Evolving MLB Defense: 1901-1929, at 2B

[leewileyfan]

Lee I don't know how you could possibly deal with this, but IIRC Buckner had gimpy ankles from before Boston. Many of his assists were pitcher flips due to his slowness rather than plays made too far from the bag (or at least that's how Bill James put it back in a mid 80s Abstract).

PV: I expected the favorable mention of Bill Buckner on defense to raise a few eyebrows. However, your recollection of a very nearly crippled Buckner and the WS disaster so indelibly captured on film merely reminds me [us] how visually oriented we are. The totality of Buckner was not captured either on the clip or in the entire context of that disastrous season in the field.

A competent fielder before he contracted a staph infection in his ankle, Buckner was moved to 1B as an adjustment to that condition. He was an excellent defensive 1B, almost from the start. My metric has him for his 1B career at .983, which would get him about +7 defense runs per season above average. A Keith Hernandez, > 1.000 would be about +14.5 DR per season above average.

I'll stand by the .983 rating for Buckner.

I don't usually like to mention other metrics or metric founders [especially the historic, much-published and of guru status]; but since they are often the mirrors I have to explain my findings to, I will make a few observations of my own about James:

1. He seems much-guided by visually captured, indelible "moments," like the unfortunate image he painted of a young Harlond Clift tripping over his own glove. He passes them on in each edition, giving them a kind of permanence.

2. I read the 1st edition of Dewan's "Bible," and was astonished at how so many of James' narrative snippets seemed to contradict or be at odds with the numbers presented. After a few added perusals, I tossed the "Bible" into the trash, feeling that I had committed no sacrilege at all.

While I can easily allow that there might be flaws here and there in my approach, I am confident of their trustworthiness. Nobody, IMHO, has yet published anything to date that I have found to be really solid or well defined.

[leewileyfan]

I have just wrapped up the edited updates on all positions [pitcher excepted] through 1964.

Since all this editing is aimed at a possible shot at publication, [heck it's been copyrighted and updated since 2004, I would love to get some feedback on the way I see the book being laid out, and some of the ideas I have [as to whether or not they'd be of interest to fans at all levels].

1. It is, first and foremost, a study of defense at each position, with an historic focus, such that players are evaluated against their contemporaries, season-by season, for starters.

2. The entire evaluation basis is base on ONLY widely published elements of defense, readily available for all seasons and positions covered [1901 through 2012], when I hope to have the project finished - I hope by Thanksgiving. The critical element of the metric is the weighting of each data input element [PO, A, E, DP, PB, WP, CS%] and the provision of a unique model for each position,as to challenge, opportunity, success and failure. The raw numbers produced at the various positions look like an unrelated hodge-podge; and that is intentional, because each position is recognized for its own unique qualities. NO position is treated as trivial, when compared to another.

3. The hodge-podge is converted into something uniform, presentable, and familiar by formulaic conversions into a range generally between .900 and 1,000, although some awful ratings may be in the low .800s and outstanding performance may "bust the template" being > 1.000.

4. The first portion of the book sets the historic background at 1901, the players, uniforms, equipment, bat, ball and glove ... the parks and their dimensions. This is woven in with the complete explanation of evaluations and how they are calculated. For each season, I have compiled a "coffee table" element, what was going on in the world outside of baseball, highlighting any events that might have impacted players' lives.

5. Then, for each season there is a table, presently listing, by position, leading players or notorious players that season, or occasionally a defensive standout who was around [perhaps] for no more than a cup of coffee. There were no defensive awards back then; so I introduce the "awards" concept, with my "bests" named in bold print.

6. When the actuality of Gold Gloves came to be, I confronted a decision. Here I'd really appreciate feedback as to the proper course of action to take;

a. I could [this first part I am doing anyhow] introduce the Gold Glove award a a new historic fact. I COULD explain that a player whose name appears in bold print, in the metric's BEST, while a player with an * is the Gold Glove winner. If a player in named in bold print, with an asterisk, there is concurrence.

b. If I get published, I favor a positive and direct presentation. I am personally AVERSE to making any negative comments about any other metrics; and I hesitate to take on the Gold Gloves as a good idea terribly implemented. Personally, yes, I see the awards as a repetitive, habit-forming exercise with some "good-ole-boy," "favorite son," or "we owe HIM one.," elements. I am very nearly convinced that GG editorializing would just distract from the positivity of the presentation.

I welcome pro and con suggestions.

7. The book, in its present draft, will note the differences in the game, the players, the fans, society, and life itself, looking back from 2012, cell phones a I-pods and all kinds of photographic apps ... ballpark MENUS and prices ... then back to 1901, with every game in the sunshine, a nary a soul in sight who appears to be talking to himself.

Anyone interested please make suggestions and comments.

Meanwhile, as I stated at the start of the thread, any questions on player performance or play comps, ask away.

[Jackaroo Dave]

Lee, even though your book is going to be crammed with numbers, it will still be a very personal outlook on baseball history. So I would not worry about negativity. I'd list the gold glovers and certainly mention fielders whose reputations aren't borne out by your research. It's not going to be a secret when your views diverge from others'. Who wants to read a book that could be summarized as 'I agree with Bill James'? Just don't make disagreements a matter of personalities and avoid drive-by critiques

In general, I think the more personal you make it, the better. Your voice is unique; let it be heard.

I like it when you include memories and anecdotes. Reading your book should be like spending an afternoon at the ball park with an old friend who's seen a lot. Two hot dogs here!

[leewileyfan]

Thanks, JD. That's the tack I'll take, with regards to awards from habit, more than current merit,

[leewileyfan]

Another searchlight signal looking for feedback. As I edit the player-position-data entry-ratings-defense runs lines for each position/player, I find myself with growing spreadsheets, which I am editing to avoid a runaway train.

From the standpoint of a prospective reader into a book on MLB and its defensive evolution [1901-2011], which would be preferable:

-longer listed entries of regular/famous/infamous/defensively noteworthy players, with primary focus on only "the best rated defensively?"

-a longer [inclusive list] of a mixed bag [perhaps "best" with average and sub-par] for a range?

-a well-contained, distilled list of "best;"

-a well contained, distilled mixed bag which always includes the best.

The manuscript will spell out, in detail, the data necessary to compile ratings independent of the book, with all the weightings, so that readers can look up any player, any season, and run the numbers at home. I do want to present factual data findings; AND I have no wish to avoid challenges, like "Is this guy's rating on Hubert Pilkington totally nuts?" However, I do not want to present data so boring as to become like reading a phone book.

Right now, each season's section that is devoted to defensive players and their ratings is in a line by line format that shows:

Position.........Team[s].........Games [based on Innings when possible].......PO.........A........PB........E.........CS% ........Rating........Player X Rating..........+/-Defense Runs

Preliminary to that statistical page is an annual page[s] devoted to current events, baseball events, player events during each season.

Any visually gifted types, with ideas about formatting and eye-catching presentation that might absorb data presentation as well, please fire away. I would be most grateful for ideas and if I ever get published will no doubt credit all contributors of any ideas that are constructive in finishing the project.

Tomorrow, I'll be doing edits, and double-checks of statistical data inputs [already done] for 1973. I am on track to polish off through 2011 before the 4th of July. I hope to devote June, July and August for securing rights to photos, illustrative diagrams, page design, color element set-up, and promotion.

Also, even though it may add a bit of work, if you have any unique [yet documented or documentable] insights into players, great defensive plays; OR current event items over the 1901-2011 period ... send 'em on and I may well use them, if they fit the context.

P.S. Questions on players and comps welcome.

[Tyrus4189Cobb]

I'm not a guy who can easily point out mathematical flaws, Lee, so as far as I'm concerned, your system is fine. I'm still interested in first base because I'm finding more and more evidence that Pujols is a top 5 all-time fielding first baseman. I'm also interested in how catchers will turn out because it doesn't seem to address how efficient catchers are at throwing out runners. Yet I understand that this is a metric of fielding efficiency so throwing out runners is a bit irrelevant.

Also, and I may have missed this because I'm still trying to wrap my head around a statistic based on reading, what do the numbers look like before you convert them to the fielding percentage form?

As for your questions, I would enjoy the book using the suggested format of longer entries based on noteworthy defenders. You can list players alphabetically or numerically (based on your metric) and have longer articles about the notable ones giving some photos, history, how they fit into your system, and even personal experiences.

If we converted your metric from fielding ability to ability in the visual arts, mine's about .003. But no harm in brainstorming, right?
-Noteworthy players get a separate "chapter" with an article-like title below a photograph. You then discuss different aspects about them like I mentioned. Other players are simply listed box-by-box including name, metric number, years played, whatever. In the box you can include a little blurb for those not defensively noteworthy but known in the game (McCovey, McGwire). The noteworthy "chapters" act as a break in between. Then you resume the other players.
-The first part of the book describes your metric with a few examples of ballplayers.
-Dedicate the right side of a page to the box format. The left side can have pictures or your notes/observations, like Buckner's crippling ankle issue. You don't have to do this for every page, though.

[Jackaroo Dave]

I'd suggest a player section with blurbs of varying lengths about players of note you have something new to say about, and an appendix with comprehensive listings. ("Comprehensive" within reason, of course.) You might find 3 levels convenient--mini-essays, notes, and data lines. Sometimes a player's story will help you with the narrative, too, so sidebars in the history section could be useful. That would be where I'd out the top ten lists, too.

But a lot of fans value completeness: "Hey, where's Coot Veal?" while others want to hit the high spots without wading through the brush. I think you want to appeal to both aspects of fandom.

[leewileyfan]

I'm not a guy who can easily point out mathematical flaws, Lee, so as far as I'm concerned, your system is fine. I'm still interested in first base because I'm finding more and more evidence that Pujols is a top 5 all-time fielding first baseman. I'm also interested in how catchers will turn out because it doesn't seem to address how efficient catchers are at throwing out runners. Yet I understand that this is a metric of fielding efficiency so throwing out runners is a bit irrelevant.

Dave: I believe it's this thread in which another poster asked for 1B evaluations, particularly because [iirc] he was hard-pressed to accept Pujols as a top defender at 1B.

I reported back with some detailed generational notations on 1B, in time categories, from 1901 to present, with my findings and evaluations. Later, at the top of this page, I was challenged on one of my "finalists," Bill Buckner. I defended the rating and explained why,

As for catchers, after many hours of wrestling with CERA [catcher ERA-type stuff], I came to a confirmed conclusion that it was a total waste of time. I also explored K/BB pitching ratios to look into the possible impact of certain catchers upon pitching staff performance; but that too seemed to generate more creative imaginings and scenarios than reality. So, after setting it aside for months, I decided to try a LESS IS MORE approach.

1. I use only Assists, DP's, PB, E, WP and Caught Stealing percentages [focus on denominators, too]; and have come up with what I believe really capture's the essence of catcher defensive contribution. Although I haven't pursued it as deeply as I'd like [YET], I have been working the WP/PB elements into a subset, relative to game -calling and battery communication.

2. In the catching subset, the raw numerical data might well look like .9809 for one catcher and 1.085 for another, totally unlike a 4.7708 or a 4.3198 that might be spit out for two competing 2B. EACH position is a unique territory [although the entire construct recognizes interplay among positions]. That is why each position has a template [of excellence, variable and evolving] against which a conversion formula can convert all the disparity into uniformity. I do explain all these steps in the introductory chapters before the year-by-year data results and comps.

For example, the catchers above are Johnny Bench and Bob Boone, in 1973. The 2B are Tommy Helms and Felix Millan, same year.

Every data input [PO, A, DP, E], as fitting, has different weighted elements, which net to the player model. That, measured to the template, and converted produces the rating.

Player X is a model, a floating variable concocted "player" at any position, every game, who has been pre-destined to be about -8 defense runs below an average MLB player at the position. If a SS has a rating of .979; and Player X is at .925, the SS will be presented as +20.2 Defense Runs better that Player X.

If a reader wanted to ask, "Well, how is that SS, compared to "average?" Player X being around -8 DR, the SS is about +12.2 DR above average.

Further, if the SS played in 110 of his team's 162 games, his playing time is 110/162 = .679 * 20.2 = +13.73 Defense Runs > Player X AND +5.73 Defense Runs saved above average SS, adjusted to playing time.

I believe that all the complexity is in the groundwork defining the system elements. Further complexity is necessitated only by sensitivity to game changes. The game has been evolving in almost every aspect over 112 seasons; and the heart of the metric is sensitivity to changes.

Hope this helps; and thanks for the ideas.

[leewileyfan]

Dave:

By the way, I ran Coot Veal for two seasons at SS:

1958 DET Games 55.8......PO 95.........A 160.............DP 30..........E 5..........Rating .935 = +3.90 > Player X; -4.1 Average
Playing time = .362, so +1.4 DR > Player X; -1.48 DR [average]

1961 WAS Games 55.8.....PO 130.......A 172.............DP 43..........E 8...........Rating .965 = +15.1 >Player X; +7.1 > Average
Playing time= .552, so + 5.32 DR > Player X; + 2.50 DR [average]

Coot, not included, in the original MS for either season, may now be an exemplar of utility roles and their sometimes surprising values, think Pokey Reese, BSX, SS, 2004.

[leewileyfan]

FWIW and in passing, the work I have been doing in editing and updating the 1970's is revealing something interesting, for me, at least. With some recent interest [in the sabermetric community] with "framing" pitchers [by catchers] ... then putting some defensive runs value to same, a few less heralded catchers MAY be giving subtle evidence that they were very good at it. Right now, my focus is on Mitterwald and May.

Anybody out there familiar with them ... or have seen them play often? Comments on their performance behind the plate are most welcome.

Working through 1978 tonight.

[leewileyfan]

Tyrus and Jackaroo Dave, in making welcome suggestions for book/stat/player presentation formats, have made references of interest in 1B. In posts exchanged on that topic, I mentioned a very favorable rating for Bill Buckner, much remembered for a critical WS error that cost the Red Sox a shot at clinching a title in 1986.

Having now just completed final edits and tweaks for 1987 data inputs, Buckner was indeed a top-notch 1b before the crippling longer term effects of an ankle staph infection impressed his lack of movement on future fan memories.

Buckner started as an OF and moved to 1B, where he excelled defensively between 1978 and 1987. Applying my metric to 1B, Buckner was a prime contender for defensive awards at 1B in 1978 [.994]; 1979 [.993]; 1983 [1.040]; 1985 [1.004], and 1986 [.994]. A fair evaluation would have placed him first at least twice.

This isn't bad when the competition includes Keith Hernandez, Chris Chambliss, Tony Perez and Wally Joyner.

As far as discussing what the raw numbers look like and how the conversion to homogeneous-looking RATINGS, I figured I'd show what the line entry for Bill Buckner [1978] will probably look like in the final product:

[POSITION] is a given, with players listed by that heading.

Player...............Team......Games.....PO......A ......DP......E......Data Sum.....Template.....Rating......Player X......Def. Runs

Buckner, Bill........CHI.........100......1,075..83.......8 5......6......2.8946..........2.950......... .994....... .925 ....... +15.66


By explanation/clarification:

1. Each data entry for each position, as to +/- value and weighting is unique to that position. The net for each player is referred to as Data Sum in the example above.

2. Player X is a model defensive player, who can be inserted at each position. He will always be around 8.4 - 9.7 defense runs below the level of AVERAGE for that position in any given season. When I list position players for a season, I always try to include at least one, who, being that range BETTER than Player X, is a solid proxy for anyone who wants to know the AVERAGE.

3. Average at 1B [1978] is about .962. Buckner, at .994, is +7.3 DR > AVG and +15.7 DR better than Player X.

4. Why Player X? There are a couple of reasons, one of which I rely on to keep me from spinning my wheels over minutiae:

a. Average is changeable, season to season and between leagues in any season. In setting up the metric, I wanted a stable, yet adaptable model that could flow with changes in the game. Each season, he is predestined to be floating at a fixed level below whatever average may be. The ratings let me know how the current crop of players is faring compared to Player X. When I see somebody in that +8 - +9 DR range, I know he's average against his peers.

b. I cannot get my head around Replacement Player. I will not knock the concept, nor the advocacy of its fans. For me [only speaking for myself], the only value I see for RP would be in the front office budget process to answer the question: What kind of player performance can I expect at each slot on the roster, IF i want to pay every player minimum MLB wage? Then I can anticipate a team and its performance at each position, and can expect to win about 48 games and lose about 114 in MLB season.

c. I do NOT put a bat into Player X's hands; nor do have have him make an occasional start or relief appearance on the mound. Player X is a fielder, probably a veteran who's lost a step or two OR a raw rookie who makes a rash play here and there but is generally reliable. Player X is an expectation level who NEVER disappoints. He his many cuts above "replacement."

d. This may seem a contradiction [but so be it]: IF Player X were to be on a roster in real flesh and blood, he might surprise with the glove then fall short with the bat: Think Jimmy Bloodworth, 1950 Phillie Whiz Kids; or Pokey Reese, 2004 BOSOX.

Now working through the 1994 season and noting the progress of Omar Vizquel at SS. Will post a year-by-year rating summary of Vizquel ratings when I wrap up 2007 in about 10-12 days. It seems his defense generates more controversy between supporters and naysayers; and I'd like to get feedback on ratings.

Meanwhile, any input, comments or player ratings questions are welcome.

[leewileyfan]

Just by way of summing up progress to date and sharing some information in the context of MLB defense as it has evolved since 1901, I jus finished updated edits on the 2000 season.

FWIW, I figured I'd list top position player results for the 2000 season, comparing them to Gold Glove Awards made for that season:

1. Players whose names are preceded by an asterisk received Gold Glove Awards for 2000.
2. Player Ratings, by my metric, are listed in [.900] form after each name.
3. If a player with a high rating [1.000] is NOT my listed top choice, it is because another player [.979] had much more playing time.

2000 Season Results:

Position.........................NL............... ..........AL.....................NL............... ..........AL

Catcher: *Matheny [.980].............*I. Rodriguez [.963]; Blanco [1.001]........Ausmus [.957], playing time

First Base: *Snow [965]................*Olerud; [.999]..........Helton [1.006]........Olerud [.999]

Second Base: *Reese [.966]...........*Alomar [.961];.........Belliard [.970].........Velarde [.995]

Third Base: *Rolen [.956]................*Fryman [.946];.......Cirillo [.965]............Glaus [.967]

Shortstop: *Perez [.986].................*Vizquel [.951]........Perez [.986]............Valentin [.984]
.................................................. .................................................. OR...Rodriguez [.976]
.................................................. .................................................. OR...Tejada [.972]

Golden Glove Awards are largely clustered by Center Fielders, since three awards are made per league per season. My approach is position specific; but here I make allowance for the GG approach:

OF: National League:........................................... ...........OF: American League:

*Jones [.979]; *Finley [.944]; *Edmonds [.964]; .......*Erstad [1.006]; *Williams [.970]; *Dye [.942]

Metric:

Jones [.979]; Hidalgo [.992]; Griffey, Jr. [.977]; .........*Erstad [1.06]; Hunter [1.008]; Stewart [.980]

[leewileyfan]

As I come through the home stretch of editing this manuscript, and having just finished the review of the 2002 season, I'll just keep on posting notes to stay focused on the wrap up, which is only a week or so away.

For 2002, I'll just stick with shortstops. My basic model for defense runs comparison, Player X, is stuck perennially at .925, which is designed to be faithfully around 8-9 runs below MLB position average. By this measure, MLB 2002 average [at SS] was about .949.

The players who got formal Gold Glove awards are indicated with an asterisk.

*Renteria [.934]
Furcal [.969]
Cabrera [.970]
Uribe [.970]
Hernandez [.973]
Rollins [.955]
Ordonez [.951]
Wilson [.988]
Cruz [.945]
Garciaparra [.956]
Clayton [.963]
Eckstein [.942]
Vizquel [.961]
Jeter [.903]
Bordick [.994]
Tejada [.966]
Perez [.962]
*Rodriguez, Alex TEX [.965]

My metric suggests that Bordick and Uribe were the actual defensive standouts at SS in 2002, taking into consideration both their metric ratings and their playing time. I'd have no argument with anyone in favor of Jack Wilson in the NL.

[leewileyfan]

Still [and always] searching for feedback, I have run into an interesting and challenging proposition, editing the last several years through 2011. That is something I'll share here, wondering if anyone else has noticed what has now struck me no less than THREE times:

First, back in 2004 and 2005 while I was compiling the data, by season, ever watchful for data feedback from the results of the numerical formulas doing their crunching. I had wanted, from the start, to do enough planning and weighing of data inputs, so that when it came to the tedious job of data entry, I'd could just kind of "tune out" and be a data entry clerk, alerted only by sudden trends in the numbers. I had expected UPWARD jumps in ratings, to warn me of things getting easier due to some changes in rules, playing conditions, or equipment improvements. What I got at one position was a notable DECLINE. That decline was among catchers.

Second, as the book manuscript was in the hands of a New York agent who wanted it [held it for eighteen months, then = NADA], I had put it on the shelf, only occasionally doing updates, AGAIN seeing that same downward trend in catching.

Finally, doing this year-by-year complete review, it strikes me again ... a notable and undeniable decline in catching.

Each time in my three experiences, the red flags started going up just before 1959. With a burst or two upward, the swoon continues, until it dawns on me that, of all the positions, catching needs a bit of fine-tuning in both templates [models of expectation] and Player X [substitute model] rating.

I have some very definite ideas of the root causes of this decline and its effects on the present game. However, I invite others who EITHER have noticed this trend OR others who disagree and think any such idea is crazy.

Either way, or in between, observations are invited.

Added note: It's not a thing that disturbs the integrity of the metric. It's more a matter of the level of catching declining, such that MLB average has been closing in on the Player X level. I believe it's the first time [over 112 seasons] that Player X will get a cut in his level of expectation, from about .925 to perhaps .910.

[leewileyfan]

There are several threads here on "Most Underrated Players," "Best at Position" and/or "Best Defenders by Position by Decade" and the like. I had no desire to hijack or drive other threads into tangents; but I believe a recent thread lends itself to study here.

The player under discussion in Lou Whitaker. late '70s into early-mid '90s and how he is regarded overall and on defense, in particular.

I have reviewed my metric data for the bulk of Whitaker's career [1978-1993]. The numerical results [in brackets] are his ratings, presented with other 2B from both Leagues [also in brackets]. In each season, players winning official Gold Glove Awards are highlighted with an asterisk.

What is fascinating for me is the number of dynamics at work DURING a player's career window:

-The player's own individual arc of play, with some blips [up and down] along the way;
-The persistent possibility of injury [Whitaker, 1988]
-The perceptions of "evaluators;"
-The rise and fall of competitors

For Clarification: Player Ratings are the products of raw input data converted into a format that RESEMBLES fielding percentage [for the sake of familiarity] ... generally between .900 and 1.000 [although many players can fall below .900 or "break the model" by exceeding 1.000 slightly].

At 2B, the MLB average player will have a rating at @ .950.

Season.........................A.L. 2B....................................N.L. 2B

1978..........................Whitaker [1.001].................*Lopes [.942]
................................*F. White [.957]

1979..........................Whitaker [.992]...................*Trillo [978]
.................................*F. White [.980].................*Flynn [.952]

1980..........................Whitaker [.973]...................*Trillo [.979]
.................................*F. White [.973]

1981..........................Whitaker [.994]...................*Trillo [.978]
.................................Gantner [1.000]
.................................* F. White [.959]

1982..........................Whitaker [.989]...................*Trillo [.965]
................................*F. White [.9 ]
.................................Bernazard [.993

1983...........................*Whitaker [.943].................*Sandberg [1.000]
..................................Grich [1.000].......................Hubbard [.990]
..................................Gantner [1.000]

1984............................*Whitaker [.955].................*Sandberg [.997]
...................................F. White [.993]....................Hubbard [.993]
...................................Cruz [.990]

1985...........................*Whitaker [.951].................*Sandberg [.980]
..................................Grich [.995].......................Hubbard .995]
..................................Gantner [.987]
..................................F. White [.980]

1986..........................*F. White [.964]................*Sandberg [.966]
..................................Whitaker [.969].................Hubbard [1.000]
..................................H. Reynolds [.998]

1987...........................*F. White [.969]...............*Sandberg [.957]
...................................M. Barrett [.988]................Ray [.970]
...................................Whitaker [.953]..................Hubbard [.999]

1988............................*H. Reynolds [.960]............Sandberg [.981]
................part-season....Whitaker [.937]................
....................................F. White [.975]
....................................W. Randolph [.977]

1989............................*H. Reynolds [.991]..........Sandberg [.957]
.....................................Whitaker [.965]...............Oquendo [.982]
.....................................F. White [.975]
.....................................Ray [.972]

1990............................*H. Reynolds [.968].........*Sandberg [.962]
.....................................J. Reed [.989]...............R. Thompson [.987]
.....................................Whitaker [.987]

1991..............................*R. Alomar [.934]..........*Sandberg [.964]
......................................Whitaker [.971]..............Oquendo [.980]
......................................Randolph [.989]
......................................Knoblauch [.973]

1992...............................*R. Alomar [.925]..........*Lind [.969]
.......................................J. Reed [.989]...............Sandberg [.984]
.......declining playing time]...Whitaker [.950]

1993...............................*R. Alomar [.953].........*Thompson [.976]
.........96 games.................Whitaker [.986].............Reed [.972]
.......................................S. Fletcher [.975]
........................................Baerga [.966]

[leewileyfan]

Jackaroo Dave [Post #28] and Tyrus R Cobb[Post #31] have made welcome suggestions in book manuscript player/ratings presentations, perhaps with pictures of selected players and lesser known snippets about them

Scanning the years [1901-2011] this raises many possibilities, like Dave Brain [1906; 1907]. The more I read about Brain, the more I learn what a topic of controversy he has been among metric creators and baseball statistical historians. Was he bad defensively? Was he merely sub-par? Or was he a very gifted defensive player, sadly under-rated by misguided measuring standards?

I have Brain as above average in 1906 and worthy of a top defender award at 3B in 1907. As my metric sees it, Brain committed lots of errors as a cost of gifted range.

If you have any players [1901-Present] you'd like to be reviewed/evaluated, please suggest them here, with whatever comments/suggestions you might also want to add.

[honus14]

If you have any players [1901-Present] you'd like to be reviewed/evaluated, please suggest them here, with whatever comments/suggestions you might also want to add.

Lee, first of all let me say that as a Cub fan, watching Bill Buckner for several years, I concluded that he was an outstanding first baseman as well, particularly given the ankle problems, although he did strongly prefer the 3-1 play to the 3 unassisted.

I was wondering if you'd take a look at Ryne Sandberg. From 1983-85 he had outstanding range factors, despite having been moved from third to second. Then, beginning in 1986, his RFs drop to near or below league, and pretty much stay there. I wondered, at the time, what happened, and no one but me ever seemed to notice. I wanted to see if your metric shows something similar, or if it may be an artifact of some sort.

[leewileyfan]

Lee, first of all let me say that as a Cub fan, watching Bill Buckner for several years, I concluded that he was an outstanding first baseman as well, particularly given the ankle problems, although he did strongly prefer the 3-1 play to the 3 unassisted.

Thanks for the feed back. Some would criticize Buckner for 3-1 plays; but it is the pitcher's job to hustle over there for just such plays, as part of the game.

RE:

: Ryne Sandberg. From 1983-85 he had outstanding range factors, despite having been moved from third to second. Then, beginning in 1986, his RFs drop to near or below league, and pretty much stay there. I wondered, at the time, what happened, and no one but me ever seemed to notice. I wanted to see if your metric shows something similar ...

Yes. In this thread [above, Post #40] I brought Ryne Sandberg into my comments regarding Lou Whitaker. About midway through that post, I get to 1983, when Sandberg is indeed outstanding, at 1.000. In a league where the average 2B is at .950, that converts to +17.4 defense rus above average over a full season of play.

As you can see, my metric numbers agree with your observations of a consistent level of excellence through 1985. While I have him at .966 [solid and above the .950 + average], there is a modest but steady decline through 1989, although he NEVER falls below average.

There is a mild bump [.962; .964] in 1990 and 1991 and an excellent performance at .984 in 1992, by which time Sandberg has hit the age of 32. In 1993 [.968] Sandberg's playing time is down to about 110 games; but he is having a superb 1994 season [.996] when he suffers a severe ankle injury which is part of his decision to retire. His is 34 at this time.

His return in 1996 has him at .956 [average + maybe a run]; but he is unhappy with his hitting which does produce 25 HRs.

Between 1987 and 1989 I believe the very modest drop off might have been, in part, due to Cub roster turnover and lack of competitive building in the organization, 1986, 1987, 1988. Perhaps the jump in 1989 helped spur the modest boost in 1991 and 1992; but that would be pure conjecture on my part.

Even with a slight decline, he never lost his defensive sharpness, even after injury, retirement and Father Time moved against him.

Thanks for the question.

[honus14]

Yikes, I missed that whole right column while I was focusing on Whitaker. Thanks.

[BigRon]

PV: I expected the favorable mention of Bill Buckner on defense to raise a few eyebrows. However, your recollection of a very nearly crippled Buckner and the WS disaster so indelibly captured on film merely reminds me [us] how visually oriented we are. The totality of Buckner was not captured either on the clip or in the entire context of that disastrous season in the field.

A competent fielder before he contracted a staph infection in his ankle, Buckner was moved to 1B as an adjustment to that condition. He was an excellent defensive 1B, almost from the start. My metric has him for his 1B career at .983, which would get him about +7 defense runs per season above average. A Keith Hernandez, > 1.000 would be about +14.5 DR per season above average.

I'll stand by the .983 rating for Buckner.

I don't usually like to mention other metrics or metric founders [especially the historic, much-published and of guru status]; but since they are often the mirrors I have to explain my findings to, I will make a few observations of my own about James:

1. He seems much-guided by visually captured, indelible "moments," like the unfortunate image he painted of a young Harlond Clift tripping over his own glove. He passes them on in each edition, giving them a kind of permanence.

2. I read the 1st edition of Dewan's "Bible," and was astonished at how so many of James' narrative snippets seemed to contradict or be at odds with the numbers presented. After a few added perusals, I tossed the "Bible" into the trash, feeling that I had committed no sacrilege at all.

While I can easily allow that there might be flaws here and there in my approach, I am confident of their trustworthiness. Nobody, IMHO, has yet published anything to date that I have found to be really solid or well defined.

lee, you seem to scoff at the mention of James' name. Just to set the record straight, James evaluates Buckner as a lifetime B defender at 1B- clearly above average. That may or may not coincide well with your ranking, but James definitely sees him as above average.

[leewileyfan]

lee, you seem to scoff at the mention of James' name. Just to set the record straight, James evaluates Buckner as a lifetime B defender at 1B- clearly above average. That may or may not coincide well with your ranking, but James definitely sees him as above average.

Big Ron:

Here is what I wrote, precisely, about Bill James: [which I do not believe suggests scoffing at the mention of his name]:

: I'll stand by the .983 rating for Buckner.

I don't usually like to mention other metrics or metric founders [especially the historic, much-published and of guru status]; but since they are often the mirrors I have to explain my findings to, I will make a few observations of my own about James:

1. He seems much-guided by visually captured, indelible "moments," like the unfortunate image he painted of a young Harlond Clift tripping over his own glove. He passes them on in each edition, giving them a kind of permanence.

2. I read the 1st edition of Dewan's "Bible," and was astonished at how so many of James' narrative snippets seemed to contradict or be at odds with the numbers presented. After a few added perusals, I tossed the "Bible" into the trash, feeling that I had committed no sacrilege at all.

Posters have often cited other metrics [published and widely known] in commenting on my metric's evaluations, pointing out their own individual reactions and/or asking me for additional input. This is welcome and as it should be.

However, there are times when a metric is quoted as gospel truth; OR, when any other metric [including my own] DARES to be in disagreement with the chosen infallible source[s]. Then, I merely speak out and tell it exactly as I see it.

I have read much of Bill James' work. I have also paid close attention to comments he has made and the way he conducts himself in interviews. I have paid particular heed to his narrative snippets about players, both in his own sabermetric studies and those of others [Dewan's "Bible;" version 1]. What I have found [often] is a snide, cynical content in the narrative that, for me, discounts the credibility of the statistical data presented.
I noted in the Dewan Defense Bible, 1, that James' comments frequently ran contrary or, at least, obliquely at variance with the stated ratings based on numbers. I tossed the thing out after two readings.

I don't believe I am being defensive [or overly defensive], when the primary and instantaneous response I get is some version of ... "But Bill James says ..." OR "How could you possibly question UZR" OR "You are WRONG because Davenport says ..."

I am pleased that James rates Buckner as well as he does. However, I do not place trust in my metric BECAUSE it comports to Bill James and his opinions. [James is NOT my source of "validation."]

If you could challenge me with a position player, whose career essentially happened within the 1901-2011 time frame, making it a position player whose defensive ratings vary widely [and perhaps generated some heated debate], I would love to receive that from you. I'd gladly look up the player and his numbers and crank them in to generate my own numbers. I'd welcome that challenge and I'd let the metric talk for me. For me, the best of all worlds is me as a data entry clerk [having done the preliminary homework] ... with the metric being its own spokesman.

[BigRon]

If you could challenge me with a position player, whose career essentially happened within the 1901-2011 time frame, making it a position player whose defensive ratings vary widely [and perhaps generated some heated debate], I would love to receive that from you. I'd gladly look up the player and his numbers and crank them in to generate my own numbers. I'd welcome that challenge and I'd let the metric talk for me. For me, the best of all worlds is me as a data entry clerk [having done the preliminary homework] ... with the metric being its own spokesman.

Okay- why don't you run Granny Hamner.

[leewileyfan]

Granny Hamner it is.

What a time to be 17 years old. It is 1944, the nation is at War; and you [Granny Hamner] are called to the Big Leagues and experience 77 at-bats, hitting .247 and getting a taste. It's hard to imagine the emotional challenges of the times for a kid to get called up, again and again, from ages 17 through 20, each time getting a diminished shot at the Major Leagues.

In 1949, you stick, at age 22.

1949: Rating .962, best in the National League among SS. This is full-time play and it is +5.7 defense runs above N.L. average at about .947.

1950: The Whiz Kids season; and Hamner is at .955 among SS, who average about .946 again, +3.5 defense runs above average.

1951: At .959, the metric places Hamner #1 again among N.L. SS - about + 7.1 defense runs above average [.940].

1952: Hamner [.956] ranks second in the N.L. behind Roy McMillan.

1953: The SPLIT season. with Hamner, now 26, playing 542.1 innings at SS, rating .920, which is -10 defense runs below average at SS [.947]. However, being only 39% of team's 2B innings, the actual cost is -3.94. defense runs.

1953: 2B seems to agree with Hamner, whose play at 2B rates .964, which would equal that of top rated 2B in the N.L. His play at 2B is equal to 57% of team schedule; and he is +4.9 defense runs above average [.950]. Adjusted for playing time, he is + 2.8 DR.

Over all, Hamner's play at the two positions nets out to -1.14 defense runs,

1954: Hamner, with 1,336 innings at 2B rates .932, essentially full-time and is - 5.2 defense runs below average.

1955: SPLIT #2, at 2B = 69.1 games, rating .908, -14.2 defense runs blow average [.949], adjusted to -6.37 for playing time.
at SS = 29.1 games, rating .910, -13.8 defense runs below average [.947], adjusted to =2.6 for playing time.

1956: Back at SS; year of severe shoulder injury: 100 games; rating .939 = -3.74 defense runs below average [.949] adjusted to -2.4.

1957: Hamner, now barely 30, plays in 117 games and rates an awful .891 at 2B. This is -20.5 DR below average [.950] and adjusted for playing time is -15.6 defense runs.

I did not pursue Hamner's career beyond this point. He seems to fit the sometime profile of the kid brought up too soon, then shuffled back and forth, who shines brilliantly for a time and then goes into a sudden decline, all to soon.

From 1949 through 1953, Granny Hamner was a very solid defensive infielder who, by the age of 26, had showed some signs that called for position shuffling [where he did very well at 2B in 1953].

What came after that was downhill.

[leewileyfan]

Expanding the Hamner comps a bit, to include some of his contemporaries at SS:

1949: Hamner [.962] is best in N.L. In A.L. Stephens [.974] and Joost [.969] are a bit higher.

1950: Hamner [.955] ranks third in N.L. behind Smalley [.980] and Marion [.975]. A.L. SS Carrasquel [.980], Lipon [.976], Joost [.969], Rizzuto [.965] and Stephens [.964] are higher. Over a complete 154 game schedule the difference between .980 and .955 would be about 9.4 defense runs.

1951: Hamner is on top in N.L. [.957]. In the A.L., Carrasquel [.979], Joost [.970] and Rizzuto [.958] have an edge.

1952: Hamner [.956] is behind Roy mcMillan [.967] and virtually tied with Solly Hemus. In the A.L., only Phil Rizzuto [.969] is higher.

1953: The first positional split season, Hamner's play at 2B is .964 in 88 games. Red Schoendienst [.974] and Rocky Bridges [.965] edge him in the N.L. In the A.L. Bobby Avila [.980] and Wayne Terwilliger [.970] rate higher.

[leewileyfan]

In another thread, a poster asked about the B-R evaluation of Tim Raines as a LF. Just keeping in practice, here's how I have Raines rated:

Season............Rating...........Player X............MLB Sea. AVG............DR>Player X..........DR> MLBLFAVG

1982.............. .952 ........... .925 ............. .947 ....................... +10.07...................+2.07
1983.............. .966............ .925.............. .948........................ +15.36...................+7.25
1984..played CF].964............ .940.............. .951........................ +11.05...................+3.00 [CF]
1985............... .955............ .925.............. .945........................ +13.36................... +5.10
1986............... .936............ .925.............. .943........................ + 4.30................... -4.00
1987............... .963............ .925.............. .946........................ +14.20................... +6.50
1988............... .966............ .925.............. .947.........................+15.40............... .... +7.25
1989............... .942............ .925.............. .949.........................+ 6.33................... -2.12
1990............... .947............ .925.............. .947........................ +8.33.................... +0.00
1991............... .969............ .925.............. .949.........................+16.72............... .... +8.62
1992...............1.002............ .925............... .947.........................+25.77............... ...+17.65
1993................ .957............ .925............... .946..........................+12.02.............. .....+ 4.02
1994................ .964............ .925............... .945..........................+14.88.............. .....+ 6.75
1995................ .948............ .925............... .946......................... + 8.67....................+ 0.33

To explain:

Player X is a model, a stable picture of defense level over a span of seasons, fixed at approximately 8 defense runs below an average MLB regular at each position. This provides a realistic level of expectation for defensive play that is considerably above what many metrics have devised for Replacement Level. Player X is a guy you might trade for or pick up on waivers. The realistic thing about the Player X level is the recurring fact, over the years I have studied [1901-Present] that during a career, a solid defender may often be closer to the Player X level in an "off" year, BEFORE entering a clear career decline slide.

Over the years covered, Raines is +176.46 DR better than Player X [including 1984, played in CF]. Compared to MLB average, this is reduced to +62.42 DR. A final adjustment would have to be made for actual innings played as a percentage of team position total each season. I didn'y go into that; but, if we consider 70% as a fair playing time estimate, Raines would be 43-44 DR above average over the seasons reviewed.