you're reading...
All-Time HoME Leaders, Uncategorized

All-Time HoME Leaders, First Base – 21-40

Today we continue through our journey around the diamond listing the best ever to play the game at each position. Well, actually, we’re still at first base. A week ago, we revealed our top-20 at the position, today it’s 21-40, and in a week we’ll name the top-20 second basemen ever. All rankings are based on Eric’s CHEWS+ and my MAPES+ lists. And we won’t stop until we give you the top-40 at every position and the top-120 pitchers.

First Base – 21-40

1B, 21-40

Where do we project the active player to finish in our rankings?

Joey Votto: Eric doesn’t address Votto because he did so last week as the 20th ranked guy on his list. For a player with six OBP titles and almost 55 WAR, Votto is criminally underrated. Unfortunately for him, he’s only hit 30 homers twice and only drive in 100 three times. On the other hand, just looking at straight WAR, he’s never had an 8-win season. But I digress. This question is about where he’ll end up. My adjusted numbers gave him 7.47 WAR a year ago. But he will be 34 this year. Imagine a graceful decline of 5, 4, 3, 2, and 1 WAR. He’d pass seven guys and move into 17th place on my list. But maybe the decline will be even less pronounced, seeing him fall to 6 WAR this year and then age gracefully. He’s now fighting George Sisler for 10th in history. I’m not saying it’ll happen, but it certainly could. —Miller

Where do our rankings diverge the most from the conventional wisdom?

It’s an easy enough call to say that it’s David Ortiz. After all, conventional wisdom says he’s going to waltz into the Hall. But I think nearly 100% of the writers for vote for Papi will do so noting his specialness or clutchiness or Papiness or all three. Ortiz deserves the Hall based a lot more on Fame than on greatness. And if you’re thinking his playoff greatness puts him over the line, well, maybe it does, though I did contribute a meandering post about the incredible amount of value you’d need to give playoff excellence to get the Sox great over the line. —Miller

It would be easy to spotlight old timers like Harry Stovey or Jack Fournier or even Dolph Camilli here. But there’s two guys on this list where I see divergence. First, no one during his career ever thought that John Olerud was a Hall-caliber player. He got attention when he took a run at .400 early in his career. He got some kudos as the slick-scooping glue that held the Mets’ infield defense together in the Piazza era. He was also known for the piano he carried on his back when he ran the bases. But his case for the Hall is actually pretty damned good thanks to a combination of good hitting and excellent fielding. Then there’s Harm Killebrew. The 500 homers pretty much starts and ends the conversation on him, and, ipso facto, he’s a Hall of Famer. Well, he was an abysmal fielder, a bad baserunner, and hit into a lot of double plays. The entire package isn’t nearly as good as the gaudy homer figures would indicate, and if you only stop at 573, you won’t agree with us.—Eric

Where do we disagree with one another the most?

Nothing to see here. There’s no meaningful disagreement.

Are there any players that MAPES+/CHEWS+ might overrate or underrate? 

 In this group, not especially. So I’d like to take just a moment to talk about how a team might have underrated a player. In 1967, the Reds moved Tony Perez to third base. He’d never played an MLB inning at the hot corner, but during his minor league apprenticeship, he’d played a vast majority of his games there. Arriving in Cincy in 1964, Perez played nothing but first base for three years, and he was slightly below average (-3 runs) in 174 games. The Reds realigned their infield for ’67, moving Perez off first base, pushing incumbent Tommy Helms to second base, putting Pete Rose to pasture in left field, and putting defensively inept left fielder Deron Johnson at first base where he could do less damage (and they were right). In 1966, these players combined for -24 runs. In 1967 they combined for -10. Bob Howsam and Dave Bristol saved themselves at least a win’s worth of runs just by putting their players where their skills made the most sense. Perez was the weakest link at -9 runs in 1967, but over the next four years, he accumulated positive value at third base.

Prior to the 1972 season, the Reds acquired infielder Denis Menke who had at one time played a decent shortstop, but whose glove faltered badly. He could hit a little, and the Reds picked him up in one of the most lopsided trades in baseball history.

November 29, 1971: The Houston Astros trade Ed Armbrister, Jack Billingham, Cesar Geronimo, Denis Menke, and Joe Morgan to the Cincinnati Reds for Tommy Helms, Lee May, and Jimmy Stewart.

Morgan alone brought back 58 WAR, with Geronimo returning 13, Menke 4, Billingham 2, and Armbrister none. For their efforts, the Astros got 4 wins from Helms, though they later flipped him for Art Howe who earned 13 WAR with the ‘Stros; 6 from Lee May who they later made part of a trade for Enos Cabell and Rob Andrews; and -1 from Stewart whom they later cut.

For Perez’s part, however, the departure of May and Helms opened holes at first base and second base. Morgan, of course, would man the keystone sack. Who would play first base? They had options on hand. Obviously Perez had played there previously and was fine. Menke might also be stationed there. He had, in fact, been the Astros’ primary first baseman the year before and played there about as well as Perez had in the past. Pete Rose had moved from left field to right field in deference to rising youngster Bernie Carbo. Cabo had been worth 4.4 WAR in 1970, faltered in 1971, and after a slow start was dealt in 1972, pushing Rose back to left. The young Hal McRae got a lot of playing time and might have been an option. George Foster was on the roster but didn’t play much and hit poorly when he did. The Reds chose to stick Menke at third base and slide Perez across the diamond.

I wasn’t there, and I don’t know what all happened, but here’s the thing. Menke was a short timer however you looked at it, and the Reds had more good hitters than the WBA, WBC, and IBF combined. But Perez looked like a first baseman and hit like one, so the move made sense. But in retrospect, leaving Perez be would have given the Reds many more opportunities and made Perez more valuable. During his time at third base, Perez earned 12 positional runs (he did appear other places from time to time). Perez left Cincinnati after 1976, but in those five years, the positional adjustment for third base was three runs a season, or 15 runs over those five years. At first base it was -45 runs. If Perez played every day, the difference between the positions was 60 runs without factoring in defense or anything else. Perez was a good first baseman from 1972–1976 and picked up 17 fielding runs. So the net of his positional adjustment and fielding was -28 runs. Perez could have been a -42 run fielder at third base over that same time and still have earned out at the hot corner.

The worst defender with 50% of his games at third base in that time as Richie Hebner who “earned” -37 runs. Bill Madlock placed with -32 runs. Dave Roberts to show at -30. Bill Melton next at -29, and the only other player over -20 was Paul Schaal at -30. Whether contemporary observers saw it or not, Perez played a good enough third base to have likely avoided that kind of disaster artistry.

And that’s just the BBREF Rfield side of the story. DRA pegs him at 8 runs instead of 17 from 1972–1976, and it likes his defense a lot at third, to the tune of 22 runs from 1967–1971. I understand why Sparky Anderson and Bob Howsam made the move they did. It seemed like a good baseball move based on the understandings of the game in 1972. Looking backwards, though, it’s possible that the difference between Tony Perez first baseman and Tony Perez Hall of Merit third baseman came down simply to a trade, a decision, and 100 years of soon-to-be-obsolete conventional baseball wisdom.—Eric


16 thoughts on “All-Time HoME Leaders, First Base – 21-40

  1. IVe always considered Will Clark ane John Olerud to be very similar players and am not surprised to see they rate about the same in your world too. Do you think will Clark was more highly regarded than Olerud during their playing careers?

    Posted by Mike Teller | March 5, 2018, 2:21 pm
    • I’d answer that somewhere between “yes” and “oh, hell yes”. Clark was a big star by his second season. Olerud was the guy who wore that strange hat and sort of made a run at .400.

      Posted by Miller | March 5, 2018, 2:26 pm
  2. Great stuff, as always, guys!

    But I’m going to go to bat, so to say, for a couple of the sluggers. Well, for one of them, I’m only going to put up the fight of Larry Walker vs. Randy Johnson in the All-Star game, which is to say, not much of all.

    I have Killebrew in, just barely, but in past systems he has been on the wrong side of the borderline. I suppose our differing opinions once again center around DRA and its evaluation of Killebrew. I use DRA, but a slightly regressed version, and it only counts for one-third of my defensive values, while I believe both of you are at 70% for non-catchers. And DRA really dislikes Killebrew, so that would account for the discrepancy. No real sweat either way.

    However, I will vigorously defend Ortiz, and not just because like Miller, I am a Red Sox fan. I have Ortiz over the line before including any bonus points, particularly postseason. And here’s why:

    The DH position, throughout its entire history, has been critically underused by every team in the American League. One adjustment I make to WAA/WAR totals is to adjust them by a 9-year rolling total of positional Wins above the Median at each position (I use median instead of average to keep the data from being swayed by positional star gluts or troughs), instead of just using the positional values from BBRef which are based on positional switching. By adjusting for positional median, I feel I get a better number for the true value of that player in that year.

    The DH has been around for almost a half a century now, And never in its entire existence has the median DH had a positive WAA for any single year, let alone rolling 9-year average (The closest ever was -.05 in 2002). The closest the rolling average ever got to 0 was -.33 in 1998, while it peaked at over -1 several time over its first decade of existence. Now I don’t even use the full DH adjustment the entire time because I did create a manual override so that counting median adjustments, the DH cannot have a higher Rpos than 1b, although they are equal in several years.

    This tells me that having an Edgar Martinez or David Ortiz, or any kind of masher who isn’t bothered by not playing in the field, provides a huge advantage over aging sluggers in the final stages of their decine phases or as a placeholder spot for regulars nursing injuries that prevent them from playing their usual positions. I know that in these days of 7-8 man bullpens, a team wants as many Zobrist-types to fill the limited bench spots to keep as much positional flexibility as possible, and thus don’t want to keep a immobile guy who just hits on the roster, but they sacrifice actual on-field value to do so.

    Over the course of his career, I have Ortiz gaining an extra 9+ wins over the median DH, which after accounting for standard deviation adjustments, give him 64,5 mWAR.

    Posted by Michael Mengel | March 6, 2018, 10:57 am
    • Hi Michael,

      Please help me understand your position. And if I bastardize it, please note that I do not mean to be critical. Are you saying that because other teams have had some of their best offensive players also have an ability to play reasonable defense or better (or they were too daft to deploy their assets optimally), that Ortiz’ value should increase? If I am understanding correctly, I think I disagree, but I can certainly be persuaded.

      Every part of me wants Ortiz to deserve to be in the Hall of Fame. I would LOVE being wrong on this one.

      Posted by Miller | March 6, 2018, 11:39 am
      • What I’m trying to do is change the value of Rpos in my system to provide a better estimation of a players actual than the position-switching estimates that BBRef uses. if you’re familiar with Dan Rosenheck’s WARP system that he created for the Hall of Merit, it it very similar to that, except he attacks replacement-level directly, while I calculate from average (median) at each position and go from there. To me, going year-by- year to compare a position’s median WAA (or 9-year rolling average thereof, in my case), and using those results to adjust Rpos creates a more accurate determination of an actual player’s value. So essentially, I start of with the same Rpos that most of the world uses, but I tweak it based upon actual real-world results to get my numbers.

        However, I do employ some manual overrides to make sure that some things that violate common sense don’t come into play. One example is DH-1b. I adjust it so that a DH can never receive an overall bigger Rpos bonus (or lesser penalty) than a 1b for that year. Another example: Unaltered, 2b in the early 1960’s was so bad that their Rpos in my system would’ve been higher than that for SS – which is clearly counter-intuitive, so I put a forced cap that the Rpos for 2b could never come within .2 of SS. Also, modern CF is so relatively deep and Corner OF so barren that corner OF would be higher in an unadulterated systems, so I place a similar floor on the gap between CF and COF at .4.

        Whether these are a result of random fluctuations or what, I’m not sure. But positional values do rise and fall over time, and IMO, we have to evaluate players in their time (and then use league strength/standard deviation to compare across eras).

        For example, while SS has always had the highest Rpos adjustment, throughout most of baseball history, it actualy underrates how hard the position is, especially from the late 1960’s to the late 1980’s. Whether it was because of the proliferation of astroturf or what, SS of this era were generally putrid and receive Rpos bonuses of up to a whole win because of the lack of depth. Thus why in my system, Bert Campaneris is in (Concepcion just out), Trammell is not just in, but is in my no-brainer territory and why Cal is in my inner-circle.

        Back to Ortiz and DHs, I don’t know whether its an inefficiency of front offices/managers, just a drop-off in league-wide available hitting talent, or something else, but for some reason, the median DH has always been below average according to an average of bWAR/gWAR. Thus, for me, the position has been literally undervalued for WAA/WAR evaluation purposes. So if your purpose is to measure actual on-field value, regardless of the ambiguities of talent, how a player would fare in a different era, etc., I feel such adjustments are necessary to reflect what actually happened. YMMV.

        I apologize if this came across as too rambling or incoherent. Feel free to continue to pester me with questions. If you’d like I can email you my excel file with all the raw data, raw 9-year rolling averages and the manually adjusted numbers that I use in my rankings to examine them for yourself.

        Posted by Michael Mengel | March 6, 2018, 2:17 pm
  3. I forgot to add, I would have Ortiz at 22 in my 1b rankings, if I included him there – just behind Jason Giambi and where Joey Votto is currently (approx. 108 PEACE +).

    Posted by Michael Mengel | March 6, 2018, 10:59 am
    • Hi Michael,

      I think I understand where you’re coming from, and to me, one clarifying line is the following:

      “I adjust it so that a DH can never receive an overall bigger Rpos bonus (or lesser penalty) than a 1b for that year.”

      If it’s a bad one, please forgive this question. Why would DH ever have a penalty as small as 1B? Is it just because of quality of players at the position?

      Then a second question. Is using medium a confounding factor here since there are so few designated hitters (none in the NL and less than one per team in the AL) compared to most other positions?

      If you can explain the first answer (so that my limited brain can understand it), that might be our answer. Otherwise, perhaps the second answer really matters. I’d be happy to think it through with you or to be told to jump in a lake.

      Posted by Miller | March 6, 2018, 2:45 pm
      • I allowed, and yes, this is completely arbitrary, but to me it makes sense, for DH to move up to the 1b level, as opposed to maintaining the arbitrary gap between 2b and SS, and CF and Corner OF, in that essentially 1b and DH are grabbing from the same talent pool (give or take), while there are positional requirements that don’t allow some 2b to move to SS or Corner OF to CF. It’s just up to teams to optimally allocate the resources between 1b and DH – always put the better defender at 1b). I’m not sure if that exactly answered your question, but I hope so.

        As to the 2nd question, I used the median only for AL teams (otherwise the median every year would be basically 0 since the NL teams don’t use the DH enough to accumulate (or lose) much WAA at DH. Plus my median numbers aren’t just for the primary person at that position for each team – it’s based upon each team’s overall DH production.

        And I doubt I would tell you to jump in a lake – peer review/healthy debate is good.

        Posted by Michael Mengel | March 6, 2018, 3:04 pm
    • WordPress doesn’t allow for me to reply directly to your comment, so this is out of place.

      Yes, this makes sense. I may not have enough math knowledge to say what I’m about to say, but with that caveat, here I go!

      I worry that the N for designated hitter might be small enough that it throws off the medium, or that it makes the median less valuable or telling, at least as compared to other positions. Also, I suspect the median is fairly similar to the mean for other positions, while for DH I suspect it is closer to the top. Maybe?

      In a point less related to math, maybe, I believe DH is a position where players are shuffled more than 1B, perhaps a lot more. Given that you look at the position and not just the player, perhaps that is throwing you off. Maybe the “DH penalty” hurts position players acting as DH more than full-time designated hitters. Maybe?

      (Please note that when I say “throwing you off”, I am presuming that you are doing something wrong. However, I do not presume that).

      Posted by Miller | March 6, 2018, 3:43 pm
      • I’m sorry it took until today to respond, but I had to work last night (restaurant manager).

        The “N” for designated hitter (15 vs. 30 for other positions) isn’t really an issue because it is really 9*N as a sample size, since I use 9-year rolling averages at the positions. In addition to providing a larger sample size, it smooths out the random year-to-year fluctuations to provide a truer sense of the relative value of positions over time.

        With regard to mean vs. median, the reason I choose median is that mean tends to get distorted by star gluts at a position (e.g. 1880’s or 1930’s 1b), whereas median isn’t affected by that and thus IMO provides a better sense of the depth at that position. And without formally doing the math, just looking at the yearly AL/MLB pages at BBRef (a few tables down is team WAA by position, with the average at each position already calculated at the bottom of the table), just a quick glance of the last few years show the average DH also having a negative WAA.

        As to the less math-related point, It really doesn’t matter. That is, maybe players who are shuffled in-and-out of DH with other positions do hit worse than those who have accepted full-time DH roles. But my system frankly doesn’t care – it doesn’t care if a position was platooned by days of the week or how playing time was allocated, it only cares about how a player did relative to his position. And Ortiz and Edgar provided huge advantages for the Sox and M’s, respectively over the DHs for other teams.

        Posted by Michael Mengel | March 7, 2018, 7:40 am
      • Let me clarify that last statement to say that my system cares about a player’s value relative to his position and that position’s strength relative to league average (thus why DH’s get a bump every year because the position is so shallow). And also, just because I cap the bump at the overall Rpos for 1b, doesn’t mean that it always gets to that point – there are still years where there is a gap between the overall Rpos for 1b and DH.

        Posted by Michael Mengel | March 7, 2018, 7:46 am
      • Well, Michael, you’ve done it. I’m out of “complaints” about your system. I understand everything you’re saying, and I don’t think I object to any of it. Next up, I have to reflect on what I’m doing. While “next up” may mean in months, I will reflect. I have to. Thank you!

        Posted by Miller | March 7, 2018, 8:28 am
  4. Great discussion, I will echo Michael’s sentiment, the work Tom Thress has done shows Harmon Killebrew and David Ortiz in a very positive light also.

    Digging this series guys, keep up the excellent work.

    Posted by Ryan | March 7, 2018, 9:33 am
    • So yesterday, I finished the absolute busiest time I will have at work through the end of my career. Seriously. I’m very lucky.

      That means I will finally read Thress in the next week or two.

      In a recent comment, Michael talked about peer review being important. There’s no doubt he’s right. I’m glad that readers like you two, commenters I’ve seen elsewhere, think work at the HoME falls into the “peer” category. We aim to get it “right”, whatever that means.

      Thank you for the continued support, Ryan, and for pushing for better, whatever that is.

      Posted by Miller | March 7, 2018, 9:45 am
      • I’d like to thank you guys for considering me among your “peers,” despite the fact that my “HoF blogging” credentials consist entirely of a bunch of Hall of Merit votes and a few comments here at the HoME.

        I never intended to cause you to rethink your evaluation process, Miller. I just wanted to present my own view, which made sense to me. But I am truly humbled that you are considering doing so on account of a few arguments I made. I am also thankful that you did pester me with questions about this part of my system so that I had to think more critically about it as well (Yay! Peer Review – and I mean that with all sincerity – I think that its good that we all come at this player evaluation problem with different approaches and not just a groupthink echo chamber mentality).

        And to be honest, I don’t feel that any part of my evaluation is truly original. I just synthesize a number of concepts I’ve seen around (mostly in BBTF HoM threads) and put my own little spin on them. Probably the biggest influences on my system were Dan Rosenheck and his own WARP methodology and your partner-in-blogging, Eric.

        So thanks again to you guys for helping me along in my evaluation journey as well.

        Posted by Michael Mengel | March 8, 2018, 8:04 am
  5. Hey, dudes, here’s my take on the DH thing. Just mine, feel free to disagree.

    + Agreed with Michael who says that teams tend to underutilize the DH in the sense that few have a full-time, hard-hitting DH.
    + There are very few career DHes since the 1973, and fewer yet who spent virtually their entire career as a DH.
    + The corollary is that any player who can play a position well enough will be deployed as a fielder, because fielding decently at a given position is rarer than hitting well in general (the pool is far smaller)
    + The other corollary is that any player who can stay healthy enough to be deployed in the field usually is because the DH is a career prolonger and has less impact on health than playing the field.
    + So most players who do a lot of DH’ing are players who have already played themselves off the field or can’t stay healthy on it.
    + These issues have usually required several seasons to manifest or become operative around age 30 or older.
    + So most players who are full-time DHes are older players outside their batting peak or prime. (Not all, of course.)
    + Some of teams’ underuse of the DH is intentional because it allows them to rest their regulars or give injured players a way to contribute without risking further injury in the field.

    But I haven’t dived deep into the actual performance of full-time DHes to know if this holds up.

    Posted by eric | March 7, 2018, 11:13 am

Leave a Reply to eric Cancel reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Google photo

You are commenting using your Google account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s

Institutional History

%d bloggers like this: