<< <i>Dallas - we will agree to disagree about Pedro/Clemens, but now I've got a question about Clark '89 vs. Mattingly '86 (which was much better than his 85 season)
Year Ag Tm Lg G AB R H 2B 3B HR RBI SB CS BB SO BA OBP SLG *OPS+ TB SH SF IBB HBP GDP +--------------+---+----+----+----+---+--+---+----+---+--+---+---+-----+-----+-----+----+----+---+---+---+---+---+ 1989 25 SFG NL 159 588 104 196 38 9 23 111 8 3 74 103 .333 .407 .546 175 321 0 8 14 5 6
Does Clark '89 win solely on OBP? If 90% of a hitter's value is OPS, as you stated earlier, where does Clark make up the differential between his .953 and Mattingly's .967? I suppose you'll say Mattingly compiled his numbers in 742 PA vs. Clark's 675 PA, but having more PA won't change either of their OPS. The only significant edge Clark holds is GIDP, which is function of PA, and that is partially offset by his 3 CS. >>
Mostly, the difference is Candlestick. I think Clark's stats are more impressive in that context (note that in OPS+, Mattingly's raw 14 point OPS advantage is reversed). GIDP is also part of it; Clark's edge is much greater than the PA ratio. And Clark's 3 CS cost him less than he gains from his 8 steals. And beyond just SB and CS, Mattingly was a truly terrible baserunner whle Clark, in his prime, was very good. Over the course of their careers, Mattingly was a much better forst baseman, but Clark was good enough in his early years that Mattingly gets no real advantage there. I think Clark's edge over Mattingly is actually pretty substantial all things considered.
This is for you @thisistheshow - Jim Rice was actually a pretty good player.
this assumes that with a runner in scoring position second a hit or walk that does not lead to an RBI is not a failure
Maddux had two strike years that would have been better than Clemens had he continued at that level for the full six months. Since they total together more than one and a half seasons worth, I wouldn't hesitate to give him enough credit for the five month season to move him the top spot for the 90s
Looking at the earlier years, Dizzy Trout is absolutely not the best from the 1940s. Williams and DiMaggio were fighting in a war and 16-year-olds and guys with one arm were playing in the outfield. It doesn't matter how good you are relative to the rest of the league when the rest of the league is made up of guys who aren't the best players. Best single season for that decade has to go to Feller
<< <i>Maddux had two strike years that would have been better than Clemens had he continued at that level for the full six months. Since they total together more than one and a half seasons worth, I wouldn't hesitate to give him enough credit for the five month season to move him the top spot for the 90s >>
<< <i>Looking at the earlier years, Dizzy Trout is absolutely not the best from the 1940s. Williams and DiMaggio were fighting in a war and 16-year-olds and guys with one arm were playing in the outfield. It doesn't matter how good you are relative to the rest of the league when the rest of the league is made up of guys who aren't the best players. Best single season for that decade has to go to Feller >>
It's way beyond the scope of what I tried to do here to see which players or pitchers were on pace for great seasons 2/3 of the way in and guess what might have happened if events had been different. It is certainly possible that Valenzuela would have the greatest season of all time in 1981 if the strike hadn't interrupted, just as its possible that Maddux' arm would have left him as the 1994 season wore on. Who knows? I don't, but I do know that what Clemens did in 1997 had more objective value than what Maddux did in 1994, so I gave him the title.
Same thing with Trout. The position that winning games in 1944 doesn't mean as much as wining games in 1940 (Feller's best) is an opinion that I can't say is wrong. But 1944 is in the record books on the same status as 1940 so Trout gets the nod from me. How would Trout have fared that year if everyone was there instead of in the Army? I don't know, so I don't penalize Trout based on an assumption that might be wrong.
<< <i>Dallas looking closer at some of your choices I would like to know why you chose Foxx's 33 season over his 32. >>
Picking the best of the 1930's was a guess; there were several seasons that were essentially the same. The park factor shifted dramatically from 1932 to 1933, turning Shibe from one of the top hitter's parks to one of the worst; I don't know what happened that year, but Foxx's numbers look very good in context. But his 1932 season, or Gehrig's '34 season, or Medwick's '37 season are all just as good. Where I had choices I thought were equally good, I picked the one that looked strange because that's the way I am.
This is for you @thisistheshow - Jim Rice was actually a pretty good player.
TheVon, there is no need for a formula, nor guesswork on how many times a player failed. Every one of Rice's plate appearances are logged into the play by play data. It is known exactly how many times a player hit in every base and out situation.
If you look at the situational batter runs, Rice gets the following for his career...
If you ignore how Rice hit with men on base, and ignore his park factor, he had 270 batter runs above league average player.
If you include how well he hit with men on base, then he has 284 batter runs above league average player. So you are correct that his men on hitting was decent, though not great.
If you include the park factor to those above, then his value would be only 190 runs above average. He is a bit unfairly penalized due to the fact that it is harder for runners to score from first or second at Fenway.
His 1978 season is 60.4 situational Batter runs NON park adjusted, and 47 park adjusted.
For comparison, Morgan's 1976 season is 84 situational batter runs, and 80 park adjusted.
I am in the camp that Fenway probably didn't affect his batting to such a high degree that these numbers bear out, but they certainly did help a good amount based on the overwhelming objective evidence.
In 1995 Major Legaue Baseball played 8/9 of a 162 game season (not 2/3), but they fit it into only 5/6 the time. With far fewer days off, a pitcher who can still take his turn in the rotation every time becomes more valuable. Because the Braves were so far ahead and because their other pitchers were so durable they didn't have to push their ace every fifth day like other teams tried. Even without projecting another month, it might be reasonable to see as helping his team in a quirky season more than what Clemens did in a normal season
I like Guidry's '78 season too. He still has the lowest ERA ever in the DH league at 1.74. He beat Carlton 208 to 182 in ERA+.
Yeah, he trailed Carlton by a good amount of innings, 346 to 276.
If you look at their pitching Runs, which factors IP as well as ERA vs. league ERA, Guidry wins 59 to 55.
In the Clemens/Maddux race, Clemens beats him 73 to 61 in pitcher runs.
No doubt the strike robbed Maddux. Dallas, sure it is possible he hurt his arm, or did awful in those missed starts. It is also possible that he threw five shutouts. Logically speaking, the most likely scenario is that he would have pitched pretty close to the rate he had that season and the two sandwich seasons. It is perfectly acceptable to assume this and give Maddux his due, just as it is perfectly acceptable to do the same with Ted Williams's missed seasons. If Maddux were missing those games due to being a wuss, then that is a different story.
Plus that fat toad Clemens was juicing.
The verdict is that Guidry and Maddux need to replace Carlton and the toad.
Dallas, logically speaking again...call me crazy, but Dizzy Trout facing the likes of Catfish Metkovich and Bud Metheny instead of Ted Williams and Joe Dimaggio, just isn't quite as impressive, and most certainly needs to be taken into account.
Regarding Feller and Maddux, we're just debating apples and oranges. If I factored in "likely" and "assume" then yes, I would have put Feller in that spot. But I didn't. I do not disagree that a list that did would be just as interesting, though. On the same note, if I knew that Clemens cheated in 1997 then I would not put him there, but I don't.
Regarding Guidry, I just disagree. Not just a little bit, either. I think Carlton completing 30 games and winning almost half of his godforsaken team's games was historically great. Add to that he was in the NL and had to bat - and hit almost .200 - and he blows Guidry away. In fact, I would need convincing that Guidry had a better year than Gaylord Perry's 1972 season or Wilbur Wood's 1971 season.
Regarding Rice, I never said he didn't have a great season in 1978. But it's the only great season he ever had among five or so good seasons and a whole lot of mediocrity, and if you think he had a better season in 1978 than Dave Parker then you'll need to show me why. Parker's OPS+ is better, he was 20 for 27 stealing bases (Rice 7 for 12), he grounded into 7 fewer DP and he won a Gold Glove. And he hit better in RISP and men on situations. Rice's only advantage is that he played every game, and I don't think it makes up for everything else.
This is for you @thisistheshow - Jim Rice was actually a pretty good player.
Dallas did you forget the the 78 yankees were 14 games back of the Sox and would have went no where had it not been for Guidry's 25 wins? Add in his shutouts and k's and sub 2.00 era and a compelling case can be made.
The 25th win by the way was against the Sox in the famous 163 rd game.
The one that is called the Dent game.
On a summers night he struck out 18 Angels.
Carlton did have an amazing season in 72.
He also did not have a Sparky Lyle and Goose Gossage in his bullpen either.
I'm not sure how you can say that about Guidry. He saved 59 runs over the average pitcher, Carlton 55, Wood 54, and Perry 53.
True, Carlton had to bat...but Carlton also had the luxury of being able to extend himself longer into the games by not having to pitch in a higher scoring environement(specifically facing a pitcher with three cupcake at bats). Being that his extra value is coming from his IP, this is a factor too.
If two pitchers are equal in value based on their IP and ERA vs. league, then I take the guy who got there by virtue of the stronger ERA+ as opposed to having the more IP because he was allowed(or not allowed) to pitch more innings, as that can be a mangers fault or decision.
None of that matters much anyway, as Guidry was responsible for saving more runs for his team, thus better.
<< <i>I'm not sure how you can say that about Guidry. He saved 59 runs over the average pitcher, Carlton 55, Wood 54, and Perry 53. >>
I'm usually wrong when I think you are, skin, but where are you getting these numbers? Using b-r.com, I get Carlton 62, Wood 63, Guidry 57 and Perry 51.
I'm using (*ERA+ - ERA) * IP / 9 or (mathematically equivalent) (*ERA+ - 1) * ER. Maybe your source and b-r.com are using different park adjustments? Or are you using something from the linear weights system with which I'm not familiar?
And I'm not offended if anyone wants to vote for Guidry - he had a hell of a season. I'm just more impressed by what Carlton did. And in those days I was a Cardinal fan until they were out of the race - usually sometime in June - and then I started rooting for my favorite player's teams, which included the Yankees since Reggie was there. Other than Yaz, I didn't really like the Red Sox much, so I was glued to my TV rooting for Guidry in game 163. And Reggie's "oops" play between first and second in the WS that year is still my all-time favorite play. Which is all to say, I have no bias against Guidry - I'd put him in the HOF twice if it would keep Jack Morris from streaking the HOF bowl - I just think Carlton's year was even more awesome.
This is for you @thisistheshow - Jim Rice was actually a pretty good player.
I just grabbed them from Total Baseball, same formula. Probably different park adjustments.
Just looked them up as to why they differ. Total Baseball has Guidry's park factor at 97, and baseball-reference at 96(both three year and single year).
The big difference lies within Carlton's. Total Baseball has his park factor at 99, while Baseball-Reference is at 104 which is the three year average. Baseball reference has the one year park factor from 1972 as 99 as well.
So using the one year park factors for the year of 1972 puts Guidry up. Using the three year average puts Carlton up.
If one believes that the defense was a big factor, then it could swing the tide a bit...though I don't think the defense factored as much on these two strikeout pitchers as some sabermatricians may believe.
<< <i>I agree about Brett; had he not been injured he was on pace for the seaon of the decade. But he was hurt, and missing 45 games left him a mile short. Still, even missing 45 games, he was about as valuable as Rice in 1978. >>
Brett played in 42 less games than Clark, but still had more HR and RBI's. How can you ignore the .390 with tons more power than Clark?
Also Brett much better in 79 than Morgan in 75(50 extra base hits is best season of decade?)
Look, nothing against George Brett, but nobody short of Babe Ruth ever did enough in 117 games to be as valuable as Will Clark in 1999. Brett came close, and he would have done it if he hadn't gotten hurt. On the if-things-were-different-they-wouldn't-be-the-same list, Brett is a runaway winner; but not on this list.
Brett in 1979 had an on-base percentage of .376, Morgan in 1975 had a .466. Morgan stole 67 bases and got caught 10, and helped his team a lot in the process; Brett stole 17 and got caught the same 10 - and hurt his team in the process. Brett hit into 8 DPs, Morgan 3. If Brett hit about 30 HR more than Morgan he would just about make up that chasm - he hit 6 more. Bett had a good year, but it wasn't the best in the AL in 1979 and it wasn't even in the same ballpark as Morgan's 1975 (or 1976) season.
This is for you @thisistheshow - Jim Rice was actually a pretty good player.
I'll be the first to admit that I don't know the right answer to this one, or if there is a right answer. Veterans Stadium was what it was in 1971, 1972 and 1973 - and it was exactly the same in all three years except for a little tinkering with the heights of the fences. B-R.com uses three-year averages and Bill James uses five-year weighted averages for park factors and I tend to think that they are correct to do something besides rely on a single year since a sample size of one year may not be nearly large enough, especially since rosters change and we're trying to measure the parks and not how a particular one-year roster hit in that park.
From 1971 to 1973, Veterans park factors went from 105 to 99 to 108 while the fences got steadily higher eah year. There is no logical explanation for that, and I think a park factor of 104 (the average) is probably a much better indicator of the kind of park Veterans was in 1972 than 99. James method would get something between 102 and 103 and that may be better still (I'm not sure how James deals with the fact that Veterans opened only one year prior to 1972).
The same thing happened, even more dramatically, to Wilbur Wood. 1971 represents a freakish dip in the park factor for Comiskey and there is absolutely no objective reason why that should have happened. I don't think it did "happen" in any meaningful sense, it's just random noise in small samples. Comiskey and Veterans were big hitters parks, and Carlton and Wood should get credit for how well they pitched in them. As you noted, Yankee Stadium was steady over that period so it doesn't matter what method we use.
So I'm sticking with Carlton, but I've learned something along the way and that's always a good thing.
This is for you @thisistheshow - Jim Rice was actually a pretty good player.
Dallas, three year park factors do have merit, but one year also have merit. What could make the vet be a 99 PF in '72, and 104 the next?
It could just be randomness, but it also could be more cold days while they were at home in '72. Just think of April and May and how it is hit or miss with the weather(probably not in Texas so much ) It is possible to catch a string of cold when at home, thus could hamper some runs.
Same with wind factor for parks. One season may find the luck(for pitchers at least) that the wind is blowing harder and in, as opposed to harder and out. Carlton himself was nearly identical at home/road in '72, as were the rest of the players who played there.
BRETT VS. CLARK
Situational Batter Runs:..........Brett 73, Clark 74. Situational Stolen Bas Runs:...Brett 1,...Clark 1
Actually, Brett did accomplish as much in 117 games as Clark did in a full season. Well, one run short. This is measured as above league average player. Both hit well in high leverage and late/close situations.
George Brett's .664 SLG% is the highest of anyone in the 70's and 80's. His OPS+ was also higher than anyone in the 70's or 80's.
But with Brett not being able to answer the bell for 45 games, it meant that a reserve had to replace him. So intertwined in Brett's value to his team, is the value of what a reserve player would contribute.
Dave Chalk was one of the primary replacments at 3B, and he checked in with a -7 situational batter runs Jamie Quirk was the other, and he had -4 situational batter runs.
You can't fault Brett for how good/bad a reserve is(that is the GM's blame), but those two guys look to represent what a typical reserve player would do... maybe a bit worse than typical.
Including the value of his replacements, you get the following value of Brett and Clark.
Brett 62 Situational Batter Runs Clark 75 Situational Batter Runs
The fact that Brett missed those games is no longer needed as a factor, as that aspect is factored right there with poor replacements.
DEFENSE: Brett had 22 Fielding Runs, Clark 9. This is BaseballProspectus, and defensive stats comes with its usual caveat. Knowing Brett outdistanced his 3B peers in fielding runs more than Clark did his 1B peers, and taking into account that Brett did it in a harder defensive position, I can certainly see that ten run deficit being made up on the defensive spectrum. This 13 runs puts them at even.
Conclusion: When you included what Brett's poor replacements did in his 45 game absence, and found out that Brett was only 13 Runs behind Clark in offensive contribution, and then took Brett's superior defense at a harder position into account, I do see Brett claiming the title as the best season of the 80's.
P.S. POST SEASON: Usually this isn't a fair thing as teammates decide whether or not you even get a crack, and there are so few games that results are a crapshoot, but Brett had a fabulous post season, Clark had one great series and one bad. Brett's post season play was a continued glimpse of his worldly season...and he made hemroid medicine famous.
I have a greater appreciation for how great Brett's 1980 season was, and if its any consolation I'll move it up to second on my list (past Sandberg's 1984 season). But that's as far as I'm going.
And I don't question that the numbers you're using are accurate, I just place a lot less credence in that system than you do. Implicit in the situational-batter-runs is the assumption that 100% of a player's "value " consists in how far that player is above average. But that's simply not true - MOST of a player's value is captured by his accomplishments that bring him to average. In other words, a player that is average has significant value, but that system says he has none. Brett was farther above average when he wasn't hurt than Clark was, and that skews the results in his favor; all of the games that Clark played that Brett did not are correspondingly undervalued.
Mathematically (and very loosely) for anyone who isn't following what I'm saying: say that the average player has the arbitrary value of 50 and that Brett's value was 90 when he played and that Clark's value was 80 when he played. I think that means that Brett was 12.5% better than Clark (90/80); the system skin is using says that Brett was 33% better than Clark (40/30). That's a HUGE difference and I don't think it's the correct way to compare two players. I'm oversimplifying and presenting only one side, but in the end this problem remains for me in that system. In this example, 62.5% of Clark's value consisted of being average and 55.6% of Brett's value consisted of being average, and I think a system that simply throws away those majorities before comparing is fundamentally flawed.
I know I'm not going to convince you of all of this in a single messageboard post - I'd lose a little respect for you if I could - but I did want to make clear that I'm not merely being stubborn. I think the stats back me up that Clark had the best season of the 1980's.
This is for you @thisistheshow - Jim Rice was actually a pretty good player.
The absolute goal is to see if 117 games of Brett at his pace, is better for a team than 160 games of Clark at his lower pace. This method nails it!
When comparing two players who are both above average, what you are saying does not come into play. It is when one is at the average baseline(or just below) where what you are saying comes into play. This happens more with career value. Currently, average would be zero runs. They are both 75 runs above average. If you were to measure them against a baseline of -15, they would still be the same, but 90 runs above. If George Brett were just FOUR runs above average for 117 games, and Clark zero above for 162 games, THEN your objections would be valid, because no matter how many games Clark played, it would seem as if Brett had more value. This is not the case in this analysis.
In this case, I gave Brett the stats of his replacements who were bad, and it shows that what he did in his 117 games + what a typical replacement player would be, actually showed that his season is better for a team, than Clark's(when adding defense). WE don't have to guess, we know.
In laymen's terms in Brett's 117 Games compared to Clark's 159, look at what they gave their teams...
Singles-...Brett 109 Clark 126 doubles...Brett 33....Calrk 38 Triples.....Brett 9......Clark 9 HR...........Brett 24... Clark 23 Walks......Brett 58....Clark 74
Outs Made Brett 298...Clark 409
That is the value of both their contributions, and is summed up more accurately in the Situational BR where it says Brett gave 73 Runs, and Clark 74. However, since Brett didn't play all the time, his team had to fill his void with replacments. The replacements playing make his team for the worse off. I gave what the poor replacements did in Brett's stead and assigned it to come up with the value of Brett's not full time role. By doing this, it eliminates your beef. It adds the -11 runs the scrubs contributed.
This method is far more accurate in the real baseball world. What we are looking to achieve is to figure out if 117 games at such a high level of Brett is better for a team than 160 games of a little lower level that Clark provided.
The only question is what can a team typical expect from replacements?
In this case, the split of George Brett, Quirk, and Chalk would win more games than the 159 games from Clark. Quirk and Chalk represent a very low level of replacement. I would say in most cases nearly every team would have somebody as good or better to use as a replacement, thus making it a pretty strong case that Brett's high level of play in the lower amount of games(when adding defense), beats Clark's. And that is the bottom line.
The situational batter runs(though not w/out a few minor problems) are far more balls on accurate as to what really occured, than what any James hypothesis can do, so that data being used is not a problem.
The absolute goal is to see if 117 games of Brett at his pace, is better for a team than 160 games of Clark at his lower pace. This method nails it!
When comparing two players who are both above average, what you are saying does not come into play. It is when one is at the average baseline(or just below) where what you are saying comes into play. This happens more with career value. Currently, average would be zero runs. They are both 75 runs above average. If you were to measure them against a baseline of -15, they would still be the same, but 90 runs above. If George Brett were just FOUR runs above average for 117 games, and Clark zero above for 162 games, THEN your objections would be valid, because no matter how many games Clark played, it would seem as if Brett had more value. This is not the case in this analysis.
In this case, I gave Brett the stats of his replacements who were bad, and it shows that what he did in his 117 games + what a typical replacement player would be, actually showed that his season is better for a team, than Clark's(when adding defense). WE don't have to guess, we know.
In laymen's terms in Brett's 117 Games compared to Clark's 159, look at what they gave their teams...
Singles-...Brett 109 Clark 126 doubles...Brett 33....Calrk 38 Triples.....Brett 9......Clark 9 HR...........Brett 24... Clark 23 Walks......Brett 58....Clark 74
Outs Made Brett 298...Clark 409
That is the value of both their contributions, and is summed up more accurately in the Situational BR where it says Brett gave 73 Runs, and Clark 74. However, since Brett didn't play all the time, his team had to fill his void with replacments. The replacements playing make his team for the worse off. I gave what the poor replacements did in Brett's stead and assigned it to come up with the value of Brett's not full time role. By doing this, it eliminates your beef. It adds the -11 runs the scrubs contributed.
This method is far more accurate in the real baseball world. What we are looking to achieve is to figure out if 117 games at such a high level of Brett is better for a team than 160 games of a little lower level that Clark provided.
The only question is what can a team typical expect from replacements?
In this case, the split of George Brett, Quirk, and Chalk would win more games than the 159 games from Clark. Quirk and Chalk represent a very low level of replacement. I would say in most cases nearly every team would have somebody as good or better to use as a replacement, thus making it a pretty strong case that Brett's high level of play in the lower amount of games(when adding defense), beats Clark's. And that is the bottom line.
The situational batter runs(though not w/out a few minor problems) are far more balls on accurate as to what really occured, than what any James hypothesis can do, so that data being used is not a problem. >>
Clark and Brett look like ties almost to me. If such is the case, I would go Brett, because he brought much more attention to the Royals then Clark did to the Giants. Meaning, more fans and more dollars. These dollars went to buy...perhaps...better equipment and players for the rest of the team. This is definitely not the Royals of today, but the Royals were strong for almost the whole 80s....Brett being there, bring in more fans/money, and helping other players want to sign there.....might be worth a couple points in this all things considered type discussion?
(I have no facts to back up anything I just said....just thinking.)
Collecting PSA graded Steve Young, Marcus Allen, Bret Saberhagen and 1980s Topps Cards. Raw: Tony Gonzalez (low #'d cards, and especially 1/1's) and Steve Young.
It is a tie, depending how much defense is viewed. If defense is viewed on the low end of defensive value, then it is a tie. If defense and position value are viewed any higher, then it is a sure win for Brett.
If it is viewed a tie, and then the Post Season is included as a tie-breaker, then Brett wins on that end too.
Very good thread. Hoopster makes a great point about the variability of weather as a ballpark factor. I think in any comparison of Carlton 72 and Guidry 78 the impact of Guidry on the pennant race has to be factored into the equation. Also, in Larry Walker's 1997 season his road numbers were off the charts and equal to his numbers at Coors. It is a shame that season is so easily dismissed as being a by product of Coors field.
<< <i>By doing this, it eliminates your beef. It adds the -11 runs the scrubs contributed. >>
See, I know I oversimplified, because those -11 runs the scrubs contributed do not eliminate my beef. They ARE my beef.
The scrubs had positive value. That is simply a statement of fact. For example, when Brett went out with an injury, the Royals didn't have to forfeit - they had third basemen who could hit major league pitching and field major league ground balls sitting on the bench ready to take his place. For another example, the scrubs hit 6 homers, 22 extra base hits in 364 ABs and hit .264 (I don't know how much of that was after Brett went out). Contrary to the conclusion of the system you are citing, they were not responsible for wiping 11 already plated runs off the scoresheet, they were responsible for driving in 41 runs and scoring an additional 26. Nowhere near as good as Brett, but the Royals never lost 5 to -1 when they played, either.
Those 11 imaginary negative runs (and scores more imaginary negative runs by the rest of the team's scrubs) don't exist. If there's a better way to say it, tell me: they simply don't exist. But in this system they are treated as if they do. And since the total runs under the system have to add up to the total runs scored by the Royals (or at least come very close), imaginary positive runs have to be introduced to balance them out. And because George Brett was the Royal who was the most above average, he gets allocated the lion's share of this imaginary booty. I know the system doesn't create imaginary positive runs explicitly, but I'm telling you that what it does is mathematically equivalent.
The one statement I've made here that I'm least sure is correct is that I don't know for sure that this system requires the "runs" it develops to add up to the runs that the Royals scored. If it does, then the system is flawed in the way that I outlined above. If it does not, then it is just nonsense since a "run" isn't related to a run; it's just a number that has no real meaning.
This is for you @thisistheshow - Jim Rice was actually a pretty good player.
<< <i>the impact of Guidry on the pennant race has to be factored into the equation. >>
Had Carlton pitched 400 innings and not allowed a single run, he would still not have been involved in a pennant race. By what logic is it reasonable to give extra credit to Guidry for pennant race heroics, when Carlton was effectively ineligible for a pennant race?
This is for you @thisistheshow - Jim Rice was actually a pretty good player.
No one said he was to be given 'extra credit' for it, just that it was acknowledged, just like Carlton is getting acknowledged for winning 27 games for a team that won 59.
And last time I looked winning was more important then losing when it comes to baseball.
The fact remains without Guidry that year the Yanks do not catch the Red Sox and go on and win the Series.
If Carlton wins 27 or 7 games the Sillies still wind up in the NL basement.
Dallas, the discussion on why BIll James needs LOSS shares is a long one, so lets wait on that.
If two players play 158 games each with IDENTICAL production offensively and defensively and in the same context, we agree they are equal? Of course!
So what if one of them then plays four more games, goes 0 for 16, and makes two errors in five chances in the field? Then what if the other one does not play due to hemroids?
Which is more beneficial to the team? The guy with hemroids is. Why?
Because even though the first guy played four more games, he did so at a rate that is condusive to losing, and his rate of play in those extra games can easily be found almost anywhere. Heck, even I could replace him and do that. So where in the world is giving him value for that??
Wheras the other guy didn't play, then his replacements must play(like Chalk or Quirk). A typical replacement player brings a certain level of play(which is well below the star player, and is even below average). This replacment level of play must be accounted in the star player's resume due to the fact that he was not able to answer the bell like the other guy. And in Brett's case, EVEN WITH A LOW REPLACEMENT LEVEL added to his team, he and a TYPICAL replacement in MLB make his team better off than Clark's team which has Clark for 160 games! That is the bottom line!
In the spirit of my post, Will Clark's 160 extra plate appearance over Brett were bascially an extra .151 batting average he added, a .262 OB% he added, and a .122 SLG% he added. Being that Brett actually hit one more HR than Clark, I took four TB away from the figure to get that. That may be the wrong way to do it, but not a biggie.
So basically, Brett and Clark both produced the same for 515 Plate appearances(Brett's total). Then Brett sat, and Clark's extra 160 plate apperances netted his team an extra .384 OPS. Does an extra .384 OPS have value? Sure, if it is being compared to Axtell playing, but that is why there are MLB player reserves and minor leagues...and typically a team could find ample replacements who could do better than a .384 OPS, which is why Brett's season, despite playing only 117 games, was better due to his overwhleming rate of play over Clark.
P.S. -The reason why they are closer in the situational batter runs is that Clark was exceptional with men on base that season. They both had similar park effects, and similar offenisve era's.
since the original message only said "hitter" I wasn't considering defensive. In that case when asking if Brett or Clark had the best season in the 80s, the easy answer is Trammell
I think we should vote on it. I choose the senator from Ill...oh wait, wrong voting box. I vote Brett.
Collecting PSA graded Steve Young, Marcus Allen, Bret Saberhagen and 1980s Topps Cards. Raw: Tony Gonzalez (low #'d cards, and especially 1/1's) and Steve Young.
We're going in circles now so this will be my last post on this particular topic. And in the end, what we're disagreeing on is more philosophical that statistical, so I don't think we're going to make any more progress.
Total Baseball and James actually arrive at similar conclusions, in a sense. Total Baseball says that the Royals third base scrubs had negative value, and balances that by giving Brett additional positive value. The net value of these third basemen is what you then equate with Brett's value. James assigns positive value to the scrubs, and therefore comes up with a lower value for Brett.
How is that the same? You, consistent with the logic of Total Baseball, speak about replacement level players, etc. Ultimately, your conclusion is that the Royals third basemen in 1980 were as good as Will Clark - Total Baseball has just assigned 100% of that value to Brett. James reaches the same conclusion - that the three Royals third basemen combined were as good as Clark. And I see no reason to disagree with that conclusion. But I'm not willing, philosophically, to give Brett ANY credit for games he never played; the entire concept of replacement level, etc. - IMO - has no place in deciding which of two players was better.
We also have the same issue we had with Carlton and Guidry: if you used multi-year park factors you'd have Clark on top, too.
So I'm declaring this a draw; as always, I've learned something.
This is for you @thisistheshow - Jim Rice was actually a pretty good player.
Comments
<< <i>Dallas - we will agree to disagree about Pedro/Clemens, but now I've got a question about Clark '89 vs. Mattingly '86 (which was much better than his 85 season)
Year Ag Tm Lg G AB R H 2B 3B HR RBI SB CS BB SO BA OBP SLG *OPS+ TB SH SF IBB HBP GDP
+--------------+---+----+----+----+---+--+---+----+---+--+---+---+-----+-----+-----+----+----+---+---+---+---+---+
1989 25 SFG NL 159 588 104 196 38 9 23 111 8 3 74 103 .333 .407 .546 175 321 0 8 14 5 6
1986 25 NYY AL 162 677 117 238 53 2 31 113 0 0 53 35 .352 .394 .573 161 388 1 10 11 1 17
Does Clark '89 win solely on OBP? If 90% of a hitter's value is OPS, as you stated earlier, where does Clark make up the differential between his .953 and Mattingly's .967? I suppose you'll say Mattingly compiled his numbers in 742 PA vs. Clark's 675 PA, but having more PA won't change either of their OPS. The only significant edge Clark holds is GIDP, which is function of PA, and that is partially offset by his 3 CS. >>
Mostly, the difference is Candlestick. I think Clark's stats are more impressive in that context (note that in OPS+, Mattingly's raw 14 point OPS advantage is reversed). GIDP is also part of it; Clark's edge is much greater than the PA ratio. And Clark's 3 CS cost him less than he gains from his 8 steals. And beyond just SB and CS, Mattingly was a truly terrible baserunner whle Clark, in his prime, was very good. Over the course of their careers, Mattingly was a much better forst baseman, but Clark was good enough in his early years that Mattingly gets no real advantage there. I think Clark's edge over Mattingly is actually pretty substantial all things considered.
this assumes that with a runner in scoring position second a hit or walk that does not lead to an RBI is not a failure
Maddux had two strike years that would have been better than Clemens had he continued at that level for the full six months. Since they total together more than one and a half seasons worth, I wouldn't hesitate to give him enough credit for the five month season to move him the top spot for the 90s
Looking at the earlier years, Dizzy Trout is absolutely not the best from the 1940s. Williams and DiMaggio were fighting in a war and 16-year-olds and guys with one arm were playing in the outfield. It doesn't matter how good you are relative to the rest of the league when the rest of the league is made up of guys who aren't the best players. Best single season for that decade has to go to Feller
<< <i>Maddux had two strike years that would have been better than Clemens had he continued at that level for the full six months. Since they total together more than one and a half seasons worth, I wouldn't hesitate to give him enough credit for the five month season to move him the top spot for the 90s >>
<< <i>Looking at the earlier years, Dizzy Trout is absolutely not the best from the 1940s. Williams and DiMaggio were fighting in a war and 16-year-olds and guys with one arm were playing in the outfield. It doesn't matter how good you are relative to the rest of the league when the rest of the league is made up of guys who aren't the best players. Best single season for that decade has to go to Feller >>
It's way beyond the scope of what I tried to do here to see which players or pitchers were on pace for great seasons 2/3 of the way in and guess what might have happened if events had been different. It is certainly possible that Valenzuela would have the greatest season of all time in 1981 if the strike hadn't interrupted, just as its possible that Maddux' arm would have left him as the 1994 season wore on. Who knows? I don't, but I do know that what Clemens did in 1997 had more objective value than what Maddux did in 1994, so I gave him the title.
Same thing with Trout. The position that winning games in 1944 doesn't mean as much as wining games in 1940 (Feller's best) is an opinion that I can't say is wrong. But 1944 is in the record books on the same status as 1940 so Trout gets the nod from me. How would Trout have fared that year if everyone was there instead of in the Army? I don't know, so I don't penalize Trout based on an assumption that might be wrong.
<< <i>Dallas looking closer at some of your choices I would like to know why you chose Foxx's 33 season over his 32. >>
Picking the best of the 1930's was a guess; there were several seasons that were essentially the same. The park factor shifted dramatically from 1932 to 1933, turning Shibe from one of the top hitter's parks to one of the worst; I don't know what happened that year, but Foxx's numbers look very good in context. But his 1932 season, or Gehrig's '34 season, or Medwick's '37 season are all just as good. Where I had choices I thought were equally good, I picked the one that looked strange because that's the way I am.
Rice had the best season in 1978.
TheVon, there is no need for a formula, nor guesswork on how many times a player failed. Every one of Rice's plate appearances are logged into the play by play data. It is known exactly how many times a player hit in every base and out situation.
If you look at the situational batter runs, Rice gets the following for his career...
If you ignore how Rice hit with men on base, and ignore his park factor, he had 270 batter runs above league average player.
If you include how well he hit with men on base, then he has 284 batter runs above league average player. So you are correct that his men on hitting was decent, though not great.
If you include the park factor to those above, then his value would be only 190 runs above average. He is a bit unfairly penalized due to the fact that it is harder for runners to score from first or second at Fenway.
His 1978 season is 60.4 situational Batter runs NON park adjusted, and 47 park adjusted.
For comparison, Morgan's 1976 season is 84 situational batter runs, and 80 park adjusted.
I am in the camp that Fenway probably didn't affect his batting to such a high degree that these numbers bear out, but they certainly did help a good amount based on the overwhelming objective evidence.
Yeah, he trailed Carlton by a good amount of innings, 346 to 276.
If you look at their pitching Runs, which factors IP as well as ERA vs. league ERA, Guidry wins 59 to 55.
In the Clemens/Maddux race, Clemens beats him 73 to 61 in pitcher runs.
No doubt the strike robbed Maddux. Dallas, sure it is possible he hurt his arm, or did awful in those missed starts. It is also possible that he threw five shutouts. Logically speaking, the most likely scenario is that he would have pitched pretty close to the rate he had that season and the two sandwich seasons. It is perfectly acceptable to assume this and give Maddux his due, just as it is perfectly acceptable to do the same with Ted Williams's missed seasons. If Maddux were missing those games due to being a wuss, then that is a different story.
Plus that fat toad Clemens was juicing.
The verdict is that Guidry and Maddux need to replace Carlton and the toad.
Regarding Guidry, I just disagree. Not just a little bit, either. I think Carlton completing 30 games and winning almost half of his godforsaken team's games was historically great. Add to that he was in the NL and had to bat - and hit almost .200 - and he blows Guidry away. In fact, I would need convincing that Guidry had a better year than Gaylord Perry's 1972 season or Wilbur Wood's 1971 season.
Regarding Rice, I never said he didn't have a great season in 1978. But it's the only great season he ever had among five or so good seasons and a whole lot of mediocrity, and if you think he had a better season in 1978 than Dave Parker then you'll need to show me why. Parker's OPS+ is better, he was 20 for 27 stealing bases (Rice 7 for 12), he grounded into 7 fewer DP and he won a Gold Glove. And he hit better in RISP and men on situations. Rice's only advantage is that he played every game, and I don't think it makes up for everything else.
have went no where had it not been for Guidry's 25 wins? Add in his shutouts and k's and sub 2.00 era
and a compelling case can be made.
The 25th win by the way was against the Sox in the famous 163 rd game.
The one that is called the Dent game.
On a summers night he struck out 18 Angels.
Carlton did have an amazing season in 72.
He also did not have a Sparky Lyle and Goose Gossage in his bullpen either.
My vote goes to Guidry though.
Steve
I'm not sure how you can say that about Guidry. He saved 59 runs over the average pitcher, Carlton 55, Wood 54, and Perry 53.
True, Carlton had to bat...but Carlton also had the luxury of being able to extend himself longer into the games by not having to pitch in a higher scoring environement(specifically facing a pitcher with three cupcake at bats). Being that his extra value is coming from his IP, this is a factor too.
If two pitchers are equal in value based on their IP and ERA vs. league, then I take the guy who got there by virtue of the stronger ERA+ as opposed to having the more IP because he was allowed(or not allowed) to pitch more innings, as that can be a mangers fault or decision.
None of that matters much anyway, as Guidry was responsible for saving more runs for his team, thus better.
<< <i>I'm not sure how you can say that about Guidry. He saved 59 runs over the average pitcher, Carlton 55, Wood 54, and Perry 53. >>
I'm usually wrong when I think you are, skin, but where are you getting these numbers? Using b-r.com, I get Carlton 62, Wood 63, Guidry 57 and Perry 51.
I'm using (*ERA+ - ERA) * IP / 9 or (mathematically equivalent) (*ERA+ - 1) * ER. Maybe your source and b-r.com are using different park adjustments? Or are you using something from the linear weights system with which I'm not familiar?
And I'm not offended if anyone wants to vote for Guidry - he had a hell of a season. I'm just more impressed by what Carlton did. And in those days I was a Cardinal fan until they were out of the race - usually sometime in June - and then I started rooting for my favorite player's teams, which included the Yankees since Reggie was there. Other than Yaz, I didn't really like the Red Sox much, so I was glued to my TV rooting for Guidry in game 163. And Reggie's "oops" play between first and second in the WS that year is still my all-time favorite play. Which is all to say, I have no bias against Guidry - I'd put him in the HOF twice if it would keep Jack Morris from streaking the HOF bowl - I just think Carlton's year was even more awesome.
I just grabbed them from Total Baseball, same formula. Probably different park adjustments.
Just looked them up as to why they differ. Total Baseball has Guidry's park factor at 97, and baseball-reference at 96(both three year and single year).
The big difference lies within Carlton's. Total Baseball has his park factor at 99, while Baseball-Reference is at 104 which is the three year average. Baseball reference has the one year park factor from 1972 as 99 as well.
So using the one year park factors for the year of 1972 puts Guidry up. Using the three year average puts Carlton up.
If one believes that the defense was a big factor, then it could swing the tide a bit...though I don't think the defense factored as much on these two strikeout pitchers as some sabermatricians may believe.
<< <i>I agree about Brett; had he not been injured he was on pace for the seaon of the decade. But he was hurt, and missing 45 games left him a mile short. Still, even missing 45 games, he was about as valuable as Rice in 1978. >>
Brett played in 42 less games than Clark, but still had more HR and RBI's. How can you ignore the .390 with tons more power than Clark?
Also Brett much better in 79 than Morgan in 75(50 extra base hits is best season of decade?)
118 RBI in 117 games
.390 Average
Only 22 Strikeouts!! Talk about a tough out!!
Brett was soooo clutch. Never forget his dramatic ALCS homer off Goose Goosage to put the Royals in the World Series in 1980.
Brett in 1979 had an on-base percentage of .376, Morgan in 1975 had a .466. Morgan stole 67 bases and got caught 10, and helped his team a lot in the process; Brett stole 17 and got caught the same 10 - and hurt his team in the process. Brett hit into 8 DPs, Morgan 3. If Brett hit about 30 HR more than Morgan he would just about make up that chasm - he hit 6 more. Bett had a good year, but it wasn't the best in the AL in 1979 and it wasn't even in the same ballpark as Morgan's 1975 (or 1976) season.
I'll be the first to admit that I don't know the right answer to this one, or if there is a right answer. Veterans Stadium was what it was in 1971, 1972 and 1973 - and it was exactly the same in all three years except for a little tinkering with the heights of the fences. B-R.com uses three-year averages and Bill James uses five-year weighted averages for park factors and I tend to think that they are correct to do something besides rely on a single year since a sample size of one year may not be nearly large enough, especially since rosters change and we're trying to measure the parks and not how a particular one-year roster hit in that park.
From 1971 to 1973, Veterans park factors went from 105 to 99 to 108 while the fences got steadily higher eah year. There is no logical explanation for that, and I think a park factor of 104 (the average) is probably a much better indicator of the kind of park Veterans was in 1972 than 99. James method would get something between 102 and 103 and that may be better still (I'm not sure how James deals with the fact that Veterans opened only one year prior to 1972).
The same thing happened, even more dramatically, to Wilbur Wood. 1971 represents a freakish dip in the park factor for Comiskey and there is absolutely no objective reason why that should have happened. I don't think it did "happen" in any meaningful sense, it's just random noise in small samples. Comiskey and Veterans were big hitters parks, and Carlton and Wood should get credit for how well they pitched in them. As you noted, Yankee Stadium was steady over that period so it doesn't matter what method we use.
So I'm sticking with Carlton, but I've learned something along the way and that's always a good thing.
It could just be randomness, but it also could be more cold days while they were at home in '72. Just think of April and May and how it is hit or miss with the weather(probably not in Texas so much
Same with wind factor for parks. One season may find the luck(for pitchers at least) that the wind is blowing harder and in, as opposed to harder and out. Carlton himself was nearly identical at home/road in '72, as were the rest of the players who played there.
BRETT VS. CLARK
Situational Batter Runs:..........Brett 73, Clark 74.
Situational Stolen Bas Runs:...Brett 1,...Clark 1
Actually, Brett did accomplish as much in 117 games as Clark did in a full season. Well, one run short. This is measured as above league average player. Both hit well in high leverage and late/close situations.
George Brett's .664 SLG% is the highest of anyone in the 70's and 80's. His OPS+ was also higher than anyone in the 70's or 80's.
But with Brett not being able to answer the bell for 45 games, it meant that a reserve had to replace him. So intertwined in Brett's value to his team, is the value of what a reserve player would contribute.
Dave Chalk was one of the primary replacments at 3B, and he checked in with a -7 situational batter runs
Jamie Quirk was the other, and he had -4 situational batter runs.
You can't fault Brett for how good/bad a reserve is(that is the GM's blame), but those two guys look to represent what a typical reserve player would do... maybe a bit worse than typical.
Including the value of his replacements, you get the following value of Brett and Clark.
Brett 62 Situational Batter Runs
Clark 75 Situational Batter Runs
The fact that Brett missed those games is no longer needed as a factor, as that aspect is factored right there with poor replacements.
DEFENSE: Brett had 22 Fielding Runs, Clark 9. This is BaseballProspectus, and defensive stats comes with its usual caveat. Knowing Brett outdistanced his 3B peers in fielding runs more than Clark did his 1B peers, and taking into account that Brett did it in a harder defensive position, I can certainly see that ten run deficit being made up on the defensive spectrum. This 13 runs puts them at even.
Conclusion: When you included what Brett's poor replacements did in his 45 game absence, and found out that Brett was only 13 Runs behind Clark in offensive contribution, and then took Brett's superior defense at a harder position into account, I do see Brett claiming the title as the best season of the 80's.
P.S. POST SEASON: Usually this isn't a fair thing as teammates decide whether or not you even get a crack, and there are so few games that results are a crapshoot, but Brett had a fabulous post season, Clark had one great series and one bad. Brett's post season play was a continued glimpse of his worldly season...and he made hemroid medicine famous.
And I don't question that the numbers you're using are accurate, I just place a lot less credence in that system than you do. Implicit in the situational-batter-runs is the assumption that 100% of a player's "value " consists in how far that player is above average. But that's simply not true - MOST of a player's value is captured by his accomplishments that bring him to average. In other words, a player that is average has significant value, but that system says he has none. Brett was farther above average when he wasn't hurt than Clark was, and that skews the results in his favor; all of the games that Clark played that Brett did not are correspondingly undervalued.
Mathematically (and very loosely) for anyone who isn't following what I'm saying: say that the average player has the arbitrary value of 50 and that Brett's value was 90 when he played and that Clark's value was 80 when he played. I think that means that Brett was 12.5% better than Clark (90/80); the system skin is using says that Brett was 33% better than Clark (40/30). That's a HUGE difference and I don't think it's the correct way to compare two players. I'm oversimplifying and presenting only one side, but in the end this problem remains for me in that system. In this example, 62.5% of Clark's value consisted of being average and 55.6% of Brett's value consisted of being average, and I think a system that simply throws away those majorities before comparing is fundamentally flawed.
I know I'm not going to convince you of all of this in a single messageboard post - I'd lose a little respect for you if I could - but I did want to make clear that I'm not merely being stubborn. I think the stats back me up that Clark had the best season of the 1980's.
The absolute goal is to see if 117 games of Brett at his pace, is better for a team than 160 games of Clark at his lower pace. This method nails it!
When comparing two players who are both above average, what you are saying does not come into play. It is when one is at the average baseline(or just below) where what you are saying comes into play. This happens more with career value. Currently, average would be zero runs. They are both 75 runs above average. If you were to measure them against a baseline of -15, they would still be the same, but 90 runs above. If George Brett were just FOUR runs above average for 117 games, and Clark zero above for 162 games, THEN your objections would be valid, because no matter how many games Clark played, it would seem as if Brett had more value. This is not the case in this analysis.
In this case, I gave Brett the stats of his replacements who were bad, and it shows that what he did in his 117 games + what a typical replacement player would be, actually showed that his season is better for a team, than Clark's(when adding defense). WE don't have to guess, we know.
In laymen's terms in Brett's 117 Games compared to Clark's 159, look at what they gave their teams...
Singles-...Brett 109 Clark 126
doubles...Brett 33....Calrk 38
Triples.....Brett 9......Clark 9
HR...........Brett 24... Clark 23
Walks......Brett 58....Clark 74
Outs Made Brett 298...Clark 409
That is the value of both their contributions, and is summed up more accurately in the Situational BR where it says Brett gave 73 Runs, and Clark 74. However, since Brett didn't play all the time, his team had to fill his void with replacments. The replacements playing make his team for the worse off. I gave what the poor replacements did in Brett's stead and assigned it to come up with the value of Brett's not full time role. By doing this, it eliminates your beef. It adds the -11 runs the scrubs contributed.
This method is far more accurate in the real baseball world. What we are looking to achieve is to figure out if 117 games at such a high level of Brett is better for a team than 160 games of a little lower level that Clark provided.
The only question is what can a team typical expect from replacements?
In this case, the split of George Brett, Quirk, and Chalk would win more games than the 159 games from Clark. Quirk and Chalk represent a very low level of replacement. I would say in most cases nearly every team would have somebody as good or better to use as a replacement, thus making it a pretty strong case that Brett's high level of play in the lower amount of games(when adding defense), beats Clark's. And that is the bottom line.
The situational batter runs(though not w/out a few minor problems) are far more balls on accurate as to what really occured, than what any James hypothesis can do, so that data being used is not a problem.
<< <i>Dallas,
The absolute goal is to see if 117 games of Brett at his pace, is better for a team than 160 games of Clark at his lower pace. This method nails it!
When comparing two players who are both above average, what you are saying does not come into play. It is when one is at the average baseline(or just below) where what you are saying comes into play. This happens more with career value. Currently, average would be zero runs. They are both 75 runs above average. If you were to measure them against a baseline of -15, they would still be the same, but 90 runs above. If George Brett were just FOUR runs above average for 117 games, and Clark zero above for 162 games, THEN your objections would be valid, because no matter how many games Clark played, it would seem as if Brett had more value. This is not the case in this analysis.
In this case, I gave Brett the stats of his replacements who were bad, and it shows that what he did in his 117 games + what a typical replacement player would be, actually showed that his season is better for a team, than Clark's(when adding defense). WE don't have to guess, we know.
In laymen's terms in Brett's 117 Games compared to Clark's 159, look at what they gave their teams...
Singles-...Brett 109 Clark 126
doubles...Brett 33....Calrk 38
Triples.....Brett 9......Clark 9
HR...........Brett 24... Clark 23
Walks......Brett 58....Clark 74
Outs Made Brett 298...Clark 409
That is the value of both their contributions, and is summed up more accurately in the Situational BR where it says Brett gave 73 Runs, and Clark 74. However, since Brett didn't play all the time, his team had to fill his void with replacments. The replacements playing make his team for the worse off. I gave what the poor replacements did in Brett's stead and assigned it to come up with the value of Brett's not full time role. By doing this, it eliminates your beef. It adds the -11 runs the scrubs contributed.
This method is far more accurate in the real baseball world. What we are looking to achieve is to figure out if 117 games at such a high level of Brett is better for a team than 160 games of a little lower level that Clark provided.
The only question is what can a team typical expect from replacements?
In this case, the split of George Brett, Quirk, and Chalk would win more games than the 159 games from Clark. Quirk and Chalk represent a very low level of replacement. I would say in most cases nearly every team would have somebody as good or better to use as a replacement, thus making it a pretty strong case that Brett's high level of play in the lower amount of games(when adding defense), beats Clark's. And that is the bottom line.
The situational batter runs(though not w/out a few minor problems) are far more balls on accurate as to what really occured, than what any James hypothesis can do, so that data being used is not a problem. >>
Clark and Brett look like ties almost to me. If such is the case, I would go Brett, because he brought much more attention to the Royals then Clark did to the Giants. Meaning, more fans and more dollars. These dollars went to buy...perhaps...better equipment and players for the rest of the team. This is definitely not the Royals of today, but the Royals were strong for almost the whole 80s....Brett being there, bring in more fans/money, and helping other players want to sign there.....might be worth a couple points in this all things considered type discussion?
(I have no facts to back up anything I just said....just thinking.)
Raw: Tony Gonzalez (low #'d cards, and especially 1/1's) and Steve Young.
If it is viewed a tie, and then the Post Season is included as a tie-breaker, then Brett wins on that end too.
Also, in Larry Walker's 1997 season his road numbers were off the charts and equal to his numbers at Coors. It is a shame that season is so easily dismissed as being a by product of Coors field.
<< <i>By doing this, it eliminates your beef. It adds the -11 runs the scrubs contributed. >>
See, I know I oversimplified, because those -11 runs the scrubs contributed do not eliminate my beef. They ARE my beef.
The scrubs had positive value. That is simply a statement of fact. For example, when Brett went out with an injury, the Royals didn't have to forfeit - they had third basemen who could hit major league pitching and field major league ground balls sitting on the bench ready to take his place. For another example, the scrubs hit 6 homers, 22 extra base hits in 364 ABs and hit .264 (I don't know how much of that was after Brett went out). Contrary to the conclusion of the system you are citing, they were not responsible for wiping 11 already plated runs off the scoresheet, they were responsible for driving in 41 runs and scoring an additional 26. Nowhere near as good as Brett, but the Royals never lost 5 to -1 when they played, either.
Those 11 imaginary negative runs (and scores more imaginary negative runs by the rest of the team's scrubs) don't exist. If there's a better way to say it, tell me: they simply don't exist. But in this system they are treated as if they do. And since the total runs under the system have to add up to the total runs scored by the Royals (or at least come very close), imaginary positive runs have to be introduced to balance them out. And because George Brett was the Royal who was the most above average, he gets allocated the lion's share of this imaginary booty. I know the system doesn't create imaginary positive runs explicitly, but I'm telling you that what it does is mathematically equivalent.
The one statement I've made here that I'm least sure is correct is that I don't know for sure that this system requires the "runs" it develops to add up to the runs that the Royals scored. If it does, then the system is flawed in the way that I outlined above. If it does not, then it is just nonsense since a "run" isn't related to a run; it's just a number that has no real meaning.
Bingo
Steve
<< <i>the impact of Guidry on the pennant race has to be factored into the equation. >>
Had Carlton pitched 400 innings and not allowed a single run, he would still not have been involved in a pennant race. By what logic is it reasonable to give extra credit to Guidry for pennant race heroics, when Carlton was effectively ineligible for a pennant race?
Carlton is getting acknowledged for winning 27 games for a team that won 59.
And last time I looked winning was more important then losing when it comes to baseball.
The fact remains without Guidry that year the Yanks do not catch the Red Sox and go on and win the Series.
If Carlton wins 27 or 7 games the Sillies still wind up in the NL basement.
Steve
If two players play 158 games each with IDENTICAL production offensively and defensively and in the same context, we agree they are equal? Of course!
So what if one of them then plays four more games, goes 0 for 16, and makes two errors in five chances in the field?
Then what if the other one does not play due to hemroids?
Which is more beneficial to the team? The guy with hemroids is. Why?
Because even though the first guy played four more games, he did so at a rate that is condusive to losing, and his rate of play in those extra games can easily be found almost anywhere. Heck, even I could replace him and do that. So where in the world is giving him value for that??
Wheras the other guy didn't play, then his replacements must play(like Chalk or Quirk). A typical replacement player brings a certain level of play(which is well below the star player, and is even below average). This replacment level of play must be accounted in the star player's resume due to the fact that he was not able to answer the bell like the other guy. And in Brett's case, EVEN WITH A LOW REPLACEMENT LEVEL added to his team, he and a TYPICAL replacement in MLB make his team better off than Clark's team which has Clark for 160 games! That is the bottom line!
In the spirit of my post, Will Clark's 160 extra plate appearance over Brett were bascially an extra .151 batting average he added, a .262 OB% he added, and a .122 SLG% he added. Being that Brett actually hit one more HR than Clark, I took four TB away from the figure to get that. That may be the wrong way to do it, but not a biggie.
So basically, Brett and Clark both produced the same for 515 Plate appearances(Brett's total). Then Brett sat, and Clark's extra 160 plate apperances netted his team an extra .384 OPS. Does an extra .384 OPS have value? Sure, if it is being compared to Axtell playing, but that is why there are MLB player reserves and minor leagues...and typically a team could find ample replacements who could do better than a .384 OPS, which is why Brett's season, despite playing only 117 games, was better due to his overwhleming rate of play over Clark.
P.S. -The reason why they are closer in the situational batter runs is that Clark was exceptional with men on base that season.
They both had similar park effects, and similar offenisve era's.
Raw: Tony Gonzalez (low #'d cards, and especially 1/1's) and Steve Young.
We're going in circles now so this will be my last post on this particular topic. And in the end, what we're disagreeing on is more philosophical that statistical, so I don't think we're going to make any more progress.
Total Baseball and James actually arrive at similar conclusions, in a sense. Total Baseball says that the Royals third base scrubs had negative value, and balances that by giving Brett additional positive value. The net value of these third basemen is what you then equate with Brett's value. James assigns positive value to the scrubs, and therefore comes up with a lower value for Brett.
How is that the same? You, consistent with the logic of Total Baseball, speak about replacement level players, etc. Ultimately, your conclusion is that the Royals third basemen in 1980 were as good as Will Clark - Total Baseball has just assigned 100% of that value to Brett. James reaches the same conclusion - that the three Royals third basemen combined were as good as Clark. And I see no reason to disagree with that conclusion. But I'm not willing, philosophically, to give Brett ANY credit for games he never played; the entire concept of replacement level, etc. - IMO - has no place in deciding which of two players was better.
We also have the same issue we had with Carlton and Guidry: if you used multi-year park factors you'd have Clark on top, too.
So I'm declaring this a draw; as always, I've learned something.
<< <i>I think we should vote on it. I choose the senator from Ill...oh wait, wrong voting box. I vote Brett. >>
I'll get out my "Brett for President" sign from that '80 season and start campaigning.