathia (7.32)
HR/9: Lester (0.59) over Sabathia (0.78)
FIP: Lester (3.09) over Sabathia (3.62)
xFIP: Lester (3.26) over Sabathia (3.86)
WAR: Lester (5.0) over Sabathia (4.0)

Sabathia has the lower walk rate, but Lester’s huge advantage in strikeouts and lower HR rate more than cancels that out. It would be hard to build a case that Sabathia has outpitched Lester based on the numbers above. However, when you look at two traditional metrics, we can see why Sabathia got so many more votes than Lester.

ERA: Sabathia (3.14) over Lester (3.26)
Innings: Sabathia (209) over Lester (182)

Sabathia has a marginally better ERA than Lester and he’s thrown nearly 30 more innings to boot, which is why he’s considered a frontrunner for the award while Lester is never seriously brought up in conversation. Even among our readership, Sabathia dominates Lester with 518 votes to Lester’s 69. While our stats show that Lester has been better, our readers prefer CC’s quantity of innings with essentially the same rate of run prevention, and don’t really seem to care that the difference is almost entirely driven by BABIP.

I would be tempted to chalk this up to the power of the narrative, where people were voting for Sabathia because the media has kept him at the forefront of the discussion, but we see this same rejection of DIPS theory in the vote totals for Hernandez, Liriano, and Lee. Even though you’re spending your Friday afternoon reading FanGraphs, most of you guys still seem to vote along the lines of innings and ERA. To me, that’s interesting.

Perhaps the divide between the traditional media and people who like nerd stats is not as big as the generally perception. In the end, both camps appear to prefer results to process when it comes to handing out awards. The only question now is whether the baseball writers agree that results can be judged without leaning heavily on Win-Loss record as a factor. If they’ve come to the same conclusions as our readers, then King Felix is line for a new crown.




Print This Post



Dave is the Managing Editor of FanGraphs.


121 Responses to “AL Cy Young Crowdsourcing Results”

You can follow any responses to this entry through the RSS 2.0 feed.
  1. Joe says:

    I think it’s great people don’t put 100% faith into approximate formula. Anyone who thinks Player A is definitively better than Player B because he has .1 more WAR is a fool. For example, a player like Matt Cain won’t get as much credit from those advanced metrics even though he’s shown over the course of his career they don’t apply to him. So why should be put absolute faith in them when it comes to something like this?

    Vote -1 Vote +1

    • Locke says:

      I’d like to hear whether Dave thinks he “lead the crowd” or not with his Felix-laden intro to this poll:

      “Of course, Felix isn’t blowing the field away like Josh Hamilton is in the MVP race. He’s had a great year, but he’s not the only pitcher having a great year. You can make an argument for pitchers other than Felix without resorting to craziness. So, I figured we’d put it to you guys. If the season ended today, who is your American League Cy Young Award winner?”

      There’s no chance he gets 60% of the vote if the poll had a neutral intro.

      Vote -1 Vote +1

      • Wally says:

        Its true enough that in polling we shouldn’t try to lead the poll takers, however I believe in this case some mild leading isn’t such a problem for a couple of reasons:

        1) People at fangraphs tend not to be the kind of folks that just go along with the crowd, will do their own research and make an informed independent choice.

        2) I also don’t believe descibing what most of the debate has been up to this point to be particularly leading. If Dave were to list of several stats that make a biased argument for one player or the other I’d agree with you, but he was pretty mild in pro-Cy statements regarding Felix, and clearly mentioned that other pitchers are having great seasons as well.

        Finally, I’m not surprised by the results. A lot of people here surely love their FIP, xFIP and WAR, but between the 3 stats no one pitcher really comes across as a clear winner. As such, it seems perfectly logical to pick the guy with the best results outside of W-L record. I chose WPA, others have looked to ERA. Doesn’t really change the outcome.

        Vote -1 Vote +1

      • Locke says:

        That’s all well and good, but it doesn’t change the fact that in his short, two paragraph intro to this poll there were 4 names written out:

        Felix’s three times
        CC’s once
        Liriano/Lee/Everyone else = 0

        I mean, READ THIS SENTENCE

        “The narrative being told at the moment is that Felix Hernandez is, by far, the most deserving candidate,”

        Needless to say, this type of thing has concrete implications on polls. It’s not even really a discussion. I guarantee if you run this poll on Baseball Prospectus with no intro, the numbers are nowhere close to this.

        I wish I could say I’m surprised Dave missed this type of thing.

        Vote -1 Vote +1

      • Wally says:

        Locke,

        First, what’s up with hostility man?

        I agree with you and fully understand the concept of leading poll questions. I just disagree with level of skew you are attributing to this particularly case. Go ahead, run a more neutral poll and see if the numbers turn out to be different. Though to do this you had better establish some idea of the mindset in your new sample group and determine it to be a rough equivalent of that found here at fangraphs. Simply going to BP is not going to be enough (where say they may defer to tERA or RA or something, where here we generally defer to (x)FIP/WAR), especially from a guy so worked up about this particular polling issue.

        Anyway, I doubt you’ll find them to be “no where close to this.” When Felix is basically in a three way tie using the more advanced metrics, it perfectly reasonable to look at the actual results as the tie breaker. Also, when you look at advanced metrics used at other sites, such as VORP or WAR on B-R.com, Felix leads by decent margin. Other readers may have used that as further evidence to give Felix the nod.

        Vote -1 Vote +1

      • Socrates says:

        Locke is unquestionably correct. The intro to the poll was very leading. I dont care if you where polling major league managers, and not lowly fangraphs readers. I think it actually influenced me. I started looking at the results. Had trouble assigning value to certain stats (by the way, I also gave more points to pitchers on playoff teams… Sorry Lester). Still at the end, I couldnt decide between Felix, CC, and Liriano. I chose Felix, maybe because of that intro.

        I work in a field where polling used very often and I am invovled in drafting probably a poll or two a week this time of year (guess the field). Anything that leading would result in a massively skewed result.

        I still think that Felix should be #1 on radars at this point. He is just so good right now.

        Vote -1 Vote +1

      • Wally says:

        Socrates,

        I’m not questioning if Locke is right or wrong regarding leading poll takers, I’m questioning just how much he seems to think the effect was.

        By your own statements it seems that you believe him to be the front runner as well, regardless of the narrative.

        I’d be interested to see the results of a non-leading poll in a similar set of poll takers.

        Vote -1 Vote +1

    • My echo and bunnymen says:

      I had this exact same argument with someone I know on facebook. I’ve done a Cy Young Watch (and at the time) Francisco Liriano led in WAR by .1. He argued against my entire system based on the fact that I choose WAR as my leading stat, though I mentioned that in the end I would choose Cliff Lee over Liriano because of his innings pitched, quality of starts, and so on and so forth. WAR isn’t perfect but it is a GREAT stat. I use it in arguments but absolute faith is easy to put into it (I put absolute faith in it) because it is a stat. It’s not lying, not deceiving, nor misleading, it’s a stat run through a formula, whether you believe it has merit, that’s another discussion.

      Vote -1 Vote +1

      • Joe says:

        My echo and bunnymen: You shouldn’t put absolute faith into anything approximate. As you said, it’s a great tool, but to use it definitively is misguided. I guarantee in 10 years the formula will be better than it is today. That being the case, how can you say with absolute confidence it is correct?

        Vote -1 Vote +1

    • Mafrth77 says:

      “1 more WAR” or between Lester and Sabatthia 25 percent differnce in the two players players value.

      Vote -1 Vote +1

  2. Joe says:

    To phrase it differently, just because Liriano has a better FIP doesn’t mean he has pitched better than Felix. There’s enough variance between the model and actual performances that this tool should be used as a rough approximate of a pitcher’s true level, not an absolute one.

    Vote -1 Vote +1

  3. Andrew says:

    I voted for Felix as I believe in giving the award to the pitcher who earned the best results- with luck, defense, and run support included. Not the one who is theoretically the best pitcher, or who may be predicted to be the best in the future.

    Vote -1 Vote +1

    • Paco says:

      Best results with luck, defense, and run support included – this sounds like W-L record to me.

      Vote -1 Vote +1

    • Bill@TPA says:

      Stats like FIP and xFIP aren’t “theoretical,” and they aren’t merely “predictive” either — “who earned the best results” is exactly what they help tell you. Where did people get this idea that they can be somehow separated from what the pitcher “actually did”?

      Vote -1 Vote +1

      • Joe says:

        Bill: It is absolutely theoretical. When attempting to classify performance independent of many variables, it’s impossible to do so without complete understanding of the system. Given that it’s based on a few of the most important underlying criteria, it can’t be said that it perfectly captures every aspect necessary to determine what “should” have happened. Until it’s refined to the point of perfection, it is only a theoretical value of a pitchers performance.

        Vote -1 Vote +1

      • dnc says:

        Thank you. This is driving me nuts.

        Vote -1 Vote +1

  4. John says:

    I guess I am curious about people’s rational for voting for Felix.

    To me Joe, it isn’t about Liriano having a better FIP (although certainly that is a part of it). It is the fact that he has a better K rate, similar BB rate and a better HR rate than Felix.

    The one big thing in Felix’s favor is his ERA and the IP advantage. Am I missing anything?

    Vote -1 Vote +1

    • Joe says:

      What you’re missing is that these guys are people who go out and pitch…again, a guy like Matt Cain has an inferior K rate and K/BB rate to a lot of pitchers with worse ERAs. That doesn’t mean he’s lucky and they aren’t. It means that the basic elements that go into pitching don’t always capture the end result. It’s not that complicated.

      Vote -1 Vote +1

    • Joe says:

      At what point do we stop calling Ricky Nolasco ‘unlucky’? Yes, there is some value to these formulas that attempt to separate out the indicators from the end product, but you’re kidding yourself if you think that they capture everything properly. Take a system id class and you’ll understand what I mean.

      Vote -1 Vote +1

      • John says:

        Sorry Joe, I just basically disagree with the idea of some unmeasurable, unobservable factor. People try to play this card a lot, Rockies fans did earlier in the year w/ Ubaldo. Tons of fans try to give justification why their guy is the exception and why he won’t regress, but generally the guy does. I am not saying that it couldn’t exist, just that I am skeptical until I see some indication it does.

        As for Cain, personally I think this is a case of a flipping a coin and coming up heads 10 times in a row.

        Vote -1 Vote +1

      • Joe says:

        John, it doesn’t matter if you agree or disagree, it’s clear enough in the results that you’re wrong. If you learn a little about system identification (the basis for what these formulas are trying to capture) you would realize the system is far more complicated than the 5 or 6 state approach that they use to fit the system. In actuality, maybe it’s a 100 or 200 state system, but that doesn’t mean you can’t get a pretty good approximation using these parameters. You can’t just point to luck when your approximation doesn’t match the results. Moreover, it would be INCREDIBLY foolish to assume you’re approximation is absolutely correct instead of allowing that sometimes you just don’t know why the input and output don’t match.

        Vote -1 Vote +1

      • B N says:

        I have to take Joe’s side on this one. I’m not saying these factors are unmeasurable or unobservable, but they ARE unmeasured and unrecorded. There is a very good chance that if you wanted to have a much improved model of pitching effectiveness, you’d have to incorporate pitching/batting trajectory data (paths of the ball, basically), lighting information, and a whole host of other important factors. And even with that information, you’d probably need to employ a game-theoretic model to see if their pitching sequencing was effective with regards to their opponents.

        Instead, we’re counting up K’s, walks/HBP, and HR. Or earned runs scored. When we want to get REALLY clever, we categorize batted ball trajectories into 3 bins. When you’re discarding huge amounts of continuous 4-D data and pulling out about 6 discrete factors? Needless to say, you’re going to lose a heck of a lot of data. Doesn’t mean it’s unmeasurable. Means its unmeasured and unanalyzed.

        Vote -1 Vote +1

      • John says:

        The outcome of 3 or 4 seasons can make it “clear in the results” that I am wrong? Sure it does. I think instead of a system ids class, you might benefit from an introduction to statistical methodology.

        Vote -1 Vote +1

      • John says:

        I am not saying that other factors don’t exist. Just that you get a more accurate picture if you ignore them rather than speculating they exist for some players. Maybe some day we can identify/measure them, but we can’t right now.

        Vote -1 Vote +1

      • John says:

        BN aren’t you kind of talking about TERA? It might not do everything you are saying, but it does do a lot. A lot of these guys w/ “unmeasurables” also are outperforming their TERA.

        Vote -1 Vote +1

      • B N says:

        Plus, if you really wanted to test if that might be the case for Cain, you could do one of two things:

        A. Test for the likelihood that Cain’s distribution of batted balls that fall in for hits is the same as the mean distribution. Same test you’d use for coin flips. Your sample size would not be 10 flips, but it would in fact be every ball that had ever been batted into play against him.

        B. Examine the distribution of pitcher’s BABIP by season, where you compare two groups: the variance between an individual pitcher’s different seasons vs the variance between all pitcher’s different seasons over that time period (maybe 5 years). If pitchers in general tend to have more similar BABIP to themselves in different years than they to do other pitcher’s seasons, your hypothesis is probably incorrect. Which states nothing about Cain specifically, but speaks about the model you’re employing.

        Vote -1 Vote +1

      • John says:

        BN why wouldn’t you simply look at tERA? He seems to continually outperform that as well as his FIP which indicates some luck, no?

        Vote -1 Vote +1

      • Joe says:

        John: I can only relate what I used in my own background, but in reality I’m sure they are quite similar. Regression analysis is regression analysis. Ignoring what you can’t observe to make your argument is extremely convenient but highly inaccurate. It does no good to try to attribute some mystical power to a pitcher like Rockies’ fans did, but it’s also silly to definitively say it’s luck that one guy is beating your projection while the other isn’t.

        B N: I agree completely. The tools are getting better and better but they aren’t (and may never be) complete. Even defensive positioning should taken into account. I’m sure the shift affects people’s hits even when you account for specific batted ball data. But to make a singular equation to handle all of that is quite cumbersome I’m sure.

        Vote -1 Vote +1

      • B N says:

        tERA is definitely getting closer. That specific approach to looking at this type of data may not end up as the end-all-be-all, but if the data is used effectively- there should be a lot less unexplained variance. In my opinion, these models will still need some time to mature. It takes a while to figure out the best way to use large quantities of data. Just look at machine vision to see how that path goes.

        Vote -1 Vote +1

      • B N says:

        And using tERA to figure out if Cain is lucky would be a misapplication. While it may be a better system in general (or at least a better concept), there is no assurance it is necessarily good for any one pitcher. If his effectiveness relies upon sequencing and/or deception, you could easily be missing a big nonlinear chunk of the situation.

        Vote -1 Vote +1

      • Joe says:

        B N: I think fielder f/x will also add to the conversation. Not only will it help to determine a fielder’s value, but we might actually be able to observe the notion of “easily fielded” hits. Specifically taking into account the leverage of a situation.

        The machine vision example is interesting, I’ve only sat in on a couple of seminars and presentations on the subject, but it’s quite an interesting subject. No matter how sophisticated the algorithms become, there are always cases where the algorithms can be fooled. Potentially not dissimilar to a guy like Cain. While they (like baseball equations) improve, it may never be possible to construct a perfect method to synthesize vision.

        On a side note, it’s interesting how much better computers are at certain things but how lost they are at others. One of the talks had to do with local memory regarding obstructed targets. This particular example was trying to track a surfer who had gone underwater and then resurfaced. As a human it’s quite apparent, but to the computer tracking a potentially amorphous gray blob with irregular movements is not so simple.

        Vote -1 Vote +1

      • John says:

        Joe: I am not saying it is luck. I am simply saying that it ends up being more misleading than useful at this point to try and incorporate it. In a legal setting, you would say that the evidence lacks probative value. It is more likely to mislead than inform. You seem to say that simply because a pitcher has outperformed or underperformed regularly that is determinative that they have something (or don’t have something) of value, whereas the evidence is only circumstantial.

        BN: I an not saying tERA is perfect, personally I prefer WHIP. But I guess I don’t understand how it isn’t an indication of luck (just not a perfect one). It looks at individual batted ball data, location and trajectory, so why would it have bias toward or against an individual pitcher?

        Vote -1 Vote +1

      • Joe says:

        John, I agree that it is potentially misleading to try to characterize the things we don’t understand particularly well. But it’s acknowledging that the methods we have aren’t perfect that opens up the conversation passed the singular number that’s often presented in these arguments. If an algorithm was right 90% of the time and it said 1 suspect was 2% more likely to have committed a crime than another suspect, would you then assume with 100% conviction that one was guilty and the other innocent? More over, not all of the advanced metrics agree. The defense attorney in this case could point to another algorithm, also correct 90% of the time (baseballreference’s WAR for example), that shows the other suspect is actually guilty.

        Now that might be a bad example because I’m sure that’s actually how the justice system works. I would imagine very rarely you have 100% confidence in any sort conviction, but it’s scary to think that people regularly do draw conclusions with absolute certainty in cases where there is anything but. The article about a year ago in the New Yorker about an arson case that was mismanaged was particularly frightening. Maybe it’s just human nature.

        Vote -1 Vote +1

      • John says:

        Well Joe: It comes down to our tolerance of error and whether we favor type I or type II errors. In the legal setting they have a bias against type I errors (false positives), because the general philosophy is that it is better to let a guilty man go free than imprison an innocent one.

        However, in this setting we don’t have any particular bias towards or against type I versus type II. Therefore, it seems to me that we should simply do our best with available evidence, even if said evidence is imperfect. It is my position, that as of right now, the path to the best conclusion is to ignore the “noise” we can’t categorize or measure, simply because that noise currently lacks “probative” value.

        Vote -1 Vote +1

      • Joe says:

        John: That’s certainly fine. My problem is the people who don’t recognize the imperfections in the analysis and then use them to belittle those who don’t agree. While I believe the analysis is certainly good enough to make certain conclusions, namely that people who vote for CC are just plain wrong, it’s not good enough to differentiate between the close cases. You originally said you did not understand how someone could vote for Hernandez over Liriano. The point of all of my posts is that choosing to be zealous over a particular metric is not fool proof and there is more room in the discussion for other points of view.

        Vote -1 Vote +1

      • John says:

        Joe, you misunderstand me. I can conceive of voting for Felix over Liriano. If I was a Seattle fan instead of a Twins fan, I admit I would be doing exactly that.

        I was honestly looking for people’s logic. It was more a matter of curiosity, not accusation. Just with 60% going to Felix, I was thinking I was missing something other than the ERA and IP, which I admit makes him a favorite over Liriano, but not such a prohibitive one. To me it seems like the narrative, not the stats have to be part of the reason for such a landslide victory.

        Vote -1 Vote +1

      • Joe says:

        John, then I apologize for misinterpreting your question. I think the first reaction I interpret over the internet is of hostility. Granted, reading your statement again it certainly wasn’t hostile. I’m used to posting something and then getting torn into by people who don’t agree.

        Personally, I voted for Felix. Given that the advanced metrics were very similar for each pitcher, I used real results (IP and ERA) as a tie breaker. It could very well be luck, but it also might be those unaccounted factors that have some people pulling their hair out. I don’t believe in “guile” or things like that helping someone pitch, but I don’t think the formulas are ironclad either. That being said, using the actual results as a tie breaker doesn’t seem terribly unfair to me.

        I’m much more receptive to someone like you who at least acknowledges the potential shortcomings of these formulas. You can certainly argue it’s better to stick to your guns and declare the winner because they rate .01% (exaggerated obviously, but still) better by a preferred metric, I just don’t have enough faith in them to do that.

        Vote -1 Vote +1

      • Mafrth77 says:

        Nolasco isn’t “unlucky”- he is one of the more extreme fly ball pitcher in baseball an is going to give up a ton of Homeruns

        Vote -1 Vote +1

  5. Andrew says:

    The CYA is results based on a pitcher’s performance in the given year..not the predicted performance going forward which s what FIP is for. Plus CC Sabathia is a huge part of the Yankees winning the division so the voters give him some credit for that.

    Felix would get my vote but I’d probably go CC 2nd and Price 3rd.

    Bill James’ CYA predictor FWIW: http://espn.go.com/mlb/features/cyyoung

    Vote -1 Vote +1

    • brownra2 says:

      The most interesting facet of this debate to me has been how often the “FanGraphs Stats” are being misinterpreted in the comments section. FIP is based on a pitcher’s performance in a given year. It is a record of what has happened, and makes no judgment regarding whether or not the K rates or HR rates that comprise its calculation are sustainable. xFIP, which normalizes HR/FB rate, is more geared toward predicting performance going forward.

      Dave mentions that the vote totals seem to indicate a partial rejection of DIPS theory. I’d argue that the vote totals indicate a partial misunderstanding of DIPS theory.

      Vote -1 Vote +1

    • John says:

      FIP is most certainly based upon performance in a given year. I would love an explanation how it is not. Certainly you can make an argument that it strips too much out, but it is based upon how a pitcher has done.

      As for the Bill James Cy Young predictor, that is worthless. It simply attempts to predict who will win based upon the criteria that sports writers used in the past. In the past (and perhaps present), the voters made poor decisions a great number of times and this simply attempts to predict future poor decisions.

      Vote -1 Vote +1

      • Luke in MN says:

        Here’s what’s meant: It’s a measure of past performance, but it’s a measure of endpoints no one really cares much about. If such a thing was possible, people would be perfectly happy with a pitcher who allowed few runs in a lot of innings, but allowed a ton of walks and struck out no one. To the extent FIP deals with endpoints people care about–i.e., runs allowed–it’s main usefulness is as a predictor.

        Vote -1 Vote +1

  6. John says:

    Andrew, if you included everything, why wouldn’t you pick CC? If you include run support, then basically you are saying Wins is what matters and Felix looks terrible compared to CC in that department.

    Vote -1 Vote +1

    • Andrew says:

      Felix has more innings and more K’s…that’s why I give him the edge.

      Vote -1 Vote +1

      • John says:

        But before you wanted to include everything, including run support? Then you give him the edge on IP and Ks, discounting a paltry win total?

        Vote -1 Vote +1

    • B N says:

      Yah, somebody better tell Felix to get out there and score some runs. That team isn’t going to win on pitching alone, you know.

      Vote -1 Vote +1

      • Andrew says:

        Oh get off it. I never said I wanted to include everything. Regardless, wins aren’t a completely worthless stat because players do have to go somewhat deep into games to get them consistently and someone like a CC or Halladay does it.

        FIP’s biggest value is that it allows you to predict regression. Knowing a player’s end of season numbers, would you rather have a guy on a one-year contract (or on your fantasy team) that you know will finish the season with an ERA 1.5 runs below his FIP or xFIP or a pitcher whose ERA is a run higher but his FIP is better than Pitcher A? Both guys pitch the same number of innings and start the same number of games

        Vote -1 Vote +1

  7. Aaron says:

    It seems like the thrust of your article is that this vote tells us more about the Fangraphs community than it does the CY race and I would wholeheartedly agree. The fact that CC received more votes than Cliff Lee makes me wonder just who the average Fangraphs reader is.

    Vote -1 Vote +1

    • John says:

      Don’t you think that baseball fans, even well educated baseball fans are still subject to great bias? Look at the Rockies fans that went nuts when someone dared say that Ubaldo was likely to regress after his great start. You think the Yankees fans that frequent this site are going to be able to look past those wins when they are casting their vote?

      My prediction/educated guess is that a lot of the guys who received fewer votes than Sabathia would outpoll him in a head to head vote. It would be interesting to see a poll with some individual comparisons like that. I am guessing more Felix Voters would go to a guy like Lee/Lester before voting for Sabathia.

      Vote -1 Vote +1

      • TFINY says:

        Actually I agree. As a Twins fan, I couldn’t look past Liriano’s FIP, even though I knew that Felix probably deserved the award more. It’s just that the rationalization that I used is more “stat worthy” with Liriano than a Yankee fan could use.

        Vote -1 Vote +1

    • Locke says:

      See my post below. Dave’s original post was extremely poorly written (if he was trying to get fair results) and he effectively turned the poll into CC or Felix, instead of: pick the CYA winner.

      Vote -1 Vote +1

    • Luke in MN says:

      I think just about anyone who voted for Liriano or Felix would also have voted Lee over CC if it came down to it. It’s just those three are going to split the stat-head vote.

      Vote -1 Vote +1

  8. AthleticsBraves says:

    Whomever voted Sabathia over Lee must not read FG much or is a Yankee fan. Lee is going to break the the all time K/BB record this year. I give a slight bonus for being on a winning team and since Felix is not far ahead of the pack in WAR, I picked Lee over Liriano.

    Vote -1 Vote +1

    • Jason B says:

      Might have helped to have a “top three” ballot instead of just picking one, like a traditional Cy ballot.

      Also, I’m sure some people picked who they thought would win, rather than who they thought *should* win.

      Finally, I’m sure there was a (small) contingent of “disagree with Dave’s (perceived) Mariner boosting/ #6org!” voters, some “I’m going to not agree with Dave just to be contrary; he’s just so SMUG!” voters, as well as a smattering of Yankees’ homers.

      Vote -1 Vote +1

  9. Jamie says:

    Baseball-Reference(or rwar) has Felix Hernandez as the #1 pitcher in the AL this year at 5.2. with Weaver second at 5.0 WAR and Liriano 3rd at 4.7 WAR.

    i don’t like fangraphs 100% DIPS WAR for pitchers. i do believe that there is some control involved in your BABIP and Strand rates.

    Vote -1 Vote +1

  10. AthleticsBraves says:

    Sorry to repeat your exact comment Aaron I didn’t see it before I started typing. I think FG is getting a lot of new readers which is great and they were probably the majority of the CC vote.

    Vote -1 Vote +1

  11. Locke says:

    I think the reason Lester got ~0 votes while CC garnered ~10% is is because anyone who can analyze the stats well enough to realize Lester has been definitively better than CC is also realizing that Lee, Liriano and Felix have been better than Lester. Hence, no votes for Lester.

    Clearly the CC votes are people who aren’t taking a deeper look, and that is meant with no snark or mean spirit, it just is what it is.

    I am, however, shocked than Felix dominated Liriano and Lee to such a degree. I think the three are much closer to an even split than any one of them being head an shoulders above the rest.

    To be honest, I’d be willing to bet $100 that Dave’s post skewed the results here. He by no means had to lead the crowd with his second paragraph if he was trying to get fair results.

    Poll taking 101 fail.

    Vote -1 Vote +1

    • John says:

      I think your first paragraphs is spot on. It was what I was trying to say above, but written much more clearly.

      Vote -1 Vote +1

    • Jason B says:

      I second that; first and last paragraphs are spot on. It should just be a poll, no introduction, with the commentary saved for after-the-fact.

      Vote -1 Vote +1

  12. Daniel Leroux says:

    One other thing to consider is that I’m sure a portion of the CC votes were done ironically- I know I was tempted to do it because of how dominant I expected Felix to be (though 60% is even bigger than that).

    Vote -1 Vote +1

  13. J. Nathaniel Sloan says:

    I voted for Sabathia because I thought the question was who -would- win the Cy Young, not who -should- win it; it’s the same reason I selected 4 years and $20MM for Jeter.

    Hernandez, Liriano, and Lee are all having better years than Sabathia, but I think Sabathia will win the award anyway because (a) he’s a Yankee and (b) he has a lot of wins.

    Vote -1 Vote +1

  14. AthleticsBraves says:

    That is probably why CC was 3rd some people thought they we voting on who they thought would win not who they thought should win.

    Vote -1 Vote +1

  15. Anon says:

    Someone said it earlier in this thread, but can we have 1st/2nd/3rd place votes like the real voting?

    This is a cool idea, I just think the results would be more interesting if we had 1/2/3 votes.

    Vote -1 Vote +1

  16. DT says:

    I voted felix because he is the most rounded candidate in terms of traditional and New age stats. Personally i would want Lee to win but his time in texas has made it hard for me. And as for Liriano feel like Fangraph’s WAR calculation and FIP/xFIP metrics are giving him a bit too much credit. As for CC, as a yankees fan..i’ve watched him pitch and he’s great. Although I feel Fan graphs might be under scoring his value due to his sudden GB tendencies and slight drop in K/9, but either way i would not give him the cy young even as a fan of his.

    Vote -1 Vote +1

  17. CircleChange11 says:

    Interesting, that at Fangraphs, the readers did not select the pitcher that leads the MAJORS in fWAR.

    This pleases me, because i strongly feel that [1] using any single metric is a mistake, [2] using any single version of WAR is a mistake.

    What this means (or should mean) is that discussions involving player values, trades, comparative performances, etc could/should be much more involved then just comparing WAR, FIP, etc.

    When all you have is a hammer, everything is a nail.

    Fill your tool box with various tools, and know how and when to use them.

    Vote -1 Vote +1

    • Jamie says:

      look at rwar. he leads the majors.

      Vote -1 Vote +1

    • Not David says:

      You find it interesting that, despite your many diatribes on the subject, you’re not the only enlightened one here?

      Vote -1 Vote +1

      • CircleChange11 says:

        Why the immediate assumption that I feel I’m the only enlightened one? If that were true, this place would be screwed. *grin*

        I talked about this subject for a couple of weeks, namely due to the same topic getting a lot of talk at Tango’s site … as it is an important subject since it was thought that FIP and fWAR would play such a role in who people were talking about as CY. I also have a high level of interest in pitching and pitching metrics.

        I find it interesting because …
        [1] The website is FG.
        [2] fWAR is based on FIP.
        [3] FIP is THE pitching metric at FG.
        [4] Many times, the thinking at FG is narrow (birds of a feather and all that).

        The majority of the readers went a different direction than what would be expected by me, given prevoius discussions on fWAR, FIP, etc.

        There are quite a few times, where anything above/below FIP is chalked off as luck, and it bothers me.

        Seriously, this would be like readers at a website called “Old School Stats” NOT voting the Wins and ERA leader as the CYA. That would be interesting too.

        I would say that getting rWAR and fWAR “averaged” is probably a pet subject of mine, because it makes too much obvoius sense (to me).

        Vote -1 Vote +1

  18. Matt says:

    It may not be an “advanced metric” but Innings is a seriously important stat to me (though I voted for Felix).

    Sabathia is valuable because he takes the mount every 5 days, and pitches 7 or 8 quality innings pretty much every time out. I don’t think he’s the most dominant but he is very consistent and reliable, which I think could be the most important feature of an ace pitcher. Indeed I would define an ace as: a starter who takes the mound every turn through the rotation, pitches deep into games, and provides his team an opportunity to win consistently.

    Vote -1 Vote +1

  19. Aaron says:

    I’m willing to conclude that the CC vote was at least in large part an artifact of a flawed setup to the original poll. That said, if the poll had come in Felix 60%, Lee and Liriano 15% each, everybody else 5%, does anyone else feel that the Felix dominance is in part because people are over-correcting for the low Win totals? As a statistically minded community are we confusing the fact that Felix is more undervalued by traditional stats than Lee and Liriano with the actual comparison of the three using the best methods available?

    Vote -1 Vote +1

  20. Omar says:

    I voted for Felix, but I’d put CC over Lester too. I mean, should we vote for Miguel Cabrera over Josh Hamilton because Hamilton’s season is largely fueled by a .400 BABIP and strong homepark advantage, whereas Cabrera has reasonable splits and a normal BABIP for his career? I mean if we don’t dock hitters for luck I don’t think we should dock pitchers either, with the exception of extreme circumstances. I find it hard to believe that Sabathia’s defense behind him is better than the defense that Lester has behind him.

    Vote -1 Vote +1

    • noseeum says:

      I would explicitly state that no one should be docked for luck in these awards. The awards are not “true talent” awards. They are “best results” awards.

      It doesn’t matter if Hamilton’s numbers are fueled by a high BABIP. It doesn’t matter if Felix’s numbers are fueled by a low BABIP. It just matters what happened. It will matter in my fantasy draft next year, perhaps, but not in my award votes for this year.

      That why I still prefer ERA to FIP when it comes to awards, or even better ERA+ in order to take park factors into account.

      Vote -1 Vote +1

  21. Rut says:

    I voted for CC because the other crowd sourcing articles where “what do you think player x’s next contract will be,” NOT “what do you think player x’s next contract SHOULD be.” So I assumed this wasn’t a normative question either. And I only realized this was not the case when Dave answered a question in the frat few comments.

    I wouldn’t be surprised if nearly all of the CC votes are the result of confusion like this, homerism, or more traditional fans voting in the poll.

    Vote -1 Vote +1

  22. Phantom Stranger says:

    A single selection ballot is absolutely the worst way to pick a candidate out of a field. To more accurately reflect the will of the electorate, a ballot where a number of candidates are rated in preference of order is the best method.

    Vote -1 Vote +1

  23. Andrew says:

    I voted for CC because I wasn’t paying close attention and I thought it was who I thought WOULD win.

    However, I’ve grown a little tired of the love for Liriano based on his rate stats. His FIP is inflated due to an extremely lucky HR% – at the end of the day that is figured way too high in there IMO. It’s ironic how much people complain about luck in stats like W-L, but when a pitcher has given up only 4 HR in 172 innings it’s not considered luck. I mean, come on, the guy had a career rate of 11% entering the season – how is his 3.3% rate not as lucky as a pitcher like Cahill with an extreme low BABIP #? To me, that’s flawed logic.

    Liriano has 19 quality starts this year, which isn’t even in the top 10 in the AL. You can make all the arguments you want against it being a silly stat, but effeciency should not be throw out in a Cy Young debate, and at the end of the day, Liriano has been nowhere near as efficient as Lee, Felix, or CC, so I eliminate him from contention. After that, Felix, to me, has been the best pitcher in the American League for the entire season – his first two starts in my excluded ;).

    Vote -1 Vote +1

    • noseeum says:

      I don’t think anyone really expects Liriano to keep the homers down this much forever, but the fact is, he has only given up 4 home runs. His FIP is not inflated. His FIP is what his FIP is. His FIP is better than his xFIP because of the low homer rate, which he’s earned, by NOT GIVING UP HOMERS.

      Vote -1 Vote +1

      • Andrew says:

        I understand what FIP is. What I’m saying is that something like HR% can be just as lucky as BABIP, which makes the argument that certain players are lucky because of low BABIP and pitchers with low FIP’s are definitely better a flawed one.

        Vote -1 Vote +1

      • noseeum says:

        Ah yeah, that’s definitely true.

        FIP and ERA for what happened, and both include luck in some form or other.

        Vote -1 Vote +1

  24. Temo says:

    I dislike that you are insinuating that rejecting WAR is rejecting DIPS. Those are two separate ideas.

    I like B-R’s pitcher WAR more than Fangraphs, and used that as a guide to make my selection.

    Vote -1 Vote +1

  25. Chair says:

    Or there just lots of Yankee fans………it’s called homerism, u know plenty of it.

    Vote -1 Vote +1

    • Captain says:

      Felix would get my vote but to say that CC was the only one that benefited from homerism in this poll would most likely not be correct. considering the lead in paragraphs before the poll and the anti-Yankee feelings of a good amount of commenters on here.

      Vote -1 Vote +1

  26. WilsonC says:

    I don’t think it’s fair to say that the vote reflects a rejection of DIPS theory. The way I look at it, there are really two very different things that come into play here when looking at DIPS theory:
    – it serves as a predictive tool a tool by focusing on the more persistent true-talent skills that are more directly in the pitcher’s control.
    – it attempts to filter team defense out of the equation to put pitchers on a more even scale in terms of past performance.

    While imperfect, and while there’s certainly differences of opinion in terms of how close the current versions of DIPS stats are to a true estimation of a pitcher’s talent level, I think most statistically minded people would agree that DIPS stats are a tremendously valuable tool when trying to get a truer picture of what a pitcher is likely to do moving forward.

    For the second part of it, there’s a lot more subjectivity involved, because we’re also looking at a difference in philosophy, rather than a difference in the trust people put in the accuracy of the stats. FIP takes the variance in team defense out of the equation, but it also assumes a more normal distribution for how balls are hit. It’s a noble goal to filter the defense out of a pitcher’s performance, but there’s a lot more to BABIP variance than how the defense played. In a lot of cases, the variance is more a result in whether the pitcher/hitter duel results in a batted ball that’s easy to convert into an out or difficult. Whether a repeatable skill or not, the fielding difficulty of a given ball in play is as much a part of a pitcher’s past performance as the number of HR he allows.

    In this particular case, you had three pitchers who were basically a rounding error away from 6.0 WAR (I think 5.9, 6.0, and 6.1 when I voted for Felix). Unless you believe that FIP is perfect, that’s way too close to reveal a difference between the three using the Fangraphs WAR alone. That means for someone who adheres strictly to the philosophy of basing a Cy Young vote off of DIPS stats, there’s three very reasonable candidates, and reasonable points could be made favoring any of the three. However, Felix had a substantial enough lead in a results-based comparison (almost a full point of WAR on B-R) that he would certainly have overwhelming support among people who are not going entirely off of DIPS.

    I have no clue what the actual breakdown is for Fangraphs philosophies, but ultimately we can look at it more as a spectrum as opposed to a stats/traditionalist divide.

    On one end, you have the approach: sort by WINS, look at Winning% and innings and reputation, call it a day.
    On the other end, you have the approach: sort by WAR, look at FIP and xFIP, call it a day.

    Sabathia’s the obvious winner by the first method, and the second method is close between Liriano, Lee, and Felix. On both sides, though, as you move toward the center of the spectrum, Felix quickly comes out as the easy choice. When you move past wins, he leads in all the other traditional categories, and once you start looking at any stats other than those included in DIPS formulas, Lee and Liriano are quickly surpassed by Felix. Throughout the middle of the spectrum, I would expect nearly unanimous support for Felix.

    As an example, suppose you have a split of 30% strict DIPS supporters, 30% who go strictly off results, 30% who use DIPS as a tool to adjust their interpretation of the results, and 10% miscellaneous (misinterpreted the poll, old school, blatant bias, etc.)

    Of the first 30%, you’d expect somewhere around 10% to go to each of Lee, Felix, and Liriano. Of the second 30%, you’d expect and overwhelming majority to vote for Felix, perhaps with a small handful leaning toward Sabathia or Price and very few votes for Lee or Liriano. The third 30% would as well have an overwhelming majority voting for Felix, since any hybrid of the first two views would likely push Felix to the forefront. The last 10% would strongly favor Sabathia. This type of breakdown would produce results similar to what we see here, despite a majority of voters who consider DIPS a valuable tool.

    Vote -1 Vote +1

    • Joe says:

      Wow, I think you perfectly captured what I had been trying to say. Well said.

      Vote -1 Vote +1

    • Jamie says:

      absolutely perfectly stated. for something like a CY Young award i’d go more results based than predictive based. rwar would have more stock than fwar for me.

      Vote -1 Vote +1

    • frank says:

      The main problem I have with FIP (when looking at past results), is the assumption more or less that all pitchers are the same when it comes to balls in play and they have ZERO control over it. They might not have a lot of control over it but even if it’s just say 20%, that’s a significant amount of potential talent measurement that is being excluded especially when folks are parsing people who have similar #’s.

      I really think you need to look at a blend of result based and modeled ‘stats’. When looking at the past, context does matter. That K with a guy on 3rd and 1 out is a lot more valuable than a K with 2out and noone on base. That HR given up when you are up 8-1 and trying to go deep into the game, is not the same as a HR in a 2-2 tie. Granted these are probably very small impacts (and I realize a lot of people disagree with the concept of a pitcher pitching to the situation), but in my view when evaluating the past they are important. Especially when you are trying to pick between pitchers that are closely matched.

      In the case of FIP there can be a lot of luck involved around HR’s (xFIP corrects for this a bit). Weather (wind, temp) can have an impact on a couple of HR’s a year, lineups (a key batter being out), the time of game (day, night twilight), and even something very rare like an inside the park HR or a HR saving catches can all start to impact FIP.

      Vote -1 Vote +1

      • DT says:

        yea i think that’s one of my problems with FIP..i mean i get that it measures pitcher in basically a neutral light…but is a grounder off a guy throwing a 88 mph sinker really same as a grounder off a guy who throws a 95 mph sinker? sure the pitcher can;t control where the ball is hit but weak grounder is more likely to find a glove than a hard hit one.

        Vote -1 Vote +1

    • batpig says:

      this is a wonderfully written post, very well said!

      Vote -1 Vote +1

  27. dc says:

    Would be interesting to see the bias in the results. I suspect, given the founders and writers, a lot of the voters are Mariners fans and are just voting their fanship.

    Vote -1 Vote +1

    • o971 says:

      Because we “Mariners fans” clearly let Dave off the hook for putting the mariners #6 in the Organizational rankings.

      Vote -1 Vote +1

  28. Rocco Rock and Johnny Grunge says:

    The Yankees have scored a total of 173 runs in Sabathia’s 30 starts, but the Mariners have scored just 99 runs in Felix Hernandez’s 31 starts.

    How can anybody acknowledge that fact and still put any stock in W-L record? It’s a completely useless number.

    Vote -1 Vote +1

    • noseeum says:

      I think there’s a legitimate position regarding wins. I disagree with it, but a friend of mine, who’s pretty SABR inclined, still thinks wins are very important for the Cy. His view was that for all intents and purposes, the Cy is the pitching MVP. I know pitchers are eligible for the MVP, but it’s very rare for them to win it.

      So if you look at the Cy that way, then you might lean towards removing pitchers from non contending teams unless their season is historic. This is generally how MVP works. You pretty much have to be on a contender, or you have to blow everyone out of the water.

      I think this is totally not the way to go, but at least it’s a somewhat legitimate criteria someone could have. And if they had it, they could drop Felix off the list immediately. I guess Lee too because he was with Seattle and then has pitched much worse for the Rangers.

      Vote -1 Vote +1

      • Yo izzgs says:

        Yeah but even in this case I would think you could pick a better stat, say Quality Starts, which is similar to but (mostly) better than Wins.

        Vote -1 Vote +1

  29. Anon21 says:

    I’m just curious why Dave thinks we need a new word for “poll.”

    Vote -1 Vote +1

  30. frank says:

    Dave, I don’t understand your analysis (and surprise?) of how pitchers down the list finished based on how the poll was setup. For example since the audience was given only one choice, those looking at K/9, BB/9, FIP etc had better options than Lester to choose from.

    I could understand your surprise if folks were asked to rank say the top 5 or top 10, but I’m surprised Lester got any votes. 10th vs 8th (or whatever place) in a specific statistic or set of statistics doesn’t matter when you are asking folks to only choose one guy.

    Vote -1 Vote +1

  31. grandbranyan says:

    In cases such as 2009 where Greinke and Lincecum were so far superior to their contemporaries W/L record & team success should not enter into the equation.

    In cases such as AL 2010 I think those factors should play more of a roll as there is much more gray area and therefore additional factors almost necessarily need to be considered.

    I voted for CC in the original poll mostly just to be contrarian and because I don’t think the writers will actually give the award to the KING, but all things considered I think Price probably deserves it the most and not just because I picked him in the 15th round for my fantasy team.

    Vote -1 Vote +1

  32. Lee Panas says:

    I don’t believe you can just look at FIP based stats. I also don’t think you can just look at runs allowed. I think the Cy Young should go to a pitcher who has the best combination of everything – results, peripherals and innings. I think Hernandez is the clear winner with innings pitched being the final thing that puts him over the top versus Liriano and Lee.

    Vote -1 Vote +1

  33. CCW says:

    This was a ridiculously executed poll. Fangraphs, please re-ask the question in a way that won’t skew the results. Thx

    Vote -1 Vote +1

  34. Eric says:

    I thought this was interesting:

    “[…]the difference between an out and a hit could have nothing at all to do with the “actual performance” of the pitcher. Imagine universe A and universe B. Universe A and universe B are exactly alike in every particular, except that in Universe B a butterfly flapping its wings pushes a baseball just out of reach of a right fielder and three runs score. […]

    One could just as easily claim that “of course you HAVE to adjust for luck, since without it, you are NOT looking at how a pitcher actually pitches – you are looking at how he pitches plus the effect of his luck. If you didn’t do that, then lots of [poor] pitchers would be winning the CYA if they happened to pitch BABIP of say, .100.”

    The real issue is that, in any given sample, it’s not at all clear to what extent things like BABIP and HR/FB% are a result of luck and to what extent they are a result of “actual performance.” […] ”

    -originally posted by Ron A in the comments thread of THE BOOK

    http://www.insidethebook.com/ee/index.php/site/member_posts/?name=Ron%20A.

    Vote -1 Vote +1

  35. Chris88 says:

    First – why should Dave have to have a neutral heading? What he said is true and people should he able to think for themselves.

    Second – WHIP, are you serious? This isn’t fantasy baseball – use a real statistic please. You are better off with something awful like ERA.

    Vote -1 Vote +1

  36. Noseeum says:

    Chris88, when it comes to Cy Young, there’s nothing awful about ERA. xFIP is worse. Sometimes it seems in the quest to determine true talent for predictive purposes people begin to think these x stats are reality. ERA is reality. How many earned runs per 9 innings were given up while pitcher X was pitching?

    That’s a heck of a lot more important than what we expect his FIP to be in the future. I’m not sure you hold this position but it’s been stated by some. I don’t think xFIP has any place in a Cy Young discussion unless it’s a discussion about who will win next year’s Cy Young.

    Vote -1 Vote +1

    • John says:

      Actually ERA is not reality. It is as much a construct as FIP or xFIP. In fact, one could argue that ERA is less a reflection of reality because it attempts to incorporate defense, but does so through excluding “unearned runs.”

      Just because a stat is traditional and been around for a while, doesn’t make it more “real.”

      Vote -1 Vote +1

    • Eric says:

      It’s about the Money had an interesting post about FIP and ERA in relation to Joba this weekend: “Defense-Independent Pitching Statistics (DIPS) have come under scrutiny lately because people believe ERA tells you what happened and DIPS tell you what should have happened. It’s a common misunderstanding for those who are not familiar with newer pitching statistics such as FIP, xFIP, and SIERA, but these stats do tell you what happened.[…]

      Because of the abstract nature of FIP (ERA is something you can easily see, whereas FIP tries to parcel out certain things which also results in the multitude of different DIPS), people think it means what the pitcher should have done or what the player might do. But it isn’t that way. It is the pitcher’s performance and only the pitcher’s performance, but it also has some predictive power because stats such as strikeouts and walks don’t fluctuate so much year-to-year (because, well, a pitcher’s ability does not vary that wildly from year-to-year). Defense and randomness can vary, which causes ERA to be unpredictive.[…]”

      http://itsaboutthemoney.net/archives/2010/09/12/joba-and-fip/

      Vote -1 Vote +1

  37. Jonathan says:

    If people voted for Liriano simply based on his FIP and xFIP then they are missing the point. All stats need to be included. He has more than 50 less innings pitched than Felix Hernandez, his 3.3 HR/FB rate is almost as hard to sustain as Cahill’s .222 BABIP, and extremely lowers his FIP and xFIP. Not to mention that Liriano has had great defenders behind him but somehow still has a .341 BABIP. Here are the defenders for Liriano (UZR/UZR150):

    Orlando Hudson: 9.9/14.9 (2B)
    J.J. Hardy: 8.4/18.6 (SS)
    Danny Valencia: 5.8/15.1 (3B)
    Alexi Casilla: 1.2/12.2 (2B) 3.4/31.8 (SS)
    Nick Punto: 5.1/33.2 (SS) 6.2/20.0 (3B) 1.9/38.7 (2B)
    Denard Span: 5.1/6.3 (CF)
    Justin Morneau: 9.8/18.0 (1B)
    Michael Cuddyer: -8.2/-18.8 (OF) -4.4/-11.3 (1B)
    Delmon Young: -9.9/-12.8 (LF)

    Granted, Cuddyer and Young have been awful, but they play the two least important spots on the field for Liriano and his almost 54% GB ratio.

    Do I think Liriano has been great in 2010? Of course, but he is not the best pitcher in AL because FIP and xFIP say he is, and I might not take him in my top 3 if I had a vote.

    Vote -1 Vote +1

  38. jo says:

    The cy young award is for what actually happended, not what may have happened, or what predictive stats say might happen.

    I mean 27ip is 3cg more than what lester has pitched.

    Vote -1 Vote +1

  39. Seideberg says:

    Joe Morgan quote today:

    “I think it’s a joke to have that kind of debate. What Sabathia has done is be the best pitcher in the AL from opening day to this point. I don’t buy into the point that if Felix is pitching for someone else he’d have more wins. They said that about Cliff Lee when he left Seattle, but he’s lost more than he’s won since he left Seattle. The name of the game is to win and he’s won. And if you’re looking at a second guy, it has to be David Price. It’s amazing to me that we have let computers define him rather than performance. His job is to win the game, not just pitch 5-6 innings. I don’t think there should be a debate between Felix and Sabathia.”

    How can someone who was so good at baseball have such little understanding of the sport?

    Vote -1 Vote +1

    • Seideberg says:

      Oh, and for the record, I am a Yankee fan who thinks Sabathia has had a fantastic season, but just not as good as Felix’s.

      Vote -1 Vote +1

    • Socrates says:

      To start, I do think that Joe Morgan is a joke when it comes to baseball analysis. That said, I do think that there is one thing that is important to evaluate that I do not believe that any stat currently takes into account. Innings pitched is important.

      You cannot tell me that Buchholz season (152 innings pitched) is even remotely close to being as good as CC’s (217 innings). Sure the innings that he has pitched have been slightly better (3.69 xFIP to 3.81 x FIP). As a matter of fact, it is not just raw innings but it is a huge advantage to a team for a pitcher to pitch later into the game. I pitcher that routinely pitches an inning less will often require an additional reliever. That 3rd or 4th reliever is not going to be very good (in most circumstances). More starts and longer outings are absolutely valuable (although I think that Cliff Lee is averaging more innings per start).

      While I guess innings are included in WAR, I think that the marginal value of pitching the 7th inning would be more than pitching the 1st inning so that should be taken into account some. In addition, I think that the situation of the inning should also have value. Was it a tie game, blowout, etc. It is my understanding that WAR still does not include that.

      I still think that King Felix and Liriano have been better than CC and more deserving of the Cy Young this year.

      Vote -1 Vote +1

      • Eric says:

        Kairos Socrates. How’s the good life?

        I actually was just reading about this at The Book:

        “Suppose you have a great pitcher, someone like CC, who gives up 3.78 runs per game, and starts 35 games, and pitches 6.8 innings per start (that’s 100 runs allowed in 238 IP).

        Suppose he goes away, and you can’t spend any more money. So, you make due with what you have. His 35 starts are cobbled together at the replacement level of 6.00 runs per game, with only 4.8 innings per start. That’s 168 innings and 112 runs allowed.

        You still need 70 innings to find somewhere in relief. Because of roster limits, and you don’t want to bring up/down too many pitchers, you start to tax your current relief corps.

        Your current relief corps, with the great starter in the rotation, threw 490 innings at 4.35 runs per game (237 runs allowed). However, without the great starter, they pitch more and are less effective. So, they need to pitch 70 more innings to make up for the loss of the starter, and instead of giving up 4.35 runs per game, they give up 4.55 runs per game. So, in the 560 innings at 4.55 runs per game, they give up 283 runs, instead of 237.

        So, let’s tally it. The 168 starter innings by the ragtag team gives up 112 runs, and the 70 extra relief innings, including the extra runs allowed by the current relievers, yields 46 more runs in relief (283-237). All in all, that’s 238 innings to makup, and a total runs of 158 runs allowed.

        That becomes the replacement level. And 158 runs in 238 innings is 5.97 runs per game.

        Now, waitaminute. The starter replacement level is 6.00 runs per game. How did we end up at the same place? […]”

        http://www.insidethebook.com/ee/index.php/site/article/how_much_value_is_going_deep_to_your_team/

        So, it seems like it does matter but since only pitching 1 inning is so much easier than pitching multiple it doesn’t make as big a change as you would think, even with a pitcher only expected to give up 3.78 runs a game.

        Vote -1 Vote +1

  40. Chris88 says:

    I wouldn’t vote for a performance-based award using a stat that includes other guys performances, but that is just me. No comment on WHIP? It still sucks, as does ERA, for any sort of individual performance evaluation.

    Vote -1 Vote +1

    • Eric says:

      Yeah, I tend to agree, this discussion reminds me of Dave’s piece from years ago (which I’m sure most of you remember, have read a few times…):
      “[W]hat are the possible events in an at-bat that can occur?

      A pitch can be thrown for a ball.
      A pitch can be thrown for a strike.
      A pitch can be swung at and missed.
      The ball can be hit on the ground.
      The ball can be hit on a line.
      The ball can be hit in the air.[…]

      Which of these six outcomes are positive for the pitcher? Called strike, swinging strike, and groundball.

      Which of these six outcomes are positive for the hitter? Called ball, line drive, and flyball.[…]

      FIP and xFIP have flaws, especially when it comes to evaluating relief pitchers, but if you’re insistent on using one number to sum up a pitcher’s contribution to run prevention, those would be your best bet.

      In this age of wonderful information, there’s just no reason to use ERA and WHIP for serious analysis […]”

      http://www.ussmariner.com/2006/08/29/evaluating-pitcher-talent/

      Vote -1 Vote +1

  41. yungmuneyholla wat it dew says:

    lol joe morgan

    Vote -1 Vote +1

  42. Chris88 says:

    xFIP uses regressed data, so that’s expected performance and not real performance either. I’m not interested in that. Felix’s era is indicitive of his ballpark and defense, is he the best run preventor in the AL? Probably not, but Felix gets the nod from me because of the inning totals. I value that durability above pure per-innings performance. Felix just so happens to have the most favorable combination of quality AND quantity. If it was 2008 Sabathia in as many innings he’d get my nod but it is not.

    The WHIP comment was Joe I think so I’d rather hear his thoughts on why Felix would make a good reliever on his fantasy team, or whatever idiocy advocates WHIP as even a stat worth mentioning.

    Vote -1 Vote +1

    • Eric says:

      I know, I still think this is Lee’s year but there’s a great case for Felix: 225.2 IP, 53.2 % GB%, 8.53 K/9, 2.51 BB/9, 0.60 HR/9…What a season.

      Vote -1 Vote +1

  43. Eric says:

    One thing that will be interesting is going forward.

    Before their most recent starts I think you could say that each pitcher went through strangely opposing and atypical stretches. Lee famously gave up 44 hits and 27 ER’s while only lasting 29.1 innings in 5 starts while Felix pitched 5 starts in which giving up 21 hits and 1 ER while pitching 37 innings (and his start before that? An 8 inning 5 hit shutout). These results count and should have an effect on how we perceive each pitcher. At the same time we shouldn’t forget the 19 starts Lee pitched before these or the 25 that Hernandez pitched. A lot can happen in a small sample and while memories and the narratives at sites like ESPN too often hinge on yesterday’s performance I think we at Fangraph’s shouldn’t yield to the temptation and should take a critical look at the whole seasons. I think we should marvel that even after missing the first month and having a surprisingly ineffective stretch of 5 starts that Lee is still in the conversation and better yet is perhaps the frontrunner for the Fangraph’s Cy Young.

    Lee leads fWAR right now at 6.3 (followed by 6.1 for Liriano, 5.9 for Hernandez, and 5.5 for Lester) and despite all his missed time has amassed 192.1 innings in only 25 starts. With Felix expected to miss starts at the end of the season (http://hardballtalk.nbcsports.com/felix-hernandez-to-get-extra-rest-cy-young-chances-take-a-hit.php) and both Lee and Hernandez more likely to pitch as they have in the larger sample size of this entire season (for example, zips expects a 2.93 FIP for Lee going forward and a 3.09 FIP for Felix) I think what many of us saw in first person earlier this year, and what many of us still see today, will prove to be true at the end of this year also: that while Hernandez is probably the best young pitcher in the game Lee has most likely staked his claim as this year’s best American League pitcher, our Fangraph’s Cy Young.

    Vote -1 Vote +1

  44. Chris88 says:

    I can agree with this. Lee has been fantastic, as have Liriano and Lester.

    For the sake of the ACTUAL Cy, I hope Felix can win out. I don’t think Lee gets the votes after his recent stretch and Felix is the only guy getting talked about enough in the M.S. to have a shot to keep the award out of CC’s hands. Perhaps Price, but I’d be just as disappointed with that selection as with a Sabathia vote.

    Vote -1 Vote +1

  45. Dave says:

    Please tell me somebody else say this today:

    Jeremy L (Philly)

    Hey Joe, I’ve been a big fan for a while. Could you please weigh in on the CC vs. King Felix for cy young debate?

    Joe Morgan
    (11:09 AM)

    I think it’s a joke to have that kind of debate. What Sabathia has done is be the best pitcher in the AL from opening day to this point. I don’t buy into the point that if Felix is pitching for someone else he’d have more wins. They said that about Cliff Lee when he left Seattle, but he’s lost more than he’s won since he left Seattle. The name of the game is to win and he’s won. And if you’re looking at a second guy, it has to be David Price. It’s amazing to me that we have let computers define him rather than performance. His job is to win the game, not just pitch 5-6 innings. I don’t think there should be a debate between Felix and Sabathia.

    Joe Morgan everybody!

    Vote -1 Vote +1

  46. Ty says:

    You’ve gotta be stupid to not even consider felix Hernandez. First off, Felix has by far the best whip in major league baseball. He also leads the majors in strikeouts. Oh and did I mention he is second in the majors in era and first in the a.l. The only thing that is hurting him is his lack of wins due to terrible run support. Everyone knows that especially people in the pacific NW. I’m a mariner fan but I’m not gonna lie, we suck. Hitting is one of the worst I have ever seen. But Felix has been consistently dominant the past couple years and this is an award he truly deserves. I agree with a 60% vote for king Felix and you probably should too.

    Vote -1 Vote +1