Lance Berkman For NL MVP
by DyingQuail
on Oct 6, 2008 1:01 PM CDT
While the playoffs finally got interesting yesterday (i.e. there won't be a four DS sweeps), HLP and I began to ponder who is deserving of the title of Most Valuable Player for the National League. Discussions we've heard recently primarily center around Albert Pujols or Lance Berkman, so I'm going to limit the discussion of MVP to those two players. In the end, you'll see that it doesn't matter that I ignored the Han Ram's or the Chipper Jones' of the NL.
We began our methodology by looking to several important statistical measures of a player's value:
The italicized categories are the categories in which a either player has the advantage. Just a quick glance at the table suggests that Pujols is clearly the dominant player this season. However, in terms of being the most valuable player in the National League, he's just not.
Why? You ask. Just keep reading.
Also, read the whole post before you vote in the poll. I may just be convincing enough to change a few minds.
Lance lost out in OBP, SLG, OPS (of course), VORP, and WPA/LI. For those of you unfamiliar with VORP or WPA and LI, I'll give you a brief overview.
VORP is a proprietary statistic of Baseball Prospectus which measures a players total offensive contribution above a theoretical replacement level at that position. It's measured in runs. So the better the player, the more runs they provide over the replacement level.
WPA:
WPA (win probability added): WPA is the difference in win expectancy (WE) between the start of the play and the end of the play. That difference is then credited/debited to the batter and the pitcher. Over the course of the season, each players’ WPA for individual plays is added up to get his season total WPA.
Hopefully the large block quote from Fangraphs gets the point across. For those of you unfamiliar with WE, it's the probability a team will win the game given the game state, the run environment, etc. So it's measuring how many wins a player adds through every action of his during a game.
LI is a measure of how many runs a run is worth, given the game state and the run environment, compared to what a run is worth at the start of a game, when each team has a 50/50 shot of winning a game. In late innings or tight games, due to leverage, the value of a single run could be worth 3 times the value of a first-inning lead-off home run. Thus, WPA/LI is a measure that neutralizes the leverage factor from the WPA. This is done to discount for clutchness or the luck of being able to enter a game state that affords such high leverage opportunity.
********************
So Berkman provided about 24 fewer runs above his positional replacement player than Pujols did. He also could only account for 4.81 context neutral wins to Pujols' 6.48 context neutral wins.
At this point, you have to be wondering how it is that Berkman could possibly be considered the NL MVP. Well, it's quite simple. Lance Berkman added more wins to the Astros than any other player did for any other team in the National League. Sure, he was helped out by the fact that he often found himself in a tight situations that allowed him to really add to his WPA total, but the fact of the matter is, when his team needed him, he clearly came through for them. Albert Pujols on the other hand, was just a tad worse than normal when it came to game changing situations for his team.
So why would we reward Pujols for failing his team in tight situations and punish Lance for coming through time and time again when the AB or the diving stop at first mattered the most? I can't think of a good reason. If the aim of rewarding someone as the NL MVP is to say that this player was the most valuable player throughout the course of a season, the only statistic which I feel adequately captures this is WPA. It is the only measure which captures the direct effect a player had on his team and it clearly indicates that Lance Berkman, is the only man deserving to walk away as your 2008 National League MVP.
0 recs |
23
comments
|
Read Related
Email
PrintComments
Someone over at SBN's sabermetric blog did a listing of wins-above-replacement for every player in the NL
And they used the newer version of the equation, which “rewards” a player with “clutch” statistics – namely BA with RISP. It also adjusted for defense. Berkman was #1.
Now, I don’t really believe in a player’s ability to be “clutch,” but I suppose if such a thing existed, it would make that player very valuable. So I guess it should factor into MVP discussions.
Puma should win it. And if not Puma, then Pujols. And if not Pujols, than Hanley Ramirez. And if not Ramirez, then Chase Utley.
by Only_A_Lad on Oct 6, 2008 2:40 PM CDT 0 recs
C'Mon
I love me some Lance, but you just can’t make this argument.
First off I’m a little dubious of this adjustment to WPA you’re making. Why should you divide by the leverage index? Beyond making the realm of statitude even more conjectural and hypothetical than it already is, the players can only be measured against the situations in which they come up. And stop trying to pretend that clutch doesn’t exist! “Discount for clutchiness” my foot!
Anyway,
Berkman hit .259 and OPSed .821 after the break. That’s 39th and 25th in the league respectively. Is that where you wish to set the threshold for MVP?
Pujols was first in both categories. I understand that Berkman was All-Universe in the first half, and it may seem that I’m kind of pre-selecting the data, but fact is Pujols was a lot better vs. Berkman’s insanity in the first half than Berkman was vs. Pujols’ insanity in the second half.
In additon to leading the NL in OPS (and batting average) in the second half, Pujols also led the NL in home OPS—and away OPS. Lance was no slouch, finishing in fourth and seventh place respectively, but the argument I’m making, basically, is that no matter which way you slice it, Pujols was better than Berkman.
OPS with runners in scoring position? Pujols 1st Berkman 2nd
OPS innings seven on? Pujols 1st Berkman 9th.
OPS vs. Righties? Pujols second, Berkman 3rd
OPS vs. lefties? Pujols first, Berkman, um, 79th.
I use OPS here because it’s easy to work with and I don’t have a BP subscription anymore, and besides, all you really need is a measure of power and a measure of reaching base. But I’m sure anyway that you’ll see these same kinds of splits for the fancy stats you have to pay for.
I love the Astros and I’m a huge fan of Lance, but you just can’t argue he was the best player in the NL this year. Coz he wasn’t.
No matter how marginalized with the numbers you want to get, he just wasn’t.
Alright, time to deport Bourn.
by rastronomicals on Oct 6, 2008 2:44 PM CDT 0 recs
I never argued for using WPA/LI. I argued for WPA.
If you use WPA/LI then you’re getting into the realm of “more conjectural and hypothetical” like you suggest.
I just used WPA…it’s the only thing that measures wins. They measure it regardless of first or second half performance.
My main question to you is: Why is that a hot performance in May isn’t as valuable as hot performance in August or September? If Berkman added the wins, he added the wins. Wins are what matters for a team. If Lance added more Wins than anyone else the National League, then he’s the most valuable player in the National League.
I don’t know if your critique now changes because you misread my argument, but I hope this clarifies all the critiques.
Finally, clutch hitting exists; however, it is not a skill. There is no year to year correlation between clutch performance year to year in hitters. I never said clutch hitting doesn’t exist, I said it’s not a skill. Next time, please read the content more throughly before you jump on your high horse.
At this point I’m wondering if you even bothered to read after the discussion of LI, because I have no idea how you could miss the point I made in the second half of the article — in which I praise Lance’s clutchness and discount the WPA/LI method.
The Crawfishboxes
A good friend of mine used to say, "This is a very simple game. You throw the ball, you catch the ball, you hit the ball. Sometimes you win, sometimes you lose, sometimes it rains." Think about that for a while.
by DyingQuail on Oct 6, 2008 3:02 PM CDT 0 recs
Why So Defensive?
I don’t have a horse and if I did it sure wouldn’t be high . . . .
I will stay with what I said: WPA/LI makes no sense. Never heard of it before you brought it up here, and I’m just offering my critique off the cuff, along the way to the Berkman, first off, you know? Maybe I was a little accusative when I say that I was dubious of the adjustment YOU"RE making. I understand you didn’t invent the stat . . . but you did introduce it here, to me, here today.
You mentioned it you explained it, I’m responding to it. I see where you say it shouldn’t hold great weight in this case, Pujols vs. Berkman, but I don’t see anywhere where you discount it in general.
Near as I can guess, “Neutral context wins” means nothing. All wins, all stats have context.
Not saying, however, that you argued for it in this particular case Berkman vs. Pujols, I can see that the Pujols guy has a higher number.
Anyway. That means I’m changing the subject to the matter at hand, after having digressed. Just like I did in my first post.
A hot performance in May IS as valuable as one in September. But my point is simply that while Berkman has the advantage during the first half, Pujols has the advantage almost everywhere else you look. May is important but so are home/road splits, 7+ splits, whatever. Look at the candidates from as many angles as you can, is what I say.
Feel free to disagree, but you can’t OPS .821 for any 2-1/2 month period and expect to be MVP. Not when another candidate hung reasonably close early.
Even the straight WPA argument is a little specious. Berkman does lead Pujols, but by 0.32, which rounded to the nearest game is . . . zero. Not to make light of things, but how might you explain WPA in the vernacular?
“How many extra games did the candidate win for his team?”
If you express the stat in those terms, you would expect a whole number answer, and 0.32 is not then statistically significant. Berkman has an advantage, yes, but I would look at the numbers and say that the two were roughly similar in that department and take no further info from it.
As far as clutch, east is east and west is west, and never the twain shall meet, but I will refer you to Bill James’ essay “Underestimating the Fog” http://www.sabr.org/cmsfiles/underestimating.pdf for an explanation by one of the leading lights that we might be missing something . . . .
Alright, time to deport Bourn.
by rastronomicals on
Oct 6, 2008 4:09 PM CDT
up
0 recs
I've read the essay and I buy the arguments
I’m just saying I don’t think it’s a skill because there’s no year to year correlation. It may exist, but it’s hard to measure and can’t be overly significant (year to year).
I terms of the rounding the WPA to the nearest win, it doesn’t matter.
What WPA is measuring, is that Lance Berkman comes up to bat with the Astros losing 3-2 in the bottom of the 8th. Hypothetically the Away team has a 71% chance of winning the game with the Astros only having a 29% chance of winning the game (these are modeled by historic data, as well as to account for the current run environment of the leagues and team). If Lance hits a solo shot, it’s a tie game and the Astros, because of the hope field advantage now hold, again hypothetically, a 56% chance of winning the game. He hasn’t won the game, but he just contributed 27% of a win or .27 WPA. So there’s no need for the rounding and Lance wins.
For the more detailed explanations, I’ll point you to Tom Tango (http://www.tangotiger.net/winshares.htm) and his three part series on the subject.
The Crawfishboxes
A good friend of mine used to say, "This is a very simple game. You throw the ball, you catch the ball, you hit the ball. Sometimes you win, sometimes you lose, sometimes it rains." Think about that for a while.
by DyingQuail on
Oct 6, 2008 4:24 PM CDT
up
0 recs
Of Course You Are Free To Disagree
and given what I know about the statstical community, it probably agress with you, and not with me but I think the interpretation I propose that over the course of a season WPA differences under .49 may not be all that significant is a reasonable one.
Like when the adjustable standings at BP says that the Dodgers third order wins are 89.6. To me, that means 89 wins that you can put in the bank.. But of course you’re free to maintain that it means 89.6. Free country, man. No-one’s even gonna ask you what six-tenths of a win might mean.
But to me, it’s the difference between merely summing numbers and trying to find a space where they will have the greatest useful meaning, the greatest, um, interface with reality.
I’ll also note that I was creating WPA graphs for this site as early as 2005 http://www.crawfishboxes.com/2005/6/10/23028/8303 so the refresher is hardly necessary.
Alright, time to deport Bourn.
by rastronomicals on
Oct 6, 2008 5:46 PM CDT
up
0 recs
How do they not have an interface with reality
They’re the aggregate probability for winning that player provided for a team. I don’t see how more real it can get like that. It’s not like saying that 10 VORP equals one 1 win. That definitely is hard to interface with reality, but not WPA.
It doesn’t matter if the team won the game, it matters that the player provided the opportunity, so there’s no need to account for the .6 wins. It is what it is.
Good, I’m glad you know that stat, obviously not everyone does. So for those not up on it, it was necessary to explicate so they can follow along.
The Crawfishboxes
A good friend of mine used to say, "This is a very simple game. You throw the ball, you catch the ball, you hit the ball. Sometimes you win, sometimes you lose, sometimes it rains." Think about that for a while.
by DyingQuail on
Oct 6, 2008 7:29 PM CDT
up
0 recs
I tend to wear Berkman colored goggles
but I can’t overlook the OPS difference + the RBI difference… If it was a close race in both OPS/RBi’s, then I could maybe give the nudge to Berkman for his WPA… but i’m not a big fan of many of these sabermetrical stats, and WPA fall in that category… so I tend to lean back to my roots of loving the OPS stat > all.
Pujols is the best hitter in baseball, and he’s not exactly a slouch at 1st base… I’d rate Berkman and Pujols about the same in the field, but unfortunately… Pujols takes the flag on offense. If Berkman keeps up his 1st half…. He’s my MVP.. But he didn’t.. and he was terrible in September.
Go 'Stros!
by Stros Bro on Oct 6, 2008 4:26 PM CDT 0 recs
The OPS difference is obvious
But it’s getting the hits that matter that counts or helping to set up the environment for those hits that count that matters (i.e. getting on base). RBIs don’t have so much to do with an individual skill as having guys that get on base in front of you, so that’s kind of a faulty argument.
The Crawfishboxes
A good friend of mine used to say, "This is a very simple game. You throw the ball, you catch the ball, you hit the ball. Sometimes you win, sometimes you lose, sometimes it rains." Think about that for a while.
by DyingQuail on
Oct 6, 2008 7:32 PM CDT
up
0 recs
Faulty Argument?? What is WPA then?
If you have a better batting average, slugging %, OBPS, then WPA is just a reflection of what others around you did when you got your hit… If Pujols was on a team that was better at getting on base, then his WPA would raise by doing the exact same thing he did this season…
In some of your earlier posts you don’t seem to believe much in clutch… well, WPA could be seen as more a characteristic of being a clutch hitter than any of the other stats. OPS, AVg etc are all pure stats that are only attributed to what YOU did as a player… not what you did in context of other players on the team..
Go 'Stros!
by Stros Bro on
Oct 6, 2008 8:15 PM CDT
up
0 recs
I don't believe in clutch skill
But in this context, I think the results of any kind of clutch hitting should be counted.
RBI’s don’t capture the contribution to winning. I was trying to point out that if we’re going to argue RBI’s, the Lance’s RBI’s were clearly more valuable the Pujols surplus.
You raise a good point about WPA. I don’t think it’s does a good job of telling you how good a player is or how he will be, but I think it does do a good job of telling you exactly how valuable to a team he was. In this instance, Lance Berkman is the most valuable.
Perhaps this will quell the fire, I don’t think that Lance Berkman was the best player in the NL, by a long shot. Clearly, Pujols was.
The Crawfishboxes
A good friend of mine used to say, "This is a very simple game. You throw the ball, you catch the ball, you hit the ball. Sometimes you win, sometimes you lose, sometimes it rains." Think about that for a while.
by DyingQuail on
Oct 6, 2008 10:03 PM CDT
up
0 recs
I tend to fall on the side of clutch hitting, even though I think there is no good
way to determine whether it is a skill or not. However, my view has been that clutch stats, like WPA, are highly relevant to MVP evaluations, regardless of whether clutch hitting is a skill or not. The MVP is based on “what happened,” not whether what happened is repeatable. If David Newhan got lucky and somehow had a 1.5 OPS and 200 RBIs, he would deserve to be MVP even though people might doubt that he would ever repeat that performance.
by clack on
Oct 7, 2008 9:06 AM CDT
up
0 recs
I don't know that I completely understand WPA
but is that taking into account advantages Lance may have in defense and baserunning?
I understand your argument, but I think you’re trying to hard to fit it for Lance. He was a monster for a stretch in the first half, but it’s hard to ignore the fact that he didn’t continue to produce down the stretch (you may not like it, but voters love to look at the old, “what have you done for me lately?”)
I think I’d have to give it to Pujols.
by littlevisigoth on Oct 6, 2008 5:18 PM CDT 0 recs
I am trying to force it on Lance
But, that’s kind of my job.
This article from Beyond the Box The Box Score includes all the defense, etc. that I’m sure you’re jonesing for. Again, Berkman comes out way ahead when you measure purely for helping his team win baseball games.
Sure Lance wasn’t as good as he was in May down the stretch, but we all knew he couldn’t be. The point is though, no one helped their team win more baseball games than Lance Berkman did this year. Whether you like the stat or not, the argument is pretty clear cut. There’s really no room for argument. I get that it’s unlikely that the BWAA and coaches will feel this way, but objectively, Lance is the NL MVP.
The Crawfishboxes
A good friend of mine used to say, "This is a very simple game. You throw the ball, you catch the ball, you hit the ball. Sometimes you win, sometimes you lose, sometimes it rains." Think about that for a while.
by DyingQuail on
Oct 6, 2008 7:24 PM CDT
up
0 recs
Thank you.
The Crawfishboxes
A good friend of mine used to say, "This is a very simple game. You throw the ball, you catch the ball, you hit the ball. Sometimes you win, sometimes you lose, sometimes it rains." Think about that for a while.
by DyingQuail on
Oct 6, 2008 7:24 PM CDT
up
0 recs
It kills me to say this - I don't want to say this. I hate being honest sometimes
Ryan Braun should be MVP.
Astros fan for life
by Joe in Birmingham on Oct 6, 2008 8:55 PM CDT 0 recs
Can we get an explication?
The Crawfishboxes
A good friend of mine used to say, "This is a very simple game. You throw the ball, you catch the ball, you hit the ball. Sometimes you win, sometimes you lose, sometimes it rains." Think about that for a while.
by DyingQuail on
Oct 6, 2008 10:05 PM CDT
up
0 recs
I can't believe you left out Win Shares...
particularly if you are trying to make the case for Berkman. Lance Berkman leads the NL in Win Shares. Pujols is No. 2. Win shares takes into account offense, clutch hitting, defense, base running, albeit not always with state of the art sabermetric stats.
http://www.hardballtimes.com/thtstats/main/?view=winshares&league_filter[]=NL
by clack on Oct 6, 2008 9:11 PM CDT 0 recs
I was weary of it
Because HBT is getting ready to do away with Win Shares for BaseRuns. I didn’t want to open myself up to that criticism.
Thought you make a great point.
Lance just flat out WON baseball games this year.
The Crawfishboxes
A good friend of mine used to say, "This is a very simple game. You throw the ball, you catch the ball, you hit the ball. Sometimes you win, sometimes you lose, sometimes it rains." Think about that for a while.
by DyingQuail on
Oct 6, 2008 10:05 PM CDT
up
0 recs
this post made me sign up with sbnation...
The fact that Ryan Howard or Manny Ramirez is going to win the MVP honestly causes me a great deal of pain. Most BBWA are going to leave Lance Berkman and maybe even Albert Pujols off their ballots. And the fact that K-Rod will probably win in the AL makes me want to die.
Why is it that national columnists are so reluctant to use advanced stats in baseball, when they frequently use QB rating as a guideline to effective QB play? QB rating has got to be the most complicated stat out there.
BTW, I kinda agree with Bill James in that I think OPS is an overrated stat and that OBP should be the most important stat in baseball…
by amaar on Oct 7, 2008 1:04 AM CDT 0 recs
Amaar
Glad you saddled up.
It makes no sense, but it’s just the name of the game unfortunately. Your QB rating point is excellent and I’d never even thought of it.
The Crawfishboxes
A good friend of mine used to say, "This is a very simple game. You throw the ball, you catch the ball, you hit the ball. Sometimes you win, sometimes you lose, sometimes it rains." Think about that for a while.
by DyingQuail on
Oct 7, 2008 10:34 AM CDT
up
0 recs
I'm going to say this one more time in case you missed it the first time...
PUMA!!!!!!!!!!!!!!!!!!!!!!
by entropic soul on Oct 7, 2008 4:04 AM CDT 0 recs
Comments For This Post Are Closed
Secondary Sidebar
FanPosts
Community blog posts and discussion.
Recent FanPosts
New FanPost
NL Central Standings
| W | L | PCT | GB | STRK | |
|---|---|---|---|---|---|
Chicago
|
97 | 64 | .602 | 0 | Lost 4 |
|
Milwaukee
|
90 | 72 | .555 | 7.5 | Lost 1 |
|
Houston
|
86 | 75 | .534 | 11 | Won 1 |
|
St. Louis
|
86 | 76 | .530 | 11.5 | Won 6 |
|
Cincinnati
|
74 | 88 | .456 | 23.5 | Lost 5 |
|
Pittsburgh
|
67 | 95 | .413 | 30.5 | Won 1 |
(updated 11.18.2008 at 4:56 PM CST)
