GM Report Card – JP Ricciardi

In December 2002 I can vividly remember calling friends and family, excited that the Phillies had just acquired Kevin Millwood.  With the imminent return of Mike Lieberthal, Johnny Estrada had become an expendable commodity and Millwood had been a key cog in the Braves rotation.  Two years later, my evaluation of the trade had changed.  Millwood had not been the answer to the Phillies pitching woes and Estrada turned out to be the lone Braves representative on the all-star team.
Of course hindsight is always 20/20 but general managers are, more often than not, evaluated by the production levels of the players they acquire and send away as well as how these production levels translate to wins.  Millwood did not meet expectations while Estrada exceeded them; therefore, it was Ed Wade’s fault for making a bad move.
With this in mind I decided to begin a bi-weekly or monthly feature evaluating general managers.  The method is somewhat of a combination of Geoff Young’s trade-tracking chapter in the 2008 Ducksnorts Baseball Annual, and Dan Levitt’s analysis of Terry Ryan at Baseball Analysts.  Win Shares is the statistic used to evaluate moves and they are assigned to all players acquired and lost during a GMs tenure.  The major difference between what I will do here and what was done in Levitt’s wonderful analysis is that he assigned Win Shares to lost players for every subsequent year; I am only assigning them for the years on the first new team they join.
For instance, in the Millwood/Estrada deal, Ed Wade would be debited for Estrada’s tenure with just the Braves.  When the Braves sent Estrada to the Diamondbacks, he then became a player lost by John Scheurholz.  Otherwise, the evaluations are pretty straight-forward.  For those unfamiliar with Win Shares, it is a statistic created by Bill James and explained in the self-titled book by James and Jim Hentzler and it measures the contribution of a player to his team’s total wins.  3 Win Shares = 1 Win; 20+ is an all-star season and 30+ is an MVP season.
To kick off this series of evaluations I chose to look at J.P. Ricciardi, GM of the Toronto Blue Jays.
Meet J.P.
A disciple of Billy Beane, Ricciardi took over the Toronto reigns in November 2001.  He replaced Gord Ash, who had more recently found himself embroiled in the Shouldergate controversy; he also hired a manager that feigned fighting in Vietnam.  The team had struggled to finish higher than third place and hoped that Ricciardi’s knack for quantifying players would pay off major dividends.
Now in the midst of his eighth season at the helm, the team is still yet to experience the success envisioned at the time of his hiring.  Sure, they finished in second place in 2006, but it did not result in a playoff berth.  In fact, they have not been in the playoffs since 1993, when some guy I have erased from memory hit a world series winning walkoff home run.
Overall Results
Before looking at each area of moves on their own, here are the overall results of his moves:

TYPE

WS ACQ.

WS LOST

NET

WINS/YR

Trades

272

378

-106

-5.89

Free Agents

327

319

+8

+0.44

Waivers

42

47

-5

-0.28

Rule V

14

2

+12

+0.67

Overall

655

746

-91

-5.06

As mentioned above, one win equals three WS.  For example, based on the free agents Ricciardi has signed, as opposed to those released or lost, the net of +8 WS equates to about three added wins.  Over the course of his six years he has added about a half-win per season in free agent moves.
Elsewhere, he has not made many Rule V moves or waiver claims, resulting in very little net Win Shares.  In trades, though, Ricciardi has bombed.  His trades have cost the Blue Jays approximately 8 wins per year.  Now this is contingent upon the traded away players performing the same way in Toronto as they did in their new destination; however, as mentioned at the start of the evaluation, whether fair or not, this is how GMs are evaluated.
Free Agent Signings
Ricciardi has received 42 WS, or 14 wins, from the free agents he has signed, starting in November 2001.  During his tenure these signings have added just about 0.5 wins per season. 
Click here to view the results for all of his free agent signings.
Of the forty signings, fourteen resulted in ten or more WS; only one, Victor Zambrano, produced negatively.  Frank Catalanotto is far and away the best signing he made, providing the team with around 17 total wins, or 4/yr.  The next highest is Gregg Zaun, previously a backup catcher who recently found himself the primary backstop for the Jays.
The highest single-season signing is a tie between BJ Ryan in 2006 and Frank Thomas in 2007.  Each had seventeen shares and contributed as much as six whole wins in the respective seasons.
Free Agents Lost
This category not only refers to players lost to free agency but also those who were released.  While Ricciardi’s 40 signings produced an aggregate 42 WS, the 29 players let go produced 47 for their new team.  Now, as I mentioned earlier, I only look at the very next team for a lost player.  Doug Davis was released and signed with the Brewers; I debit Ricciardi all of Davis’s WS while on the Brewers.  Once he joined the DBacks, he becomes Brewers GM Doug Melvin’s “problem.”
Of the 29 lost or released, five produced WS totals of 30 or more: Esteban Loaiza (30), Kelvim Escobar (51), Carlos Delgado (31), Chris Carpenter (48), and Doug Davis (36).  Looking at the yearly averages: Loaiza (15/yr), Escobar (13/yr), Delgado (31/yr), Carpenter (12/yr), Davis (9/yr).
Click here to view the results for all free agents lost/players released.
Trades
Ricciardi has made 29 significant trades from 2002-2007; trades that resulted in at least one win share on either his, or the other, side.  A trade was considered insignificant if nobody made the major leagues or both parties summed to 0 WS.  Overall, his trades have been the worst facet of his moves.  The players acquired produced an aggregate 272 WS–91 wins–which comes to +15 wins/yr.
The players lost, however, produced 378 WS for their new clubs.  378 WS = 126 wins = -21 wins/yr.  Though rounded a bit, he brought in 15 wins/yr with these trades but lost 21 wins/yr.  The net of -5.89, or -6 really leaves a significant stain on his Toronto resume.
The best trade pulled off involved getting Eric Hinske and Justin Miller in exchange for Billy Koch on 12/7/2001.  Koch played just one year with Oakland, bringing in 19 WS; Hinske and Miller combined for 65 WS.
He also made two really bad trades that, on their own, account for much of the net loss.  Both trades involved ridding the Jays of major league commodities for prospects that never cut the mustard.  The first, completed just six days after the Hinske deal on 12/13/2001, saw Luke Prokopec head to Toronto in exchange for Cesar Izturis and Paul Quantrill.  Prokopec contributed 0 WS in a brief 2001 stint while Quantrill and Izturis combined for 66 WS from 2001-2006.  The other one, completed almost a year later on 12/15/2002, saw Jason Arnold join the Jays while Felipe Lopez headed to Cincinnati.  Essentially the same story, Arnold contributed nothing while Lopez produced 43 WS from 2003 to 2006.
In terms of trades, commenter Darren pointed out that certain players were being double-counted; he was correct and these are now fixed.  What he meant can be explained in the Bobby Kielty deals; the Jays traded Shannon Stewart for Kielty mid-2003 and I counted Kielty’s one half-year with the Jays and Stewart’s 3+ years with the Twins.  In the end this gave Kielty 4 WS for JP and Stewart -39 WS against JP.  This was not correctly done on my part because Kielty was traded the next year for Ted Lilly.  At that point, Stewart’s WS with the Twins should have stopped and it would then be Kielty vs. Ted Lilly.  So, the Stewart-Kielty would be +4 vs -9 and then the Kielty-Lilly would remain the same.  Otherwise, it would be Stewart counting against Kielty even though the K-Man was not there anymore.  This did not happen too often in the trade log but I did make the corrections reflected in the results.
Click here to see the results for all players acquired and lost through trades.
Waiver Wire
Another way to acquire free talent or get rid of the undesirables is the waiver wire.  Ricciardi was essentially even in this acquisition aspect, bringing in 42 WS and giving away 47.  His most productive waiver claims were Pete Walker (12) and Frank Menechino (11).  Of players he lost to waivers, Scott Eyre produced 19 WS for the Giants and Bruce Chen chimed in with 16 for the Orioles.
Click here to see the results for his waiver moves.
Rule V
Ricciardi’s Rule V selections and losses were often than not returned; in other cases, they simply never amounted to anything.  The only three Rule V picks that were significant resulted in 14 WS gained and 2 WS lost.  Though a small sample this happened to be his best area.  Corey Thurman gave him 4 WS in 2001 and Aquilino Lopez gave him 10 in 2002; Matt Ford contributed 2 WS to Milwaukee in 2002.
Position Evaluation
Another interesting way to analyze his moves is to look at how he fared by position.  Perhaps he had a knack for finding relievers but struggled to sign quality shortstops.  Here are the results:

TYPE

WS ACQ.

WS LOST

NET

WINS/YR

SP

125

267

-142

-8

RP

172

122

+50

+3

C

76

13

+63

+4

1B

41

32

+9

+0.5

2B

22

89

-67

-4

SS

33

94

-61

-4

3B

166

17

+149

+9

OF

20

112

-92

-5

These numbers are much more rounded than the overall results but you can see Ricciardi has fared best with third baseman and worst with starting pitchers and outfielders.  In fact, 14 of those 20 WS for outfielders belong to Matt Stairs; most of the other OFs he acquired did nothing. 
Conclusion
I hope this shed some light on what Ricciardi has done and how it effected his team’s success.  There is still room to improve the system and one such facet I am considering would be to compare the lost players to their replacements; for instance, Orlando Hudson was traded away but how did he compare to the incumbent second baseman?  Perhaps he would not count as much against Ricciardi when we see Aaron Hill’s numbers. 
Until we have a bunch of these analyses conducted we cannot rank the GMs but, based on Win Shares, Ricciardi certainly will not be amongst the leaders as he has cost his team about five wins per season with his transactions.
I am still deciding who the next GM for this should be, so if anyone has thoughts, leave them in the comments section.  I’d prefer it be a somewhat current time frame and, whoever you pick, also specify the team; don’t just say Pat Gillick, say Gillick with the Mariners or Gillick with the Blue Jays, etc.

About these ads

20 Responses to GM Report Card – JP Ricciardi

  1. darren says:

    Eric,
    I really like the idea of this kind of analysis, but you need to rethink your criteria for including a player in the Win Shares counting against a GM. I can probably illustrate this best with a couple of examples (albeit one that will only show up here if/when you include 2008):
    The Jays traded Shannon Stewart for Bobby Kielty in mid-2004, then moved Kielty for Ted Lilly after the season. For 2005-2006, you counted both Kielty’s and Stewart’s Win Shares against JP, and only Lilly’s for him. This doesn’t really add up. After 2004, you can think of it as if Kielty’s numbers should count both for (from the Stewart trade) and against (from the Lilly trade) JP’s Win Shares. Essentially, they should cancel out. Your number are (from 2005-2006) evaluating a hypothetical Stewart and Kielty for Lilly trade, which isn’t what happened.
    Just to show that this is a common problem, the Jays traded Orlando Hudson and Miguel Batista for Troy Glaus, then two years later moved Glaus for Rolen. However, according to this system, JP currently has Rolen’s numbers counting for him, but Glaus’ and Hudson’s counting against him. That makes no sense whatsoever. He obviously would not have traded Hudson and Glaus for Rolen, but that’s what this method is essentially evaluating going forward. The only sensible thing is for Hudson and Batista to count against JP for 2006-2008, for Glaus to count for him for 2006-2007 and for Rolen to count for him for 2008.

  2. darren says:

    Oops, the Stewart/Kielty/Lilly deals were actually in 2003, so there were actually 3 seasons (2004-2006) where Stewart and Kielty were double-counted. Kielty had 23 Win Shares over those 3 seasons that should not have been counted against JP.

  3. Darren, I understand what you’re saying and will take it into consideration for the future articles. I’m still harnessing the system. The other area I’m thinking about is comparing players traded to their positional replacements. For instance, Orlando Hudson was traded away but how much better were his Win Shares than the incumbent 2B, etc?
    As far as the examples you gave above, the way I did it here was count each transaction as its own entity. So, to correct it I would look at each year instead of the overall, which should make the results more accurate.
    Kielty for Stewart during the 2003 season, so we look at Kielty’s +4 in 2003 and Stewarts -9.
    Then, prior to the 2004 season, Kielty to Oak for Lilly. Stewart’s WS would then stop because the player linked to him in the previous transaction is now gone. In 2004 it would be Lilly’s +16 against Kielty’s -5; in 2005, Lilly’s +4 against Kielty’s -10; in 2006, Lilly’s +12 against Kielty’s -8.
    So, the way it is currently, Kielty for Stewart would be +4 against -39, while Lilly for Kielty would be +32 to -23.
    With the correction, Kielty for Stewart would be +4 against -9 with the Lilly trade staying the same. Instead of losing an aggregate -26 WS from both of these, it would be +4. -5 from the Stewart trade and +9 from the Lilly deal.

  4. The Rolen example you gave would now work as follows:
    1) Hudson and Batista for Glaus… in 2006 it is Hudson and Batista vs. Glaus. In 2007 it is Hudson vs Glaus because Batista joined a new team… he therefore would no longer count against Ricciardi.
    2) In 2008, Hudson is no longer counted against the Blue Jays because the Glaus for Rolen changes it to simply Glaus vs. Rolen. So, if/when this year is included, the Glaus for Rolen ends the Hudson relationship.
    This would apply solely to trades, though, where the Kielty like example makes sense. For instance, in the Prokopec for Izturis and Quantrill example, just because Prokopec never pitched again or accrued any WS does not mean Quantrill and Izturis should not be counted anymorel; they should because they were producing for a new team and Prokopec was not traded away for a new replacement.
    The free agents, waivers, and rule v are not effected by this so those results stay the same… it just requires some change when dealing with players traded for/away who are subsequently traded for/away by their new teams.
    I’m making the corrections to it now and updating soon.

  5. darren says:

    Your correction makes sense, although I still don’t completely agree with its principles. Personally, I still don’t like that, in this example, Stewart stops counting as a negative just because Kielty is no longer with the team. To me, the long-term judgment of those two deals should be Stewart vs. Lilly, not Kielty vs. Lilly. By dropping Stewart from the count after 2003, it’s like the Stewart-Kielty deal no longer matters from 2004 onwards, when it really should still be part of JP’s evaluation.
    I completely agree with you about what you’re saying about the Prokopec deal and how it should be evaluated. I would actually like to see Quantrill’s and Izturis’ numbers included in all subsequent years, not just the team they went to immediately after the Jays. If we’re ultimately going to have a nice way of comparing GM’s, you’d have some guys get lucky if they dealt away a good player who happened to move from their subsequent team to another team soon after their first move.
    I really like the idea of looking at it positionally. One of my first reactions to this was that many deals aren’t made solely on the basis of the how good the players are, but also to address organizational needs by dealing from organizational strengths. For instance, the Jays did give up Dave Bush as part of the Overbay deal, but their rotation now includes 5 guys who are almost unquestionably better than Bush (3 of whom are younger), so in a sense they really didn’t lose much at all when they gave him up.
    As for the free agent/waiver/rule 5 stuff, certainly what I was saying before doesn’t impact it.I can see it generally skewing towards guys who have more money to spend. Again looking at JP, it’s hard to blame him for losing Carlos Delgado’s production – he really didn’t have a choice. On the other hand, Brian Cashman would get credit for 2 Win Shares (not a lot, but it’s something) for signing Carl Pavano, which was without a doubt a contract that should reflect massively negatively on Cashman, not positively.
    Moreover, though of course we don’t know all the details either, indications are that before 2007, JP made every attempt to resign Lilly (and try to sign Gil Meche too) but they turned him down. It’s a bit harsh to blame a guy for a player’s prerogative.
    Having said all that, it’d be silly not to look at free agent signings at all. What about not subtracting un-resigned players, and evaluating signings on a per dollar basis?

  6. darren says:

    Further to my point on free agents lost, what’s the difference between not signing a free agent who used to be on your team, and not signing a free agent who from another team? If you ask me, there isn’t one.

  7. darren says:

    Oops, that should have read “not signing a free agent who was from another team”

  8. Looking at dollars per Win Above Replacement or dollars per Win Share is definitely something to keep in mind, and it would bring in the team need factor.
    Your last two comments I’m honestly not too sure what you’re saying… I looked at both guys he re-signed that were previously on the team (like Zaun for instance) and free agents from other teams. And free agents were included in this analysis but evaluated solely on Win Shares. If you’re suggesting adding the monetary factor, it’s definitely a possibility.
    Ultimately, if a guy was signed in 2003, became a FA after the season then re-signed in 2004, it ultimately got counted as one signing because it would not really make a difference other than add as a transaction.
    Also, I don’t care about not being able to re-sign players or blaming/not blaming someone for what they couldn’t do… this would also be absolved in the positional replacement. They lost Delgado, who had an MVP-caliber season but perhaps the replacement 1B performed well, too. I’m just interested in how the players performed and how this contributed to wins.
    The reasoning behind only looking at a player on his initial new team (Quantrill, Izturis example) is because I’m going to be doing this for a lot of GM’s, not just Ricciardi; due to this, there’s a circular effect where players belong to different GMs as moves are made. Once Izturis and Quantrill left the Dodgers they were no longer the Dodgers GMs responsibilities but rather that of the GM of the new team they joined.
    The reasoning, more in-depth perhaps, for the Stewart-Kielty revision, is that it isn’t necessarily a long-term evaluation in the way you described it above. Kielty was a player acquired and then lost. If Stewart is A, Kielty is B, and Lilly is C, then trade one is A for B… then B was traded for C… we can’t really jump to A vs C without acknowledging the B factor. We can look at the whole result of Stewart vs Kielty, which overall would be 39 vs 4… and then Kielty vs Lilly, which would be 23 vs 32.
    I don’t agree with judging Stewart’s 39 against Lilly’s 32 because it ignored Kielty, who contributed for the Jays and for his new team. If we look at the difference in each trade and compare them, we get -35 for the Stewart/Kielty and +9 for the Kielty/Lilly.. which is what I had from the start.

  9. darren says:

    Sorry if I wasn’t clear about the resigning thing. What I meant more or less comes down to this: there’s just no difference between letting a free agent go and failing to sign another team’s free agent. With that in mind, I think it’s pretty arbitrary to subtract a free agent’s production from his previous team’s GM’s cumulative Win Shares recrod.
    Even if you disregard money available and team need, it wasn’t only JP that didn’t sign Carlos Delgado in 2005 – there were 28 other GM’s that missed out on him. However, Delgado’s Win Shares are subtracted only from JP’s total merely because Delgado happened to be an ex-Jay.
    I do agree that if you compare production at each position before and after a deal then you’re probably getting a good idea of how a deal really worked out. The Hudson/Batista-Glaus deal is a good example. How much did the Jays lose in 2nd base and relief pitching production, and how much did they gain at 3rd base?
    I see where you’re going on the “circular effect” but I would still maintain that wherever a player goes in the future, he’s still a GM’s loss. If/when you look at Josh Byrnes, why should his record benefit from the fact that the Jays only held on to Glaus for 2 years? The Jays decision to move him should have no bearing on whether or not it was a good deal from the Diamondbacks’ perspective.
    As for the Stewart-Kielty-Lilly thing, I’m not suggesting that Kielty be ignored. His time with the Jays should count. JP lost Stewart’s 39 (as per above, I’d actually add his years beyond when he left Minnesota too), gained Kielty’s 4 in 2003 and Lilly’s 32 afterwards.
    The fact that you’re getting 39-4 in the Stewart-Kielty deal should indicate a problem in the evaluation method merely from common sense. Maybe the Jays came out behind in that deal, but it was nowhere near that level of one-sidedness.
    If you want to accurately evaluate each deal on its own, take the player’s entire future, so Stewart-Kielty would be 39-27 (+ whatever they gained after they departed their next teams). Kielty-Lilly would be 23-32 (++ again). What ends up happening is that Kielty’s numbers from 2004 and beyond end up on both sides of the ledger, so they cancel out.

  10. I definitely see your points and you’ve given me a lot of good stuff to think about for the next one!
    For consistency’s sake, though, and no disrespect towards you, I am going to stick with just debiting WS from the initial new team, not the entire future. In the Glaus example, he would still be Byrnes’ loss, as you mentioned, but Glaus’s production on the Cardinals would not be as much Byrnes as Ricciardi. It could be weighted somewhat.

  11. darren says:

    Thanks Eric, I appreciate your replies, it’s been really interesting thinking about this and going back and forth. I’m looking forward to the next one.
    As for who the next one should be – I know it would probably be more work, but what about someone with a pretty long track record like Kevin Towers? It might also be interesting to see how his more recent deals stack up with his older ones – in other words, has he gained anything from his years of experience.

  12. Darren, it’s funny you mention Towers because an evaluation of him is in part what made me interested in conducting these analyses. In the 2008 Ducksnorts Baseball Annual, Geoff Young looked at every single trade Towers made and evaluated them via Win Shares… he used the format I’m using here of just debiting WS for the tenure on the initial new team.
    He also only used trades, not signings or anything else.
    I’d probably pass on him as the next one just for that. Someone suggested Walt Jocketty to me as well as Billy Beane. Both would be interesting.

  13. brent in Korea says:

    I think that it should be noted that the first year was spent cutting costs. They Jays lost a lot of over paid talent that year. Also, it hurts JP that a pitcher who couldn’t cut it in the AL East can succeed in the NL. I think a summary of what the GM’s did and offering analysis would be more useful than comparing Win Shares. What I mean is that you should bring more context into the argument.

  14. Brent, definitely. I think I got so caught up in compiling the data that more context would have helped.

  15. Jun says:

    This really doesn’t work. Ricciardi was brought in to slash payroll and most of his early trades were forced/based on that. Quantrill was a cash dump. Stewart was in his prime but unable to be resigned. Looking purely at WS any GM that has to slash payroll for a few years is going to look like crap in the short run.
    As well, if you’re going to look at FA win shares, you have to take into account the financial aspect (although J.P will probably get slaughtered there). Break even on WS but save 50 million and you’re the greatest GM of all time but wouldn’t get any credit using this type of analysis. Not resigning Delgado was the right move for the team, but J.P. gets punished for it, whereas someone like Cashman who can sign whoever (*cough* Clemens *cough*) or take on bloated salaries like Abreu when other teams can’t afford them is going to look like a genius.

  16. Clutch Hitter says:

    JP has been sensational in the trading department, the only real knock on him is that almost none of his free agent signings have worked out thus far.

  17. anonymous says:

    As a note, Esteban Loiaza was not signed as a free agent, he was actually traded for Darwin Cubillan and Michael Young (with Texas).
    I think if you add in the Young factor alone, this is going to destroy JPs trades even more.

  18. Logan says:

    Loiaza for Cubillan and Young was not a JP trade. It was a Gord Ash trade. It was probably one of the worst trades in the history of the game but you can’t blame JP for that one cuz that was Ash’s doing

  19. All of the transaction were literally taken directly from Baseball-Reference’s transaction logs. Yeah, the financial aspect i what I’m harnessing for the next one.

  20. Jake says:

    I think this is a great project, and it’s certainly well grounded. I hope you don’t overcomplicate the adjustment for domino trades, but agree with the need to tweak it. Not so sure there’s value in measuring by position, as I don’t see any position-related science here. A great deal that gives up a great 1B and shortstop for a great RF and 2B is still a great deal, especially when you have a great 1B ready for promotion, but how would that look in a positional breakdown? And yes, I’d love to see the Jays’ Gillick done one day.
    Also love to see Billy Beane. So much was made in Moneyball of how well Billy and crowd analysed college stats, that one wonders why a team with so many high draft picks has managed to come away with so little over the years. My theory is that Billy is a great trader, and his speciall skill is trading a commodity one year early rather than one year late. I’d love to see what your analysis reveals.
    The Beane thing might bring in duration or longevity as an element. Look not only at what he got in deals like Mulder and Zito, but how much longer he gets the benefit.
    I have to add some Jay comments. The guy who mentioned that JP was brought in to slash payroll doesn’t know that Gord Ash’s average payroll was almost 30% lower than JP’s has been. Dumping Delgado’s $18MM salary was a great idea, even if it was at the end of the contract. But JP had raised such a ruckus about how Carlos’ salary was ruining the Jays that Delgado was not coming back at any price, and he signed for $4MM with the Marlins. Dealing Stewart was a good idea, not so much because of the contract, but because he was showing signs of breaking down to the point where the contract was a losing deal. So while I wouldn’t fault JP for dealing him when he did, I’d want to know why he signed him back this year.

Leave a Reply

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

WordPress.com Logo

You are commenting using your WordPress.com 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 )

Google+ photo

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

Connecting to %s

Follow

Get every new post delivered to your Inbox.

%d bloggers like this: