New voters, please click the link.

1987 Ballot

1) Wade Boggs - #2 last year for me. Boggs versus Gwynn is all positional value. They have nearly equal bats and gloves compared to their position. Gwynn's edge in baserunning is evened out by Boggs playing 3B. The difference between Boggs and Gwynn is rounding error.
2) Tony Gwynn - I've seen a lot of Gwynn as the best NL player on prelim ballots but no #1 votes yet
3) Roger Clemens - best pitcher again and has a reasonably good argument for #1. Top 3 are essentially tied.
4) Alan Trammell - just below the top three, postseason is unremarkable
5) Ozzie Smith - Trammell is the better hitter and Ozzie the better baserunner and defender. Value is pretty close.
6) Frank Viola - raw scores just below Straw but gets a postseason boost
7) Dale Murphy
8) Darryl Strawberry - best hitter 1987
9) Eric Davis
10) Bret Saberhagen
11) Tim Raines - rate production doesn't look any better than anyone ahead of him
12) Mike Schmidt - old man makes another ballot
13) Jimmy Key

14-20) Pedro Guerrero, Bob Welch, Jack Clark, Mark McGwire, Cal Ripken, Doyle Alexander, Orel Hershiser

Correction - I just found a copy/paste error in my spreadsheet that affected all the hitters.

1) Roger Clemens - that makes it back-to-back for me
2) Wade Boggs
3) Tony Gwynn
4) Alan Trammell
5) Ozzie Smith
6) Frank Viola
7) Dale Murphy
8) Darryl Strawberry
9) Bret Saberhagen
10) Eric Davis
11) Tim Raines
12) Jimmy Key
13) Mike Schmidt

14-20) Pedro Guerrero, Bob Welch, Jack Clark, Doyle Alexander, Orel Hershiser, Mark McGwire, Cal Ripken

1987 Final Ballot

1. Wade Boggs, 3B, Boston Red Sox: no one else was in the same class offensively- 174 OPS+ led league by 10; 154 runs created led league by 17
2. Alan Trammell, SS, Detroit Tigers: based on the generally accepted premise that MVPs come from teams in playoff contention, Trammell was the clear choice in '87 over George Bell. If I were ever to go with my heart over the numbers, this would have been the year.
3. Dale Murphy, RF, Atlanta Braves
4. Tony Gwynn, RF, San Diego Padres: Murphy narrowly edges Gwynn based on slightly better defense, +11 to +7
5. Paul Molitor, DH/3B, Milwaukee Brewers
6. Eric Davis, CF, Cincinnati Reds
7. Tim Raines, LF, Montreal Expos
8. Jack Clark, 1B, St. Louis Cardinals: a great offensive season (league leading 176 OPS+) dragged down by poor defense (minus 7 fielding runs)
9. Jimmy Key, P, Toronto Blue Jays: the best pitcher in either league with 164 ERA+ in 261 innings
10. Roger Clemens, P, Boston Red Sox: close to Key with 154 ERA+ in 281 innings; Key edges out the Rocket with slightly better WHIP (league leading 1.05 for Key to 1.17 for Clemens)
11. Mark McGwire, 1B, Oakland Athletics: great rookie year sneaks onto the bottom of the ballot
12. Mike Schmidt, 3B, Philadelphia Phillies: the old man is still pretty good
13. Darryl Strawberry, RF, New York Mets: a great year for NL outfielders; Darryl is the 5th on the ballot and sometime outfielder Clark shows up as a first baseman

14. Frank Viola, P, Minnesota Twins
15. Dwight Evans, RF, Boston Red Sox
16. George Bell, LF, Toronto Blue Jays
17. Ozzie Smith, SS, St. Louis Cardinals- neglected to include him in my prelim
18. Don Mattingly, 1B, New York Yankees
19. Pedro Guerrero, LF/1B, Los Angeles Dodgers
20. Brook Jacoby, 3B, Cleveland Indians

Chris - top NL pitcher?

Ok,I tweeked my system a bit from the prelim. I'm taking 4 player rating categories, bWar, fWAR,Win Shares and WARP2 and then grading each player by where they rank-10 points for first place,9 for second, 8 for third and so on. Win Shares doesn't rate pitchers to well so I'm giving them an additional 10% plug in WS(My prelim ballot had no WARP2 and used a .25% plug),then adding the rankings to reach a winner.
So here we go, comments and critiques are always welcome.Comments on the players can be found in the prelim

1-Wade Boggs
2-Alan Trammell
3-Tony Gwynn
4-Roger Clemens
5-Eric Davis
6-Dale Murphy
7-Ozzie Smith
8-Tim Raines
9-Darryl Strawberry
10-Frank Viola
11-Jack Clark
12-Bret Saberhagen
13-Ted Higuera
14-Mark McGwire
(3 tied at 15,Ryan,Welch and Schmidt)


ETA-I'd like to thank the guys at the yahoo HOM group for letting me join their group and access the WARP2 data

Gack-Just remembered that WARP2 doesn't include pitchers.Teach me to try and do balloting during a series game. That changes things a bit.Let's try this again-this time w/o WARP2.

1-Alan Trammell
2-Wade Boggs
3-Roger Clemens
4-Tony Gwynn
5-Eric Davis
6-Tim Raines
7-Dale Murphy
8-Ozzie Smith
9-Frank Viola
10-Jack Clark
11-Darryl Strawberry
12-Bret Saberhagen
13-Ted Higuera
14-Mark McGwire
2 tied at 15, Ryan & Welch

Chris - top NL pitcher?

Orel Hershiser at 22. In the prelim, I mentioned that he was the best of a weak bunch but that comment got cut when I remembered to include Ozzie Smith.

1987 final ballot
My usual combo of 4 different WAR(P) sytems and 2 different Win Share systems.
No playoff bonus. 10% catcher bonus.

1 Clemens, Roger 9328
2 Trammell, Alan 9220
3 Boggs, Wade 9043
4 Gwynn, Tony 8269
5 Raines, Tim 7996
6 Davis, Eric 7983
7 Murphy, Dale 7679
8 Viola, Frank 7632
9 Smith, Ozzie 7504
10 Clark, Jack 7307
11 Strawberry, Darryl 7247
12 Saberhagen, Bret 7223
13 Key, Jimmy 7161

14 Molitor, Paul 6827
15 McGwire, Mark 6458
16 Schmidt, Mike 6393
17 Hershiser, Orel 6345
18 Higuera, Teddy 6138
19 Welch, Bob 6068
20 Guerrero, Pedro 5998

1987 Final Ballot

Batters: start with RA (using value added runs), adjust for park, position and defense (average of TZ and DRA) Convert adjusted RAA to wins. Add 60% of normal Runs above replacement to get WARR (wins above reduced replacement)

Pitchers: start with RA (using value added runs) adjust for quality of opposition, park, team defense and role (reliever or starter) Convert adjusted RAA to wins. Add 60% of normal runs above replacement to get WARR (wins above reduced replacement)

1. Eric Davis 8.42 WARR
2. Roger Clemens 8.32 WARR
3. Dale Murphy 7.71 WARR
4. Bret Saberhagen 7.23 WARR
5. Alan Trammell 7.13 WARR
6. Mike Schmidt 7.10 WARR
7. Frank Viola 6.93 WARR
8. Wade Boggs 6.86 WARR
9. Tim Raines 6.66 WARR
10. Bob Welch 6.62 WARR
11. Tony Gwynn 6.53 WARR
12. Darryl Strawberry 6.37 WARR
13. Orel Hershiser 6.35 WARR

Rest of the top 20
13. Randy Ready
15. Jimmy Key
16. Charlie Liebandt
17. Teddy Higuera
18. Ozzie Smith
19. Jack Clark
20. Rick Sutcliffe

1987 MMP ballot. Differences from the consensus are explained by:
(1) WS are weighted more heavily than WAR; (2) Pitchers are generally rated
on the high side (3) Very high rates can (within reason) overcome a lack
of playing time (Molitor, Davis, Clark), (4) Players are rated, in part, relative
to others at same position (Molitor up; glut of RF Strawberry, Gwynn and
Murphy unfortunately down).

1. Alan Trammell
2. Roger Clemens
3. Jimmy Key
4. Paul Molitor
5. Frank Viola
6. Tim Raines
7. Wade Boggs
8. Ozzie Smith
9. Eric Davis
10. Bret Saberhagen
11. Jack Clark
12. Daryl Strawberry
13. Tony Gwynn

14-21. Orel Hershiser, Dale Murphy, Kirby Puckett, Willie Randolph,
Charlie Liebrant, Pedro Guerrero, Mark Langston, Mark McGwire

For my ballot, I've tried to stay as true to my 1987 opinions as I can, although new information can be used to shuffle it around a little. Everyone is closely bunched, and I could see many possible orders.

1. Wade Boggs. Start with the usual Wade Boggs profile and add some home runs. What's not to like?
2. Tony Gwynn. Start with the usual Tony Gwynn profile and give him Raines-like OBP and SB. What's not to like?
3. Alan Trammell. The peak year of his career, and yes, he does belong in the Hall of Fame.
4. Roger Clemens. Proving that his 1986 was no fluke.
5. Eric Davis. Just a breathtaking ballplayer. Would rank higher if he could have stayed healthier. And any kid trying to copy his stance and swing probably had to give it up as unworkable.
6. Tim Raines. Would rank higher if the greedy owners hadn't cost him April.
7. Dale Murphy. Who knew that he wouldn't have any more great seasons?
8. Jack Clark. The scariest hitter in the majors. Downgraded for not staying healthy and for not being a very good defender.
9. Ozzie Smith. Still Ozzie with the glove and a pretty darn useful offensive player.
10. Frank Viola
11. Bret Saberhagen
12. Darryl Strawberry
13. Mike Schmidt

Top NL pitcher is Bob Welch.
Paul Molitor comes very close to making this list despite only playing 118 games (and scoring nearly a run a game).

Final ballot: see prelim for comments, no postseason credit but some credit for playing for Division contending team
1. Trammell
2. Boggs
3. Gwynn
4. Murphy
5. Davis
6. Raines
7. Dw Evans
8. Clemens
9. Key
10.J Clark
11.O Smith
12.Molitor
13.Viola

My ballot uses Win Probability Above Replacement to rank players. To calculate this, I first take context-neutral WPA (WPA/LI) to measure batting and pitching contributions. Then, I apply each player's batting runs-to-WPA/LI ratio and apply it to their fielding runs and positional adjustment. For instances in which the batting runs to wins ratio is exceptionally small or large, I adjust the fielding runs to wins ratio at 10:1. I then add in a replacement level adjustment for playing time (WAR - WAA). Finally, I consider postseason performance by including WPA weighted by playoff round.

The 1987 ballot is very tight. One win separates the top five players, so it's not like I can say with certainty that Mike Schmidt is the MMP, even though he gets my first-place vote. It's really just measurement error and you can make a case for any one of the top five.

Trammell and Smith were among the leaders before considering playoff performance.

1987 Ballot

1. Mike Schmidt - 8.3 WPAR (4.4 O / 1.7 D / 2.1 R)
2. Dale Murphy - 8.2 WPAR (5.3 O / 0.6 D / 2.3 R)
3. Tim Raines - 8.0 WPAR (5.6 O / 0.1 D / 2.3 R) - Adjusted for collusion lockout.
4. Eric Davis - 7.9 WPAR (5.1 O / 0.9 D / 1.9 R) - If he was more durable, he would have won the MMP.
5. Wade Boggs - 7.4 WPAR (4.6 O / 0.8 D / 2.0 R)
6. Alan Trammell - 7.2 WPAR (4.9 O / 0.8 D / 2.1 R / -0.6 POST)
7. Frank Viola - 7.0 WPAR (4.4 P / 2.2 R / 0.4 POST)
8. Roger Clemens - 7.0 WPAR (4.6 P / 2.4 R)
9. Tony Gwynn - 7.0 WPAR (4.6 O / 0.1 D / 2.3 R)
10. Jimmy Key - 7.0 WPAR (4.7 P / 2.3 R)
11. Randy Ready - 6.4 WPAR (3.5 O / 1.4 D / 1.5 R) - Another MMP-caliber season on a per-game basis.
12. Ozzie Smith - 6.3 WPAR (2.2 O / 3.1 D / 2.3 R / -1.3 POST)
13. Darryl Strawberry - 6.3 WPAR (5.6 O / -1.5 D / 2.2 R)

Voters - just for politeness please mention your top NL pitcher. We're fairly evenly split between Welch and Hershiser and there are people keeping track of such things.

Five different first place vote getters ties the record.

EDIT: Top NL pitcher is Orel Hershiser (#14 on my ballot) - 5.9 WPAR (-0.6 O / 1.5 D / 2.7 P / 2.3 R).

Here's my final ballot. See comment #63 of the discussion thread for details. My raw system puts Tony Gwynn fairly comfortably off-ballot (probably top 20 or so). I moved him on-ballot in deference to other systems that really love him and also to get better pitcher-nonpitcher balance on my final ballot.

1. Alan Trammell
2. Roger Clemens
3. Darryl Strawberry
4. Ozzie Smith
5. Frank Viola
6. Wade Boggs
7. Dale Murphy
8. Teddy Higuera
9. Mike Schmidt
10. Eric Davis
11. Tony Gwynnn
12. Bret Saberhagen
13. Orel Hershiser

Here's the top hitters in 1987 in the NL terms of RPA runs produced above avg.:

Top Hitter Pos Age Team Runs
Jack Clark 1B 31 St. Louis 57.83
Darryl Strawberry OF 25 New York 45.67
Eric Davis OF 25 Cincinnati 38.37
Pedro Guerrero OF 31 Los Angeles 36.92
Tony Gwynn OF 27 San Diego 34.94
Dale Murphy OF 31 Atlanta 32.21
Will Clark 1B 23 San Francisco 30.41
Kal Daniels OF 23 Cincinnati 29.43
Tim Raines OF 27 Montreal 29.37
Randy Ready IF 27 San Diego 25.36
John Kruk 1B 26 San Diego 24.30
Howard Johnson 3B 26 New York 24.22
Von Hayes 1B 28 Philadelphia 22.08
Mike Schmidt 3B 37 Philadelphia 20.97
Bill Doran 2B 29 Houston 19.41
Tim Teufel 2B 28 New York 17.99
Andy Van Slyke OF 26 Pittsburgh 17.87
Mike Aldrete UT 26 San Francisco 16.47
Jerry Mumphrey OF 34 Chicago 15.30
Andre Dawson OF 32 Chicago 15.02
Keith Hernandez 1B 33 New York 14.74
Billy Hatcher OF 26 Houston 14.14
Leon Durham 1B 29 Chicago 13.46
Candy Maldonado OF 26 San Francisco 13.38
Bob Brenly C 33 San Francisco 13.37
Ozzie Smith SS 32 St. Louis 13.37
Kevin Mitchell 3B 25 SF/SD 13.13
Tim Wallach 3B 29 Montreal 13.13
Lenny Dykstra OF 24 New York 12.78
Ryne Sandberg 2B 27 Chicago 11.66

Here's the top pitchers in the NL in 1987 per RPA runs produced:

Rick Reuchel 13-9 38 Pitt./SF 26.68
Tim Burke 7-0 28 Montreal 20.75
Nolan Ryan 8-16 40 Houston 18.23
Mike Scott 16-13 32 Houston 16.80
Andy McGaffigan 5-2 30 Montreal 15.18
Dennis Martinez 11-4 32 Montreal 14.26
Zane Smith 15-10 26 Atlanta 14.08
Bob Welch 15-9 30 Los Angeles 13.90
Mike Dunne 13-6 24 Pittsburgh 13.74
Dwight Gooden 15-7 22 New York 13.44
Dave Smith 2-3 32 Houston 13.17
Kent Tekulve 6-4 40 Philadelphia 12.06
Pascual Perez 7-0 30 Montreal 11.29
Ron Robinson 7-5 25 Cincinnati 11.09
Rob Murphy 8-5 27 Cincinnati 10.42

Here's the top hitters in the AL in 1987 per RPA runs produced:

Top Hitter Pos Age Team Runs
Wade Boggs 3B 29 Boston 51.85
Mark McGwire 1B 23 Oakland 44.22
Dwight Evans 1B 35 Boston 39.87
Alan Trammell SS 29 Detroit 37.61
Kent Hrbek 1B 27 Minnesota 31.40
Paul Molitor UT 30 Milwaukee 31.18
Brian Downing DH 36 California 29.16
George Bell OF 27 Toronto 27.31
Darrell Evans 1B 40 Detroit 26.70
Don Mattingly 1B 26 New York 26.33
Rickey Henderson UT 28 New York 26.10
Danny Tartabull OF 24 Kansas City 25.50
Ken Phelps DH 32 Seattle 25.24
Wally Joyner 1B 25 California 23.09
Kirby Puckett OF 27 Minnesota 22.50
Brook Jacoby 3B 27 Cleveland 21.74
Kirk Gibson OF 30 Detroit 20.24
Mike Greenwell UT 23 Boston 19.30
Carney Lansford 3B 30 Oakland 19.19
Chet Lemon OF 32 Detroit 18.93
Larry Sheets OF 27 Baltimore 18.03
Alvin Davis 1B 26 Seattle 17.55
Matt Nokes C 23 Detroit 17.04
George Brett 1B 34 Kansas City 16.06
Tom Brunansky OF 26 Minnesota 15.51
Robin Yount OF 31 Milwaukee 15.37

Here's the top pitchers in the AL per RPA runs produced:

Top Pitcher W-L Age Team Runs
Jimmy Key 17-8 26 Toronto 33.13
Roger Clemens 20-9 24 Boston 32.64
Bret Saberhagen 18-10 23 Kansas City 24.71
Teddy Higuera 18-10 29 Milwaukee 24.28
Doyle Alexander 14-10 36 Detr/ATL 19.85
Charlie Leibrandt 16-11 30 Kansas City 19.61
Floyd Bannister 16-11 32 Chicago 19.33
Charlie Hough 18-13 39 Texas 17.50
Danny Jackson 9-18 25 Kansas City 16.54
Mark Langston 19-13 26 Seattle 15.96
Jose DeLeon 11-12 26 Chicago 15.79
Frank Viola 17-10 27 Minnesota 15.37
Tom Henke 0-6 29 Toronto 15.32
Richard Dotson 11-12 28 Chicago 15.10
Bill Long 8-8 27 Chicago 14.83
Dennis Eckersley 6-8 32 Oakland 12.88

caiman - can you finalize that and rank your top 13? I don't want to guess at how you weight leagues or pitchers versus position players.

Final ballot, comments in prelim:

1) Alan Trammell
2) Tony Gwynn
3) Wade Boggs
4) Eric Davis
5) Tim Raines
6) Roger Clemens
7) Frank Viola
8) Jimmy Key
9) Dale Murphy
10)Paul Molitor
11)Randy Ready
12)Ozzie Smith
13)Jack Clark

Sorry, forgot NL top pitcher - at the time, back in '87, I thought Nolan Ryan.

Today I would probably say Hershiser (but Nolan had a really outstanding year and got royally screwed on absolutely criminal run support).

No ballot yet from
John Murphy
bjhanke
Yardape
Johnny Fora
Voxter
John DiFool2

Official 1987 MMP Ballot (no credit for postseason efforts):

1) Alan Trammell: Best ML player and shortstop.
2) Wade Boggs: Best ML third baseman - could be #1 instead of Trammell.
3) Jack Clark: Best NL player and ML first baseman.
4) Eric Davis: Best ML center fielder.
5) Roger Clemens: Best ML pitcher.
6) Tim Raines: Best ML left fielder.
7) Ozzie Smith: Best NL shortstop.
8) Tony Gwynn: Best ML right fielder.
9) Mark McGwire: Best AL first baseman.
10) Paul Molitor: Best ML utility player.
11) Darryl Strawberry: Best ML baseball name.
12) Dale Murphy: Best ML surname once again.
13) Tim Burke: Best NL pitcher.

This is Brock Hanke's ballot. I used my normal method of taking DL's lists by Win Shares and WAR, doing the ordinals, adding and then making the list. One of the reasons I've been doing it this way is that I wanted to see just how bad the disconnects between Win Shares and WAR are. I don't know where they are now, but as of 1987, they are catastrophic and getting worse with each passing year. My 11th place is Jack Clark, who is 5th by Win Shares and nowhere near the WAR ballot line. 12th is Bret Saberhagen, who is 6th by WAR and nowhere near the ballot in WS. Then it's Mark McGwire, who is 7th by WS and nowhere near the WAR ballot. This is mind-boggling. I can't come up with more than ten agreements better than those? The worst case of all, although he would not have made the ballot no matter what, was Robin Yount. Yount is 18th in Win Shares, and second-to-last on DL's entire LIST by WAR (59th place). If this gets any worse, I will have to abandon the method due to an utter lack of agreement on the ballot.

At any rate, here are the names, such as they are:

1. Alan Trammell
2. Wade Boggs
3. Tony Gwynn
4. Roger Clemens
5. Tim Raines
6. Eric Davis
7. Dale Murphy
8. Darryl Strawberry
9. Frank Viola
10. Paul Molitor
11. Jack Clark
12. Bret Saberhagen
13. Mark McGwire

Hi DL. You can simply see how any player on my 4 lists (above) are ranked by listing them in run value order. However, as per your request, I'll list the top 13 in 1987:

1. Jack Clark 57.83 runs.
2. Wade Boggs 51.85 runs
3. Daryl Strawberry 45.67 runs
4. Mark McGwire 44.22 runs
5. Dwight Evans 39.87 runs
6. Eric Davis 38.37 runs
7. Alan Trammell 37.61 runs
8. Pedro Guerrero 36.92 runs
9. Tony Gwynn 34.94 runs
10 Jimmy Key 33.13 runs
11. Roger Clemens 32.64 runs
12. Dale Murphy 32.21 runs
13. Kent Hrbek 31.40 runs

Note: Just gotta get one thing off my chest. I am astounded by the nonsense of the positional advantage given to certain players on this list. Yes. shortstop is a more important defensive position than is first base or DH yet, on average, the DH position is positionally equal to shortstop on defense when evaluating individual players. That may sound crazy but it is statistically correct. Comparing any position to any other, for the purpose of evaluating a player's defensive positives or liabilities, is WRONG. It is statistical nonsense. The ONLY ACCURATE evaluation of defense is within the position. Every position is totally neutral. Some players are above average at that position. Some player below average AT THAT POSITION. Rickey Henderson was a great defensive left fielder. Often he was the best defensive player on his team, based on the advantage he gave them at his position OVER OTHER PLAYERS AT THAT SAME POSITION. I could care less that someone, without any statistically tested methodology, has declared shortstop to be a more important defensive position than first base or DH. Even 'Big Pappi', at DH, (by definition!!) is the equal on defense to the average shortstop. I should also note that defensive ratings of individual players cannot be correct if these ratings are not, at minimum, stadium adjusted.

caiman - thanks for ranking them. I can't really count it otherwise.

It is not the difference in average value that varies by position but the difference in positional replacement value. Due to degree of difficulty SS and C are more scarce than 1B and LF.

No ballot yet from
Yardape
Johnny Fora
Voxter
John DiFool2

There was a prelim from Voxter

It is not the difference in average value that varies by position but the difference in positional replacement value. Due to degree of difficulty SS and C are more scarce than 1B and LF.

Right. Anybody can "play" DH, but Big Papi at short would be a disaster-and-a half.

Anyone have an e-mail for Voxter? Can you ping?

Balloting is closed. Voxter's prelim wouldn't have changed much.

DL: replacement for SS is MUCH EASIER than replacing, say, David Ortiz at DH. The drop-off in production from Ortiz to his replacement will be much greater under most circumstances. There are lots of mediocre SS's out there that are not that much worse than many regular SS's. Of Course, that does not apply to players such as Hanley Ramirez, but he's not the norm. Likewise for the Catcher position. The great hitter, at whatever position, provided they are not absolute horrid on defense, cannot be easily replaced, even at first base. In fact, the difference between the top group of starters at first base and the average firstbasemen is usually much greater than for SS or catcher. I think that you have positional scarcity exactly backwards.

Not really. To replace the DH you can rotate in the 1B or LF and make it a job sharing arrangement. To replace the SS you get a futility infielder hitting .200 if you're lucky. Replacement value is pretty well established at this point and it's not my idea. You can use theoretical or measured (worst 3 regular players at a position) to make the determination.

DL: I really do not care about what has been "established". I think a lot of it is bogus. The difference in output between the premier first basemen and the non-premier first basemen is much larger than for SS's. The player you 'rotate in' is usually too far below the production of the regular. That is not the case at SS simply because the premier ss's are so much poorer producers on offense. Most teams, unlike the Bosox, do not have a premier hitter on their bench to 'rotate in' when one of their top hitters is lost.

I should point out that Trammell made the top of my list without any positional adjustment. None. Zero. Nada. Zilch. He's there based on a combination of his hitting and his defense. That combination beat everybody else with the exception of Wade Boggs, whom I have mentioned above could have easily taken the top spot on my ballot himself.

The only positional adjustment I give is to catchers, since their offensive numbers are affected by the wear-and-tear they experience throughout a season.

Replacement value is pretty well established at this point and it's not my idea.

It's one of the few things baseball people and sabermeticians have never seriously disagreed on.

RN: Where do you get the idea that I am proposing that Big Papi play SS? That is a bogus argument. The question is how do you replace 'big papi';s total production vs. replacing the SS's total production Offense + defense). 'Big papi" play's at DH, which is, by definition. a zero value on defense. Therefore, it is defensively neutral. At shortstop, half the players are, by definition, BELOW average on defense. Therefore those SS's are worse defensive players than is 'Big Papi" when calculating the overall defense of a team. Therefore assigning an automatic plus to a SS over a DH is complete nonsense.

RN: Where do you get the idea that I am proposing that Big Papi play SS? That is a bogus argument. The question is how do you replace 'big papi';s total production vs. replacing the SS's total production Offense + defense). 'Big papi" play's at DH, which is, by definition. a zero value on defense. Therefore, it is defensively neutral. At shortstop, half the players are, by definition, BELOW average on defense. Therefore those SS's are worse defensive players than is 'Big Papi" when calculating the overall defense of a team. Therefore assigning an automatic plus to a SS over a DH is complete nonsense.

I don't know who this RN is, but I'll answer your question anyway. ;-)

Therefore, it is defensively neutral. At shortstop, half the players are, by definition, BELOW average on defense. Therefore those SS's are worse defensive players than is 'Big Papi" when calculating the overall defense of a team. Therefore assigning an automatic plus to a SS over a DH is complete nonsense.

No, no, no and no. By that logic, I am a better fielder than half the shortstops and I don't even play professional baseball (which is good, BTW :-). This is one of the big mistakes of using liner weights-type metrics for baseball analysis, BTW. Here's another one: a HOF-caliber player for 20 seasons can become an average-or-worse player by similar metrics if some idiot owner kept him on the roster for 30 more players as a below-average player. Does that make sense to you? Not to me. Yes, that's an extreme example, but it illustrates the silliness of using these types of metrics without context.

OK DL & John Murphy:

First I identified John Murphy as 'RN' in my last post. My mistake.

Now let me get down to actual numbers to prove my point:

I just printed out my listing of the top 15 players in both the AL and the NL at First base and at SS. Included are the rating for those players (who should be bench warmers)that are below the top 15. I will list, one by one, in each league the difference in RPA (total offense + defense)rating belween each of the 15 and the top rated bench warmer at that position.

NL First Base top bench player RPA is .118. NL top SS bench player is .103

First base Shortstop
1. .177 - .112 = 65 point loss .140 - .103 = 37 point loss
2. .151 - .112 = 39 .135 - .103 = 32
3. .144 - .112 = 32 .133 - .103 = 30
4. .144 - .112 = 32 .131 - .103 = 28
5. .139 - .112 = 27 .121 - .103 = 18
6. .132 - .112 = 20 .120 - .103 = 17
7. .130 - .112 = 18 .120 - .103 = 17
8. .128 - .112 = 16 .117 - .103 = 14
9. .127 - .112 = 15 .117 - .103 = 14
10. .127 - .112 = 15 .114 - .103 = 11
11. .126 - .112 = 14 .111 - .103 = 8
12. .126 - .112 = 14 .110 - .103 = 7
13. .122 - .112 = 10 .109 - .103 = 6
14. .120 - .112 = 8 .105 - .103 = 2
15. .118 - .112 = 6 .105 - .103 = 2

In every case above, the replacement player at SS loses less points! The difference at the top sot is the biggest of all.

Now for the AL comparison:

First Basemen

Sorry, I hit the wrong button! And sorry for the scrunched numbers! The previous post has 6 columns. The first 3 are for the first basemen. the second three are for the shortstop.

Sheesh! Let me redo the list. I jus saw that I list the top bench player as .118. Not!! it was .112 for first base in the NL.

Here's the NL comparison in better format, in order of top to 15t rated at that position:

NL First Base top bench player RPA is .112. NL top SS bench player is .103

1. First Base is .177 - .112 (the replacement player) for a drop off of 65 RPA points

1. Shortstop is .140 - .103 for a drop off of 37 points

***

2. First Base is .151 - .112 for a drop off of 39 RPA points

2. Shortstop is .135 - .103 for a drop off of 32 points
***

3. First Base is .144 - .112 for a drop off of 32 RPA points

3. Shortstop is .133 - .103 for a drop off of 30 points
***

4. First Base is .144 - .112 for a drop off of 32 RPA points

4. Shortstop is .131 - .103 for a drop off of 28 points
***

5. First Base is .139 - .112 for a drop off of 27 RPA points

5. Shortstop is .121 - .103 for a drop off of 18 points
***

6. First Base is .132 - .112 for a drop off of 20 RPA points

6. Shortstop is .120 - .103 for a drop off of 17 points
***

7. First Base is .130 - .112 for a drop off of 18 RPA points

7. Shortstop is .120 - .103 for a drop off of 17 points
***

8. First Base is .128 - .112 for a drop off of 16 RPA points

8. Shortstop is .117 - .103 for a drop off of 14 points
***

9. First Base is .127 - .112 for a drop off of 15 RPA points

9. Shortstop is .117 - .103 for a drop off of 14 points
***

10. First Base is .127 - .112 for a drop off of 15 RPA points

10. Shortstop is .114 - .103 for a drop off of 11 points
***

11. First Base is .126 - .112 for a drop off of 14 RPA points

11. Shortstop is .111 - .103 for a drop off of 8 points
***

12. First Base is .126 - .112 for a drop off of 14 RPA points

12. Shortstop is .110 - .103 for a drop off of 7 points
***

13. First Base is .122 - .112 for a drop off of 10 RPA points

13. Shortstop is .109 - .103 for a drop off of 6 points
***

14. First Base is .120 - .112 for a drop off of 8 RPA points

14. Shortstop is .105 - .103 for a drop off of 2 points
***

15. First Base is .118 - .112 for a drop off of 6 RPA points

15. Shortstop is .105 - .103 for a drop off of 2 points

I hope the above is more readable! Every last player at first base would have been harder to replace!

Here's the AL comparison:

AL First Base top bench player RPA is .113. NL top SS bench player is .102

1. First Base is .167 - .113 (the replacement player) for a drop off of 54 RPA points

1. Shortstop is .141 - .102 for a drop off of 39 points
***
2. First Base is .158 - .113 for a drop off of 45 RPA points

2. Shortstop is .139 - .102 for a drop off of 37 points
***
3. First Base is .157 - .113 for a drop off of 44 RPA points

3. Shortstop is .125 - .102 for a drop off of 23 points
***
4. First Base is .152 - .113 for a drop off of 39 RPA points

4. Shortstop is .124 - .102 for a drop off of 22 points
***
5. First Base is .146 - .113 for a drop off of 33 RPA points

5. Shortstop is .121 - .102 for a drop off of 19 points
***
6. First Base is .135 - .113 for a drop off of 22 RPA points

6. Shortstop is .118 - .102 for a drop off of 16 points
***
7. First Base is .134 - .113 for a drop off of 21 RPA points

7. Shortstop is .118 - .102 for a drop off of 16 points
***
8. First Base is .132 - .113 for a drop off of 19 RPA points

8. Shortstop is .113 - .102 for a drop off of 11 points
***
9. First Base is .128 - .113 for a drop off of 15 RPA points

9. Shortstop is .112 - .102 for a drop off of 10 points
***
10. First Base is .125 - .113 for a drop off of 12 RPA points

10. Shortstop is .110 - .102 for a drop off of 8 points
***
11. First Base is .125 - .113 for a drop off of 12 RPA points

11. Shortstop is .109- .102 for a drop off of 7 points
***
12. First Base is .118 - .113 for a drop off of 5 RPA points

12. Shortstop is .108 - .102 for a drop off of 6 points
***
13. First Base is .117 - .113 for a drop off of 4 RPA points

13. Shortstop is .107 - .102 for a drop off of 5 points
***
14. First Base is .114 - .113 for a drop off of 1 RPA points

14. Shortstop is .106 - .102 for a drop off of 4 points
***
15. First Base is .14 - .113 for a drop off of 1 RPA points

15. Shortstop is .105 - .102 for a drop off of 3 points

Only at the bottom of this list (#12 on) is the difference favoring the SS, but only because the difference between the starter and the replacement is so negligible since they are almost interchangeable.

Answer to John: Yes, you are the equal to any DH on defense. So am I. So is any one-year old. They don't play defense. Neither do you or me or the one-year old or the DH. Therefore we don't count on defense. We're a big fat zero on defense and so is the DH. The shortstops, however, do play in the field and the only way to measure the quality of their defense is against other SS's, not DH's or you or me. The average defensive shortstop is exactly equal to the DH to you and to me because they are average which means they are rated at zero effect on defense. Not to understand this means that you don't understand the very basis of statistical analysis.

I've never even heard of RPA before now but I'm guessing it's solely an offensive stat. The problem with replacing a player (other than DH), is that you're replacing both his offense and his defense. It is more difficult to find a player who can play a passable shortstop than a passable first base. That has a huge effect on replacement value.

Did anyone notice what happened in the post-season? Where would the Bosox and the Cards have been without David Ortiz and Carlos Beltran? Sheesh..... Yeah had the Bosox, e.g., lost Stephen Drew, instead of., e.g., David Ortiz, they'd have been in less trouble? Yeah, sure.......

That is why I was flabbergasted by the idiocy of the Kansas City front office when they traded Wil Myers. It was a typical Tampa Bay smart front office move, taking advantage of a stupid front office.

Chris: RPA is all inclusive. It takes in offense and defense. I understand that you do not know about RPA. I have not published in almost 2 decades, but I was 'Moneyball' before Billy Beane was "Moneyball'. Please check out the recent article on me on the Grantland website entitled "Saber Rattler" or the article on me, from several years ago, from the ESPN archives entitled "The Rise and Fall of Mike Gimbel"

I should note that the 1987 list does only include offense for the hitters. That is because I have not evaluated the defense for that year and am using the scaled down data to make all the data the same from 1900 to 2012. I would expect Wade Boggs and Dwight Evans to move up the ratings because they were very good defenders at their position.

Is your position that below average players have negative value?

Yes

You may want to consider including defensive value for 1988 and beyond.

Please check out the recent article on me on the Grantland website entitled "Saber Rattler" or the article on me, from several years ago, from the ESPN archives entitled "The Rise and Fall of Mike Gimbel"

I read both articles. Both were very good reads. I found the Scott Cooper part interesting--hard to believe he was a two-time All-Star despite being a far cry from Boggs. Angelos should've hired you: reason #183 that he is an idiot. The only good thing Angelos ever did was spend on the free agents that made the team so good in '96/97, its not surprising that he'd fall in love with that decision and scoff at the idea of parting with those players

Chris: I'd love to do even more than the defense for 1988. The problem is that I have too many things to do. So little time to do them. There are play-by-play data files available for 1988. Some day I'll do that analysis, but I am trying to get my new website updated. Lots of work, not much time and I have political and union matters to take care of as well and I have two videos on Youtube that are not baseball related but are important in both political and scientific matters.

Tubbs: I really thought that I had an agreement to advise the O's, but Angelos vetoed the GM's request to hire me. Then he fired the GM a few months later!

Caiman: I understand time constraints, especially if one is a jack of many trades. Yet I wonder if this isn't a case in which perfection is the enemy of good. You might not have the time to do as thorough a defensive evaluation as you'd like but the solution shouldn't be to simply ignore defense altogether. You end up with a partial picture and, conceivably, flawed results.

Chris: Yes, the results are flawed in my historical RPA ratings. I have previously stated that I have used a stripped-down version of my RPA formula for the historical data. Therefore, these are flawed results but, I believe, that, on the whole, they are a decent result. I just accepted the fact that most of the historical data is not available. Only the stripped down data is available, at least prior to about 25 or 30 years ago when detailed play-by-play data has begun to be available. My main purpose was to get people thinking in terms of the actual, statistically important data. Thinking in terms of runs, not statistical anarchy such as WAR and all its variants. Runs is the measure of value in any baseball game. The measure of value of any player or pitcher is how many runs better or poorer is that player than the median player. It is that simple, but how to measure that run value is what is somewhat complicated. I do want to improve and expand my research. I just am only one person and I have many 'balls in the air'. Perhaps too many 'balls in the air' that require my attention. For that I truly apologize for any flaws in my ratings but I ask for your indulgence in consideration of my other responsibilities.

Aw man, I bombed out on the final. Sorry, guys.