Quantcast

scared babyIt occurs to me that I haven’t really been chapped by a manager decision in quite a while (Mike Quade used to do it to me regularly), but I was pretty aggravated last night. With one out in the bottom of the 8th, Jeff Samardzija came to the plate to hit in a 1-1 game. He was pitching a fantastic game, had the only Cubs hit of the night to that point, and is the staff ace. I can understand wanting to send him out there for another inning. I didn’t agree with the decision, given that Samardzija was already well over 100 pitches, and you really need a strong batter in that situation. Still, I accepted it.

But then, in the 9th inning, Samardzija was allowed to throw another 20ish pitches, pushing his total to 126, and was not pulled even after walking two batters with relievers warm and ready (thankfully, a well-timed double-play ball saved us from a potentially even more uncomfortable conversation).

I am sure Samardzija would have been angry to have been pulled (for his part, he isn’t worried about pitch counts), and I know his stuff still looked good. But there’s an obligation to protect the players from themselves at times, and, knowing what we know about pitcher arm injuries, it almost always a terrible idea to allow a pitcher in that situation to throw more than 115 pitches, let alone 126.

Why was Samardzija left in so long?

He was performing well, and he gave the Cubs a good chance to go into the bottom of the 9th with the game still tied, so there’s your primary reason. But what else did manager Rick Renteria offer after the game?

As he told the media – several times – when they asked the obvious question, “We wanted that game for him.” The implication was pretty clearly that Renteria wanted to see Samardzija finally get his first “win” of the year.

126 pitches for a shot at a “win.”

When will the pitcher W/L stat’s reign of terror come to an end? It’s already been dumbing down baseball discourse for far longer than it ever should have, and now it’s threatening bodily injury to the Cubs’ best pitcher?

I am, of course, being a little tongue-in-cheek here, and I’m not going to overdramatize a single outing (even though there is credible research that indicates allowing a pitcher to go over 120 pitches in a single start – even once – can be one of the best predictors of future arm injuries). Pitchers come in different shapes, sizes, and abilities, and Samardzija does seem like the kind of guy who could handle elevated pitch counts a little more regularly than other guys.

The point here is simply that, if you’re going to allow Samardzija to throw too many pitches, at least allow him to throw too many pitches for better reasons than to get him a mark on the back of his baseball card – a mark that tells us nothing about how well he pitched. The proof? Samardzija has pitched as well as any pitcher in baseball this year, and he still doesn’t have one of those wins. By pitcher W/L record, Samardzija is tied at the bottom as the worst pitcher in baseball.

Your instinctual reaction to that fact shouldn’t be, “Man, he’s been unlucky,” or, “Man, he deserves a win!” – your reaction should be: “Holy crap, what kind of POS stat would suggest that Samardzija is tied for being the worst pitcher in baseball so far this year? That must not be a very good stat.”

Consider this: if Welington Castillo singles in the 9th inning last night, Samardzija gets a “W.” That was after he’d left the game. Nothing he did could change at that point. This all-important W/L stat was riding entirely on things outside of Samardzija’s control. Castillo didn’t get a hit. Samardzija didn’t get a win.

Is his start suddenly different? After he’d already left the damn game?

In every argument since ever about the pitcher W/L stat, there are pretty much only two defenses to its continued use, and both are completely bogus.

(1) “Wins and losses are the most important thing in baseball, so how could it not be an important stat?” Yes, team wins and losses are the most important stat of all. Unfortunately, they are completely different from the pitcher W/L stat, which is as artificial as any other stat about an individual player. Except that this stat doesn’t actually tell us how well the player performed – a guy can throw 9 innings, give up no earned runs and no hits and strike out 20 and still get a “L”. A guy can throw 5.0 innings and give up 40 earned runs and still get a “W.” How quickly would you defenders of W/L jump all over a sabermetric stat if it was that crappy at evaluating player performance? You rail on WAR because it’s inaccessible and opaque, and yet you defend a stat that is on its face and by definition a misleading descriptor of a pitcher’s ability?

(2) “Yeah, it’s not great, but it’s a good short-hand for how well a guy has pitched in his starts.” First of all, no it’s not. Second of all, if that’s all you’re looking for, there are way better stats out there – Quality Starts, for example.

The pitcher W/L stat is a terrible stat, made only more terrible by how pervasive its “importance” is in our current baseball culture.

Stop it. Stop caring about that stat, and start caring about stats that matter. You know, stats that tell you what your eyes have told you all season long: that Jeff Samardzija is pitching like a beast, and no bogus stat is going to change that fact. He doesn’t need to get a “win” to justify his awesomeness this year.

  • pinkstonaa

    I wasn’t upset that they brought back Samardzija for the 9th – at that point, he had retired 12 consecutive batters – but I was upset that they didn’t start the inning with anyone warming up. A lot of people were throwing around the “pitchers hit a sharp downturn at 120 pitches,” but he came into the inning at 107, so I don’t think there was any problems thinking he could go in and get one or two outs, but RR seemed content to let him go the whole inning, no matter what.

  • Illini Cubbie Fan

    I know that I am definitely in the minority here, but I personally don’t mind the “W/L” stat. Although it doesn’t reflect on the performance of a pitcher in any way, it does reflect how the team plays. Isn’t it possible that players play better when certain pitchers are on the mound? That has to have some sort of value, in my opinion.

    • roz

      “Isn’t it possible that players play better when certain pitchers are on the mound?”

      Sure, but team wins and losses don’t tell us anything about that. You can play a horrendous game and still win, and you can play a great game and still lose. Wins and losses tell you nothing except which team had more runs when the game ended. Anything else that you try to take from them is just speculation that is better understood through a multitude of different statistics.

    • Chad

      No it doesn’t reflect how the team plays as much. If a team scores 8 runs but the pitcher gives up 9 and gets the loss what does that say about the team? Or a team scores 6 and are up 6-3 in the 9th and then the closer walks them full and then a new reliever gives up the grand slam, what does it say about the team?

    • http://www.bleachernation.com Luke

      Isn’t it possible that players play better when certain pitchers are on the mound?

      It is possible.

      What does that have to the with pitching W/L stat?

      If that is what interests you, devise a stat to measure it. That would be pretty interesting. But pitching W/L does not measure it at all.

    • DocPeterWimsey

      Possible? Yes. Is there credible evidence for this? No. Run support for individual pitchers fluctuates substantially from year-to-year.

  • pinkstonaa

    Strange thing is, losses for relief pitchers is a pretty telling stat.

    • Chad

      Not necessarily. As stated above if it is a 6-3 game and a reliever walks them full then a new reliever gives up the GS to lose he gets the loss. Is that necessarily true of his performance? I don’t know. It’s about as useful as the hold stat.

      • pinkstonaa

        That’s probably the exception that proves the rule. I think in general, if a reliever has a lot of losses that means he is giving up leads and not having a good year.

      • pinkstonaa

        For example, the L Justin Grimm took last night was a deserved L.

        I think in the long run, W-L stats tend to sort themselves out. Like MatthewP notes below, it means something when a pitcher accumulates a certain threshold of wins over a career.

        • http://fullcount1544.blogspot.com FullCountTommy

          Felix Hernandez, one of the best, if not the best pitcher of our generation only has at least 15 wins one time in his career. Wins are useless.

          • MatthewP

            I think that; as annoying it is that a person refuses to open their old-school brain to the concept of advanced metrics, so it is equally annoying when those who embrace metrics completely disregard any statistics that were used prior to their implementation.

            As much as anyone would like to argue otherwise, wins, losses, ERA, BA, HR’s and RBI’s all have some value, little as it may be. No need for complete polarization. To me that’s a tiresome internet trait that’s become ever more prevalent.

            • Voice of Reason

              So, you like the stats that say the Cubs should be a .500 team?

              While I think some of the stats are cool to look at, games are won and lost on the field, not on paper.

        • Chad

          If you had to assign a loss to someone yes Grimm probably deserved it, but how about the offense being so bad and not letting it come to Grimm in 12 innings. So the team deserves the L, not sure why it has to be assigned to a pitcher. Why don’t we give W/L to each and every player based on how they perform that day. Even if you go 4-4 and hit for the cycle and your team loses you get a L, but if you are 0-5 with 5 k and your team wins you get the W. It’s silly. W/L should be for teams only not individual players.

          • pinkstonaa

            That’s a good argument.

        • http://www.bleachernation.com Luke

          If a pitcher accumulates a lot of wins, it means he played for a long time on a lot of good teams.

          • Brocktoon

            And was a good, durable pitcher(though goes hand in hand with playing for a long time)

            I get that this is a standard fish in the barrel thread, but people get so up in arms whenever some 2002 sabermetrics topic like sac bunting or pitcher wins come up that they need to shout down everyone else in the thread.

          • bnile1

            What is missing in this conversation is perspective.

            Wins are not a useless, stat but as with any stat you have to understand the caveats that go into using it. It is not designed to be a performance stat. ERA, WHIP, SO etc are all performance stats, and are much better for evaluating performance. Wins are also completely meaningless for relievers for obvious reasons (blown leads and rally etc). It also cannot be looked at in a vacuum (see shark this year). Unfortunately people tend to ignore these things when they look at it.

            Wins are designed to be an outcomes measure. What happens when this guy goes to the mound. There are other things that are reflected in this stat. Consider a few situations.

            1: Condition. If a pitcher on say the Cubs were to have the wind blowing out in all his starts. Would his ERA etc be adversely affected, and could he have a higher ERA than another pitcher who was not as “good” but always pitched with the wind blowing in. In this case the Wins would indicate that pitcher A did a better job of pitching in tougher condition(and that pitcher B may not be as good as his number indicate).

            2: Innings- Guys who go later in games regularly tend to get more wins(and losses). But shouldn’t a guy going though the line up again have a stat that reflects that. Wins in an imperfect way do.

            3: Situation; IF a team scores 6 runs, should your starter try to get a shut out, or try to keep the team from scoring 5 while going as deep into the game as possible. There are guys who tend to be better than the situation, they seem to have a knack for giving up runs when it won’t hurt them and not giving up runs in close games. That’s not luck

            4: Personal contribution. Good defenders and guys that can handle the bat(NL), and not make mistakes(wp, hbp etc) tend to win, Guys who don’t tend to lose

            5: “unearned runs”. Some guys respond to an error by getting the next guy out, some guys tend to fall apart

            6: Magic- Some guys in certain years have the magic that when they go to the mound the team wins, some guys are cursed(see shark). If you are going to bet, and pitcher A has a record of 12-1 and pitcher B has a record of 1-12, who are you betting on? Did you ask for the ERA before you answered? To the degree that you did not, that demonstrates the value of a Win.

            Now see the caveats above but in general pitchers with a lot of wins have most if not all of these traits, and guys who did not win probably had some of the issues above. No it’s not a perfect stat, and a guy who is 6-5 is not necessarily better than a guy who is 5-6, but ALL OTHER THINGS EQUAL, if the numbers are equal or close, then the guy with the most wins probably has done the better job overall. NO the guy with the most wins is not necessarily the best pitcher, and a guy with no wins may still be pitching well, but as part of at pattern it is useful.

            Now would I ever condone leaving a guy in for 126 pitches just to get a win?? NO. The wins should come naturally and not be considered of such importance that you modify your decisions a manager, PERIOD.

            • http://www.frenchrocks.net Ian Afterbirth

              Very interesting and well thought out.

              I remember being a kid and finding out how the W/L stat was computed for pitchers and thought it made no sense.

              I was a smart kid.

            • ssckelley

              Excellent point!

    • pinkstonaa

      Change “Strange thing is” to “Playing devil’s advocate.”

  • Senor Cub

    I am all for bringing Quade back. I don’t know squat about RR and his track record but I know how he calls games and I was extremely annoyed that Shark was still in the game at 120+ pitches last night. Dusty memories are beginning to come back. I don’t think he makes it to the end of the year.

  • jh03

    ::slow claps::

  • JulioZuleta

    I said it last night in the EBS. That DP bailed RR out. I was convinced at the time (and even more so after reading his comments) that he would have left Shark out there to get the third out if he had say, struck Viciedo out. That was infuriating.

  • WGNstatic

    The one positive thing I will say about the W-L statistic is that it does have some “sentimental” value. I don’t disagree with anything Brett, or many others, have said derisively about this statistic, but to me, there is room in my statistical lexicon for outdated counting statistics with ties to the history of the game.

    • http://www.bleachernation.com Brett

      As a conversation piece … I say, whatever. God bless, go forth and have fun.

      Just don’t you dare evaluate a pitcher on it!

      • KHRSS

        I doubt anyone evaluates a pitcher on their win/loss record.

        • IA_Colin

          Have you seen Jon Heyman’s articles?

  • MatthewP

    It’s shouldn’t be as important as it still is, but I don’t like the word “useless.” It has some value in showing the relative greatness of many players over the course of their careers.

    Nobody disagrees that pretty much all of the pitchers in the history of baseball over 200 wins were great, and you never hear that say… Warren Spahn was a mediocre pitcher who benefited from those “powerhouse Boston/Milwaukee Braves” teams. They had some great seasons. Some seasons he won about 1/3 of their games.

  • CubbieBubba

    MLB should get rid of the Team W/L record. After all, statistics show the Cubs are so much better than their actual record anyway. And it doesn’t even factor in the quality of top rated, unproven, draft picks!

    • IA_Colin

      Go back to your bridge. Trollololol. Not even close to the same argument. Try again.

  • Coach K

    The Cubs have cost Shark 8 wins this year alone! #GarzaLogic

  • Rebuilding

    Some thoughts about last night’s game:

    * I have no problem with Renteria going to Grimm when he did. Grimm is a better pitcher than Wesley Wright and has shown more this season. Also, Grimm can be stretched out 2-3 innings in an extra inning game. It didn’t work out, but I have no problem with the move.

    * I’m a big supporter of replay in all sports, but baseball’s replay is damn near worthless. It seems that the league has taken the stance (probably because of the mystifyingly powerful umpire’s union) that you need not only DNA evidence, but a picture of the murderer with the weapon over the body. I have seen more blown calls not reversed than reversed this year.

    * Leaving Jeff Samardzija in for over 120 pitches is inexcusable and indefensible. I’m sure Renteria got a call from the FO last night much like he must have gotten one after the “5 bunt game”. I understand his job is to win games (and it’s highly debatable if that was the right call given 2 straight walks), but that can’t be at the expense of the long term goals of the organization. If Samardzija blows out an elbow that is a catastrophic blow to this rebuild effort – whether we were going to resign him or trade him. Right now Jeff Samardzija’s right arm is most valuable thing the Cubs have. To do it in order to get him a “win” just compounds the stupidity

  • Matthew

    Absolutely spot on, Brett. I couldn’t help but read faster and faster because I agreed with every word more and more.

  • EQ76

    Wins and Losses did used to be a valuable stat back in the day when pitchers pitched more innings.. Dudes used to hurl 9 innings per outing pretty consistently and obviously wins and losses were very appropriate stats to help judge the success of a pitcher.

    Having said that, baseball is all about history. The W-L stat is a big part of history. There’s no use of getting rid of it all together, but I believe most people have enough common sense to look at the wide variety of other stats and be able to see the true value of a pitcher. Sometimes I feel like people are arguing something that’s really not as big a deal as it seems.

    I for one have always, even back in the 80′s, put more stock in ERA and K/BB ratio than wins and losses… not with the advanced stats, we have so many more ways to judge the value of a pitcher.

    • EQ76

      ** Now with the advanced stats, not the word “not”

  • Transmogrified Tiger

    I didn’t really have a problem with Samardzija pitching the 9th.

    - He’s a big guy who’s been free of injury trouble, so there’s no specific concerns with Shark’s health outside of “more pitches are worse, generally speaking”
    - He had an extra day’s rest prior to last night’s start
    - His stuff/mechanics held up into the 9th, which is by far the most important point
    - He had been on cruise control for several innings, retiring 11 in a row prior to the 9th. All pitch counts are not created equal, and Samardzija’s 107 pitches entering the 9th had not been high stress. He hadn’t pitched from the stretch in an hour+ at that point.

    As for the W/L stuff, of course it’s terrible and needs to go away. It’s tangent to the real reason though, and that’s that Samardzija has been incredibly unlucky and Renteria was trying to give him something to feel good about. Renteria didn’t leave Samardzija in the game simply so that he could have a 1 next to his win count, he left him in so that Samardzija could achieve something that was important to him. So rail on W/L for being so ingrained that Shark sees it as a carrot, but that’s several levels removed from the actual decision being made at that point(which again, wasn’t egregious for a bunch of reasons).

  • AndPrintIt

    Assuming Samardzija can make it through July, any “future arm injuries” will be some other team’s problem. I would be more concerned about his longevity if he were more interested about being with the Cubs long term. As he seems most interested in squeezing every last dollar out of the Cubs (and the Cubs seem unwilling to value him more than Edwin Jackson), perhaps the Cubs should be positioning him as a “workhorse” to maximize the return on a trade. If he blows out an elbow for Toronto, send Anthopoulos a fruit basket and move on.

    Callous? Yes. Cold-hearted? Absolutely. But these are not players. They are not even people. They are assets. If an asset will not be part of the rebuild, the team needs to maximize its value to bring the best return of assets in a trade.

    Thank you, Theo et al for teaching me how to dehumanize people … er, resources … and to remove the emotion that I was foolishly investing in a collection of assets. The culture at Wrigley has definitely changed.
    And the force is strong with this one.
    Play ball!

    • http://www.bleachernation.com Brett

      If I’m a team interested in trading for Samardzija (for this and next year), high pitch counts with the Cubs will do absolutely nothing to entice me. It would give me concerns.

      • Edwin

        Especially if higher pitch counts now lead to ineffectiveness and/or injury in the next month.

      • AndPrintIt

        On the flip side, if the asset does not go deep into games (which probably means a lower pitch count), one might question the asset’s durability or ability.

        Consider the pitch count leaders for 2013:
        Verlander
        Shields
        Wilson
        Lester
        Wainwright
        Dickey
        Tillman
        Shark
        Darvish
        Kershaw

        Should the Cubs shy away from any of these guys because of their high pitch counts? I hope not. (Rumor has it Shields and Lester might both be available next year. And they are both currently in the top 10 for highest pitch count this season.) A rotation of Verlander, Shields, Lester, Wainwright, and Edwin Jackson might be good enough to contend for a wild card.

        Good pitchers gonna pitch. And that is going to raise their counts.

        • http://www.bleachernation.com Luke

          There is a difference between high pitch counts and too high pitch counts.

          • Brocktoon

            I’m confused, did none of the top 10 guys in pitch count have “too high” pitch counts?

  • MightyBear

    Two comments on the win stat being absolutely useless (it is). 1. One of the northeast “experts” when Nolan Ryan was up for the HOF made the comment “He’s a .500 pitcher.” Its true he was 320 and 280. He also played for the Angels, Astros and Rangers when they weren’t very good. He also had over 5000 K’s, 7 no hitters, 19 one hitters and was hitting mid 90′s when he was in his 40′s. He was probably the most dominant right handed pitcher in the history of baseball and this guys knock on him was he was a .500 pitcher. If he had been with great teams, he would’ve won 500 games. 2. A few years ago a pitcher with the Rangers won 20 games. He was 20-11 and his ERA was 5.5. The Ranger team averaged 8 runs a game during his starts. If the Cubs had scored 8 runs in the Sharks 7 starts, he’d be 7-0. The wins/loss stats are useless. As Casey Stengal used to say “You have to be one Hell of a pitcher to lose 18 games in this league.”

    • Spriggs

      Thank you MightyBear. Nicely stated.

      I do find it entertaining that after all the lectures from Lenny and JD about W-L records being a useless measure for a pitcher (and I agree), what is the first stat they mention as a new pitcher comes into the game?

      • Edwin

        Old habbits die hard, I guess.

        • DocPeterWimsey

          If you meant to write “habits,” then, yes. If you meant to type “Hobbits,” then only if they have Rings of Power.

          :-)

          • Karl Groucho

            And in a surprising twist, Bruce Willis cast as Bilbo for The Hobbit Three: Old Hobbits Die Hard

            • Edwin

              Thirteen dwarves. One dragon. The odds are against Bilbo Baggins… That’s just the way he likes it.

          • Edwin

            Ha. I thought something looked odd after I posted that.

      • Voice of Reason

        Spriggs,

        Point well taken on the first stat the announcers give. BUT, it’s universal and can help reveal if the pitcher has received support from his offense. If a guy is 0-10 with a 1.95 era then that’s a tell. It’s also a tell if the reliever is 12-3 with a 1.95 era.

        • http://www.bleachernation.com Luke

          It isn’t universal. You will have to look very hard to find the last time I referred to a pitchers win/loss record.

    • Brocktoon

      “If he had been with great teams, he would’ve won 500 games.”

      Speaking of Garza math.

  • MattM

    I don’t know if this has been discussed yet, but I’m not happy about how RR manages the bullpen. I don’t mean how he picks pitchers but how he never has two guys up throwing ready to come in in situations! That loss to the cardinals for instance: Rondon had pitched two games in a row and was not used to that load yet. Let him pitch I’m fine with that but have a lefty and a righty up just in case. Because he didn’t he HAS to leave Rondon in way two long and we lose the game and Rondon’s confidence is shaken.

    He pulled the same shit with shark yesterday. Yes, the double play worked out, but have other guys up just in case!!!! In Grimm’s case as well….Have guys up and ready to go!!! Why is it that every other manager does this but the one the Cubs have? Seems like we always get the wierd managers that do unproductive things for us. When are going to get a Maddon!!!

  • KHRSS

    I think people here exaggerate with how the W/L record of a pitcher is perceived. It’s been years(decades probably) since it’s been used as an indicator of how good a pitcher is.

    • http://www.bleachernation.com Luke

      It was being cited last summer as the criticism of the Cubs drafting Zastryzny. He had a terrible win loss record, so, the critics said, he must be a terrible pitcher. And then the stopped looking and moved to something else.

      Pitching W/L are still used by some as the definitive measure of a pitcher’s ability and value. Sadly.

      • Jon

        The good news is this year…Zastryzny has terrible individual numbers so the criticism may resume!

        • roz

          I don’t like how giddy you seem to be about that, but yeah, Zastryzny is not looking good this year.

        • http://www.bleachernation.com Luke

          True, but not because of his W/L record.

    • Patrick W.

      One of the theses of this post is that it was used LAST NIGHT.

  • Webb

    The guy is leaving in a month, was the most productive offensive player on the night, and has gotten no help from the rest of his team to win games he pitches in. If he wants to swing a bat in the 8th he gets to swing a bat in the 8th for me. I mean, what do you tell that guy in the dugout?

    “I’m sorry, you retired 12 batters in a row, allowed one unearned run, broke up a no-hitter, scored the tying run, and have been let down by the rest of your team in every game you’ve played this season, but we are going to go ahead and not allow you to try and finish this game against your wishes and instead turn to a relief core that has had a propensity for failing in must-win situations for more than a month.”

    ….PS. We don’t plan on securing your future for the next 5 years like the rest of your peers either.”

    I really enjoy reading this site, but this post and the “Comparing rebuilds” posts read as though our collective frustration is getting the better of us enjoying the game for what it is.

    Who cares what Jose Abreu is doing? Could the Cubs have rebuilt in a different fashion? Absolutely (The starting pitching was certainly there). They didn’t.

    So we wait, enjoy the progress of misters Rizzo, Castro, Castillo, Lake and Olt, and stop worrying about what other people think or how wrong it was to let a player gut it out because he damn well earned it more than anyone on the roster.

  • Aryk

    Maybe we could start a movement where we get the manager to bring the same pitcher into the game to get the final out of the 5th inning every time the team is winning. we could have the first 50 game winner in history. Give him a few years and we could get him more wins than Cy Young.

    Unfortunately the problem here is entrenched in baseball DNA, when these pitchers goto arbitration I am sure that their W-L record isn’t worthless or when they are discussing a contract or when winning the Cy Young, the list goes on and on for why a pitchers W-L record isn’t worthless. It is a completely worthless stat (and has been proven as such) but until it stops being the first thing used to compare pitchers to other pitchers it will remain significant.

  • TSB

    If a team has a winning record, but a pitcher on that team has a losing record, doesn’t that reflect on his ability?

    • http://fullcount1544.blogspot.com FullCountTommy

      No, not even a little

    • DocPeterWimsey

      Much (if not most) of the time, it reflects low run-support for that pitcher. That in turn represents bad luck in opposing an inordinate number of plus-starters in his starts.

      Dennis Eckersley had that on the 1984 Cubs. Despite a plus offense, they seemed never to score for him.

  • KHRSS

    Almora walked again today. 1-2 with a double, so far.

    • http://www.bleachernation.com Luke

      Progress!

      • KHRSS

        Bryant was taken out after one at bat. They didn’t say why on the game feed when they mentioned it.
        DOes anyone have more info on this?

        • Pepitones wig

          maybe got called up?

        • Brocktoon

          Stomach flu.

          • J. L.

            Are you serious? Don’t toy with our hearts, man!

            • Brocktoon

              Jesse Rogers (yeah I know) said as much

    • Spoda17

      He must have read Luke’s article…

  • mosconml

    Pitcher wins don’t matter to analytical nerds like us, but they do seem to matter at least a bit to the player. And I can understand that — it’s a tangible way of saying “you did well enough to lead your team to victory today”. It’s not a good stat but I bet Samardzija would be at least a bit less frustrated if he had a couple wins.

    And I don’t think the Quality Start is much better, or at least it’s flawed. Pitching 6 innings and giving up 3 runs is a quality start? That’s a 4.50 ERA. No thanks. I’d prefer it be at least 5 innings and no more than 2 ER.

    • JacqueJones

      Quality start is a lot better because it indicates how often the pitcher has “given his team a chance to win” on average teams score between 4-5 runs in a game in the mlb. If you give up 4.5 runs per game, you have given your team a reasonable chance to win, so i think that stat actually means something.

  • Spoda17

    I have been scratching my head about RR’s decisions all year. This is yet another example. Makes absolutely no sense.

    Brett good point about future trade partners, hell we have talked about Rondon’s high pitch count at NC State as a warning sign.

    I think RR might get a call from Theo regarding this. Has anyone noticed that RR has pretty much stopped sac bunting..? I think he may have been redirected from that philosophy.

  • Jon

    People that continually bash on RR crack me up. When you will you realize that you a manager cant’ make chicken soup out of chicken salad!

    No manager can win with the garbage given him, not to mention, Castro and Rizzo’s development re-emergence has been awesome this year. THE MOST IMPORTANT THING!

    • Head and Heart

      No one is bashing him for wins and losses. Regardless of the players on the roster he can still make moves that make sense logically. The Cubs would likely still find a way to lose. But that’s not really the point.

  • J

    Well, Cubs still bunted last night with Bonafacio and runner on second. They scored the tying run, but essentially gave up a chance of getting a second run. Either way, RR’s bullpen use – especially with 8 relief pitchers – is more puzzling than his love of the bunt.

  • http://obstructedview.net Myles

    Here’s a hypothetical question.

    You have your choice of either of these two pitchers. You literally don’t know a single thing about either pitcher save one; you don’t know their age, FIP, K%, anything. You only know one thing: one went 13-7 last year, and the other one went 7-13. Which one do you choose?

    If you choose one at random, then you can truly say that wins and losses have no value as a stat. If you choose the 13-7 pitcher, you are intrinsically assigning a value to wins and losses. If you choose the 7-13 pitcher, I don’t know what to tell you.

    I can absolutely agree that wins and losses are not a very good stat, and for the reasons you mention. They’ve been long superseded by other, more refined stats. However, there is at least SOME intrinsic value, and I don’t think that anyone would seriously look at the above hypothetical and say that they’d just choose one at random.

    Lastly, you offer quality starts as a “way better” substitute for wins and losses. This isn’t really true; quality starts suffer from the same problem that wins and losses do. You can go 6 IP, allow 14 runs (3 earned), and get a “quality start,” which is seemingly just as valuable as Kerry Wood’s 20-K one-hitter. If you are going to rail on wins, you pretty much have to rail on quality starts too, which are better than wins and losses but still pretty bad. I’d suggest Game Score as a slightly better alternative (though it suffers from the many of the same problems as QS do).

    • Norm

      I would choose someone my scouts have seen.
      Or choose someone I do have information on.
      Otherwise you’ll be stuck with Felix Doubront (11-6 in 2013) over Cole Hamels (8-14 in 2013)

      • http://obstructedview.net Myles

        This is not the question. I’ll repeat it.

        You have your choice of either of these two pitchers. You literally don’t know a single thing about either pitcher save one; you don’t know their age, FIP, K%, anything. You only know one thing: one went 13-7 last year, and the other one went 7-13. Which one do you choose?

        • Darth Ivy

          your question is not answerable because you can’t pick only based on those data.

          • http://obstructedview.net Myles

            It is absolutely “answerable.” You could answer, for instance “I’ll take the 7-13 pitcher.” or “I’ll take the 13-7 pitcher.” or “it makes no difference.” It’s a hypothetical question.

            • Darth Ivy

              not for me. I couldn’t chose between two guys only based on those data.

        • Norm

          I’d let my underlings decide because it really is a coin flip.

    • Edwin

      Knowing nothing other than 13-7 and 7-13, it makes no difference which one you choose. Without other numbers to add context, you have no way of knowing who is the better pitcher.

      • Brocktoon

        It’s an incredibly silly exercise, because nobody would ever be stuck only knowing those numbers, but the odds are in favor of the 13-7 pitcher being better than the 7-13 pitcher.

        • http://obstructedview.net Myles

          This is what I’m trying to get at. Of course you’d never only have these numbers, but the number itself clearly DOES have some value. Does is provide a ton of use? No, absolutely not. Is it useless? No way; it’s a (very) crude measure of how valuable a player is to his team.

          • Brocktoon

            But considering we have access to all those other numbers, it is relatively useless.

            W-L can be used as a crude career measure to show a starting pitcher was both good and durable, and that’s about it.

          • Norm

            It’s useless.
            You don’t know if the 13-7 pitcher played for a 115 win team and the 7-13 pitcher played for a 65 win team.

            It’s useless.

            And one more time, it’s useless.

      • http://obstructedview.net Myles

        This is demonstrably false, unless you believe that pitcher winning percentage has no correlation (or a negative correlation) with things like FIP or ERA or FRA or K-BB% or wOBA against (the stats you’d rather have, to be sure). As wOBA against goes up, a player’s winning percentage goes down. It’s also a relatively strong correlation.

        • Edwin

          Yeah. Well…ok.

        • Norm

          If this is true:
          As wOBA against goes up, a player’s winning percentage goes down.
          fine…but the opposite is NOT true:

          As winning percentage goes down, wOBA against goes up.

          Not true.

    • Karl Groucho

      The problem with this, right, is that we don’t only have those things. Why was W-L a popular statistic at one point? Because, in a vacuum, it has the sort of intrinsic appeal you’re referring to. But we’re not in a vacuum, and baseball has progressed beyond hamfisted “statistics” that show next to nothing.

      • Karl Groucho

        (I guess it’s not ostensibly a statistic; it is a statistic, so pardon the scare quotes. It’s just not a particularly useful one.)

      • http://obstructedview.net Myles

        This is precisely my point, though. There is an appeal to wins and losses. They have an intrinsic appeal. They are worth SOMETHING. The amount that they are worth is open to interpretation (I would come down on the side of “not very important at all”). Wins add a non-zero about of information about a pitcher. My hypothetical is an attempt to show that.

        • Karl Groucho

          “Wins add a non-zero a[m]ount of information a pitcher.”

          This is wrong, right? They add a non-zero amount of information about a pitcher *if we’re starting from nothing*. But we’re not starting from nothing, and given what we know today they are detrimental to analysis (and, as we saw with RR last night, potentially how tactical decisions are made on the field.)

          • http://obstructedview.net Myles

            This is where you get into the actual philosophical debate of “pitching to the score” and “grinding out wins,” and I definitely don’t have a definitive answer to whether or not these sort of things exist. They seem likely to exist, if only faintly (and I mean really, really faintly). I think it would be really, really interesting to see whether or not there is a correlation between a players difference in wOBA from low to high leverage situations and a players winning percentage (which would have to be regressed somewhat to account for what a pitcher’s winning percentage would be on a “true-talent” level). That’s pretty far out of scope for a silly argument on a message board.

            • Karl Groucho

              Nothing’s too far out of scope :)! Though the burden’s on you to substantiate that one — and if you can find or cook up something cool I’d be glad to hear about it!

            • Edwin

              slow day at OV? Not enough 49er’s draft talk?

              • http://obstructedview.net Myles

                I didn’t feel up to writing a whole post of content, and yet here we are, with more wins spilled over a stupid statistic than I’ve probably ever spent on any article in my entire life.

                • http://obstructedview.net Myles

                  I meant words, not wins. It infects me!

                  • Edwin

                    Wins. They’ll hurt you, every time.

            • Brocktoon

              I’ve seen a couple write-ups on “pitching to the score” when Jack Morris zealots were arguing for him to the HOF. There was nothing there.

    • Karl Groucho

      And no, QS does not have the same problem as W-L. It has the problem that it is not a particularly nuanced stat — you either have a QS or you don’t — and for that it’s not worth a whole lot. But unlike W-L, it is not in any way influenced by the performance of your team; so for someone who wants a “how did a pitcher do in one game” stat it at least has to tell you *something* about the pitcher’s performance.

      • http://obstructedview.net Myles

        Doesn’t a win tell you something about the pitcher’s performance? Perhaps, that he left the game after 5 or more innings with a lead that was not surrendered?

        • Jon

          A pitcher could leave a game with a 1-0 lead, or a 15-14 lead, so the “win” stat tells me just about as much as a piece of bellybutton lint.

          • Darth Ivy

            you can analyze the composition of that lint and know the type of clothing that person wears. Then, it’s possible to make other sound assumptions based on that clothing, like what the FBI does.

            You can find out a lot based on someone’s bellybutton lint.

            • Karl Groucho

              The fabric of bellybutton lint is highly correlated with a person’s income, actually. I hear Joe Ricketts only has cashmere bellybutton lint ;)

              • Darth Ivy

                see. Bellybutton lint is way more valuable than W/L records

          • Brocktoon

            Well considering there have been thousands of pitcher wins given out while giving up 0 runs over 5+ innings compared to 0 while giving up 14 runs over 5+ innings, I’m not so sure I’d go with your lint crystal ball.

            • http://obstructedview.net Myles

              Thanks, I was about to post something to that effect.

            • http://www.friendly-confines.com hansman

              The problem isn’t that W/L doesn’t tell us anything it’s just that it paints with such a broad brush that Bob Ross couldn’t get it to look good.

              Hell, it probably isn’t even a brush. It’s like using a cheap paint sprayer to paint the Mona Lisa when you have DaVinci standing right there.

              • http://obstructedview.net Myles

                So you do agree that W/L has some value, and are merely arguing about the degree to which it has value?

                • Brocktoon

                  It’s marginal value is roughly zero though.

          • http://obstructedview.net Myles

            And a quality start tells me that a pitcher only had 3 (or less) earned runs, one of two types of runs that are allowed.

            In 2000, Mark Mulder allowed 9 runs (2 earned) over 6.2 innings. He surrendered 10 hits and 2 walks. Quality Start.

            In 1964, Whitey Ford allowed 10 runs (3 earned) over 6 innings. He surrendered 15 hits, including 4 home runs, and he struck out 0. Quality Start.

            People overstate how poor wins and losses describe pitchers. If they were truly meaningless, than why do they show a correlation if just about every statistic that is meaningful?

            • Karl Groucho

              I don’t think anyone is really saying QS is a good stat, just that in its definition it tells you something about a pitcher: he went at least x innings and gave up at most y earned runs.

              A W, for an SP (let’s not touch on its ridiculousness w/r/t RPs), only tells you *about his specific performance* one thing: that he went at least x innings.

              • http://obstructedview.net Myles

                Brett:

                ‘(2) “Yeah, it’s not great, but it’s a good short-hand for how well a guy has pitched in his starts.” First of all, no it’s not. Second of all, if that’s all you’re looking for, there are way better stats out there – Quality Starts, for example.’

                I include QS merely to say that if you are looking for better pitching metrics, there are WAY better ones than QS. If you are moving from wins to something else, why stop at the next worst thing down the road?

                • http://fullcount1544.blogspot.com FullCountTommy

                  Because with some archaic thinkers, baby steps are the best way to handle things

                • http://www.bleachernation.com Brett

                  Quality Starts might be the next worst thing down the road, but “down the road” is about 800 miles and 15 bathroom stops from W/L. If you’re looking for a quick and dirty, per-start figure, that’s the lesser of the two evils – again, by 800 miles.

                  • http://obstructedview.net Myles

                    Why are you taking a bathroom stop every 50 miles? How fast are you going?

                    • http://www.bleachernation.com Brett

                      That’s how – eh hem – shitty the W/L stat is.

                    • http://obstructedview.net Myles

                      Ok, you win this round.

                • Karl Groucho

                  Ah, I didn’t really read Brett’s whole piece, guilty face. To your point, though: I don’t think the answer to “QS isn’t progressive enough” is “let’s be more regressive!”

                  I also think, though I’m not speaking for myself at all, that people like the notion of quantifying how well a pitcher did in a given game. (Probably out of some intuition that maybe he’d “play to the context” or something.) If that’s your bag, then QS does tell you more than W-L, see e.g. what we’ve been posting here.

                  Though I’d just recommend looking at xFIP to see how well a pitcher has done and IP to see over how many innings he’s sustained it.

                  • http://obstructedview.net Myles

                    I’m not trying to be more regressive. I personally think that winning percentage is a garbage stat that is pretty much useless. I am a) trying to play devil’s advocate here, and b) saying that while it’s pretty much useless (because other stats are a lot better), it is definitely NOT “absolutely useless.” It seems to be if we are trying to pursue a better stat than wins and losses (an admirable goal, and one easily achieved), we should aim higher than quality starts, which tells us more about a pitcher but still not nearly enough. Even vanilla Game Score is a better measure of how a pitcher is doing in a given season.

                    • Karl Groucho

                      I don’t mean to make an ad hominem point there — just that if QS isn’t good because it’s not accurate enough, turning to W-L isn’t the answer! (See the correlation of each to advanced metrics in the CommunityGraphs link I posted below.)

                      And devil’s advocate or no: if the only value in W-L is that it correlates weakly to better statistics, it doesn’t have any intrinsic value. Just look at the better statistics and skip the noise-filled middle step.

                    • http://obstructedview.net Myles

                      Oh, I wasn’t trying to accuse of you that, and I DEFINITELY am not trying to say that W/L should ever be used in any serious context. My whole point is just that if someone says “hey, that pitcher is 14-6 on the year,” that statement DOES have some value.

                      As to whether or not the stat itself has some intrinsic value, I kind of just made a comment earlier on it. I’m not sure whether it does or not, and I think it’d be really interesting to see a study on it.

                    • http://www.friendly-confines.com hansman

                      “My whole point is just that if someone says “hey, that pitcher is 14-6 on the year,” that statement DOES have some value.”

                      Some value…not much but some.

                      We know that he appeared in at least 20 games but don’t know if he was on a crappy offensive team so 14-6 is good or if he was on a good offensive team so 14-6 isn’t that good.

                      I think that is the whole problem folks have with W/L…2/3rds of the stat depends on things that are completely out of the pitcher’s control and with determining a player’s talent/abilities/production, you should try to remove factors out of their control as much as you can.

            • Brocktoon

              It’s a fairly incidental correlation. Look at this top 10:

              Pete Rose
              Carl Yazstremski
              Hank Aaron
              Rickey Henderson
              Ty Cobb
              Eddie Murray
              Stan Musial
              Cal Ripken
              Willie Mays
              Barry Bonds

              Now that stat seems to correlate really well with some of the best players in baseball. The stat is games played.

              • http://obstructedview.net Myles

                Good players play in more games than bad ones. If you told me one person played in 5000 career games and one person would play in 50 games, which one would you assume is better, or had the better career?

                • Brocktoon

                  Like I said it’s incidental. It tells you who was likely better, but it doesn’t tell you much.

                  • http://obstructedview.net Myles

                    Like I said, it doesn’t tell you much, but it tells you something. Don’t misinterpret my argument to be that W/L have a lot of value. My argument is that they have SOME value. They are not useless.

                    • http://www.friendly-confines.com hansman

                      Then why wouldn’t you use IP?

                    • http://obstructedview.net Myles

                      IP is another, different, stat, that while not very useful, does actually tell you something about a player’s value both on an individual and career basis.

                      Also, my previous sentence has entirely too many commas.

        • http://fullcount1544.blogspot.com FullCountTommy

          There is only one thing that a win tells you about the pitcher’s individual performance: He went 5 innings. That is it

          • Chad

            Only if he’s a starting pitcher. If he’s a reliever it just means he was in the game (or just finished pitching before another member of his team threw a pitch) when his team took the lead.

            • http://fullcount1544.blogspot.com FullCountTommy

              Duh brain fart. Just shows even more how much of a failure the stat is in measuring the pitcher’s individual performance

              • Darth Ivy

                It’s really annoying when a closer comes in to get a save, blows the lead, but then gets the W because his team won it in the next inning (unless it’s the Cubs, then I’m just happy they got the win)

              • Karl Groucho

                Yeah hah — it’s such a ridiculous stat as applied to an RP I didn’t even think of it that way! I haven’t seen anyone banging the W-L-as-useful-for-RP-analysis drum though.

          • Karl Groucho

            ^This

            You’re certainly not, for what it’s worth, wrong that there is correlation between W and more advanced statistics; nor that there is correlation between QS and more advanced statistics.

            This read might be of interest to you: http://www.fangraphs.com/community/is-using-wins-quality-starts-the-answer/

            • Karl Groucho

              But if at the end of the day we’re saying “look W-L isn’t all bad, it correlates to good metrics!” then, uh, just look at the good metrics :)

          • DocPeterWimsey

            And that’s true only for a starter: for a reliever, it only means that he was still considered the pitcher when a lead was taken. Whether this was a blown-save/hold (i.e., a “vulture” win) or a case where he shut the door while his batters caught up, or he faced one batter to end the inning in a tie game that got untied, it’s all the same.

            This is where a “win-probabilty added” stat is better than wins, holds or saves: how many times does a reliever leave after having improved his team’s chances to win?

            For starters, the old “win share” statistic would be a big improvement.

            • http://fullcount1544.blogspot.com FullCountTommy

              They called me the snake in college, I had a tendency to give up 1 run leads, only to go on and get the win myself ;)

            • http://obstructedview.net Myles

              Oh, I’m more-or-less willing to punt on reliever wins. I should have clarified, and that’s my fault. Relief wins are so much more situational as to be less valuable by probably an order of magnitude.

              • TWC

                But if you’re willing to disregard (“punt” on) reliever wins, seems like you must disregard SP wins, too. Is there any other baseball stat that only applies to some players, not others? I can’t think of one. Seems like it’d be a pretty useless stat if it weren’t able to be universally applied, no?

                • http://obstructedview.net Myles

                  Can hitters accrue saves? Doesn’t a starting pitcher have to acquire wins in a different way than relievers (by pitching at least 5 innings)?

                  • TWC

                    “Can hitters accrue saves?”

                    Yeah, if they’re given a chance to pitch, they sure can.

                    “Doesn’t a starting pitcher have to acquire wins in a different way than relievers…”

                    Yeah, they do. Which furthers the absurdity of a W/L stat for pitchers.

                    • http://obstructedview.net Myles

                      And why do stats have to be universally applied? Samardzija is having a terrible season this year, his on-base percentage is only .118! Since he also bats, that stat must apply equally to him!

                      That statement is absurd, and so is the statement that you can’t meaningfully differentiate starting pitcher wins and losses and relief pitcher wins and losses.

                    • TWC

                      “Samardzija is having a terrible season this year, his on-base percentage is only .118! Since he also bats, that stat must apply equally to him!”

                      It sure does. Samardzija is having a terrible season — at the plate. Is that really in question?

                      Determining W/L differently for relievers and starters is as absurd as calculating batting average differently based on which inning you’re hitting in.

                    • http://www.friendly-confines.com hansman

                      “And why do stats have to be universally applied? Samardzija is having a terrible season this year, his on-base percentage is only .118! Since he also bats, that stat must apply equally to him!”

                      Maybe that is an argument for the DH in the NL.

                    • http://obstructedview.net Myles

                      I guess my question to you would be why is it as absurd as calculating batting average differently based on the inning. It seems reasonable to me, especially because wins for starters has the added value of displaying a modicum of longevity.

                    • TWC

                      “I guess my question to you would be why is it as absurd as calculating batting average differently based on the inning.”

                      Because if the *definition* of the stat is different based on when it occurs during the game, or the position being played (RP/SP), then the stat has no value, because it isn’t universal. Some stats are calculated (BA, wOBA), and others are granted based on a specific set of conditions (saves, HRs). But no other stat has qualifiers associated with it (*he* has to pitch 5 innings to qualify for this stat, but *that guy* only has to make ONE pitch to qualify). If there’s no consistency, there’s no value.

                      “It seems reasonable to me, especially because wins for starters has the added value of displaying a modicum of longevity.”

                      That seems a pretty weak last grasp at trying to make a pitcher’s W/L numbers relevant.

                    • http://obstructedview.net Myles

                      Again, I am genuinely curious as to why the “universability” of a stat has some standing as to whether or not the statistic is valid. Why does this rob the stat of all of it’s value?

                    • http://www.friendly-confines.com hansman

                      How much weight would you put in HR totals if they moved the fence into the infield dirt for pinch hitters?

                      That is what is happening with reliever wins. They only have to face 1 batter to qualify for a win. Starters have to face a minimum of 15.

                      Really, it’d be like moving the fence to somewhere within bunting range. (Assuming an average OF wall depth of 425 feet, it’d have to be moved to 28 feet from home for pinch hitters)

  • tfilarski

    When did a pitch count of 100 become the standard of when a pitcher should be taken out? Let’s just take every pitcher out at 50 pitches to “save his arm”

  • Titan

    As long as we’re calling out bad stats, I wouldn’t lose a second of sleep if I never heard about winning a regular season series again (sorry, Brett). Namely, I don’t care how a team amasses their win total at the end of the season, just was it enough to make the playoffs or not? Stats like these seem to imply that a particular team can “turn it on” when the match-up requires it. The distribution of wins probably has more of a stochastic element to it. Furthermore, a team could win 5 of 9 games taking the only one game of a series, sweeping the next series, and winning only one in the next series. In this manor, a team could lose 2 of every 3 series and still having a .555 winning percentage. Like pitching wins, I think these stats are important to fans because they help build a personal narrative which often does not correlate to team wins or winning championships.

  • fossilhippie

    Win totals matter lots to a pitcher if his contract has bonus clauses for them, which an awful lot do have.

    • Brocktoon

      Zero pitchers have contracts with bonus clauses for wins.

    • http://www.bleachernation.com Brett

      Actually, I believe you cannot tie bonuses to counting stats like that, per the CBA – only stat-neutral things like games played, games finished, time on the DL, awards, etc.

Bleacher Nation Privacy Policy and Terms of Use. Bleacher Nation is a private media site, and it is not affiliated in any way with Major League Baseball or the Chicago Cubs. Neither MLB nor the Chicago Cubs have endorsed, supported, directed, or participated in the creation of the content at this site, or in the creation of the site itself. It's just a media site that happens to cover the Chicago Cubs.

Bleacher Nation is a participant in the Amazon Services LLC Associates Program, an affiliate advertising program designed to provide a means for sites to earn advertising fees by advertising and linking to Amazon.com.

Google+