Reds - Signed Phillips
Cincinnati Reds - Signed 2B
Brandon Phillips to a 4-year, $27 million
I'm thinking of two second basemen. One of those second basemen is Brandon Phillips. The other is a mystery player. Phillips is 27 for the 2008 sason, the mystery player is 28. Phillips will cost the Reds $27 million for the next 4 years while the mystery player might not make a single year of Phillips's salary over the new contract. Phillips and the mystery player are both roughly average defensively at 2nd. Phillips and the mystery player, when combining translations and major league numbers, you get the following performance by OPS+ for their careers:
Age BP M.P.
2000 80 COLLEGE
2001 98 COLLLEGE
2002 44 65
2003 86 78
2004 88 89
2005 64 101
2006 88 85
2007 105 117
Lest it be a question of major vs. minor performance, the Mystery Player has played nearly a full season in the majors and has a 106 OPS+ in the majors, while Phillips, in his best season, managed a 105. Mystery Player is absolutely free for the Reds to acquire.
Now, if you've been paying attention at all, it isn't hard to figure out that Mystery Player is Jeff Keppinger, also on the Reds. Brandon Phillips is a good player, with enough power and speed to balance his hacking nature, but the Reds have two good second basemen, which would have made the one with the pretty homer total a tremendous trading chit to fill a hole. Outside of the signing of Cordero, it has to be a pretty disappointing offseason for the Reds - they failed to sell high on Phillips as they should have and despite the amazing quartet of Bruce/Cueto/Votto/Bailey, they made absolutely no move to either design the roster to force Dusty to play them or to trade them for someone who Dusty will play. The contract itself isn't too unreasonable, however, so the loss is mainly opportunity cost.
2008 ZiPS Projection - Brandon Phillips
----------------------------------------------------------------------------------
AB R H 2B 3B HR RBI BB SO SB BA OBP SLG OPS+ DR
----------------------------------------------------------------------------------
Projection 568 87 154 26 2 21 79 39 94 22 .271 .325 .435 94 1
2009? 572 86 154 26 3 22 86 37 93 22 .269 .323 .441 95 1
2010? 578 89 155 26 2 22 87 38 93 19 .268 .323 .434 93 0
2011? 556 84 146 24 2 20 79 37 87 15 .263 .318 .421 87 -1
----------------------------------------------------------------------------------
Opt. (15%) 587 105 176 33 4 27 103 46 87 27 .300 .360 .508 119 4
Pes. (15%) 525 57 128 21 1 15 65 29 99 15 .240 .286 .370 66 -4
----------------------------------------------------------------------------------
Top Offensive Comps: Tony Cuccinello, Granny Hamner
Dan Szymborski
Posted: February 16, 2008 at 01:57 PM |
131 comment(s)
Related News:
Cincinnati
Reader Comments and Retorts
Go to end of page
Statements posted here are those of our readers and do not represent the BaseballThinkFactory. Names are provided by the poster and are not verified. We ask that posters follow our submission policy. Please report any inappropriate comments.
Unsure where "hasn't shown he can play every day and hit" is being answered with what is clearly part time duty on his AB totals, but whatever floats your boat Dan.
Jeff Keppinger At-Bats
2007: 481
2006: 510
2005: 255 (injured)
2004: 505
That's full-time duty.
No, I didn't say that those were Keppinger's Major League at-bats, I said those were the at-bats that proved Keppinger could play in the Major Leagues. The determinant factor of whether or not one can play in the Major Leagues is whether a player is a good player or not, not whether he earned some magic Major League Pixie Dust.
Not at all. Phillips is considerably better with the glove and much more durable. And while it's easy to call '07 a career year for Phillips, Weeks' only real success at the MLB level was for two months of last year. He's a nice player but not an "easy choice" over Phillips.
Weeks broke his wrist when he got hit by a pitch. I don't think there's a durability issue on either side.
Weeks is also two years younger than Phillips.
Weeks' only real success at the MLB level was for two months of last year.
He is a second baseman with a career OPS+ of 98. Discounting his 12 AB callup in 2003 he's been successful in every single major league season. If we exclude "seasons" of less than 50 ABs, Phillips's second best season at the plate is worse than all three of Weeks's reasonably full seasons in MLB.
as a hitter yes, as 106 said, with the glove no.
Weeks has career numbers of .249/.357/.410/98
Phillips is at .262/.306/.419/85
even in Phillips career year of 2007 Weeks had a higher OPS+
Weeks is also 2 years younger
Weeks has some power, is a good baserunner, will take a walk, but he does not hist for good averages
To be clear, I was talking about hitting (which is the context in which Weeks was brought up). The defense favors Phillips, certainly, but if we're looking at the whole picture I would still much prefer Weeks because he is going to hit well enough to move to an easier defensive position - and I think he's athletic enough to make a reasonable centerfielder should Milwaukee decide to move him.
Keppinger is .309/.367/.439 in the majors and .321/.374/.420 in the minors for his career. Given that they are only 1 year apart in age I doubt any system would project Phillips to have higher rate numbers in 2008 than Phillips.
Phillips is likely the better glove man though.
Cnt Player OPS+ HR SB Year Age+----+-----------------+----+--+---+----+---+
1 Barry Bonds 205 34 39 1992 27
2 Barry Bonds 188 42 40 1996 31
3 Hank Aaron 179 44 31 1963 29
4 Larry Walker 178 49 33 1997 30
5 Willie Mays 174 35 38 1957 26
6 Barry Bonds 170 40 37 1997 32
7 Barry Bonds 170 33 52 1990 25
8 Jose Canseco 170 42 40 1988 23
9 Howard Johnson 169 36 41 1989 28
10 Jeff Bagwell 168 43 31 1997 29
11 Barry Bonds 168 33 31 1995 30
12 Ken Williams 165 39 37 1922 32
13 Jeff Bagwell 162 42 30 1999 31
14 Darryl Strawberry 162 39 36 1987 25
15 Vladimir Guerrero 160 39 40 2002 26
16 Eric Davis 155 37 50 1987 25
17 Barry Larkin 154 33 36 1996 32
18 Bobby Bonds 151 32 30 1975 29
19 David Wright 150 30 34 2007 24
20 Ellis Burks 149 40 32 1996 31
21 Dale Murphy 149 36 30 1983 27
22 Tommy Harper 146 31 38 1970 29
23 Willie Mays 146 36 40 1956 25
24 Bobby Abreu 145 30 40 2004 30
25 Howard Johnson 145 38 30 1991 30
+----+-----------------+----+--+---+----+---+
Cnt Player OPS+ HR SB Year Age
+----+-----------------+----+--+---+----+---+
26 Bobby Bonds 142 39 43 1973 27
27 Bobby Abreu 141 31 36 2001 27
28 Raul Mondesi 140 30 32 1997 26
29 Vladimir Guerrero 139 34 37 2001 25
30 Ron Gant 139 32 33 1990 25
31 Alex Rodriguez 136 42 46 1998 22
32 Bobby Bonds 136 37 41 1977 31
33 Alfonso Soriano 135 46 41 2006 30
34 Howard Johnson 133 36 32 1987 26
35 Bobby Bonds 133 31 43 1978 32
36 Carlos Beltran 132 38 42 2004 27
37 Bobby Bonds 131 32 45 1969 23
38 Alfonso Soriano 129 39 41 2002 26
39 Ron Gant 127 32 34 1991 26
40 Alfonso Soriano 126 38 35 2003 27
41 Sammy Sosa 121 36 34 1995 26
42 Jose Cruz 119 34 32 2001 27
43 Jimmy Rollins 118 30 41 2007 28
44 Shawn Green 116 35 35 1998 25
45 Dante Bichette 112 31 31 1996 32
46 Sammy Sosa 111 33 36 1993 24
47 Alfonso Soriano 109 36 30 2005 29
48 Preston Wilson 109 31 36 2000 25
49 Raul Mondesi 108 33 36 1999 28
50 Brandon Phillips 105 30 32 2007 26
+----+-----------------+----+--+---+----+---+
Cnt Player OPS+ HR SB Year Age
+----+-----------------+----+--+---+----+---+
51 Joe Carter 104 32 31 1987 27
I thought there were additional concerns other than the wrist. It looks like he's had thumb issues, though he's healthy coming into ST this year.
The Milwaukee Journal-Sentinel's Tom Haudricourt reports Milwaukee Brewers 2B Rickie Weeks is pain-free entering spring training. "This is the first time in three years I've had a rehab-free offseason," said Weeks. Weeks (wrist, thumb) tore a tendon in his left thumb in 2005 and then had wrist problems for two years. "I think it's (the wrist) pretty much good to go right now," said Weeks. "I've been able to do everything I need to do this winter."
No, actually it was more like "Keppinger's shown for years that he can play every day and hit that near-magic .290 in the majors."
No, actually it was more like "Keppinger's shown for years that he can play every day and hit that near-magic .290 in the majors."
Which he has. He's hit .309 in 466 MLB PA, .334 playing everyday in over 1000 AAA PA, and .338 playing everyday in 400ish AA PA. That's a player that's demonstrated he can play everyday in the majors and hit .290.
Don't try arguing with them.
I think Keppinger and Phillips are an interesting contrast- as shown by their corresponding MLB and minor league rate stats- Keppinger has always outhit Phillips- at virtually every age and level- including the majors (which is why literally every projection system has Keppinger better by rate stats)
and yet...
Phillips clearly has the physical ability to be so much better...
If any of the two is going to have a Rollins' 2007 season it's clearly Phillips, on the other hand it's more likely that Kepp has a Freddy Sanchez 2006 season than Phillips ever has a season like Rollins just did...
Since a large number of players have played in any given minor league as well as the major leagues, it is possible to project how well a player will play in the future (in the majors) on the basis of minor league numbers with roughly the same accuracy that you can on the basis of major league numbers. That is, you can correlate how well players did in the minors with how well they did in the majors, and since we have an enormous number of players to work with, the translations are quite accurate. (As these things go.) I believe that Bill James first demonstrated this in the 1985 Abstract, which is a good introduction to methods that translate minor league performances into major league projections.
Kepplinger's minor league performance has been terrific. Because lots and lots of players have played at both for the Reds/Mets/etc farm teams and in MLB, the projections based on minor league data are roughly as accurate as the projections based on major league data. Projections from Kepplinger's minor league data show him to project to be a very good MLB player in the future. And the methodology used to derive these projections is about as good as methods used to predict Phillip's future MLB performance on the basis of MLB data.
So now, the argument is that since Kepplinger is good, cheap, and with little trade value (because translations from minor league numbers are under-appreciated), and Phillips is good, somewhat more expensive, and with much more trade value, why not trade Phillips for something good, and play your other good second baseman?
Because
1: The potential downside is much worse- If a GM does that, and Keppinger is a bust- well the Gm will be regarded as one huge dumbass and probably just shortened his tenure quite a bit (unless his owner cares more about saving money on salaries than anything else)
2: The downside of not making the move is much less, keeping Phillips, letting Kepp rot on the bench, won't raise barely an eyebrow from the MSM- plus if Phillips turns back into a pumpkin you have a handy replacement.
So ideally you'd like to correct for this bias but that will require having a good model to predict selection that doesn't rely on the same variables you're using to predict performance ... and I don't know that we have any.
You might be able to make a case that the selection bias is minimal if you could show that the guys who don't get a real shot actually perform, in the aggregate, at about what you'd expect. The issue here, of course, is selection bias in the other direction. When one of the Phelps-types gets called up and struggles in their first 30-50 PA, they're going back down. If they do well, they get more playing time and are not in your low-PA sample. We'd expect the low-PA group to have worse than expected production.
If you could get this detailed, the best sample might be the one where the team had basically no choice but to play the guy because of injury to the starter and see if those guys performed to expectation (kind of a quasi-experimental setup).
Anyway, I don't think we will be able to statistically show that teams are making mistakes in cases like that. The players who never get a shot may indeed have serious flaws that don't show up in AAA but would in ML. The ones the teams deem "have what it takes" may indeed have what it takes.
Or to translate to snark: the idiots running the experiment that tells us how good players like Keppinger really are at the ML level are the same idiots who don't give players like Keppinger a shot.
Until we get some GMs with a truly scientific mindset -- I hypothesize that Keppinger will not consistently produce in MLB despite his MLEs, let's test that hypothesis! -- we'll never really know. :-)
-----------------
Many of the "pro-Phillips" posts are based solely on last year's numbers. If Phillips repeats his 2007 performance, he's a pretty good player -- depending on what his true defensive value is, maybe very good. I don't think anyone here disagrees with that. There's just no good reason to think he'll repeat that performance, especially with the bat and maybe the glove given, as far as I know, there wasn't any evidence before this year that he was a top defensive 2B (and defensive numbers bounce around a LOT from year to year so it is definitely possible for an average guy in some metric to lead the league in that metric in any given year).
----------------
But there is a real question about what the Reds could get in trade for Phillips. I tend to agree with Dan in this thread -- remember, in the intro, Dan even says the contract in itself isn't unreasonable -- but it's hard for him to, on the one hand, argue the Reds can fill a significant hole by trading Phillips while, on the other hand, writing this:
Even if I accept "proven" which I don't really, on what world is Brandon Phillips a proven valuable commodity? Prior to 2006, he was thought of as a failed prospect that the Reds picked up on the scrap heap with a .735 OPS in AAA and he had all of an 88 OPS+ in 2006.
Well, if Phillips is going to bring value in trade, it needs to be this world where he's considered a proven valuable commodity. Granted, all the Reds needed was for him to be that in the "world" of at least one GM. Unfortunately for them, that turned out to be their own GM.
And no, it's not that unfortunate. But you just shouldn't sign these arb buyouts after a guy's career year. Make him repeat it. Would that make him more expensive next year? Actually, probably not or at least not by much. Look at those contract comps -- I believe most if not all of those contracts were signed after the player established a given performance level and they still aren't making more than Phillips. The Reds are paying him as though he has established 2007 as his level of production. The Reds didn't need to commit 4 years to Phillips and they shouldn't have. And that's regardless of whether they should have traded him and started Keppinger or not. But it's not a big enough deal to really restrict them.
Now, who could the Reds have traded him to? One option presumably was the Cubs who were hot after a 2B for no particularly good reason and have decent young starters to trade -- and the Cubs might well see Phillips as a good leadoff option to move Soriano down. But teams shy away from trading within their division. The Mets needed a 2B before resigning Castillo, maybe they could have gotten one of the pitchers who went to the Twins. Given how much they liked Hunter, the White Sox might have gone for Phillips with some high-risk, high-reward starting prospects in return. The Padres might be another possibility. I don't see a deal here that blows me away. So maybe if I could fill a couple holes -- good 4th OF (Griffey insurance) plus a good reliever (something like Murton and Gallagher perhaps).
And of course, there is an easier answer to Affeldt as default 5th starter than trading Phillips -- namely having done a better job in finding minor-league FAs, NRIs, Josh Fogg (who they just signed), maybe giving Lohse a call, etc. Krivsky still seems to have some major flaws and trading Phillips might have helped paper over one of them but it wouldn't change the fact that he doesn't seem a very good GM.
As to Dunn, I don't think I'd resign him if I was the Reds. It's not even that I don't think he'll age well -- I don't think he will, but I mean with regard to his post-31/32 career while he should be fine for, oh, 2008-2012 at least. But with his defense in LF and Votto at 1B (assuming Dusty sticks with him), he's just a bad investment for the Reds. Of course, I don't know if the Reds have any non-Bruce OF on the way and Junior will be off the books soon, so I could possibly be talked into it were I the Reds' GM.
You'd be surprised, some of these projection systems are really advanced.
People are quick to dismiss Keppinger's 2007 as a fluke, while in the same breath declaring Phillips the best 2B ever based on, um...2007.
You're falling into the same trap that the MSM usually does...wowed by impressive milestones like 30/30 and ignoring the actual data.
The thing is that MLEs don't miss on just one side - translations say a lot of bad things, too. Teams also aren't selecting players based on the accuracy of the MLEs, they're basing it on whether the players are good or not. As such, if teams possessed this ability, we'd see other things happening. We'd see players that should exceed their MLEs being called up. We'd see an ability to recognize which MLB numbers won't be predictive for future MLB play. We'd also see the failure rate for AAA->MLB promotions be lower than AA->AAA promotions because of teams showing a greater selectivity of promoting players for play (comparatively few players are ever labelled a AA.5 player and not given a promotion if they play very well). And if translation didn't work, it would also mess up the distribution of ability, completely screwing up the geometric distribution as players get worse.
Obviously MLEs, like MLB numbers, aren't perfect. One problem is that it's typically a linear model, which it really shouldn't be. I've spent about 80 hours in the last month reformulating my translations using roughly a million major and minor league plate appearances to a non-linear model with Statistica. Previously, there was a very small correlation (under .10) between the percentage of at-bats that were major league and the accuracy of the future projections - I've got that down to a hair above 0 now, suggesting that any issue with translations weren't philosophical ones but just too simple a model.
The composite projections posted here awhile back have Kepp at .303/.362/.417
and Phillips at .276/.321/.459
CBS Sportsline: Kepp: .280/.346/.408
Phillips: .280/.325/.476
That appears to be the outlier
and having my Roto league on CBS Sportsline for a few years, and being familiar with their projections- I'd take Marcel any day
Kepp's career MLB line is .309/.367/.439 he hit .332/.400/.477 last year
Phillip's career MLB line is .262/.306/.419 and he hit .288/.331/.485 last year.
Why is Kepp going to do much worse than his career line and Phillips so much better?
Phillips K/d 3X as often as he walked.
Kepp walked more than he K'd - which he had consistently done in the high minors.
#120 is making a joke about a mistake in the quoted portion.
ouch
Out three weeks. I guess this is why Krivsky wouldn't trade Phillips -- he knew an injury was imminent!
You must be Registered and Logged In to post comments.
<< Back to main