PDA

View Full Version : Analyzing Bowden's Trades by Win Shares



D-Man
07-10-2002, 07:46 PM
Win Shares

A win share is basically one-third of a win. "Why one-third of a win?" you might ask. "Why not one-tenth of a win or a half of a win?" Well, because we can confidently say that a player who adds one-third of a win is significantly better than another. For instance, we can confidently say that a player who has 12 win shares this year is better than another who has 11. This difference has statistical significance.

It's pretty easy to think of a player's win shares in the following manner--a bench player or reliever would routinely get 1-10 win shares per year, a regular player would usually get 10-20, an all star 20-30, and an MVP candidate would get 30+. For comparison, Barry Bonds had 54 win shares in 2001. . . I think it goes without saying that that doesn't happen very often. Additionally, win shares evaluation is a fair to both starters and relievers, middle infielders and sluggers, players from the 1920s and 1990s, etc. Overall, it is a nice tool to use in evaluating everything related to baseball.

Some people have problems with win shares, but I've found it largely interesting and a great tool.

Bill James invented the concept and has a couple of books out on the topic if anyone is interested.

Evaluating Trades

Evaluating trades is tough to do, even when using win shares as a tool. The world is an infinitely complex place, and no trade occurs in a vacuum. There are plenty of other factors to consider, such as payroll implications, roster depth, areas of organizational strength and weakness, free agency, clubhouse politics, etc. There are limitations to what win shares can offer. But they do provide a good starting point for the discussion.

GMs can be evaluated on other criteria, as well. They are responsible for adding players into the system through the draft and free agency; they also are also responsible for overseeing the player development, and ensuring that the scouting/development guys are on the same page. So the trading aspect is only one part of the GM's job description.

Note that I only have the # of win shares through 2001, so they can't be evaluated through this year.

Evaluating Trades with "Long Legs"

It's hard to evaluate trades when players re-sign with teams after they are traded. That seems to happen a lot. For instance--
Dave Burba (CLE)
Paul O'Neill (NYY)
Dan Wilson (SEA)
Mike Remlinger (ATL)
Bret Boone (CIN)
Ed Taubensee (CIN and CLE)
Griffey (CIN)
Shaw (LA)
Norm Charlton (SEA)

So how do you fairly analyze these guys in trades that re-sign with clubs? Is it fair to count O'Neill's 9 years of win shares in New York against the Roberto Kelly trade that happened in 1992? I don't think so. I figured the best way was to do it was to not include their free agency years. If anyone wants to correct me about who was a free agent and when, please feel free to indicate that. I have a feeling that my rough approximates for free agent status are wildly incorrect.

Here's who I counted and for how long-
Dave Burba (CLE through 2001)
Paul O'Neill (NYY only through 1994)
Dan Wilson (SEA through 1999)
Mike Remlinger (ATL )
Bret Boone (CIN through 1998)
Ed Taubensee (CIN through 1997 and all CLE)
Griffey (CIN through 2001)
Shaw (LA through 2001)
Norm Charlton (SEA through 1996)

Likewise, I didn't count guys who were shipped off to third parties (e.g., Fordyce's years in BAL counting against the Reds' totals). I'm simply evaluating how well the Reds did, not how well they did against their competitors. Plus, that would require a lot of time that I just don't have to give to this mini-project.

Bowden's Trade Record Overall
By my count, he has added +169 win shares by trading in his nearly 10 years on the job. That means that his trading alone has added ABOUT 6 WINS TO THE REDS' TOTAL PER YEAR WITH HIS TRADING ALONE. For sake of comparison, Bowden has brought in about 17 win shares per year; last year, the Reds' best player (Sean Casey) only had 18 win shares. . . So Bowden, from up in his office, was better for the club than any other Red was that actually played on the field. Think about that--this total is absolutely amazing.

I don't see how anyone could say that his trading ability is anything other than a prime asset of this team.

Bowden's Best and Worst Trades

Without further ado, here are Bowden's five best trades by the win shares method:
1.) John Smiley and Jeff Branson for Danny Graves, Damian Jackson, Jim Crowell, and Scott Winchester (1997, +54 win shares). There's little doubt about this one. Smiley's arm broke before he could offer any help for the Indians, and Branson served a middle-infield backup for a couple of years. Graves, on the other hand, became an all star and Jackson was used to acquire Greg Vaughn for the Reds' 1999 run.

2.) Dmitri Young for Jeff Brantley; and subsequently, Dmitri Young for Mike Kelly (1997, +45 WS). Dmitri was actually traded to the Reds twice after the 1997 season because he was picked up in the expansion draft by the D-Rays (can you believe he was traded for Mike Kelly??). Both the Cards and D-Rays gave up on him, meaning a tip o' the cap to Bowden is well deserved on this one.

Tied-3*.) Ed Taubensee for Ross Powell (1994, +33 WS). Because of the evaluation method outlined above, I only included Taubensee's 1994-97 campaigns because I think his first walk year with the Reds was 1997 (please correct me if I'm wrong). This doesn't include Taubensee's 16 WS in 1998 or 15 WS in 1999, meaning this trade was probably a lot better than it appears.

T-3*.) Hector Carrasco and Scott Service to KC for Jon Nunnally and Chris Stynes (+33 WS). This might seem like eons ago, but both Nunnally and Stynes looked like potential all stars in 1997. They contributed a whopping 22 WS combined in only a half-season in Cincinnati. Nunnally eventually fell off the face of the earth, and Stynes became a nice utility guy.

5.) The Shaw/Griffey family of trades (+30 WS). Why is this a family of trades? It's easier for me to organize it that way. The guys in the original trade have come and gone. As a result, I chose to evaluate at the trade like this: 1998-2001 Jeff Shaw for 1998-2001 Reyes, 1998 Konerko, 1999 Cameron, and 2000-2001 Griffey. That's one heck of a trade, even if Griffey has been hurt for the past few years. If you add in the 2001 Pokey Reese/Reyes trade for Gabe White and Luke Hudson, this is an outstanding series of talent upgrades. The Reds obviously came out ahead in this group.

Five worst:
T-1*.) Bobby Ayala and Dan Wilson for Bret Boone and Erik Hanson (1993, -30 WS). This one comes as a bit of a surprise to me. But the Reds gave up an outstanding reliever and all-star catcher for a all star 2b and one-year rental. If Hanson had stayed a Red for longer than a year, this likely would've been an even trade, since the Boone/Wilson part of the trade was essentially a wash.

T-1*.) Reggie Sanders, Damian Jackson, and Josh Harris for Greg Vaughn and Mark Sweeney (1998, -30 WS). This one looks a lot worse on paper because the Reds gave up two guys who became regulars and only got one everyday player back. Damian Jackson wasn't going to play over Pokey Reese in 1999, so the Reds dealt from an area of organizational depth. Having both Vaughn and Reese (42 combined WS) playing in 1999 rather than Sanders and Jackson (30 WS) was a huge plus for the Reds that year; as a result, I'm not sure if this loss presents an accurate read on this trade.

3.) Paul O'Neill for Roberto Kelly (1992, -23 WS). Obviously, it could've been much worse if I counted all of O'Neill's season in NY.

4.) Brook Fordyce for Jake Meyer (1999, -16 WS). This really wasn't a big loss because the Reds had catching depth in 1999 (Taubensee and Johnson at the major league level, plus LaRue was fresh of his batting title year in AA). Moreover, Fordyce was out of options. Fordyce had a nice 1999 (12 WS) for the Sox, but other than that, there's not really much lost here.

5.) Gabe White to COL for Manny Aybar (2000, -15 WS). This was just a dumb, dumb, trade. Since White is back in Cincy, I guess all is well that ends well. FYI, the guy who the Reds traded for Aybar later in 2000, Jorge Cordova, looks like an interesting power reliever in AA.

The fact that Bowden only has four trades where he lost more than 10 win shares in telling. Bowden gave up more talent in only two of his five worst deals (the trading away of White and O'Neill). Add in the fact that in two of those three deals, the big difference makers were guys who weren't going to play in Cincinnati (Damian Jackson and Brook Fordyce) but who ended up getting a shot elsewhere. I think this is pretty strong evidence that Bowden is an astute trader.

Moreover, Bowden's only bad trades happened awhile ago (O'Neill, Boomer Wells, and Ayala/Dan Wilson) with the exception of the Gabe White to COL deal. And the Reds managed to get White back pretty cheaply.

Salary Dump Trades (-5 WS)
No real talent was lost here (Stynes is the exception), and the Reds couldn't afford these bloated salaries.

Chris Stynes for Donnie Sadler and Michael Coleman (2000, -8 WS).
Ron Villone for Jeff Taglienti and Justin Carter (2000, -1 WS).
Tim Belcher to the White Sox for Johnny Ruffin and Jeff Pierce (1993, +10 WS).
Michael Tucker to the Cubs for Ben Shaffar and Chris Booker (2001, -6 WS). Shaffar was used to pick up Jose Silva this year.

The Boone/Neagle Family of Trades (0 WS)
Bret Boone and Mike Remlinger for Denny Neagle, Michael Tucker, and Rob Bell (+3 WS). This one had the potential to be a huge deal for the Reds, but now Bell has crapped out, and Tucker and Neagle are both gone. Interestingly, Remlinger actually outperformed Neagle (24 WS to 17) during the 1999-2000 span.

Denny Neagle for Drew Henson, Jackson Melian, Brian Reith, and Ed Yarnall (2000, -3 WS). Many people think this was a bad trade, but I think it would've worked out much better if Reith hadn't been rushed to the big leagues in 2001 (after he had only had a handful of starts above A ball). He has struggled with his confidence and consistency ever since his first promotion. I see this clearly as player development issue, NOT a trade issue.

Michael Coleman and Drew Henson for Wily Mo Pena (2001, no effect).

Rob Bell for Ruben Mateo and Edwin Encarnacion (2001, no effect).

This original trade is now responsible for adding Pena, Mateo, Reith, and E. Encarnacion to the system. Those are some of the best players currently in the Reds' farm system. This trade family might be in the positive column in a few years.

The O'Neill/Burba Family of Trades (+46 WS)
Like the Boone/Neagle family, this series of trades has added a bunch of players to the system, so the original trade is not nearly as bad as one might think initially. Through a series of trades, Roberto Kelly eventually brought the Reds Sean Casey. This trade chain, believe it or not, has actually added wins to the Reds' total overall.

Paul O'Neill for Roberto Kelly listed above.

Kelly and Robert Etheridge to ATL for Deion Sanders (1994, -3 WS).

Deion Sanders, John Roper, Scott Service, Dave McCarty, and Ricky Pickett for Dave Burba, Mark Portugal, and Darren Lewis (1995, +27 WS)

Dave Burba for Sean Casey (+22 WS). A lot of people have been down on Casey lately, but if the Reds had gone with Konerko instead of Casey in 1999, the Reds simply wouldn't have made a run (Casey had 23 WS that year, Konerko had 14).

The Wells Trades (-15 WS)
David Wells to BAL for Curtis Goodwin and Trovin Valdez (1995, -9 WS). This is case of Bowden being Bowden and going after the guy with wheels who can't hit.

C.J. Nitkowski, Dave Tuttle, and Mark Lewis to the Tigers for David Wells (1995, -6 WS). This was a lot better than it appears, because Nitkowski served as an OK middle reliever, but you can get relievers pretty much anywhere. No real talent loss.

Relievers: Coming and Going (+31 WS)
Taubensee to CLE for Jim Brower and Robert Pugmire (2001, +7). Brower to MON for Bruce Chen (2002, too early to tell).

Larry Luebbers, Mike Anderson, and Darren Cox to the Cubs for Chuck McElroy (1994, +7). McElroy to ANA for Lee Smith (1996, -2).

Chris Hammond to FLA for Hector Carrasco (1993, -4). Carrasco to KC detailed above.

Lenny Harris to NYM for John Hudek (1998, +2). Hudek to Braves for Mark Wohlers (1999, +4). Wohlers to Yankees for Ricardo Aramboles (none, but the payoff could be big.)

Norm Charlton to SEA for Kevin Mitchell (1993, +6 WS). I added all of Charlton's years in SEA (1993-1996); if you exclude the final two years (did he re-sign with SEA as a free agent?), then this trade looks like one of JimBo's best (+26 WS).

B.J. Ryan and Jacobo Sequea for Juan Guzman (1999, -1). This one is much more one-sided than it appears; Ryan stinks as a middle reliever, and the Reds used the supplemental pick from the loss of Guzman in the draft to get Dustin Moseley, who is one of the club's top pitchers in the minors.

Tim Belcher to SEA for Roger Salkeld (1993, -5).

Andre King to NYM for Mike Remlinger and Luis Ordaz (1995, +14). Remlinger to ATL detailed above.

Johnny Carvajal to MTL for Gabe White (1997, +14). White to COL described above. White back to CIN in the Pokey Reese/Dennys Reyes to COL for Gabe White and Luke Hudson trade. This final trade looks better every day.

Willie Greene/Bichette Family of Trades (+4 WS)

Willie Greene for Jeffrey Hammonds (1998, +10). Jeffrey Hammonds and Stan Belinda for Dante Bichette (2000, -2 WS). Bichette for Chris Reitsma and John Curtice (2000,
-4). This final trade will start looking a lot better after 2002.

Other Assorted Trades (+27 WS)

Mark Sweeney to MIL for Alex Ochoa (2000, +16 WS). Ochoa to COL for Todd Walker and Jason Jennings (2001, +7 WS). That's a nice flip trade, turning a spare part into an everyday 2B.

Juan Castro for Kenny Lutz (2000, +4 WS).

Dmitri Young for Luis Pineda and Juan Encarnacion. Encarnacion has replaced Young's production at a much cheaper rate, and the Reds have added another power reliever for the bullpen in Pineda. I think the Reds have won this one hands down.

Hector Mercado for Reggie Taylor.
Chris Booker for Jose Silva.

<small>[ 07-10-2002, 05:57 PM: Message edited by: D-Man ]</small>

Boss-Hog
07-10-2002, 09:51 PM
Fabulous analysis, D-Man. A very interesting read.

Boss

Cooper
07-10-2002, 10:58 PM
Good stuff. Wonder what things would look like if you added in salary info. (it would take forever). Is a deal worth a lot more if there is the same production for 20% of the salary?

Seems like Bowden has done a lot of these kind of deals which would make him only more valuable.

The infatuation with speedy, defense oriented players is his only weakness re: player evaluation. It drives me crazy, but it's very marginal kind of stuff-on the whole it's not important. I'm glad we have Bowden-we could have the guy in KC (and then we'd all lose our minds).

Doc. Scott
07-10-2002, 11:10 PM
Stunning. D-Man, my hat is off.

Let this be an endorsement for Bill James' New Historical Baseball Abstract and his Win Shares books.

NDRed
07-11-2002, 02:38 AM
Wow D-Man impressive stuff. Worthy of archives. Thanks for taking the time to do all that work. Great Read!

<img border="0" alt="[Beer Me]" title="" src="graemlins/beerme.gif" />

oregonred
07-11-2002, 04:32 AM
great post -- you D' Man... <img border="0" title="" alt="[Wink]" src="wink.gif" />

I'd forgotten about a couple of those trades long ago. Man, Jimbo stays busy if nothing else.

Talk about thorough, how'd you like to be known in life as a guy traded for Juan ".200 OPS" Castro...

Juan Castro for Kenny Lutz (2000, +4 WS).

<small>[ 07-11-2002, 02:41 AM: Message edited by: oregonred ]</small>

backbencher
07-11-2002, 11:49 AM
Thanks, D-Man. That's good stuff.

I am not sure that I agree with the Win Shares methodology for ranking trades, primarily because, as you say, there are a lot of other variables. I need to think about it more.

One thing that is very clear, though, is that Bowden rarely gives up talent, and almost never gives up talent that he could keep. There just are not a lot of Reds "regrets" floating around the league.

RFS62
07-11-2002, 12:03 PM
Outstanding post, D-Man.

Thanks for doing all that work. Great read.

D-Man
07-11-2002, 12:40 PM
Thanks everyone for the kind words. It actually was a lot of fun looking back at all this stuff. One of the things that amazed me was the sheer number of trades that Bowden has made. I actually didn't include all of them, because there were numerous other trades that were of little or no significance. It was just a waste of my time to look into them.

Backbencher: I actually agree with your comments and your concerns. I think that Win Shares *is* limited in scope, and that's why I included those statements up front. I didn't want everyone to think this is a case in which I am saying that this is the final score on his trading history. Rather I wanted this to be a starting point, whereby we can say, "OK, here's what the objective data says." And if it doesn't make sense, then let's try to explain it another way, which I tried to do as commentary for several of the deals. I am pretty confident that Win Shares is a nice tool to get an accurate read on about 80% of these trades.

I am less confident about the "family" trades, however, where certain guys were brought in then shipped off a year or two later. These deals are quite complex, and perhaps I am doing a disservice by trying to simplify (oversimplify?) them to a certain degree. We all recognize that Bowden will be judged on the big deals that he makes rather than the Castros-for-Kenny-Lutzes types. Even so, despite these limitations, I thought a Win Shares analysis would be a good place to start the discussion. I am certainly open to suggestions for ways of refining the analysis; that's why I posted it.

I had a few additional random thoughts about the trades--

*I didn't really comprehend the sheer vastness of relievers that Bowden has brought in (and shipped out). And they all have been more effective after coming to Cincinnati. Someone else on another thread mentioned that Bowden should get credit for utilizing one of his organizational strengths--in particular, the Gullett School for Rehabs, Retreads, and Failed Pitching Prospects (TM)--and I think I have to agree. There are tons of decent relievers available on the waiver wire and available as minor league league free agents; despite this impediment, but Bowden is able to exchange these guys to shore up areas of organizational weakness (and acquire the occasional toolsy outfielders as well :) ).

*Flip trades. Bowden is able to get a guy at a minimal cost and turn him around for something useful. Ochoa, Damian Jackson, and Dmitri Young are all examples of this who were brought in cheaply and shipped out for something more useful.

*Bowden seems to subscribe to the Branch Rickey approach, "it's better to trade a player a year early rather than a year late." So many of these guys were dealt away at the perfect time before they became too ineffective or too expensive (or both). Taubensee, Bichette, Steve Parris, et al., are prime examples of this. Timing is crucial to trades, and Bowden seems to have a good feel for the timing.

<small>[ 07-11-2002, 10:44 AM: Message edited by: D-Man ]</small>

Hoosier Red
07-11-2002, 12:49 PM
"
posted 07-10-2002 05:46 PM
--------------------------------------------------------------------------------
Win Shares

A win share is basically one-third of a win. "Why one-third of a win?" you might ask. "Why not one-tenth of a win or a half of a win?" Well, because we can confidently say that a player who adds one-third of a win is significantly better than another "

Okay help me out, you still didn't explain how one player is assigned a "win-share"

Great read, just need some clarification.

Cooper
07-11-2002, 12:54 PM
Bowden knows how to build a bullpen out of spare parts. This is a huge strength of his -he rarely pays for relievers because he knows that if you look around you can find them growing on trees.

He doesn't worry about the "veterans" and "proven closers" stuff.

Amazed at how many teams can't build a bullpen because they are unwilling to go down to triple A and find guys on the cheap.

D-Man
07-11-2002, 01:10 PM
</font><blockquote><font size="1" face="Verdana, Arial">quote:</font><hr /><font size="2" face="Verdana, Arial">Okay help me out, you still didn't explain how one player is assigned a "win-share"

Great read, just need some clarification.</font><hr /></blockquote><font size="2" face="Verdana, Arial">I'm sorry--I didn't explain this well enough. Sorry for the confusion.

First, you start with the team's number of wins. The Reds last year had 66 wins. To find the number of win shares per team you can multiply the wins by three; the 2001 Reds, for instance, had 198 win shares to divide among all the players on that team. Some of those wins can be attributed to hitting, some to pitching, and the rest to defense (fielding).

Bill James has devised hundreds of formulas to determine how many of those wins can be assigned to each player on that team. First he calculated how many wins a certain player created by his hitting, then how many he created by his pitching, and then how many he created by his defense. A player will get the same number of wins shares if he is on a bad team as he would on a good team, but his teammates would get fewer on a bad team.

To provide an example, here's what the 2001 Reds win shares looked like:
2001 Reds (66-96)
Hitting 105.3 win shares (WS)
Fielding 28.1
Pitching 64.6

Casey 18 WS
Griffey 14
Boone 13
D. Young 13
Graves 11
Dessens 10
Dunn 10
LaRue 9
Sullivan 9
Brower 8
Ochoa 7
Reese 7
Walker 6
L. Davis 5
Larkin 5
Rivera 5
Stinnett 5
B. Clark 4
Guerrero 4
Riedling 4
Tucker 4
Jennings 3
Mercado 3
Reitsma 3
Acevedo 2
Cromer 2
MacRae 2
C. Miller 2
Rijo 2
Wohlers 2
R. Bell 1
Castro 1
Nichting 1
Piersoll 1
Reyes 1
Selby 1

Hoosier Red
07-11-2002, 01:36 PM
All right D-man, you made a big mistake here. You've gotten me to the blissful point of interest and confusion.

"A player will get the same number of wins shares if he is on a bad team as he would on a good team, but his teammates would get fewer on a bad team."

How is this possible, somebody has to be getting screwed here. Who gets the same number of win shares on a bad team and who doesn't.

For example.
Does Sean Casey(PLAYER) get 18 ws on a good team or a bad team, but Ken Griffey Junior(TEAMMATE) gets less because he's on a bad team.
or
does Ken Griffey Jr.(Player) get 14 WS whether on a good team or a bad team, but Sean Casey(Teammate) was limited to 18 because he was on a bad team.

Isn't everyone a player, and a teammate?

cincinnati chili
07-11-2002, 01:48 PM
D-man: I agree. One of the most interesting posts I've ever seen on this board.

I strongly recommend you send this to Rob Neyer, the Reds front office, people at the Enquirer, and Bill James himself. It certainly took some work, and you should be acknowleged for it.

I think Win Shares are a good, but flawed analytical tool. My problem comes in the fact that all values are positive. Players don't get a negative score if they have a season substantially below replacement level.

Ayala, if memory serves, hurt the Mariners badly in at least one of his years. The Mariners would have been better off picking up some righty at random off the waiver wire. So if win shares were tweaked to account for this fact, I don't think that the Wilson and Ayala trade would have been so bad for the Reds.

westofyou
07-11-2002, 01:49 PM
</font><blockquote><font size="1" face="Verdana, Arial">quote:</font><hr /><font size="2" face="Verdana, Arial"> Casey 18 WS </font><hr /></blockquote><font size="2" face="Verdana, Arial">Imagine that?

Not only that but Sean Casey led the Reds in Runs Created Per game last year (6.12) and was 61st in ALL of BASEBALL. (that folks is in the top 18% or so)

50th the year before and 28th in 1999.

So while he has dropped he's still not.

"Worthless"

"Giving empty numbers"

or should he

" Be traded for a bad pitcher"

"Benched"

Yeah, yeah, Casey is in a year long slump that has effected his power numbers.

In fact he's well on his way to his worse year ever (much like O'Neill in 92)

But careful what you wish for. (well we know most of you wish for a guy who hits HR's at will playing 1st)

Because if this team trades Casey for a Dave Burba (and trade him at this point and you'll get that sort of ilk) who is dog meat in 3 years and Casey plays 10 more years it might not be that good of a deal.

D-Man
07-11-2002, 03:37 PM
</font><blockquote><font size="1" face="Verdana, Arial">quote:</font><hr /><font size="2" face="Verdana, Arial">Originally posted by Hoosier Red:
<strong>All right D-man, you made a big mistake here. You've gotten me to the blissful point of interest and confusion.

"A player will get the same number of wins shares if he is on a bad team as he would on a good team, but his teammates would get fewer on a bad team."

How is this possible, somebody has to be getting screwed here. Who gets the same number of win shares on a bad team and who doesn't.

For example.
Does Sean Casey(PLAYER) get 18 ws on a good team or a bad team, but Ken Griffey Junior(TEAMMATE) gets less because he's on a bad team.
or
does Ken Griffey Jr.(Player) get 14 WS whether on a good team or a bad team, but Sean Casey(Teammate) was limited to 18 because he was on a bad team.

Isn't everyone a player, and a teammate?</strong></font><hr /></blockquote><font size="2" face="Verdana, Arial">Sorry once again for the lack of clarity.

Casey would get 18 win shares if he was on the Braves, Expos, Reds, or on whatever team. His performance is NOT dependent on what his teammates do.

Casey's fellow TEAMMATES would get fewer win shares if, say, the team only had 66 wins as opposed to 97 wins because the teammates wouldn't be contributing as much. This is just another way of saying that good players tend to make contributions that help teams win.

Take the 2001 Reds and compare them to the 2001 Braves:

Braves (88-74) Reds (66-96)
Hitting 104.1 WS 105.3 WS
Fielding 46.4 28.1
Pitching 113.5 64.6

The reason the Braves won 88 games and the Reds only 66 should be quite evident from the above. The Reds actually had better hitting than the Braves in 2001, but the fielding and pitching were A LOT worse. If you take Casey and put him on the Braves, he would still get 18 win shares, but the TEAM as a whole would be better because the Braves had much better fielders and pitchers than the Reds did in 2001. The better fielders and pitchers made the Braves a winning team, just as the lack of good fielders and pitchers made the Reds a bad team in 2001.

Does this make more sense now? Sorry about the lack of clarity in my writing.

Hoosier Red
07-11-2002, 03:43 PM
Beautiful, thanks for the lessons professor.

D-Man
07-11-2002, 03:53 PM
westofyou:
In case you were interested, I posted his win shares below (and Konerko's, for a point of reference).

Sean Casey
1999: 23 WS
2000: 17
2001: 18

Paul Konerko
1999: 14
2000: 15
2001: 17

Keep in mind that an all star player generally gets 20-30 win shares per year, and an everyday player usually gets 10-20. . .This appears to me that Sean Casey is a pretty good player; not a great player, but a good one. He hit his peak in 1999, but I wouldn't say he is spiraling downward into oblivion, as others have suggested that he is. I tend to think that his decline in 2002 is either, a.) health related; or b.) another data point in his tendency to have Jekyll-and-Hyde in-season splits.

backbencher
07-11-2002, 04:05 PM
D-Man

I have a follow-up to Hoosier Red's question about distinguishing between players and their teammates (plus, I like to see this thread toward the top).

Suppose two teams are IDENTICAL statistically, and each of their component parts is identical statistically, with the only difference being a handful of wins. Say, a team with all the attributes of the 2001 Reds but with two extra 1-run wins and an extra 2-run losing margin in a blowout. All team ERAs, BAAs, OPSes, etc. would be equal. The hypothetical Reds team would have 6 extra win shares to distribute. Who gets them? Couldn't Casey, or whomever, get an additional share?

Forgive me if I misunderstand the methodology. I have not read the book.

D-Man
07-11-2002, 04:18 PM
</font><blockquote><font size="1" face="Verdana, Arial">quote:</font><hr /><font size="2" face="Verdana, Arial">Originally posted by cincinnati chili:
<strong>Ayala, if memory serves, hurt the Mariners badly in at least one of his years. The Mariners would have been better off picking up some righty at random off the waiver wire. So if win shares were tweaked to account for this fact, I don't think that the Wilson and Ayala trade would have been so bad for the Reds.</strong></font><hr /></blockquote><font size="2" face="Verdana, Arial">Here are Ayala's win shares totals from 1994-1998:
94: 10 WS
95: 7
96: 4
97: 9
98: 0

And here are his raw numbers from those years:
94: 56 innings, 76 Ks, 2.86 ERA, ERA is 71% better than league average
95: 71 innings, 77 Ks, 4.44 ERA, ERA is 10% better than league average
96: 67 innings, 61 Ks, 5.88 ERA, ERA is 16% LOWER than league average
97: 97 innings, 92 Ks, 3.82 ERA, ERA is 19% better than league average
98: 75 innings, 68 Ks, 7.29 ERA, ERA is 36% LOWER than league average

Based on this basic data, I tend to think that win shares does a pretty good job of assessing how many wins Ayala contributed, with his 1996 season as the only possible exception. I haven't done all the math, but there could be other possible explanations for why Ayala got 4 win shares in 1996--maybe he was unlucky (unusual # of hits on balls in play), maybe there's a ballpark effect that distorts the raw numbers, maybe other relievers let all his runs score (i.e., he was hurt by the rest of the bullpen). I assume James has run the numbers on all this stuff.

</font><blockquote><font size="1" face="Verdana, Arial">quote:</font><hr /><font size="2" face="Verdana, Arial">Originally posted by cincinnati chili:
<strong>I think Win Shares are a good, but flawed analytical tool. My problem comes in the fact that all values are positive. Players don't get a negative score if they have a season substantially below replacement level. </strong></font><hr /></blockquote><font size="2" face="Verdana, Arial">True. And I can understand your concern. Ayala's 1998 was indeed horrific, and in your view, he should've been penalized for that.

But one issue I've always had with replacement level statistics (which won't preclude me from using them in the future, anyway :) ) is: how do you come to a final conclusion as to where the "true replacement level" is? Personally, I haven't seen anything that has left me 100% confident with it.

Plus, I like the fact that win shares uses simple numbers. :)

<small>[ 07-11-2002, 02:22 PM: Message edited by: D-Man ]</small>

D-Man
07-11-2002, 04:35 PM
</font><blockquote><font size="1" face="Verdana, Arial">quote:</font><hr /><font size="2" face="Verdana, Arial">Originally posted by backbencher:
<strong>D-Man

I have a follow-up to Hoosier Red's question about distinguishing between players and their teammates (plus, I like to see this thread toward the top).

Suppose two teams are IDENTICAL statistically, and each of their component parts is identical statistically, with the only difference being a handful of wins. Say, a team with all the attributes of the 2001 Reds but with two extra 1-run wins and an extra 2-run losing margin in a blowout. All team ERAs, BAAs, OPSes, etc. would be equal. The hypothetical Reds team would have 6 extra win shares to distribute. Who gets them? Couldn't Casey, or whomever, get an additional share?

Forgive me if I misunderstand the methodology. I have not read the book.</strong></font><hr /></blockquote><font size="2" face="Verdana, Arial">So you're asking, what if the # of team wins doesn't match with the actual performance of that team's players? I.e., what if a team got more wins than it otherwise could've expected, and how do you distribute those extra wins?

Well, to be perfectly frank, I don't know yet!!! I didn't read the entire explanation section of how the win shares system is allocated (the explanation of the long method is more than 100 pages). I only got throught 25 or so pages before I decided to skip ahead to the "random essays" part of the book (I always enjoy James's essays). I've also spent a lot of time looking through the various lists and charts. . . I'm kinda working my way backwards. :)

Perhaps this will serve as an impetus for you to buy the book!

<small>[ 07-11-2002, 02:37 PM: Message edited by: D-Man ]</small>

gonelong
07-11-2002, 04:36 PM
Does anyone have a Win Share link? Did a little search with no luck.

As far as Casey is concerned, I am concerned that his production (Win Shares, HR, RBI, however you measure it) is dropping as his salary is increasing. His 2002 WS numbers can't be to impressive either, showing a clear downtrend since '99.

I'd be curious as to his rank in Win Shares for 1B across the league during his tenure. I'd bet he'd be a league average or below.

The big question is how easily can his Win Shares be replaced? Thats the only question you have to answer IMO. According to VORP it shouldn't be too hard to find a replacement for his based on this years stats.

1999 VORP 42.3
2000 VORP 21.3
2001 VORP 18.9
2002 VORP 3.5

Useless? No. Replaceable? Yes, and most likely at a substantial cost savings.

GL

<small>[ 07-11-2002, 02:51 PM: Message edited by: gonelong ]</small>

backbencher
07-11-2002, 04:42 PM
</font><blockquote><font size="1" face="Verdana, Arial">quote:</font><hr /><font size="2" face="Verdana, Arial"> So you're asking, what if the # of team wins doesn't match with the actual performance of that team's players? I.e., what if a team got more wins than it otherwise could've expected, and how do you distribute those extra wins?
</font><hr /></blockquote><font size="2" face="Verdana, Arial">Yes, exactly. I phrased it differently because I am just not enough of a sabrehead to buy into this portion of it:

</font><blockquote><font size="1" face="Verdana, Arial">quote:</font><hr /><font size="2" face="Verdana, Arial"> than it otherwise could've expected </font><hr /></blockquote><font size="2" face="Verdana, Arial">I probably will buy the book.

Cooper
07-11-2002, 04:43 PM
Ayala's 96 season was spent in a hitter's park -a good hitter's park. That may make his stats league avg. after you include the park adjustment.

Chili: I've read alot of the stuff on Baseball Primer and Fanhome and I still am not sure that the folks that advocate for negative values are correct in their thinking. True replacement values (as D-Man alluded to) seem to move all over the place.

Also, in the case of Ayala -how do you know when to send him down to bring up a replacement level player? Seems as if you can only do that once you know where replacement levels have been set...after the fact. It would be silly to send him down to bring up another player who would be replacement level quality until you know what those values are. Ayala had, at that point, 2 good seasons in a row- how do you know when to throw in the towel and bring up the replacement level player? Based on what values? Good god, I sound anti-sabermetrics today, but i think I'm making some sense.

D-Man
07-11-2002, 04:51 PM
</font><blockquote><font size="1" face="Verdana, Arial">quote:</font><hr /><font size="2" face="Verdana, Arial">Originally posted by gonelong:
<strong>Does anyone have a Win Share link? Did a little search with no luck.

As far as Casey is concerned, I am concerned that his production (Win Shares, HR, RBI, however you measure it) is dropping as his salary is increasing. His 2002 WS numbers can't be to impressive either, showing a clear downtrend since '99.

I'd be curious as to his rank in Win Shares for 1B across the league during his tenure. I'd bet he'd be a league average or below.

The big question is how easily can his Win Shares be replaced? Thats the only question you have to answer IMO.

GL</strong></font><hr /></blockquote><font size="2" face="Verdana, Arial">I don't think each player's win shares are published anywhere on the Internet. I think that if this data becomes available, we'll probably have to pay for it because Stats Inc. will probably have proprietary rights to it :(. I think Bill James mentioned in an online chat he had on ESPN.com that Stats Inc. was thinking about doing a CD that you could use to crunch raw data.

I can't seem to find the link to that chat, though. . . I'll keep looking.

Here's a link to Rob Neyer's column on the book:
<a href="http://espn.go.com/mlb/columns/neyer_rob/1366976.html" target="_blank">http://espn.go.com/mlb/columns/neyer_rob/1366976.html</a>

Here's a link to the book, which is where I got all the data from:
<a href="http://www.amazon.com/exec/obidos/ASIN/1931584036/qid=1026413651/sr=8-1/ref=sr_8_1/002-7259200-3177641" target="_blank">http://www.amazon.com/exec/obidos/ASIN/1931584036/qid=1026413651/sr=8-1/ref=sr_8_1/002-7259200-3177641</a>