In Response to Re: Multi-Year Contracts of the Epstein Age -How did they work out?:


From fangraphs, meaning of WAR: “If this player got injured and their team had to replace them with a minor leaguer or someone from their bench, how much value would the team be losing?”

And then:

Offensive players – Take wRAA and UZR (which express offensive and defensive value in runs above average) and add them together. Add in a positional adjustment, since some positions are tougher to play than others, and then convert the numbers so that they’re not based on league average, but on replacement level (which is the value a team would lose if they had to replace that player with a “replacement” player – a minor leaguer or someone from the waiver wire). Convert the run value to wins (10 runs = 1 win) and voila, finished!

Still totally arbitrary. So, in determining Scutaro's WAR, we look at Lowrie I'd assume. But now to figure Lowrie's WAR, we look at Scutaro? Or Iglesias? Does it take into account blue chippers who are probably better than the starter right now? I understand how it works quite well, I just don't see any real value in it. wRaa is another biased stat. It's comparable to RBI, weighted heavily towards middle of the order hitters. And UZR is far too arbitrary for my liking. So you take two metrics I don't see much value in, throw in some more confusion, divide by something arbitrarily decided by a mathemetician, and...voila? No thanks. Talk about reinventing the wheel.

There are no stats you can look at to determine a player's worth without watching said player. They can give you an idea, but the eye test is the end all, be all. It's easy to identify a good fielder when you see one (Ellsbury), and it's easy to identify a hitter who doesn't provide much (Drew) beyond artificially inflated metrics.

There's no need to try and combine all this nonsense, as I said before, for 100 years fans had no problem identifying good, bad, average or other players, but now people are trying to be smarter than everyone else and it just doesn't work.

I can't wait for people to apply this to football, and talk about a LBs UZR, or a RBs WAR. No, they won't, for some reason the geeks (I don't mean that in an offensive way at all, I consider myself a stat geek) flock to baseball.

It's impossible to find a stat to compare players across the board when the face different pitchers, play in different ballparks, hit in different spots in widely different lineups...it's like trying to predict when and where lightning will strike. You can punch in all these numbers, trends, stats, weather conditions, elevation, humidity, whatever you want. It doesn''t work.

As far as UZR...I'd rather not look at it. What about the SS who plays on a team with longer IF grass, and has mostly Lowe-type grounball pitchers? At the end of the year, he may have 100 more chances than the guy with FAR greater range. Fldg% is only good for figuring what the guy will do if he gets to it. I don't want the SS who gets to everything, and throws it into the 3rd row either. Again, eye test. I watch Jeter and see he has no range. I watched Pokey Reese and saw him get to things I didn't think possible. You have to watchm take your heads out of the textbooks and watch the f r i g g i n games.

In Response to Re: Multi-Year Contracts of the Epstein Age -How did they work out?:


If I had a gun to my head, and was allowed only ONE pitching stat to judge all pitchers...without a doubt, it'd be WHIP. Do guys get on base with regularity against this pitcher? All I need to know.

In Response to Re: Multi-Year Contracts of the Epstein Age -How did they work out?:
Moon you basically are making the point here that all defensive stats are subjective I think. And they are. A fielder in Fenway Park will tend to get more errors than he might in other cities because the official scorers are tougher, so I get what you are saying about FLD%, which should be the least subjective stat of the bunch.

In baseball we love to count and measure everything. SABR attempts to take every facet of baseball and give it a metric to determine performance instead of subjective evaluation. But in that process when it comes to defense subjectivity IMO needs to be applied to create the stats. That doesn't say that the stats derived from that exercise are useless but they are at best a blunt instrument and not the exacto knife that most offensive stats are. I believe Epstein wasn't just talking out of the side of his mouth last off-season when he said the RS don't live and die by UZR and have reservations about the metric. He never said they don't look at but they don't take its result as a conclusive measurement of defensive prowess.  

The statistics are far from perfect, I understand. Was Drew worth more than Ortiz? Maybe, maybe not. Quantifying defense in a static number is improbable. But if looking at the game from a general manager's point, it's something that must at least be attempted. UZR and WAR are steps in the right direction, in a part of baseball stats that will always have slight inaccuracy. However, completely disregarding the stat right now is the wrong way to go, WAR takes into play a huge variety of values and is something good to view if looking for a quick estimation of performance.

I too think Drew was given 1 year too many, I think he was well worth every year, but I'm also aware that is the baseball market. Free agents don't get paid what they're worth every year of the contract, you hope you don't overpay too much, but you're aware that you're going to have years in the contract with decline.

It's never apples to apples in baseball. It's fruit salad to fruit salad; some has more oranges, others more grapes, still others more apples, and some lack any strawberries, but it's all fruit salad.

True and false at the same time. I understand your point, but we're talking about the WAR stat here. I don't think you can compare players of different positions by the WAR stat, because each position is valued differently by WAR.

The Sox, at times, have outsmarted themselves trying to project a player do something totally different in Fenway than they've ever done before because it's here now. That is rarely the case. Gonzalez isn't even doing anything different, just balls that would be in play in Petco are off or over the wall in Fenway, and we're talking two extremes in terms of ballparks, everyone saw this coming. Only the Sox did with a few others, and they've usually been wrong.

I think you're mixing the word SOX with the word MEDIA. I don't remember ever reading the Sox signing someone based off what they'd do at Fenway, this projections are always made by the media.

In Response to Re: Multi-Year Contracts of the Epstein Age -How did they work out?:


That's probably a fair assessment, I just kind of assume the thinking is similar given the contracts. I should, however, know better than to assume anything.

In Response to Re: Multi-Year Contracts of the Epstein Age -How did they work out?:
I agree. In fact in the Epstein era if anything it would easier to make the case they have ignored Fenway as impact factor a few times.

JD Drew's spray chart before he came here said that Fenway would not amplify his game and would mute his HR totals slightly, it has. Crawford is going to be the same IMHO. Coco Crisp being a switch hitter with pull power was muted by the yard. Texieria IMO never would have been as successful here as he has been in NY, a switch hitter with a dead pull upper cut swing. You can't build your team ignoring you play 81 road games but the other 81 at home do come into play big time.

Now the RS under Gorman and Sullivan put too much emphasis on RH SLG without looking at spray charts to see what it would really mean. No one player ever proved how faulty that could be than Tony Armas who hit to fly balls from LF center to right center.

Adrian Gonzalez may be the first guy they have acquired in a while that they may have looked at the spray charts and said to themselves "wow and 81 games in Fenway...". But mostly what appealed to them is the ability to hit to all fields with power and to hit anywhere in the count by adapting to the count. He is (personality quirks and defense aside) as Manny as you can get. They wanted A-Gon because he could hit in the Grand Canyon with an 0-2 count on him IMO. 

In Response to Re: Multi-Year Contracts of the Epstein Age -How did they work out?:
[QUOTE]In Response to Re: Multi-Year Contracts of the Epstein Age -How did they work out? : I've been fighting this battle ever since WAR/UZR became the new popular stats. I'm surprised Moonslav, who's posts are often good, hasn't come to their defense yet. Everything, well, almost everything about both stats are completely arbitrary. For WAR...who decides what a replacement player is worth? How do you figure what any single player is worth in terms of wins in such a team oriented game as baseball? Is a replacement player on the Sox the same as a replacement player on the Pirates? And UZR? OH my god, let's draw a completely arbitrary circle around what someone has decided is the territory a fielder SHOULD cover. What?! I know baseball is a staticians dream game...but there's already enough stats to pour over, no need to try and create something new...ever heard of reinventing the whell?





My point exactly WAR and UZR are flawed stats, how people put any form of credibility into these is beyond me.  Sometimes i think they are for fans that don't watch the games and don't really know whats going on with the team. 

In Response to Re: Multi-Year Contracts of the Epstein Age -How did they work out?:



Its a much better way of evaluating a player than WAR.  It is probably the best method of individually judging a pitcher as well. 

My point exactly WAR and UZR are flawed stats, how people put any form of credibility into these is beyond me.  Sometimes i think they are for fans that don't watch the games and don't really know what's going on with the team.

The concept of UZR is pretty simple, draw a circle, check how many balls are caught in that circle by a player, if the player misses a large percentage more in that circle than another player, he has less range, regardless of how fast he runs. It's main problem is not taking into account ballpark differences in my opinion.

Its a much better way of evaluating a player than WAR.  It is probably the best method of individually judging a pitcher as well.

No it isn't, look at John Lackey and Clay Buchholz of last year. One had an ERA in the twos the other had an ERA in the fours. However, look at their BABIP. Lackey's was incredibly high, he had poor defense behind him, Buchholz's was incredibly low he had better defense and more luck. Lackey also put in 50 more innings than Buchholz. This is important, more innings, less work for your bullpen, less work for bullpen, more rested bullpen, stronger bullpen. None of this is factored into ERA. All of this is factored in WAR.

In Response to Re: Multi-Year Contracts of the Epstein Age -How did they work out?:
WHIP is best and particularly in evaluating relief pitchers. A reliever can come in with two on, two outs give up three hits, two walks, two runs and get pulled for another pitcher. Pooped the bed but his ERA is zero but his WHIP will tell you just how bad he was.  

ERA does have value with starters. It will tell you on average how much run support a pitcher needs to win. It tells you relative to WHIP how good a pitcher is with runners on. Dice K in 2008 had a higher WHIP but a lower ERA. He was tough to score on. But the pitcher with a lower WHIP and similar ERA is the better pitcher. There isn't a magic bullet stat for pitching but WHIP is the best stand alone and W-L the most worst.

Just my takes 

The problem with UZR is how arbitrary it really is.  Ellsbury is no where near as bad as UZR tries to make him out to be.  And other well rated players are no where near as good.  I could go on and on about the issues with UZR.  But then again I could go on and on about the flaws in WAR.

How about going on and on about the flaws of Fldg% and Gold Glove Awards?

As for Jacoby, UZR has him as +5.5 in the OF since 2008 (-0.2 in CF). Fielding percent has him as #2 in MLB since 2008.

If Jacoby is really #2 in fielding, why hasn't anyone from the Fielding Bible ever voted him in thir top 10 CF'er in any year?

I'd say UZR is highly flawed, but more accurate than using Fldg% as your guide.

I think if yu look at the top 10 UZR players at each position over a 3 year sample size and compare it to the top 10 Flg% guys or the Gold Glove vote, you'll agree that UZR's top players are better fielders. Yes, there will be some eye-openers, as there is with any method of judging. UZR should never be th only metric or stat used. Several stats, metrics and systems like the Fielding Bible all point towards Jacoby being about an average (or a tiny bit below) CF'er from 2008-2011.

Still totally arbitrary. So, in determining Scutaro's WAR, we look at Lowrie I'd assume...

You assumed wrongly. Maybe you should first try to understand how WAR and UZR work, before blasting them as totally arbitrary.

There's no need to try and combine all this nonsense, as I said before, for 100 years fans had no problem identifying good, bad, average or other players, but now people are trying to be smarter than everyone else and it just doesn't work.

There has always been heated debates about what players are better than others. For years BA, HR and RBIs were it. Now, we realize OBP is a higher indicator of converting into wins than BA. We realize that RBIs are not any more important than the player who leads off andscores runs by getting on base a lot and using speed to increase his chances of scoring. We find that OBP and Slg% mean more to a team's success than HRs.

Some people weight defense more than others. Some count speed more than others. Some positions on the fild are more importnt than others. No one stat can capture a plaer's true value. WAR attempts to combine many differet factors into one rating or score. Yes, it is not perfect. I have some big disagreements over some of their numbers. I think the defense of OF'ers are counted too highly. I think guys like Zobrist and Crawford re "over-rated" by WAR and UZR. However, when I look at the top 20 players by WAR over a large enough sample size, I think the are pretty close at nailing it dead-on. Their top 20 is certainly a better indicator of good all around players (players judged on more than just one area) than any single stat you can come up with.

WAR andUZR are just tools. They are just one measurement of a player's ability and skillset. They never should be used in isolation, just as Fldg% or BA should be.

If I had a gun to my head, and was allowed only ONE pitching stat to judge all pitchers...without a doubt, it'd be WHIP. Do guys get on base with regularity against this pitcher? All I need to know.

I agree, especially for relief pitchers. ERA is still improtant for starters, but I think WHIP should be used at least equally with starters as well.

The amount (oe percent) of inherited runners allowed after a pitcher leaves the game can highly effect ERA. Plays that should have been made on defense, but were not called errors also effect ERA (and WHIP to a lesser degree).

Take  look at Wake's ERA as a starter this year. It is 4.29. However, on many occasions his runners were allowed to score by those who took over in relief. I am certain it is a much higher percent than other starers on the team. He also has had several runs score of plays that should have been made but were not. I documented them on a Wake bashing thread a while ago, and I found that he could have allowed 10 ERs as a starter instead of 20 with just a little luck. His ERA could easily be 2.50 or 3.00. Looking a Wake's WHIP shows he has pitched well. He is currently 2nd on the team in WHIP among pitchers with more than 35 IP (7 pitchers).

Its a much better way of evaluating a player than WAR.  It is probably the best method of individually judging a pitcher as well. 

Judge for yourself...

Best pitchers by WAR from 2009-2011 combined:
1) Halladay
2) Verlander
3) C. Lee
4) Z. Greinke
5) Felix H.
6) Lincecum
7) Sabathia
8) D. Haren
9) J. Johnson
10) Lester

Best pitchers by ERA frm 2009-2011:
1) Wainwright
2) Feliz H (aided by pitcher's park)
3) Halladay (clearly he should be #1)
4) Pineda
5) Johnson
6) J. Garcia
7) Kershaw
8) Lincecum
9) T. Hanson
10) J. Santana

In Response to Re: Multi-Year Contracts of the Epstein Age -How did they work out?:


One thing to remember is that Epstein "Has" to spend the money because of the impatience of the fan base today. In a different market/budget, Epstein would have not traded for A-Gon (with the big extension) and not sign Crawford and waited for Rizzo and Kalish to be ready. But that might have meant building for 2012, rather than 2011.

Look at the outcry two years ago when the payroll was something like $140M and he didn't have a big signing. The fanbase was up in arms.

Theo is sort of forced to spend the money and FA contracts can be a crpshoot and players typically are overpaid. But having the money, doesn't guarantee success. Look at the WS drought the Yankees had in the 2000s. Look at the Mets and some of the other higher-payroll teams. None have had the success the Red Sox have had.

So in any evaluation of contracts, you have to factor in the on-field success with the dollar amount.

Since OPS has become a "hot stat" of late, why not look at the opponents OPS vs our pitchers. This goes beyond WHIP, since WHIP counts a single as equal to an HR.

As a starer:
1) Beckett  .498 (WOW!)
2) Dice-K    .646
3T) Wake    .703
3T) Aceves .703
5) Buchholtz .706
6) Lester       .711
7) Lackey     .835
 
As a reliever (30+ PAs):
1) Hill        .349
2) Bard     .553
3) Albers  .589
4) Oki        .643
5) Aceves .666
Papelbon  .684
Atchison  .804
Wakefield .818
Wheeler    .867
B Jenks     .868
 

Albers has a .589 OPS against? I wouldn't have guessed.

For reasons that 10K has articulated, I do not use WAR in my player valuations. The defensive metrics are quite good and are congruent with my defensive evaluations early on after watching players. The way player value to market are measured is the fatal flaw of WAR.

For reasons that 10K has articulated, I do not use WAR in my player valuations. The defensive metrics are quite good and are congruent with my defensive evaluations early on after watching players...

Except for Jeter, right? 

2002   0.9
2003  -6.0
2004  -0.7   (GG)
2005  -13.4 (GG)
2006  -7.7   (GG)
2007  -17.9
2008  -0.4
2009   8.0   (GG)
2010  -5.4  (GG)
2011  -2.4
Average since 2002: -5.0 (-70.5 in Range)

Cue: goalpost movement number 112,347.

Prevaricatingly mistating the premise #1,28282828282837474747474747

UZR can be use to bash Jake but not Jete. I get it.

In Response to Re: Multi-Year Contracts of the Epstein Age -How did they work out?:

No, claiming a HOF SS was "never ever a good defensive SS" isn't "getting it".

You are the same dim bulb who didn't see Ellsbury's CF defensive issues in the fall of 2007 and the early spring of 2008.

I don't like GG voting. In fact, I don't like much voting, including All Star, it's too much a popularity contest. I was shocked, though happy, they at least got the CY right last year with Felix and his low win total.

I will concede, Moon, that the one point you've been most consistent on, is that they probably should be viewed as one small part of the whole evaluation...I just have a hard time doing it.

A gold glove voting here or there might have the order wrong. But, even if a guy wins 5 gold glove votings from other teams' managers and coaches that doesn't translate to "never ever was a good defenisive player".