I'll agree with 3rdand15 and Shade on one thing. I would take Reyes over any leadoff hitter in the game at this point. Not Damon though. I'd take Ichiro over Damon in a heartbeat. He is better than Damon in every facet of the game. I'll take king's side that I would take Ichiro over Dunn in a heartbeat as well. As far as the OPS thing, I think you guys are arguing apples and oranges. King is saying it's not the be-all-and-end-all of the argument. 3rd is admitting that, just saying that it is a very realistic stat to use to begin to gauge a player by. Maybe take a step back and not get frustrated with each so quickly. You both seem to be saying the same thing in different ways, and misinterpreting each other. (That's not really the same as apples and oranges, I know, I just couldn't think of anything more witty right now.)
I take Dunn over Ichiro in a second. Dunn is a run producer. If Ichiro walked like Dunn did, and had like a .460 OBP, then fine, a leadoff man in Ichiro with his speed and defense, he might be better than Dunn, but as it is now, there is no way Ichiro is better than Dunn. And OPS is just a clean, easy stat that people can understand and the numbers make some sense. If I said VORP or RC/27 or WARP3 or something, people would freak out. I don't even understand those fully yet.
The whole arguement about who you'd take as a corner outfielder is silly because obviously Dunn would be a horrible option in RF and Ichiro's defense would be wasted in LF. dunn gets a big advantage in offensive numbers and Ichiro gets the nod in the defense and baserunning ability. It's a pretty close call IMO and I guess it depends on the need of the team and whether you believe Ichiro's skills are starting to diminish (especially power) or if he just needs to get out of Seattle.
Taking the player as a whole you could argue for Ichiro over Damon, but it's not true that Ichiro is better than Damon at every facet of the game, since Damon is a vastly better power hitter (his slugging percentage is more than 100 points higher, and he has 60 extra base hits this year, compared to a pretty pathetic 28 for Ichiro). Ichiro is of course far far better in the field, and while Damon is an okay baserunner, Ichiro is better. Ichiro is also a much better fielder than Dunn, of course, and Dunn strikes out too much, but I would certainly rather have Dunn on my team (as FTC noted, however, I couldn't play him in RF, so I'm assuming I have a decent enough rightfielder). Despite the fact that "chicks dig the long ball," people seem to continue to undervalue power hitters, when every available statistic says that extra base hits are much more valuable than singles in terms of producing runs. This is the kind of thing that had the Phillies fans undervaluing Mike Schmidt for years. Of course it would be nice if Dunn struck out as little as Ichiro, and had as high an average, but then he would be one of the best players ever (like juiced Bonds). This also relates to what IMO is the overvaluation of the "productive out" (ones that move up baserunners). Of course productive outs are better than unproductive ones, but if you look at those run expectation tables that 3rd linked to (I've linked to others as well recently), you can see that a productive out is in a real sense a little bit worse than the "average" play in a game; that is, the average number of runs scored before a productive out is a little more than the average number afterwards. So, for example, teams have scored an average of 0.93 runs from that point on on when there is a man on 1st and no one out, and an average of 0.75 when there is a man on 2nd and 1 out, showing that the "productive out" has actually hurt. Others: man on 2nd, 0 outs, 1.15, man on 3rd, 1 out, 0.99; men on 1st and 2nd, 0 outs, 1.58, men on 2nd and 3rd, 1 out, 1.44. You see this pattern repeatedly if you look at those tables. This is why the sacrifice bunt is usually not a good play, unless the batter on deck is better than the batter at the plate (which is why bunting with the pitcher is perfectly sensible). Of course there can be individual circumstances where such a play might be desirable, but on average these productive out plays don't really leave a team in a better position than the one in which they started, despite the fact that "baseball men" are in love with them. I certainly don't feel that they can in any way count for more than the benefits that come from being a power hitter. What I think people have to remember is that comparing a productive out to an unproductive one ignores the possibility that something good will happen if you hit away (this also argues for having a hitter in your lineup batting second who naturally hits the ball to the right side, so that they're not automatically "giving themselves up" if they hit that way).
I think alot of it has to do with the "specialization" and compartmentalization that has become more and more prevalent in baseball. People feel like bullpen pitchers have to have clearly defined roles and this is rarely deviated from. For example, the "setup man" will come in the 8th and pitch to the 3-4-5 hitters with a 1 run lead, while the "closer" will pitch to the bottom of the order in the 9th with a 3 run lead and the closer will get the save. Logic would say you would want your best pitcher pitching the 8th but this is never done(unless the closer is pitching the 9th as well). This past offseason we heard about how the Yankees needed a leadoff hitter, and that they had to go after Damon, despite the fact that Jeter was a very good(and better then Damon) leadoff hitter in 05. In this same vein I think that way too much is made of having a guy that makes "productive outs" hitting second. Ignoring the fact that they play totally different position, Adam Dunn and his 200 K's hitting 2nd in the Mets lineup would undoubtedly produce more runs then Lo Duca's .300+ average, productive outs, and 30 K's. The bottom line is that better hitters in your lineup are going to result in more runs, even if they don't fit the "mold".
I think the choosing between Ichiro and Dunn is a more case to case basis, if your team needs a cleanup hitter or more power they will obviously take a younger Dunn, just like if you need a leadoff hitter with defense and speed you would take Ichiro, this reminds me of the trade the White Sox did years ago when they traded Carlos Lee away for Podsednik because they saw they had power but needed a player to start up their offense, it all depends on the teams philosophy
And as a fan you have to hope that your GM isn't the kind of guy that makes deals like that(on the White Sox side). That was, is and will always be a terrible move. Obviously you can't attribute all of this to the Lee/Podsednik swap, but the White Sox runs scored dipped 124 runs from 04 to 05. Of course, they allowed 186 runs less then they did in 04, won 99 games and won the WS, so the trade was proclaimed a "win-win". If the trade was made after the 03 season you could excuse it, as Podsednik had a pretty good season in his first full time action in the majors, and Lee had a bit of a down year for him. Chicago apparently disregared the putrid 04 season Podsednik had.
Okay, Damon is a better power hitter, I'll give you that. I was thinking in terms of what the Yankees would be better off with. I would easily put Ichiro in centerfield in Yankee Stadium, and have no problem with it. He can cover at least as much ground as Damon, and has a significantly better arm. Plus, as a leadoff hitter, Ichiro would score more runs than Damon, and be much more of a steal threat. That would open up the pitcher to mistakes that guys like Jeter and Abreu would absolutely abuse. Don't get me wrong, I was one of the people who was happy from day one to get Damon, but to get Ichiro, I would dump him in a heartbeat. I definitely don't underestimate power hitters. If you have the choice between 9 Giambis, or 9 Ichiros in your lineup, you go with the Giambis. You will get more overall run production in the end. However, I think people are underestimating the value of a "for average" hitter. If you have a lineup of 5 Ichiros, and 4 Giambis, and you alternate through the lineup with each, you are going to score a lot of runs. As for your numbers, they are totally skewed, and I'll quickly explain why. Of course there will be more runs scored with less outs, and less for each number of outs after. It only makes sense. With one out, you have one less out to play with. It's the difference between Aaron Guile making the third out, or Damon getting another at-bat. While I see the point you are trying to make, it is totally off-base. A sacrifice bunt is usually not a good play? Even when you are down one, the tying run is at second, one out, and the guy on deck has the ability to get a base hit? I think you are falling too much in love with the modern day long ball game. At it's foundation, baseball is still much more a stretegic game than say, football. I would say baseball is to chess, as football is to Monopoly. In baseball, as in chess, one small move can mean the difference between a win and a loss. In football, as in Monopoly, the key is to choke your opponent into submission.
Why would Ichiro score more runs? They have similar OBP(Damon's is a tad higher) and Damon's extra power(60 XBH vs 28) more then makes up for 14 SB's . But 9 Giambis will still be better then the 5 Ichiros and 4 Giambis. Again, this isn't something that can be debated or "skewed" these are facts. These are huge sample sizes and you can see the numbers are fairly consistent from year to year. It's been noted that there are situations when you have a poor hitter at the plate and a much better hitter on deck a sacrifice can be the right play. But for the most part when you start sacrifice bunting you're hurting your ability to score runs.
This is another problem with your argument, you are basing this on who fits the Yankees better, and with the Yankees who already have Arod, Giambi and when they get back Sheffield and Matsui you can argue that they have 4 cleanup hitters in their lineup and they would really have more of a use for more speed and defense then even more power, but if you are starting from the beginning you have to choose a cleanup hitter over a leadoff guy
I'm assuming the example you gave above is a typo (you meant no one out, not one out), since that situation is certainly not a good time to bunt, since there is virtually no difference in scoring a run with two outs and a man on second or two outs and a man on third, and it's very unlikely that a runner on second will be doubled off. I've already had this conversation a couple of weeks ago, and I don't really want to get into it all again, but I don't see what you're saying at all. Put aside all of the "unique" characteristics of an individual play - those obviously change from play to play, are absolutely impossible to quantify, and are certainly very subjective. The point of the game is to score runs, and the expected runs scored tables show how much of that is done. It is simply empirically true that fewer runs are scored on average from the point where a runner is on first with no one out than when a runner is on second with one man out. This has been true going back at least until the 1950s, when George Lindsey constructed the first of the expected runs tables for a paper in the journal Operations Research, and is found consistently year after year, whether it is a pitching era or a hitting era. What this implies is that if you have an average hitter up, and an average hitter on deck, the situation after a successful sacrifice bunt will be worse on average than before it. This also doesn't take into account that many sacrifice bunts are unsuccessful. The same effect is true for a runner on 2nd, runners on 1st and 2nd, and situations with 1 out. This is not my opinion - this is based on the actual numbers of runs scored in games. The situation can certainly be worse after a player swings away, but for an average hitter, that is not the case on average. The cause is exactly what you said - you have one less out to play with. That is exactly why "productive outs" are far less productive than people give them credit for - you only have 2 outs to play with (on the third the inning is over), and you just gave one away. Note that these are just averages, and the specific characteristics of the hitter and the player on deck also matter - obviously bunting with a pitcher is a good idea, because the pitcher has a much higher probability of making an unproductive out. You might make the same argument with a rookie hitting .224 with fewer than 100 at bats followed by a 12-year veteran with a lifetime .290 average (of course, Guiel isn't a rookie, and he hit .294 last year in KC). That wasn't what this discussion was about, however - it was about the value of "productive outs" when comparing decent (or much better than decent) hitters like Dunn, Ichiro, Lo Duca, Lee, Podsednick, etc. For average players productive outs put the team in a slightly worse position in terms of runs scored; for batters of higher quality, they are even more wasteful, since those batters have that much higher probability of doing something much more productive. If you have a regular position player who is bunting all the time because he has a high probability of making an unproductive out, I would say the solution is not for him to bunt, but rather to get a better position player.
Obviously I agree with the second paragraph above, but it must be noted that the first statement is impossible to prove empirically, since no such lineup has ever existed. People have used computer simulations that suggest what you're saying, but that is different from the expected runs tables, which are based on actual game results.
Well yes, when you're talking about "5 Ichiro's and 4 Giambi's" such a discussion has to be theoretical. But I think the point is valid that it's not possible for run production to improve when you replace 5 hitters with less productive hitters, and the other 4 remain constant.
There is a problem here. Of course when you make an out, ANY out, the chances of scoring decrease. But I don't think anyone is saying that moving the runner over is better than getting a hit, which would over value it. It IS however much better than striking out, which is why I'd rather have a guy with theoretically the same stats as Dunn, only half the K's. That player would have a MUCH higher value than Dunn. In your example, yes, the expected runs go down from .93 with a man on first and no outs to .75 with a man on second and one out, it goes down to .56 with a man still on first and one out. That's a huge difference. Your other examples- man on 2nd, 0 outs, 1.15, man on 3rd, 1 out, 0.99 man on second, 1 out, .75; men on 1st and 2nd, 0 outs, 1.58, men on 2nd and 3rd, 1 out, 1.44, men on 1st and 2nd, 1 out, .95. I don't think there is an overvaluation of "productive outs", because no one in their right mind wants an out over a hit. They DO want a ground out to move the runners over a strikeout every time though.
120 of Dunn's 163 K's this year have come in situations where it is not possible to make a "productive out", the whole strikeout thing is way overblown. To do a complete analysis of players ability to move runners up you'd have to go through an entire seasons PBP logs, which I don't think anyone is prepared to do.
"Productive outs" is a fuzzy concept. If a player is trying as hard as he can to get a hit, and ends up hitting a ground ball to second that moves the runner up (rather than striking out), there is no question that this is helpful to a team. It's much easier to see the usefulness of these expected runs numbers if you consider an overt strategy to "give oneself up," like the sacrifice bunt. There is a runner on 1st with no one out. A successful bunt results in a man on second with one out. At that point, the expected number of runs that will score in the inning is 0.75. If it's a failed bunt, there's still a man on first, now with one out, with an expected number of runs of 0.56. I'll keep things simple by ignoring the possibilities of bunting for a hit and bunting into a double play (those are relatively rare events). Say the bunter is a pretty good bunter, and is successful 85% of the time. Then, by calling for a bunt, the expected number of runs after the play is over is (0.75)(.85) + (0.56)(.15) = 0.72; the first part comes from the bunt being successful, the second from it not being successful. Now, what if you have the batter swing away. We'll keep things simple and assume that in that case he either goes to first on a walk or a single (and the runner goes to second), or he has an unproductive out. Note that a decent player is probably more likely to get an extra base hit or a long single than to hit into a DP, so I'm actually giving the swing away position an unfair disadvantage (for a thorough analysis we would need to have the chances of each possible outcome, but these calculations show the basic principles). If he gets on base, the expected number of runs scored is 1.58; if he makes an unproductive out, it is .56. Let p be the chance that the batter will, in fact, get to first base. Note that this is basically his OBP. The expected number of runs after he is up, if he doesn't bunt, is (1.58)(p) + (.56)(1 - p); the first part comes from him getting on base, the second from him not getting on base. The sacrifice bunt is only a better strategy (that is, leads to a higher average number of runs scored), on average, if (1.58)(p) + (.56)(1 - p) < .72. Solve the inequality above, and you see that a successful bunt is only more effective than swinging away on average (assuming an "average" batter on deck) if p < .157; that is, if the OBP is less than about .160. Of course, no position player has an OBP anywhere near that low, which is why the sacrifice bunt (other than with a pitcher) is rarely a good play. An analysis of "productive outs" would involve comparing the distribution of results (strikeouts, walks, singles, doubles, etc.) when a player isn't trying to do productive outs to the distribution when he is. This could be done based on play-by-play data (as 3rd noted), but I don't have that data (actually, a PhD student and I are working on a paper on baseball, and whether there is evidence for momentum within games, and she is getting such data from retrosheet, but I don't have it yet). The sacrifice bunt analysis above, however, suggests that a player should definitely not try too hard to give himself up (since that approaches a sacrifice bunt), because that will actually hurt the team; that's why a lefthanded hitter has an advantage, since he can try as hard as he can and still get the runner over on an out a lot more easily than can a righthanded hitter. The other part of the analysis that would be needed would be to look at how often the situation even comes up. I don't have the data to construct it completely, but even Jose Reyes has only gotten on base 88 times this year with the bases empty, and no doubt some of those were with two outs, so how many times has Lo Duca come up in the "productive out" situation anyway (he has more than 460 plate appearances)? (I recognize that this is not completely the story, since Reyes could have singled in a runner from second with no one out, thereby being on first with no one out, but you see the point, I hope.) This is already too long, but one other thing - I am not claiming that average runs are the only way to make these decisions. The usefulness of these tables, however, is that they give a manager or player the ability to see that there had better be lots of special circumstances going on if plays like the sac bunt are going to be used, since there are strong arguments against them on average.
I don't know if I can even follow this thread any more, hahaha. Too much math involved for me. Seriously though, devil said what I was trying to say, better than I said it. Of course, if given the choice between a single and a ground out that moved a runner, I'd take the single every time. As far as my comment of a runner on second, one out, and a sacrifice, I meant what I wrote. It is the right time to bunt. Your tables can say anything they want, I would still call for the play, especially if it's a fast guy on second. You can argue that the runner may be advanced by an infield out, but that isn't a sure thing. If the ball gets hit to the shortstop, the runner on second goes nowhere. That's an unproductive out. The bunt will almost always move that runner to third, since if the third or first baseman picks it up, he won't stop the runner going to third. The pitcher has a chance, but more often than not, he is better off throwing to first for the sure out. The catcher has the best chance of stopping the runner, only because he can look him off before throwing to first. At that point, you have your tying run 90 feet away. Yes, there are now two outs, but there are other factors that come into play. A wild pitch ties the game. A bloop single ties the game. An error ties the game. Math can solve many issues in the universe. I don't think baseball is one of those things. From my days playing, to my time coaching my kids, I've seen some crazy things happen. That's what baseball is about. If it were all about statistics, Mo Rivera would never blow a save. ARod would never slump. Jeter would create a run any time a runner was within 180 feet of home plate. But that's not baseball. Baseball is about matchups. It's about a certain guy coming up in a certain situation. Thinking about it now, Giambi was a terrible example to use. He gets on base too often, and doesn't strike out often enough. A team of 9 of him would be huge, simply because when 2 or 3 of them had a HR in the game, they would score 6-8 runs. On the other hand though, there would probably be a ton of errors, and you would never see that team catching balls on the warning track, those would be doubles. A team of Ichiros would grind out games. They would rattle pitchers, since they would always have an eye on the baserunner. If you have a team full of high-average players, regardless of power, that team is going to score a lot of runs. It won't be very flashy, but at 6a.m., the paper only shows W or L for the night before. Now if you mix those two types of players, like I tried to say, you have a great lineup. You have baserunners at least once an inning, and you are going to get homers with runners on, more often than not. If baseball was played by robots, then yeah, I would want a team of Giambis. But it's played by human beings, and so many situational things happen in baseball that it's impossible to confine success to a series of statistics. Stats are of course very useful. However, you have to look at the whole picture.
I said in another thread a couple of weeks ago that I'm always surprised that people who will accept the benefits of statistical analysis in meteorology, clinical trials, finance, marketing, economics, etc., etc., are so resistant to the idea that it can be used in exactly the same way in sports. I will not attempt to try to convince you of my position, but I will point out that what I bolded above is precisely what statistics does not say. It is these analyses that attempt to model and understand all of the randomness that you are talking about, and to try to separate that which is signal (Rivera is a great pitcher, Pujols is a great hitter) from that which is noise (hard-hit balls go right at fielders, dribblers make it through the infield). There is no other way to do it that has the slightest hope of finding what is real, and what is just a fluke, and there is nothing about baseball that is any different from any other endeavor in that regard. The analysis I did here is naive, because I only have simple data, and I was just trying to illustrate the way the statistics can be used. Using play-by-play data you can of course take into account things like matchups, ballpark effects, fatigue effects, and so on - situational effects that matter. This is exactly what STATS, Inc., does, for a lot of money, for all of the teams in major league baseball. You can also sort out what effects don't matter, and are just luck. I would be stunned if you would find many people who would agree that you should bunt with a fast man on second and one man out, unless you have two really pathetic hitters coming up to bat. I certainly can't recall it ever happening this year in all of the Mets games I've watched with Reyes on 2nd. A runner doesn't score from 3rd with two outs too often - it only happens about 1/3 of the time. Runners score from 2nd with one out much more often (more than half the time, I would estimate, although I don't have the data), and they score from 2nd with two outs almost as often as they score from 3rd with two outs. Wild pitches are rare. Errors that stay in the infield are rare. (Maybe the point is that you're thinking of coaching kids, where wild pitches and dropped throws are common, but in the major leagues they're not.) Even if they occur, then it's man on 3rd or 1st and 3rd with two outs, and you still have a chance to score. Bloop singles (and all of the other kinds of hits) score the runner from 2nd anyway with two outs, and so would a poor throw by an infielder that gets past the first basemen or a missed catch by an outfielder, since the runner is running on contact with two outs. It is poor strategy to play for a rare event in favor of a common one.
Don't get me wrong, I am not saying stats don't have a place in the sport. That would be flat out ignorant. If stats didn't matter, then we wouldn't hear insane things like "This guy hits this pitcher once every 3 at-bats, during night games, when the temprature is between 68 and 72 degrees." (Yes, that was silly, but the exaggeration illustrates the point. My writing has nothing to do with my kids. I simply used them as examples. I think part of my "problem" is that I am too old school for the modern sport of baseball. I still see the game in the more pure pre-strike form. The sort of time when your power was locked into the 4,5 positions, and sometimes, if you were really lucky, in the 3 or 6 holes. A guy like Rickey Henderson was an abberation, rather than the rule. Pitching today cannot keep up with modern day hitters. There's too much film to study. And let's be honest, steroids does far more for a hitter than a pitcher. Powerball is the game today. If I can get over my prejudices, perhaps I can see it from a different angle. I don't think that will ever actually happen though. When I played, it was just different. Not in an "I walked to school barefoot, in the snow, uphill, both ways" sort of way, just different. Baserunners were critical to a team's success. A game ending 7-6 was jokingly referred to as a football game. Now a score like that is a slow night. I hope I am not offending you with my comments about mathematics. I'll admit, I hate math, and always have (which everyone I know thinks is strange, considering I am a programmer *shrug*) That doesn't mean I don't believe it has it's own important place in the world. Actually, I think math, in its various forms, is the basis for everything else. Anyway, I just wanted to let you know it's nothing personal. BTW: Wasn't this an ARod thread? :breakdance: