2009 NL All-Star Ballot

After covering the AL this afternoon, let’s take a look at my NL All-Star ballot.

Catcher: Brian McCann (ATL), Jesus Flores (WAS), Yadier Molina (STL)

McCann and Flores are both performing well in limited action, while Molina is the best defensive catcher in baseball and a decent hitter to boot. Lots of competition here, as you could make a case for John Baker, Carlos Ruiz, or Russell Martin, but I’m happy with the three above.

First Base: Albert Pujols (STL), Adrian Gonzalez (SD)

I went with 12 pitchers to satisfy the one player per team approach, so I had to lose a position player somewhere, and first base is where I decided to skip out on the third guy. Joey Votto‘s DL stint hurt him, and there’s not much separating the group that contains Prince Fielder, Ryan Howard, and Nick Johnson, so I went with none of the above.

Second Base: Chase Utley (PHI), Orlando Hudson (LA), Brandon Phillips (CIN)

The first two here are easy as well, while Phillips edges out guys like Freddy Sanchez due to superior defense.

Shortstop: Hanley Ramirez (FLA), Yunel Escobar (ATL), Ryan Theriot (CHC)

The Braves should be thanking their lucky stars that the Padres didn’t take Escobar and run over the winter.

Third Base: Ryan Zimmerman (WAS), David Wright (NY), Chipper Jones (ATL)

Three pretty easy picks here. Sorry, Casey Blake fans.

Outfield: Raul Ibanez (PHI), Mike Cameron (MIL), Justin Upton (ARI), Matt Kemp (LA), Carlos Beltran (NY), Ryan Braun (MIL)

Starting Pitchers: Tim Lincecum (SF), Johan Santana (NY), Wandy Rodriguez (HOU), Josh Johnson (FLA), Chad Billingsley (LA), Dan Haren (ARI), Zach Duke (PIT), Ubaldo Jimenez (COL)

Relief Pitchers: Jonathan Broxton (LA), Heath Bell (SD), Ryan Madson (PHI), Rafael Soriano (ATL),

The Just Missed guy: Scott Hairston. Finally living up to some of the hype he got as a minor league hitter, and handling himself just fine in the outfield to boot. But the NL has a lot of really good outfielders, and Hairston just can’t quite live up to the bar set by the rest of those guys.

Final team tally:

Four: Atlanta, Los Angeles
Three: New York, Philadelphia
Two: San Diego, Washington, Arizona, Florida, Milwaukee, St. Louis
One: Chicago, Houston, Cincinnati, Colorado, Pittsburgh, San Francisco




Print This Post



Dave is a co-founder of USSMariner.com and contributes to the Wall Street Journal.


77 Responses to “2009 NL All-Star Ballot”

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

    Good list overall, the only place I would differ is I think Javier Vasquez should be in. I know the Rockies need someone in there though, so I say substitute Hawpe for Kemp, then Vasquez for Jimenez.

    Vote -1 Vote +1

    • BIP says:

      Hawpe’s defense is putrid enough that he’s within spitting distance of replacement level despite the career year he’s having offensively.

      Vote -1 Vote +1

      • Kevin S. says:

        Within spitting distance? He’s 0.8 WAR through two months. If league-average is 2 WAR for the year, league-average through two months would be 0.67 WAR. Hawpe’s been an above-average player, hardly “spitting distance” from replacement level.

        Vote -1 Vote +1

      • Trevor says:

        I did forget to look at their defense though, and now I see that Hawpe is dragged down considerably. Ok, substitute Iannetta for Molina, then Vasquez for Jimenez. Iannetta and Molina both sit at 0.8 WAR.

        Vote -1 Vote +1

  2. B says:

    “while Molina is the best defensive catcher in baseball”

    Curious as to how you evaluate defense at C. Is there a metric you use as your primary indicator, or based mostly on scouting evaluations, or some combination or metrics and what baseball people are saying?

    As a Giants fan, I’d send Cain over Lincecum. Cain’s been having a better season so far – I know Lincecum’s FIP is awesome while Cain’s is pretty average, but when it comes to stopping runners from scoring, Cain has done a better job of it so far this year {see: ERA (same defense behind them)}, even if it’s mostly because of an unsustainable strand rate. That said…Lincecum did win the Cy Young last year, and it’s not hard picturing him with the better stats by the end of the year (and he has star appeal!), especially given their respective FIP’s, so obviously it’s not a crime against humanity or anything to go with him over Cain…

    Vote -1 Vote +1

    • Teej says:

      You just did a really good job of explaining why Lincecum has been better than Cain. :)

      Vote -1 Vote +1

    • tom says:

      Lincecums been better than every pitcher not named Greinke this year, sooooooooo…..no.

      Vote -1 Vote +1

      • pm says:

        Johan Santana

        Vote -1 Vote +1

      • Kevin S. says:

        *this* year, pm. Tiny Tim’s second in WAR among starters.

        Vote -1 Vote +1

      • B says:

        These are all based on FIP, though, which doesn’t actually include anything about preventing runners from scoring. Obviously I understand the point of FIP, but if you’re basing this completely on what a player has done in the half-season before the All-star game, it doesn’t seem FIP is a great stat for this. Even if Cain is getting lucky by getting hitters out with runners on – he’s still getting those hitters out with runners on, preventing them from scoring. He may not continue to do so, but so far, he has.

        Vote -1 Vote +1

      • tom says:

        to pm:
        Prove it. What website do you think your on? I don’t know how anybody can even try to say that Santana has been better than Lincecum this year or last year. FIP, xFIP, WAR all agree with me.

        Vote -1 Vote +1

      • don says:

        Johan and Lincecum are having just about the same season as far as “true outcomes” go, except Lincecum’s HR/FB rate is ridiculously low. Sure, Johan’s only doing ‘better’ insofar as stranded runners are concerned, but they’re really pretty close.

        Vote -1 Vote +1

  3. Trevor says:

    I think you argued against yourself there, by pointing out Cain’s unsustainable strand rate. Lincecum is better in just about every other category than Cain.

    Vote -1 Vote +1

    • B says:

      So just because he can’t sustain it doesn’t mean we shouldn’t give him credit for accomplishing it up to this point?

      Vote -1 Vote +1

      • Daniel says:

        Yea, I don’t understand penalizing pitchers for unsustainable peripherals but not hitters. Jesus Flores has a .400 BABIP (in all of 90 ABs). Wright has like a .460 or something

        Vote -1 Vote +1

      • B says:

        I think a lot of people don’t necessarily fully understand stats and realize why something like FIP works so well. FIP is a measure of the most influential factors that determine the end result – not the end result itself. Over a large enough sample, the factors out of a pitchers control will even out (there are still a few factors not accounted for a pitcher controls but they’re relatively minor), causing FIP to pretty accurately describe the end result (which is why it’s such a good predictor). Where I think most people get confused is in smaller samples (like half a season), the end result can, and often does, drastically differ from the FIP measurement and they forget the end result is what actually matters…

        Vote -1 Vote +1

      • Trevor says:

        I’m not saying you can’t give him credit for it, but Lincecum is just so much better in every other meaningful category; K rate, BB rate, HR/9, WHIP, not to mention the large BABIP differences (Cain’s is low, meaning he’s been a bit lucky, while Lincecum’s is pretty high). You’re using one stat, Cain’s strand rate, to say he’s having a better season, but he is actually not.

        Vote -1 Vote +1

      • B says:

        Actually I’m using ERA, which I think is fairly reasonable given they have the same D behind them. The strand rate is the primary explanation for how Cain’s ERA has been so good. You should read my previous comment about stats like FIP vs. the actual end result.

        My real argument is this: every fan has a different opinion on what makes an all-star. When looking at production so far this year, Cain has been better. Every other factor fans take into account favors Lincecum. So the more a fan favors production this year, the more likely they should be to favor Cain instead of Lincecum.

        Vote -1 Vote +1

      • Trevor says:

        I’m not sure if you understand FIP. FIP, as I understand it, is a measurement of what their ERA SHOULD be given their peripherals and rates. It’s not saying what it WILL be. The reason it differs is mainly either luck or quality of defense. Since they have the same defense behind them, the reason for the differences in FIP in this case is luck. Their rates agree with that assumption, given Cain’s low BABIP and high strand rate. Lincecum’s BABIP is pretty high, which means he’s been unlucky and his ERA should be lower. Cain’s has been low, meaning his ERA should be higher than what it is.

        That’s what FIP is saying, that his peripherals and rates point to a much higher ERA than what he has. But Cain has been very lucky this year, leading to his low ERA. His strand rate is 13% higher than his career average. In time his ERA should regress to normal.

        BTW, I’m not saying that your opinion is wrong, just that I do not agree with it.

        Vote -1 Vote +1

      • B says:

        I have a very good understanding of what FIP is, I assure you. Those rates you mentioned are the influential factors that cause the end result I discussed (K%, BB%, HR%). FIP is what the ERA should be once the other peripheral factors like strand rate that regress to some mean actually regress to that mean. So far this year, the sample isn’t big enough for that to have happened yet.

        Essentially, you’re arguing that we should credit a player for what their end result should be, rather than what it is. In this case, the ERA/ERA+ is the end result, and that’s where Cain has been better.

        Vote -1 Vote +1

      • Trevor says:

        Except my point is that ERA/ERA+ is not the only way to value a player’s performance, nor is it even the best. Almost every other stat points to Lincecum being better, and you have to use the whole picture to evaluate performance.

        Vote -1 Vote +1

  4. B says:

    Still hoping to get a response to my question about catcher defense!

    Vote -1 Vote +1

  5. Evan says:

    I’d love to know your all-star selection philosophy, Dave.

    Are you going for the best players, the guys having the best half-season, or the biggest stars?

    I tend to select the biggest stars – it is the All-”Star” game, after all.

    Vote -1 Vote +1

  6. JG says:

    No Franky Rodriguez? Madson instead? Are you going out of your way to have a non-closer relief pitcher, I don’t get it?

    Vote -1 Vote +1

  7. ThatsRobbery says:

    Flores is out 3 months apparently. Who replaces him now? Id say Martin at catcher or Votto for another 1st basemen

    Vote -1 Vote +1

  8. Eric says:

    Miguel Tejada should be on that list.

    Vote -1 Vote +1

    • Cooper Nielson says:

      Yeah! Explanation?

      Tejada leads all NL shortstops in batting average, hits, doubles and RBI. He’s second to Hanley in HR, OPS, SLG and — believe it or not — OBP, despite a comically low 6 BB. (He has also only struck out 13 times.)

      Seems to me he is clearly the second best hitter among NL shortstops (and much closer to first than third). Is his defense really that bad?

      Vote -1 Vote +1

  9. Preston says:

    If I’m actually looking at building the best team for the All-Star Game, I’d definitely put another first baseman on there – probably Votto, Howard, or Fielder, mainly to function as a pinch hitter – rather than either a third 2B or SS. Otherwise, I’m not sure you’ll have a big time power hitter off the bench to use as a pinch hitter in the late game – you have to figure Adrian Gonzalez will sub in at first in the middle of the game, and the same is probably true of Braun. Other than that, you’re maybe looking at the last of your third basemen, but as a manager I’d feel a lot better having a Prince Fielder ready to pinch hit than having Brandon Philips as a defensive replacement for Utley and Hudson, both strong defenders in their own right who are probably better hitters than Philips.

    Vote -1 Vote +1

    • JH says:

      I’m assuming Dave only used 2 1B because he assumes Pujols will play the whole game since it’s in STL. That would leave Gonzalez as your LH power bat off the bench in the Fielder role. I do agree with Preston though, that I’d rather have Fielder than Fontenot.

      Vote -1 Vote +1

  10. Andrew says:

    B you are confused. When you are referencing FIP in your posts, you are really talking about xFIP. I know you do not fully understand FIP because:

    1) You confused FIP with xFIP and;
    2) You continue to use “same defense” in your argument but FIP is not only equivalent to ERA but tells you how well a pitcher pitched regardless of his fielders.

    To argue Cain has been better, or even CLOSE to as good as Lincecum so far is crazy.

    Vote -1 Vote +1

    • B says:

      Wow, Andrew…I don’t even know what to say. Not once in any of my posts did I even say anything specific enough about FIP for you to determine I was talking about the wrong stat. Anyways, I was talking about FIP. That answers that question.

      Secondly, your second point shows you clearly lack any understanding of statistics. I guess the pretty name they gave FIP means it 100% accurately describes the pitching component of ERA, right? Now, a person who knows what they’re talking about can make a decent argument that even though the defense behind them is the same, that doesn’t necessarily mean the defense factor of ERA is cancelled out, and frankly, they would be right. However, I’m comfortable enough with the fact that the same players are behind them to ignore this, as the effect is probably pretty minimal and I don’t think we have a good way to measure the difference at this point in time, anyways.

      What it comes down to is the pitcher who lets in less runs is doing a better job. Obviously ERA/ERA+ has its weaknesses as Trevor pointed out earlier, the biggest being how it divides credit (all credit going to the pitcher, none to defense/ballpark, etc.). In this situation (comparing 2 pitchers same team), with this small of a sample, though, it’s a better judge of how they’ve done because it measures how many runs they’ve allowed – with most of the significant weaknesses mostly cancelled out.

      Again, as I discussed earlier, FIP is a measure of the most significant factors that lead to the result, and over a large enough sample this is better because it accurately describes the result while taking the defense factor out (and ballpark to some degree), but we aren’t dealing with a large enough sample.

      Cain ERA+ 192
      Lincecum ERA+ 147

      Vote -1 Vote +1

      • whofox says:

        I think that I just disagree with you about what a pitcher’s job is. You seem to be saying, correct me if I’m wrong, that a pitcher’s job is to limit runs crossing the plate. That seems reasonable but in the end I just don’t agree that that is a pitcher’s job. I think that a pitcher’s job is to strike out lots of batters, give up as few BB and HR as possible, and then hope for the best. Lincecum has done those first three things far better than Cain so far this year (to be honest, though, I’m not quite sure how well he has been “hoping for the best”) so I would definitely choose Lincecum. In the end, I just think we disagree about what a pitcher’s job is.

        Vote -1 Vote +1

      • B says:

        Well, the thing is runs are what determine who wins and loses. Obviously there’s been good work done in baseball to figure out what factors within a pitchers control cause runs to be scored/saved, and as you said, FIP (with K’s, BB’s and HR’s) has proven to be a great stat for determining this, so to a degree I can see where you’re coming from. There’s just a step missing in your logic that recognizes that runs are what ends up determining the winner.

        I just want to note that just because FIP proves to work well doesn’t mean those are the only 3 factors within a pitcher’s control. Batted ball profiles have shown to be influential (hence tRA), and I’m sure there are other factors within a pitchers control that we discount simply because they’re pretty inconsequential.

        Vote -1 Vote +1

      • Teej says:

        I don’t understand how you can say that the sample is too small to use FIP, but you have no problem using ERA. It’s the exact same small sample size, and on top of it you’re using an inferior measurement, so it’s an even more unreliable assessment of performance.

        You can look at virtually any stat other than ERA and figure out that Cain has pitched worse than Lincecum this year. You’re just sticking to the only one that tells you otherwise, under the incorrect notion that comparing two guys on the same team accounts for a good chunk of ERA’s many shortcomings. And that’s fine if that’s how you want to vote for the ASG. To each his own. Just say that you’re voting based on ERA, and that’s that. If you value results over process, who am I to say you can’t do that? But when you try to argue that ERA is a measure of how well a pitcher has performed, that’s when you’re going to hear howls.

        Vote -1 Vote +1

      • B says:

        Well if you want to have a discussion about how much ERA’s shortcomings are negated by comparing two pitchers on the same team, I think that could be an interesting discussion. I just hope anyone who wants to discuss this actually have a strong stat background so they know what they’re talking about.

        The reason small sample size isn’t an issue for ERA in this case is ERA is directly measuring how many runs a pitcher let up. That’s simply a fact – it happened, sample size doesn’t matter. The problems with ERA as a measurement all have to do with how much of it is credited to the pitcher, and sample size doesn’t affect any of the biases here, at least in terms of talking about one year of performance. The pitcher is pitching for one team, so his defense and home ballpark don’t change (or generally change very little) and that bias is still present throughout the whole year, regardless of how big your sample is.

        FIP, on the other hand, is trying to estimate ERA based on K’s, BB’s, and HR’s. However, we know ERA is affected by many other attributes, most of which regress to some mean. Large samples cause these attributes to regress to that mean, and is why we need them for FIP to work well.

        “If you value results over process, who am I to say you can’t do that”

        Here is where we agree. My argument to begin with was a fan voting for the ASG based on 2009 results up to now should vote for Cain over Lincecum (though we don’t vote for pitchers, anyways). The more you value the results so far this year, the more likely you should be to vote for Cain. Lincecum is much more likely than not to outperform Cain the rest of the year, as we know from the other measures in his favor.

        I guess when you say the word performance, we define it differently? I see a players performance as the results they’ve achieved. So Cain has gotten a disproportionately high number of hitters out with runners on base and isn’t likely to continue to do that – he still has accomplished it, it’s lead to him preventing runs from scoring, which won games.

        Even though Lincecum’s peripherals have indicated he should have (and on average would have) helped the Giants more than Cain – turns out, in the real world for this 2009 season, Cain has caused opponents to score less runs than Lincecum has. This means Cain has contributed more to the Giants winning games than Lincecum. That’s the case for Cain.

        Vote -1 Vote +1

      • Teej says:

        I guess when you say the word performance, we define it differently? I see a players performance as the results they’ve achieved.

        It appears so. When I look at how a pitcher has performed, I’m looking at what percentage of guys he’s striking out and walking, and whether he’s doing other things like keeping the ball on the ground. I see that as the performance. It’s more indicative of what a player actually did on the mound than looking at the runs the team allowed while he was on the mound.

        Cain isn’t putting batters away like Lincecum. He’s walking more than Lincecum. He’s not making batters swing and miss. That’s performance as I see it. I find that to be a lot more indicative of how well he pitched than the timing of the hits he allowed, how his defense handled balls in play, funny bounces and good/bad luck.

        I understand you’re saying you’re voting based on what happened, not what’s likely to happen going forward. But I’d still maintain that what happened was Matt Cain pitched worse than his teammate.

        Vote -1 Vote +1

  11. Andrew says:

    FIP
    Lincecum: 1.94
    Cain: 4.17

    tRA
    Lincecum: 2.59
    Cain: 4.19

    tRA+
    Lincecum: 146 (tRA is 46% better than league average)
    Cain: 112 (tRA 12% better than lg avg)

    To show how far apart they are.

    Vote -1 Vote +1

    • kris says:

      I’m not sure if you’ve noticed, but once something has happened, it no longer matters as to what should have happened. I’m not arguing Cain as a better pitcher than Lincecum, but I am arguing the complete and utter dismissal of luck when evaluating ‘history’

      Just because someone got lucky doesn’t mean we discount what happened!

      The All Star Game isn’t about who will be the best going forward, duh. If someone has only allowed 1 Runner to cross the plate due to a record setting .050 BABIP and 99% strand rate — Guess what? They got lucky, it’s unsustainable, but it damn well happened.

      It would be terribly presumptuous for any statistician to claim that he knew so much about advanced baseball statistics, that he could determine what was random chance and what was skill. If you can do this with certainty, please write a paper. At this point, all we can do is posit a best guess. Sure, it may be right 95 percent of the time but it’s not perfect.

      When people think the formulas they’re working with are perfect, they forget too many important things about the nature of baseball.

      Vote -1 Vote +1

    • aweb says:

      Cain – 18 runs in 66.1 innings
      Lincecum – 27 runs in 71.2 innings

      Lincecum should be expected to outperform Cain from here on, due to the things you said, (and numerous others), but Cain has been on the mound while runs were not scoring. It’s the end result that matters to some, not the “deserved end result”. Cain – lucky. That’s fine for me, but then I’m happy watching the lucky first halfers in the all-star game, rather than the probable best players. I don’t care a whit if “it counts” or not.

      Vote -1 Vote +1

  12. whofox says:

    It seems like you have to consider Reynolds at 3B instead of Chipper. At the plate you can’t argue with Reynold’s production and while I know his defense stinks Chipper has been worse this year and Wright has been pretty bad, too. To me, part of the All-Star game is giving recognition to those guys who aren’t going to be HOFers but are maybe having career years.

    Vote -1 Vote +1

  13. Brian Recca says:

    I understand what B, aweb, and kris are trying to say but I’m having a hard time accepting it. I am a Giants fan and I love what Matt Cain has done even if it is lucky but I know that while Cain has a nice ERA Lincecum has just been completely dominant at time. I haven’t really seen that yet from Cain. Because the All Star Game counts I want the players who have been absolutely dominant in there to pitch. Guys like Santana, Greinke, Verlander, and Lincecum.

    It’s a shame Bengie Molina hit a rough patch he could have been a candidate for a third catcher. Also there is no way Ryan Madson should be in there over Fransisco Rodriguez.

    Vote -1 Vote +1

  14. Craig says:

    No Trevor Hoffman on the list? He has been nothing short of amazing this year with 13 saves and 0 runs. Also just 1 walk and 6 hits vs. 14 K’s. I expect though that he will be on the actual all-star team because well he is Trevor Hoffman.

    I applaud your Mike Cameron pick though, many people, brewers fans included, do not just how good he has been for the team.

    Vote -1 Vote +1

  15. philosofool says:

    I’m glad to Wandy Rodriguez on this list. It’s nice to see an under appreciated guy make the list. (Not that there aren’t other under appreciated guys on this list, just that I’ve thought he was a solid pitcher for several years and he never got enough respect. He also had that obscene home/road split for a couple seasons.)

    Vote -1 Vote +1

  16. Chris says:

    I’m surprised to see Hoffman left off this ballot with the season he’s been having. I think I’d have him on there over one of Soriano or Madson.

    Vote -1 Vote +1

  17. Berkmaniac says:

    To echo the sentiment of a couple other posters: No Miggy Tejada??!!

    Vote -1 Vote +1

  18. Andrew says:

    I don’t understand it. People advocating Cain over Lincecum admit Cain has been lucky but ask why can’t we give him credit for it anyway? Because he had nothing to do with it – hence why you and I call it lucky. Cain’s low LOB% is not a result of any skill, but rather luck. And no, I do not want to give someone credit for something they did not have control over. Which is where FIP and tRA come in. They take into account, better than most other stats out there, all that a pitcher DOES have control over. As in, they show how the guy actually pitched.

    It’s nice that Cain has allowed fewer runs than Lincecum but that’s not because he has pitched better, it’s because he’s been far luckier. And as I said before, I do not want to give credit where credit is not due.

    Also B, this sentence right here, “…causing FIP to pretty accurately describe the end result (which is why it’s such a good predictor)…” shows you do NOT understand FIP. FIP is not a predictor, xFIP is.

    Vote -1 Vote +1

    • B says:

      Um…FIP is the same as xFIP except one normalizes HR rates and one doesn’t. Someone here needs a basic stat class…

      It’s a stretch to call it all luck. Can you say for sure he wasn’t throwing good pitches when he needed to to prevent runs? It may be most of the difference is the result of situational pitching – a hard issue to argue either way, on one hand most of the time the pitcher isn’t able to throw those pitches and over a large sample the probability of throwing those pitches (whether it be more spin, right pitch, better location/velocity, or whatever) will be what it should be. So if the actual results are different than the expected results based on the probability of the pitch being effective – sure that’s a lucky result, but hard to say the pitcher didn’t have control over the pitch he threw.

      Teej made some good points earlier, and I don’t really have any disagreement with him. My problem is with those like Andrew who lack any real understanding of stats and want to dismiss results in favor of what they believe “should have happened”. The game is played to win, and sometimes what should happen doesn’t, so I encourage everyone not to lose sight of the results that matter…

      Vote -1 Vote +1

  19. Andrew says:

    I think the best way to sum this all up is Cain supporters are being far too results oriented.

    Vote -1 Vote +1

    • B says:

      There’s only one Cain supporter that’s voiced an opinion on this thread…

      As I’ve said multiple times now, by the way, if you want to pick a Giants All-star based on who’s had the best results so far this season, Cain’s the pick. He’s done more to help the Giants win games than Lincecum has.

      Vote -1 Vote +1

      • Kevin S. says:

        WAR would seriously disagree with you. Lincecum’s done more to help the Giants win games. Cain’s *had* more help, depending on what you want to chalk it up to (luck, defense, situation…). Ultimately, you’re still giving him credit for things he shouldn’t be given credit for.

        Vote -1 Vote +1

      • B says:

        I’ve been over this in the thread. I’ll give you the short version. Cain is responsible for allowing fewer runs to score than Lincecum. That helps win more games. To think that FIP covers everything a pitcher is responsible for is silly. I’ll quote fangraphs on the subject, “Now, we’re not claiming that FIP captures everything a pitcher is responsible for.” Some things aren’t included, like situational pitching, for instance. Most of these things are either relatively unimportant, or regress to a mean with a large enough sample. The sample is not big enough, they haven’t had a chance to regress to the mean, and these things have a tangible effect on the Giants winning and losing.

        Similar to my example earlier, take a couple of high leverage situations. Pitcher throws really good pitches he normally misses and gets out of them unharmed. Is that luck? Sure you could argue that, as the real outcome was much better than the expected outcome. Did he have control over that? Well, he threw the pitch. You could have two of these situations that lead to two Giants wins. That’s pretty tangible results right there. FIP won’t give much credit for it, but that’s ok, it’s not designed to, these things work themselves out over time, which is why FIP works well.

        So much for short version…

        Vote -1 Vote +1

      • Kevin S. says:

        Wow. If you’re going to act like you’re talking to a toddler, you might want to actually respond to what was said. I didn’t mention FIP in this post. Not once. I used WAR, and I brought up the fact that a pitcher is not solely responsible for the number of runs scored while he’s on the mound, a point you completely ignored. It’s an extension of why wins are a terrible way to evaluate a pitcher… because so much goes into a team winning a game beyond what a pitcher contributes, it’s silly to give him sole credit for it. Now, with runs, a pitcher does have greater control, but not complete control. So really, stop acting like the insufferable know-it-all and respond to the points that are made, not the made up ones you already have an argument against.

        Vote -1 Vote +1

      • B says:

        Now, I know WPA has issues, especially in who gets credit for the WPA each play. But just to highlight the tangible effects Cain’s success on the mound has had this year for the Giants, I’ll throw this out there:

        Cain WPA:
        1.52
        Lincecum WPA:
        1.02

        Vote -1 Vote +1

      • Kevin S. says:

        Yes, WPA does have it’s issues. One such issue is, again, it gives the pitcher credit for things that the pitcher didn’t do. Another is the fact that it was create to placate idiots who believed in things like “pitching to the score” and “raising ability in the clutch.” Nobody who understands statistics takes it seriously as a measure of performance. A team’s win probability at any point in a game is a good thing to know, but tracking an idividual’s WPA is kind of worthless.

        Vote -1 Vote +1

      • B says:

        I assumed you were using Fangraphs win values which is why I made the points about FIP. I apologize for the tone it’s just the things I didn’t address were things I discussed earlier. I hear you on wins, and it definitely applies to ERA, too. My earlier premise was since we’re comparing two pitchers from the same team – the same defense is behind them and they should roughly have the same ballpark effects (though I’ve quoted ERA+ up above) – that should minimize most of the problems with ERA. That’s essentially why I feel it’s a good measure in this situation, while FIP isn’t because of small sample problems, and the fact that things arguably in the pitchers control (like situational pitching) aren’t measured by FIP and are heavily in favor of Cain and directly resulted in better production for the Giants so far.

        Vote -1 Vote +1

      • B says:

        “Another is the fact that it was create to placate idiots who believed in things like “pitching to the score” and “raising ability in the clutch.””

        True, and I called situational pitching “in the pitchers control” really for lack of a better term. I’m having trouble finding the proper way to say what I want to say, but it’s something along the lines of I find it hard not to give credit to a pitcher for performing in the clutch. Obviously it’s not like they raised their ability or anything, and over time it will regress to their normal production since it’s unsustainable…but in small samples it does make a difference when it comes to the team winning games, which is the ultimate goal.

        Vote -1 Vote +1

      • Kevin S. says:

        And I apologize for coming off so prickly. I thought your tone was patronizing, but perhaps I’m being overly sensitive. Your assumption that because Lincecum and Cain play before the “same” defense means it’s homogeneous is an inherently faulty one; performance and lineups over such a relatively small sample are certainly subject to fluctuation. I asked elsewhere (perhaps upthread, I’m too lazy to check) if it was possible to get a team UZR behind a given pitcher. I’m not willing to accept the situational pitching postulate over a defense/randomness theory without something beyond speculation to back it up.

        Vote -1 Vote +1

      • B says:

        Not completely happy with my explanation, wanted to get it out there before you jumped all over my situational pitching comment…

        Basically, even though situational pitching is unsustainable, in this case it did happen. Cain has achieved a high strand rate (primary reason his ERA is where it is), and it has helped the Giants win more games. I think any of us will be hard pressed to figure out how much of it was from things in Cain’s control (good pitches in the situation that he normally doesn’t make, for instance) compared to how much of it was simple luck (like hard line drives that happen to go right at someone). In the end I have a hard time simply dismissing this performance when it’s happened. Especially dismissing it on the guise of “just luck” or whatever else when some of it was the result of things a pitcher was in control of in that situation.

        Vote -1 Vote +1

      • Kevin S. says:

        No, I get that there is such a thing as clutch performance. A-Rod just grounded into a double play with the bases loaded; he failed to get the job done, but that doesn’t make him a choker. It’s clutch ability that earns derision. I think my issue is you’re attributing the Giants’ difference in run prevention with Cain on the mound as opposed to Tiny Tim to Cain coming through in big situations, when there’s just as strong a likelihood that fielding differences were responsible.

        Vote -1 Vote +1

      • B says:

        I do agree with you about the faults in the defense assumption. If I were doing real research on the subject, it’d definitely be something I’d have to delve into and come up with evidence for that claim. That said, I’m comfortable assuming the differences are small enough to ignore because I think the theory behind it is sound (same defense cancels each other out), and frankly, I don’t think there would be a reliable way to measure it, anyways. Maybe if we had a huge sample, but then the issue would be moot because Cain/Lincecum’s ERA/FIP/tRA would closely match each other and we’d all be in agreement.

        As for situational pitching vs. defense/randomness…I credited situational pitching for most of the difference based on Cain’s peripherals. His BABIP is .277 compared to a career .281, so that doesn’t appear to be the explanation. His strand rate is 87.8%, compared to 74.7% for his career. Lincecum’s BABIP is .369, so for him it’s definitely a huge factor why his ERA and FIP don’t match.

        Vote -1 Vote +1

  20. Andrew says:

    http://www.hardballtimes.com/main/fantasy/article/2009-lips-era-leaders1/

    Out of the 25 pitchers listed Lincecum is third while Cain is nowhere to be found. For those unfamiliar with LIPS ERA check out this article:

    http://www.hardballtimes.com/main/article/dips-lips-and-hips/

    FIP doesn’t account for HR/FB or include IF FB% (which are effectively as good as strikeouts as almost 100% of infield flies become outs) while LIPS does.

    Vote -1 Vote +1

  21. Andrew says:

    B you continue to attack my knowledge of stats when it’s clear I know more about the ones we’re using than you. You don’t even seem to realize that there is a stat which defines how a team is fielding behind a pitcher. It’s called DER, but it’s not perfect because it can be affected by BABIP, LD%, and whether or not a pitcher induces ground or fly balls (ground balls are turned into errors more often than fly balls). FWIW Cain’s DER is much higher than Lincecum’s, but it can be explained by Lincecum’s inflated BABIP and LD%, and the fact that he induces many more GBs than Cain.

    You continue to talk about FIP as if it has more sample size problems relative to ERA, but that’s simply not true. The part of it which is subject to regression are the home runs, and xFIP solves that. We can look at FIP to see how a pitcher HAS pitched, and then his xFIP to see if his HR/FB and LOB% rates are due for regression. In other words, FIP is no more subject to sample size issues than ERA is.

    Cain’s BABIP with runners on: .159
    Cain’s BABIP with RISP: .105

    He should not be rewarded for unsustainable BABIP rates (which he has no control over) resulting in fewer runs scored. Simple as that.

    Kevin: League average is 7.3% and some pitchers consistently over/under-perform that number.

    Vote -1 Vote +1

    • B says:

      My attacks were more over your grasp of statistical principles, not your knowledge of different baseball stats.
      “The part of it which is subject to regression are the home runs, and xFIP solves that.”

      When measuring the runs a pitcher is responsible for (our goal for the day), there are sources outside of FIP’s equation that “regress to the mean” to make FIP more accurately reflect those runs. In fact, you brought one up yourself. “He should not be rewarded for unsustainable BABIP rates.” Those BABIP rates regress to the mean, and as they do, FIP does a better job of explaining the runs the pitcher was responsible for. These runs, by the way, are an objective number we know (pretty easy to measure the number of runs scored when a pitcher is on the mound). The real issue is dividing responsibility for them, and that’s where sample size comes in. There’s not really a whole lot to say without repeating myself…so I guess I’ll just let it go?

      Vote -1 Vote +1

      • aweb says:

        Such a heated debate for what comes down to – some of us don’t have any issues with rewarding pitchers who have gotten luckier than others. Yes, Lincecum, from all reasonable measures, has done more to prevent runs this year than Cain has. But Cain has been luckier, as evidenced by unreasonably low BABIP in key situations – it doesn’t bother me at all giving him a trip to the all-star game for that. That makes some of us strange in the view of others, perhaps, but that’s all there is to it.

        The All-star game is an exhibition game – managers still treat it that way, as do players who drop out with minor injuries. I honestly don’t care if the very best possible players are selected, and I’m fine with rewarding flukey first-halfs. That’s how the all-stars have usually been selected. Of course, by the time it comes to picking the all-star staff, it’s quite likely Lincecum will look better than Cain, even by the basic “runs scored while he was pitching”, results-based method. All-star games are like post-season awards to me (and others, although I can’t speak specifically for anyone here) – they reward what has happened, lucky or not.

        Vote -1 Vote +1

  22. Andrew says:

    Also, B, since you seem to have such a problem with regression issues, read up on LIPS and then please tell me why it’s flawed in telling us Lincecum has been much better than Cain this year.

    “…think any of us will be hard pressed to figure out how much of it was from things in Cain’s control (good pitches in the situation that he normally doesn’t make, for instance) compared to how much of it was simple luck (like hard line drives that happen to go right at someone).”

    Exhibit A, again:
    Cain’s BABIP with runners on: .159
    Cain’s BABIP with RISP: .105

    Sorry but I couldn’t help myself.

    Vote -1 Vote +1

    • B says:

      It seems LIPS refers to late inning pressure situation? Not sure how that’s relevant, so why don’t you give a go at explaining the stat, what it’s uses are, what it’s strengths and weaknesses are, and the theory behind it, and why it’s relevant in this specific situation.

      Vote -1 Vote +1

      • Andrew says:

        Why should I bother doing you a favor? All you have done so far is bash my knowledge of stats and ignore everything that goes against what you’ve been saying (Cain’s BABIP with RISP, for example). If you are too lazy to educate yourself, then that’s your problem not mine. I provided you a link above – click it.

        Vote -1 Vote +1

      • B says:

        Well…fair enough.

        Vote -1 Vote +1

  23. jpd says:

    Hey, Andrew, I’m gonna have to say, you sound like a huge jerk.

    Exibit A:
    “It,s clear that I know more about the ones we’re using than you”

    sorry, I couldn’t help myself

    and, I support B. actual results over what should have happened. How can you argue against actual results, as fluky as they may be? what actually happened is always more important.

    Vote -1 Vote +1

  24. k burrell says:

    mccann & chipper come on guys vote for them

    Vote -1 Vote +1

  25. stlfan says:

    I’m biased as a Cardinals’ fan, but no love for Joel PIneiro (top ten FIP in NL) and Chris Carpenter even though the game is going to be in STL?

    Vote -1 Vote +1

  26. James says:

    I don’t understand all the stupid argument about who’s better: Cain vs. Lincecum. The argument should be why Cain vs. everyone else who’s not Lincecum. Cain should be a lock for a an NL starter spot, but for some ludicrous reason this retarded page doesn’t even list him. How can you leave off a guy how is in the top 3 in Wins, ERA and Winning %? Disgusting.

    Vote -1 Vote +1

Leave a Reply

Your email address will not be published. Required fields are marked *

You may use these HTML tags and attributes: <a href="" title=""> <abbr title=""> <acronym title=""> <b> <blockquote cite=""> <cite> <code> <del datetime=""> <em> <i> <q cite=""> <strike> <strong>