Whatever you might think of Derek Jeter—with 72 career Wins Above Replacement, averaging the two sources—he’s rarely been the best player in his league for even one season. In 1998 and ‘99, perhaps. And if you’re really careful, you might find another season in which he deserved the MVP Award. But even in his best seasons, he’s always had some competition, while Wagner usually didn’t. Strictly by the numbers, Wagner wasn’t just the greatest shortstop who ever played the game; he was one of the five greatest players, period. You can look it up.
...Still not ready to give up on Derek Jeter? Well, here’s how you get him past Cal Ripken and perhaps even Honus Wagner ...
First, you give him a big dollop of extra credit for the Yankees’ success since he arrived in the majors back in 1996. And he deserves some extra credit. There’s an intangible benefit to his steadiness over all those years, and his postseason numbers are right in line with his regular-season numbers ... which of course is impressive because he’s typically faced better pitching in the postseason.
Second, you assume that every single sophisticated defensive metric is just flat-wrong; that he’s been not a poor defensive shortstop for most of his career, but that instead he’s been average, at least.
If you want to award Jeter 10 extra Wins Above Replacement for his intangibles and his postseason play, and make him an average defensive shortstop, you’ve got him at roughly Cal Ripken’s level. And if you want to assume that the roughly 90 years separating Jeter from Wagner practically invalidates Wagner’s performance, then ... Well, then you can make the argument that Derek Jeter really is the best shortstop ever to play baseball.
I can’t make that argument. But to a lot of people you and I know, maybe even some of our friends, it’s not crazy.
Reader Comments and Retorts
Go to end of page
Statements posted here are those of our readers and do not represent the BaseballThinkFactory. Names are provided by the poster and are not verified. We ask that posters follow our submission policy. Please report any inappropriate comments.
Why should we worry about a small difference in the rate of decline for foreign-born pitchers? How large can these samples be? Why would we want to look only at BABIP, especially when rising K rates change who is putting balls in play?
So lets ignore any evidence that might contradict your feelings?
We just did that. But apparently, you didn't like the answer.
We did? Where? What we did was the final word on this topic? What we did was the start not the end.
Ah, now I see the game....
I think you're playing solitary.
a) You couldn't model deadball scoring all that well from counter stats. Between huge numbers of errors, automatic use of one run strategies (and the lower DP rate that comes from this -- and the fact that there are few hard hit balls) and kamikaze baserunning, the actual run scoring is ... well it's not random, but it's nothing close to today's game.
There's always some kind of play on, and the stats don't cover how successfully any given team or player executes them.
b) Batter strikeouts are in fact significant. You can't really model team runs scored without taking this into consideration. Best I can tell they serve as a crude proxy for how well a team executes all of those plays mentioned above.
For decades the Reds did not allow their minor league teams to use the DH. The idea was that it would give their pitchers an advantage. Feel free to check this yourself but I was unable to find any evidence that this resulted in any improvement at the major league level.
Oh they had good hitting pitchers from time to time, but often it was from guys who started hitting at the major league level. And they often had utterly dreadful pitcher's hitting.
What I don't understand is your apparent refusal to accept the war year results, which do appear to show that pitcher hitting can be a useful benchmark for assessing other changes. Same with the disparity between AL and NL pitchers in the first two decades of integration -- the results are pretty much exactly what we'd expect to find given the talent difference in the leagues. One can accept these as valid findings while still remaining skeptical about how stable pitcher hitting has been over 110 years.
Rk Yrs From To Age1 Alex Rodriguez 6 1996 2003 20-27
2 Cal Ripken 4 1983 1991 22-30
3 Ernie Banks 4 1955 1960 24-29
4 Derek Jeter 2 1998 1999 24-25
5 Nomar Garciaparra 2 1998 2000 24-26
6 Robin Yount 2 1982 1983 26-27
7 Jim Fregosi 2 1964 1970 22-28
8 Lou Boudreau 2 1947 1948 29-30
Most seasons 5+ WAR
Rk Yrs From To Age1 Barry Larkin 7 1988 1999 24-35
2 Cal Ripken 7 1983 1991 22-30
3 Alan Trammell 7 1980 1990 22-32
4 Derek Jeter 6 1998 2009 24-35
5 Nomar Garciaparra 6 1997 2003 23-29
6 Alex Rodriguez 6 1996 2003 20-27
7 Ernie Banks 6 1955 1960 24-29
8 Pee Wee Reese 6 1942 1954 23-35
9 Ozzie Smith 5 1985 1989 30-34
10 Robin Yount 5 1980 1984 24-28
11 Jim Fregosi 5 1964 1970 22-28
Rk Player OPS+ WAR/pos Rfield PA From To1 Honus Wagner 150 134.5 85 11748 1897 1917 H
2 Alex Rodriguez 144 104.6 -14 10634 1994 2011
3 Arky Vaughan 136 75.6 21 7722 1932 1948 H
4 Hanley Ramirez 132 29.3 -28 3757 2005 2011
5 Nomar Garciaparra 124 42.5 24 6116 1996 2009
6 Ernie Banks 122 64.4 54 10394 1953 1971 H
7 George Davis 121 90.7 146 10178 1890 1909 H
8 Lou Boudreau 120 56.0 118 7024 1938 1952 H
9 Vern Stephens 119 43.5 -2 7241 1941 1955
10 Joe Cronin 119 62.5 28 8840 1926 1945 H
11 Hughie Jennings 118 46.4 60 5648 1891 1918 H
12 Derek Jeter 117 70.4 -143 11155 1995 2011
13 Barry Larkin 116 68.9 28 9057 1986 2004 H
14 Robin Yount 115 76.9 -46 12249 1974 1993 H
15 Toby Harrah 114 47.1 -96 8767 1969 1986
16 Ed McKean 114 40.0 -48 7626 1887 1899
17 Jim Fregosi 113 46.1 4 7403 1961 1978
18 Luke Appling 113 69.3 41 10254 1930 1950 H
19 Cal Ripken 112 89.9 179 12883 1981 2001 H
20 Carlos Guillen 111 26.3 -35 5277 1998 2011
21 Ray Chapman 111 27.4 -26 4591 1912 1920
22 Jack Glasscock 111 58.7 149 7552 1879 1895
23 Alan Trammell 110 66.9 75 9376 1977 1996
Well, my main problem is that if increase in the level of play were that steep, long careers would be impossible. How could Honus Wagner have a 145 OPS+ in 1899 and a 156 12 years later? How could babe Ruth hit 219 in 1919 and 218 in 1931? Ted Williams 234 in 1941, and a mere drop to 190 20 years later? Aside from a couple of small peak seasons, Don Sutton was a pretty consistent 110 ERA+ pitcher. He had seasons within a few points of that in 1966, 1968, 1974, 1976, 1982, and 1986. In addition, he had 9 other seasons between 94 and 121 in that time frame. If the level of increase is so steep as to make Honus Wagner close to replacement level in 100 years, how could Don Sutton be so consistently above average for 20 years?
Ty Cobb
Eddie Collins
Tris Speaker
Home Run Baker
Shoeless Joe Jackson
Honus Wagner
Nap Lajoie
Sam Crawford
Clyde Milan
Ed Konetchy
The 40th best positional player from 2006 to 2009 was Casey Blake, Brian Roberts, and Aaron Hill. 30th was Placido Polanco.
Say a fictional player with time traveling ability named Honus Wright is a 175 OPS+ in 1912. Assume that the rate of league improvement is constant (no evils of war, segregation, or sudden expansion). Improved competition reduces his OBP and SLG each year by multiplying by .997.
After 10 years, he's at 167, after 15 he's at 163. Now most people who hit 175 and are still playing 15 years later will be a lot worse than 163 simply because of old age. In reality we don't know how much of the typical observed decline is aging and how much is league improvement. In truth the two effects cannot be separated. Anyway, that same rate of decline makes Honus Wright an average player (104) by 2012.
But what if the rate of decline is .995 instead of .997? Then he's at 162 and 155 ten and 15 years later. Not much difference than before, it would be hard to see the difference within a player's 15-20 year career. But at this rate of decline the player is now only a 67 OPS+ after 100 years.
That's the real problem with timelining for such long periods, it's way too sensitive to assumptions. With this pitcher hitting model it could be built with different assumptions, all that seem very reasonable and not a big deal over short time periods. Extending it out to 100 years though, and the tiny difference in assumptions mean that we could project Honus to be anywhere from a 65 OPS+ to a 140.
Does that count the transformer shot? I saw him hit one out off Juan Berenguer when Reggie was with the Angels.
But that's the potential value of trying to figure out pitcher hitting, and whether it's changed over time. It can potentially help us separate aging from improved competition. And even if we can never reach a consensus view of pitcher hitting over long periods of time, it may be a useful tool for answering other questions. The deline in quality in the war years is one example. The difference in league quality 1947-1972 is another.
And just for the record, this methodology does NOT result in Wagner being labeled a replacement player in today's game. Depending on how you adjust the hitting stats, it pegs Wagner at an OPOS+ of between 107 and 116 (I mistakenly gave a lower # earlier in the thread). It puts Ruth at 159-170, and Mantle at 149. One can be skeptical of these precise estimates, of course, but they are certainly all plausible. And they are also pretty consistent with Gassko's entirely different methodology based on individual players' decline over time, and broadly consistent with the shrinking variance in performance. The more that different methodologies tell the same story, the more confidence we can have in the results.
Up to a point. If different methodologies are all based on the same set of underlying assumptions, the methodologies will likely tell the same story - but if the underlying assumptions aren't valid, the story may still be incorrect.
-- MWE
You must be Registered and Logged In to post comments.
<< Back to main