The New “Moneyball” Approach
It is not very controversial to state that “Moneyball” was a divisive book. Michael Lewis wrote some things that can only be interpreted as denigrating to the scouting community, painting a picture of an out-of-touch collection of old men being replaced by smarter, better analysts. It should have been no surprise that people who considered themselves scouts, or had a lot of respect for the profession, were offended by some of the stuff Lewis wrote.
I wonder how different the book would be if it written today, though, because we are currently in the midst of a market correction based on statistical analysis agreeing with long held scouting beliefs. Defense is at a premium while high strikeout sluggers are struggling to find offers, and this charge is being led by the “smart teams” that Lewis would espouse are doing things the right way.
The Mariners focus on defense under Jack Zduriencik is a well known story by now. But, they aren’t the only ones heading that direction. The Boston Red Sox signed Mike Cameron to replace Jason Bay and have made their interest in Adrian Beltre well known. The A’s signed Coco Crisp and currently have an outfield with three center fielders penciled in as starters. Defensive specialists Adam Everett, Alex Gonzalez, Jack Wilson, Placido Polanco, and Pedro Feliz have all signed, while the guys who provide value with their bats are still sitting on the market.
The teams that use statistical analysis the most are doing what their scouts have been recommending for years. Stats geeks are validating the insights of scouts. If Lewis was following the game right now, documenting stories from inside a “smart” front office, the tone would have to be dramatically different, even if the point was still the same – good teams spend money on undervalued assets.
Timing really is everything. That Lewis chose to write the book when on base percentage was undervalued created a division between stats and scouting that simply would not exist if the book was written today. With the new found appreciation for defense and its place in a player’s total value, stats and scouts agree more than they disagree at the moment.
Perhaps the subtitle for the sequel to Moneyball should be “Why The Fat Scout Was Right All Along”.
27
That’s the one criticism I had of Moneyball, that it painted scouts as total idiots. A nicer tone may’ve sold the idea of the book more.
That being said, a little bit of a bridge too far that high-strikeout sluggers aren’t going to get love? They still get way more than they would have 10 years ago, and teams like them more than, say, the typical fan who would quickly have you believe that Mark Reynolds is the worst player ever by citing his high strikeout rate. Still, it’s obvious right now that Moneyball ironically turned defensive specialists from overvalued to undervalued, and smart teams are feasting on it.
“That’s the one criticism I had of Moneyball, that it painted scouts as total idiots. A nicer tone may’ve sold the idea of the book more.”
That’s been a failing of Lewis’s writing style in almost all his books: he’ll go to painful lengths to put his subject on a pedestal, usually denigrating any counterpoints as completely useless.
A nicer tone may’ve sold the idea of the book more, but Lewis wasn’t really trying to sell the idea. He was trying to sell a book, and staking out an extreme position that generates plenty of controversy is a great way to do that. The people who are actually advancing the ideas — at least the folks who are making progress — tend to be far more even-handed. But they aren’t in the business of selling boat-loads of “must read” sports books to a general audience, and it didn’t do them any favors to be lumped in under the flag Lewis claimed to be flying.
Defense is the new black, but I wonder whether its measurement is as reliable as OBP. UZR appears to fluctuate randomly form season to season, and the Red Sox haven’t been able to find cheaper replacements at SS on the market.
UZR doesn’t “fluctuate randomly.” Defense has small sample size issues. You need at least 3 years of data for anything useful, and remember, defense can (and does!) slump, too. An outlier bad UZR number doesn’t mean the system is broken. All-stars hit .200 for a month every now and again.
At least 3 years for UZR. Can that even be used in a professional sports setting? Is there enough data to determine an outlier?
Isn’t this analysis of ‘Moneyball’ based more on what people think the book is about then its actual content? In one chapter Lewis does discuss the A’s use of advanced defensive metrics to explain Johnny Damon’s value as a CF (ironic though it may seem now) and he outlines the rise of STATS Inc systems for evaluating defense, which play a major role (if not the major role) in the design of teams like the Mariner’s. While it is true that Beane’s use of OBP is a central point in the book, it is hardly the only market inefficiency he exploits. It is just the only one anyone ever cares to discuss.
Also, Derek Jeter won yet another gold glove at SS, despite being average at best with his glove. The always stat savvy Red Sox named Jacoby Ellsbury their defensive player of the year while he was hated by UZR and Youk and Pedroia were just incredible. Clearly the “old-school” scouting world is not entirely in line with the SABR community on defense and that is helping to create the undervaluing. How else can you explain Jack Z getting more than a bag of balls for Yuni?
>>>How else can you explain Jack Z getting more than a bag of balls for Yuni?
easy, dayton moore.
Well,gold gloves are determined by the media, not scouts. I think scouts have always known that Jeter has below-average range, no? (Though he has gotten better, and I think his UZR reflects that).
What I have heard from Yankees fans is that Jeter is playing back further which would increase his range.
They’re not determined by the media — that’s the MVP and Cy Young Awards. The Rawlings Gold Glove Award, like the Silver Slugger, is voted on by coaches and managers.
I think this goes back a little more than just Moneyball. Bill James was writing about this stuff years before, and Beane had already been a GM for 5 years before Moneyball came out. James was hired by the Red Sox about when Moneyball came out. Alderson had also used saber stats when he was GM. Bill james had the Ken Phelps All-Star team and the yanks traded Buhner for him in the late ’80s. Even Branch Rickey used obp, Casey Stengal was pretty well aware of player’s platoon splits, etc.
Lewis had books to sell and was slightly hyperbolic. Sure, lots of teams were and still are clueless, but it is not as if no one understood at all that there is a difference between baseball skills and pure athletic ability or the relative price to pay for those skills other than Billy Beane. I am pretty sure Davey Johnson knew what he liked in a player in the and had read James in the early ’80s.
http://miscbaseball.wordpress.com/2009/12/04/davey-johnson-the-mets-and-sabermetrics/
I do like how the A’s put so much emphasis on OBP for years, which in Moneyball appeared novel.
Then you remember that back in the FORTIES, Branch Rickey was pretty much saying this, and building a team in the exact same light as Beane was in 2002.
It’s a point I wish Lewis highlighted in his book, to show the bullshit factory MLB really was for years. It really is amazing how the media can get people to believe anything. While we all love speed in the leadoff spot here, (Raines being a great example), it’s by no means the definition of a leadoff hitter (Earle Combs is one of the great leadoff men of all time, stole 96 bases for his entire career yet still scored a run in 18.2% of his PA’s). Then they see how exciting Maury Wills is and BAM, get a track star in the leadoff spot.
Maybe go back to little league when you’re old coach said “a walk is as good as a hit” when you were 3-1 in the count and looking to crush one. That implies OBP > BA. Yet high BA – average OBP guys are loved, while medium BA – high OBP guys get trashed (perfect example, Ellsbury vs. Drew).
Now, we battle over defense. Teixeira wins a gold glove and spends 2009 getting a hero worship. Stat nerd points out that his fielding metrics are just okay (mind you, not even bad, just not anything to write home over), and doesn’t imply in any way that Teixeira isn’t a great player, and BAM shitstorm.
Kind of a random post, but it draws to what I think my favorite part of Moneyball is; it taught us, and others, to question what we hear. Gone are the days that Joe Morgan can advocate benching Manny Ramirez for Juan Pierre based on batting average, or can say Orlando Hudson and Robinson Cano’s defense is on par with Chase Utley, or even Dustin Pedroia. And now that defense seems to be the new undervalued asset, we can do what Bill James has wanted for years, which is make baseball “look like baseball” again. Only now, it’ll be less overrun by idiocy.
Great post.
Of course, Joe Morgan still advocates for stupid things. That won’t end until he is fired. So, those days are not completely gone. Its just that we don’t need to believe it and we can create websites pointing out how terrible of a broadcaster he really is.
Fantasy Alpha, You’re right about UZR fluctuating too much from season to season, but unfortunately, this is the best we have right now. I don’t think a single player gets enough chances in a season to create a suitable sample size, but an aggregation of the last four years seems to suffice…at least for now I guess. I’m not sure we’ll ever have a defensive metric as cut and dry as OBP.
I curious what the next statistical trend will be. Any thoughts?
The next revolution could be middle players, players who are positives in both offense and defense but are extremely good in neither area. All above average but never eye popping. While the extremes fluctuate wildly maybe some team will look to the middle? Or maybe its too damn early for me to be writing
I play a baseball card game called IBL, and I’ve found that when making a team, two factors work together that I think are not especially valued:
1) Platoon splits are huge — more platoons should exist.
2) Utility guys are extremely handy (and sometimes can solve multiple platoon problems).
Eric Hinske FTW?
I discussed this with Delmon Young and Jason Kubel with the Twins. If they were platooned at DH, it would improve the team’s offense and with getting Young out of LF, it would help the outfield defense as well. Unfortunately, it would mean acquiring a hitter to replace Young in the outfield then…
http://weareoffthemark.wordpress.com/2009/11/13/if-i-managed-the-twins-for-a-season/
Agreed. If the Phillies could platoon Howard with an average hitting/fielding first baseman, look out. Will never happen of course.
I completely agree. As a Met fan, I’ve been saying all off-season that I would prefer them to sign a platoon bat for 1B and LF to pair with Murphy and Pagan respectively, but this is mostly because I don’t care for most of the available free agents. Normally, I think the Mets should go after the best player available because of the comparative financial flexibility.
I think platoons and players who have position flexibility are extremely valuable to teams with owners that value cost-certainty as much as winning. There is a great deal of bang-for-your-buck potential there.
Has anyone ever successfully quantified the true value of speed on the game of baseball? In other words, has anyone been able to express the impact that speedy players (SB’s, taking, extra bases on base-hits, etc.) have had on the game of baseball. I think a player like Jose Reyes (and a team like the Angels of the past few years) would be looked at differently if some of these things were not only quantified but weighted properly. Thoughts?
Platoons will return once managers own up to the fact that — on all but the very best staffs — the more pitchers you have (and use), the more guys you are giving a chance to blow the game for you.
I think what you forget about with UZR is that it’s measured based on the average. Runs Created, as an offensive example, is calculated with 0 as a baseline.
Let’s say player A has two seasons where the average wRC is 65. Season 1, he creates 70 runs, Season 2, he creates 60.
Compare to his fielding. Season 1, he posts a -5 UZR (or maybe 15 runs above a uselessness baseline). Season 2, he’s +5/+25.
What looks more different to the naked eye, 70 to 60, or -5 to +5?
UZR’s problem, imo, isn’t fluctuations, it’s just that the sample sized required for it to be fully utilized takes a good 3 seasons, and by then a fielder’s ability could change drastically. But in terms of snapshot performances, imo, UZR does its job.
No idea if there is something as paradigm-shifting as OBP was to AVG for the value of a hitter. I think what will happen is proving conventional baseball wisdom. Focusing on defense does just that.
Quantifying the value of speed? Why does the sabre community hate Juan Pierre but everyone in baseball loves him?
Because he’s small and singles-hitting and one of the nice guys in baseball, aka someone everyone wants to succeed.
Except us cause we’re stubborn asshole downers who like Brian Giles.
Well personally I think the most undervalued stat is not a stat in the traditional sense at all – it is the salaries that players get.
I think clubs will always have a tendency to sign big name players to big money contracts, but I think teams quite often overlook the value of:
1) moderate contracts for OK but not great players (i.e. $1-$2M a year for someone like Jeff Keppinger or someone in that ballpark). 25 players on a team X $2M/year is only $50M.
2) calling up guys from the minors instead of signing veterans. I still think this is way underdone in MLB and feeds into (1); moderate contracts.
There is value in signing someone great in theory but I think it’s so overdone right now, that it’s so easy to get a $25M contract, that there is something wrong with this.
I agree with most of this post, but just because defense is en vogue right now doesn’t mean it’s any more or less valuable than it was in 2002. Neither is OBP.
A lot of GMs jumped at players with “old-player skills” after Moneyball, but that left “young player skills” undervalued. If this trend keeps up, then eventually we’ll get the rise of OBP and the three-run homer again.
It moves in cycles.
Exactly. What Dave misses here isn’t that people thought defense was “unimportant” (maybe some did, but I don’t think this was really the general opinion). It’s that Beane, a guy on a really tight budget, couldn’t pay for the five-tool All-Stars. What he could afford was a team full of OBP guys, and many of them were lacking in defense. It was more of a necessary sacrifice than anything.
For the past couple years, the A’s have shifted towards defense – the new undervalued commodity (partially because people jumped on board the OBP train, and partially because only recently have we been able to actually measure it properly). The Mariners have done a great job of this.
At some point, defense will get expensive again and the trailblazers will have to find new undervalued commodities (actually, the A’s have found another: young bullpen arms. It’s completely asinine to give replacement-level pitchers $11 million when those types of relievers are a dime a dozen). It’ll be interesting to see what that will be (I’ve always thought guys who don’t hit a ton of homers, but have HUGE doubles/triples numbers have been undervalued as power hitters).
I agree, Billy is up to something. Take this article:
http://bleacherreport.com/articles/45811-billy-beane-should-abolish-the-closer-strategy
and this one:
http://sports.yahoo.com/mlb/blog/big_league_stew/post/Morning-Juice-Beane-A-s-experiment-with-all-re;_ylt=AvbFq45i6e2sq4in9ViJu6ARvLYF?urn=mlb,193908
and what do you have?
If rosters sizes are ever expanded even a little, this looks to me to be a place where you can deploy undervalued talent.
Dave,
The “moneyball” approach was never espousing one statistical trait over another in absolutes, it was always and is still about arbitrage. OPS was undervalued, Bean saw that and pounced. He was able to get high on-base guys for relatively cheap because the market was undervaluing getting on base.
I think the same thing is happening now. Smart teams are realizing that the cost benefit analysis makes defense the smart play right now. GM’s can get talented defensive players on the cheap, before the market corrects.
It really isn’t about scouts vs. geeks, it’s about maximizing value.
I think Dave’s aware of Moneyball’s central premise
Obviously, which is why this entry seems strange to me. “Smart” team builders take advantage of undervalued talents. Obviously the characteristic that is undervalued will fluctuate as other teams copy cat.
The shift to defense being the en vogue tool doesn’t invalidate the books central premise or its tone.
The core of and pin-up for the high OBP A’s was Giambi, and he was a pre-Beane pick-up. Tejada was not a high obp guy, but he wa a good player. Another Alderson pick. Chavez? Alderson. grieve? Alderson. Hudson? Alderson. Matt Stairs? Alderson. (of course Billy was in Alderson’s front office, so he gets some credit)
Picking the college age pitchers ready to contribute soon to that team with high draft picks (Mulder and Zito) before the core of hitters got too pricey were more important Beane contributions than finding undervalued OBP hitters. It is not as if people didn’t know that a lot of Tony Phillip’s value, say, was in the walks before he came back to the A’s.
The idea that Beane created a winner on the cheap in Oakland by exploiting a market that undervalued obp seems a little overplayed. And of course there are always idiot GMs to exploit. I just don’t think the market was as inefficient as people think it was.
I understand your point that it was market inefficiencies, wherever those were, and not obp, which was the emphasis. I just wonder if even that inefficiency was as great as people think it was.
Dave Cameron wrote:
“this charge is being led by the “smart teams” that Lewis would espouse are doing things the right way.”
Matt S wrote:
“Clearly the “old-school” scouting world is not entirely in line with the SABR community on defense and that is helping to create the undervaluing. How else can you explain Jack Z getting more than a bag of balls for Yuni?”
One of these things is not like the other.
Well, both sides agree defense matters. One side is focused on figuring out who is good at it; the other likes Web Gems on ESPN.
The true message of Moneyball was not that OBP is king, as many (Joe Morgan, I’m looking at you) seem to believe. The true message of Moneyball was that “smart teams” use stats and all the other tools at their disposal to find undervalued assets, as you point out. At the time, OBP was undervalued. Today, fielding is undervalued by many. It’s not just the M’s and the Sox either. The Rays made it to the World Series w/ young talent (via the scouts) and terrific defense.
As a Cards fan, I’m glad that Bay is not even an option for their LF opening. Their backup plan, should they fail to sign Holliday, seems to be a player much cheaper. Rather than spending multiple millions on a productive offensive player like Bay whose value is reduced dramatically by his poor defense, the team seems to see someone like DeRosa or Nady as nearly Bay’s equivalent b/c of the relative difference in defense. Neither of those 2 will ever be confused w/ Carl Crawford in LF but they won’t be confused w/ Adam Dunn (who also doesn’t seem to be on the Cards’ radar) or Brad Hawpe either.
Although Bay is a mediocre defender, UZR was adjusted to show he actually had a good 2009 in the field, not the really bad one UZR initially had.
That’s all just journalism though. If you have a technical story, one of the most prevalent ways to box it up and sell it is as an underdog story. Think about all the “science” stories you remember as a kid — Columbus, Galileo, etc. An interesting story needs an antagonist, and in Moneyball it happened to be the scouts. Without the scouts, the story lacks arc.
Anyway, I thought the main takeaway of the book wasn’t that OBP or OPS was the elixir of baseball, but that Billy Bean was searching for market inefficiencies and undervalued players. I think that’s why the book was so interesting from a more general perspective, reminding you to not just get caught up in the market and trends, but to think critically about value in any market.
I like the way Game Of Inches phrased it
“the spirit of Moneyball was not about on-base ability and slugging. It was about exploiting market inefficiencies so as to maximize the difference between runs creation and runs prevention. Back in 2002, when the raw tools of offense were undervalued, the focus of the underrated market value game was OBP and OPS. In this new era of baseball, however, “[t]he newest, cheapest commodities are defense and athletic players.” And it’s not just Billy Beane who has taken notice.”
I’d argue one of the other market inefficiencies that’s been exploited since OBP is steal %. The Rays and Phillies have both exploited this one, and others are beginning to. Steals were over-rated in the days Moneyball was written, and can still be today. But a player’s ability to steal without getting caught at a significant rate adds lots of value whether or not he adds lots of counted steals or not.
Did a little on the spot number crunching, came out with a multiplicate unbiased regression equation of:
y = 0.002644x + 0.6433, where x = years since 1980, and y = stolen base rate, for the American League. Adjusted R-squared of 64.67%, Coefficent of variation of just 2.48%. American League, as evidenced, has been improving his stolen base rate.
However, it doesn’t take a math genius to see this pattern:
1981: 62.15% success
1994: 68.95% success
2009: 73.98% success
For years, though, the AL success rate kept hovering at 68%-71%. 2006-09 have been the four best years for SB’s in the AL. It looks like teams are starting to understand that your star base stealer going 40 for 50 on the season is better than 50 for 75.
I think the real issue here is the speed of information. Wasn’t Mr. Beane acquiring fast defensive players for the last few years and then seeing the team not score runs? That everyone is claiming to look for defense is indeed “the new black” as mentioned above. In reality the total situation is must more varied. Half the teams just aren’t spending money. A bunch of them are negotiating thru the press by bad mouthing Bay or Holliday or someone else about their defense. The real problem is that getting quality players thru free agency is almost impossible. There’s maybe 3-4 per off-season because teams sign players into their 30′s. And sabermetrics says you can’t sign anyone over 30 to a contract of over 3 years. GM’s are using this thinking as a heavy weapon and it’s sloooowwwwwiiiinnnggg down the market. And way way over valuing young players. There’s safety in statistics too. We need to study what trends sabermetrics creates and whether they have the positive influence we think they do. I think they over value potential greatly to be honest. But I’m a Cubs fan ands live in fear of what Jim Hendry will do next.
I wonder what the Orioles want for Felix Pie?
I feel your fear.
I think we’re now seeing Batting Average being undervalued. LA Angels are a testament to that. They had the best team batting average, but were below average in BB%, yet they were the 2nd highest scoring team in baseball. Putting the ball in play is the best way to score runners. Period.
The Angels were also third in OBP and fifth in SLG. Not that hits are bad, but AVG is clearly not the best way to measure offensive production.
Scoring Runs can be broken into 2 aspects: getting on base and moving the runners on base home.
The first aspect is represented by OBP. The second is represented by ISO (SLG-AVG).
Where does Batting Average come into the equation? OBP. A .400 OBP, whether by hits or walks, is fantastic, but BA is such a volatile stat that a BB% heavy OBP is a much more favorable skill set than a BA heavy OBP. BABIP depends on fielding position, fielder skill, level of contact, wind, etc. BB depends on a hitters ability to recognize pitches. It’s just that simple
More on BB% and its relationship to wOBA (and PA):
http://gameofinches.blogspot.com/2009/11/is-walking-in-baseball-old-mans-skill_25.html
I think we’re now seeing Batting Average being undervalued. LA Angels are a testament to that. They had the best team batting average, but were below average in BB%, yet they were the 2nd highest scoring team in baseball. Putting the ball in play is still the best way to score runners.
“Perhaps the subtitle for the sequel to Moneyball should be “Why The Fat Scout Was Right All Along”. ”
No it shouldn’t/
Couple of points:
1. I see the phrase “you need 3 years of UZR” to really determine a player’s defensive ability. As a guy who works with and analyzes data, this is a general rule of thumb for any basic data analysis. Still, it is often common to have some notable variability among individuals in a population sample if you only have 3 years of observation data for each individual. I would like to actually see some analyses to see if the “3 years of UZR” rule really holds up or if adding a 4th and 5th year year of observation for each individual really make a dramatic impact on the overall population numbers.
2. If ‘defense’ is the new ‘OBP’, then it is likely that there are a number of teams that have created new proprietary metrics/methodologies to further evaluate players defensively. I would be very curious to see how they have expanded on existing metrics (e.g., UZR) or if they have created entirely new metrics.
3. What fascinates me more about the current rage over defense isn’t scouts vs. saber but how opinions are formed & linger among the general media and fanbase regarding the defense prowess (or lack therefore of). As a Phils’ fan, Utley is a great example. He was ridiculed as a poor ‘defensive player’ for years and if anything he is actually seems to be overrated now as a defensive player.
He’s overrated because his UZR is good? I don’t get it.
I can’t help but laugh whenever I hear about Moneyball, when I do I remember quotes such as this:
“I like it,” says Toronto GM J.P. Ricciardi flatly. “I think they’re on to something.”
“I think it’s great,” adds Oakland GM Billy Beane.
“The most innovative MLB front office minds agree that Theo Epstein is doing the right thing in Boston.”
“We get afraid to do something because conventional wisdom tells us it can’t be done.” said Ricciardi.
It was prior to the 2003 season, and what were the Moneyball guys all talking about?
They were talking about Theo Epstein attempting to re-invent the wheel with his infamous moneysaving closer-by-committee strategy.
How did Theo’s plan work out?
Two months into the season Theo made a panic trade for BH Kim who eventually became the Red Sox closer.
Later in the season Kim lost the closer job to Scott Williamson, who pitched well but apparently not well enough to be brought into Game 7 of the ALCS as Pedro was left in to give up the lead.
Less than two months later, Theo lavished free agent closer Keith Foulke with an enormous contract … and the Red Sox have had a designated closer ever since.
But how could Bill James, Theo Epstein, JP Ricciardi and Billy Beane all be so wrong about the same thing?
The same way pitching wins are now thought of as irrelevant, RBI are now thought of as meaningless, and WAR is used to justify leaving Chris Carpenter entirely off this year’s NL Cy Young ballot.
Really? One half-hearted attempt at using a closer-by-committee approach torpedoes that idea forever? FAIL.
“Well I was shootin’ dice down on the corner and the guy gave me even money and let me take 1, 2, 3, 4, *AND* 5! You know what happened? I rolled a 6!! What a rotten deal! Damn I’m a sucker.”
And, if you honestly think that (a) pitchers’ win totals are the single best metric for measuring pitching prowess; (b) runs batted in are the single best metric for measuring hitting prowess; and/or (c) a reasonable case couldn’t be made to vote Carpenter the 4th best pitcher in the NL, and/or a reasonable case couldn’t be made for voting Vazquez (or Haren) a top 3 pitcher in the NL, then…
MASSIVE FAIL… *computer starts smoking* *screen blinks out*
It’s not that there’s *no* value in wins, or runs driven in. That’s never been a SABR contention. More wins, and more RBI’s, are better than not. I’ll take more of each for my beloved Jays, please.
It’s that methods for measuring past successes, and/or measuring future probabilities of success, have advanced just light years beyond these simple measures. *Adjusts horn-rimmed specs n’ pocket protector*
For the last time, the 2003 Red Sox “bullpen by committee” was NOT the model proposed by Bill James.
The model is to have your relief ace pitch not in the 9th of a tie game, or a small lead, but to pitch in the most important situations. The Red Sox did not have a good reliever, so they went out, got Keith Foulke, and hey, remember how Keith Foulke was used during the the 2004 ALCS?
Now THAT is the relief ace model.
Look at that beginning of 2003 bullpen. You’re going to blame a SYSTEM for its failing? You are a simpleton of the highest order if that’s the case.
And btw, Hillenbrand was traded for Kim. Hillenbrand was a player the Red Sox wanted out of Boston anyway. That was not a desperation trade. Quit re-writing history.
What’s sad is that this won’t be the last time someone has to make this point.
Here’s a number: 17. The # of losses credited to the Red Sox bullpen in 2009. This was tied for the lowest in MLB with the Yankees.
But instead, we go back to one opening day game vs. Tampa Bay in 2003 and crucify years of Bill James’ research as a result of Alan Embree and Chad Fox combining to suck one inning.
Apparently a botch job in one game carries more weight than the 567 blown saves in MLB in 2009 doing it the way it’s always been done.
Rest assured the current defensive metrics will become obselete in the near future once the new technology that measures outfielders’ skill sets such reaction time, route-taking, throwing efficiency, accuracy and other very objective measures take hold. I read that that these machines will make an appearance in 2010 in something like one or two ballparks. Watch and see how transformative this technology will be in measuring real skill and the time it takes to execute.
I think stats like UZR are big joke. If you want to know how good someone is, spend some time watching the player’s approach and ability to efficiently adapt to whatever challenge develops.
My favorite decade of baseball is the 80s, not just because the Cards went to 3 WS, but because (as Bill James described) teams won in so many different ways.
Once the “steroid era” hit, every team seemingly changed their methodology to accumulating as many home run hitters as possible, and the players figured out that more HRs ment more $$$ … AND the fans wanted to see high scoring games. Steroids + “Coors Field” = a more NFLlike MLB, which non-purists (casual fans) favor.
Now, that the steroid era is *cough* over, and combined with the reality that not every team can have a 15M player at every starting position, some teams are looking for the bargains as you describe. The bargains are (and probably have always been) good defensive players. When you can get a good defensive player that also gets on base, then double jackpot. If he can run, even better.
I grew up with the “Whitey ball” Cardinals and they seemed to stockpile speedy outfielders (to the tune of not having a place for Lance “1-dog” Johnson) just because they understood that you can get by with average pitching if you have great defense all the way around, and I am doubtful that there has ever been as good of an IF defense as the 82 Cardinals. If you add in OF, then the 85 or 87 Cards are tough to beat (defensive numbers alone). That’s how you make it to the series with one guy hitting near or slightly above 20 HRs.
The thing about strong defensive guys is that they are more limited in number (currently) than the offensive players. Very few prospects get signed because of defense, and teams likely are not patient with light hitting defensive guys unless they’re base burners. We are much more enamored with the guy that hits 20 minor league dingers and strikes out 25%+ of the time. Not sure modern fans want to go back to the 60s and enjoy 3-2 games on a regular basis.
As others have mentioned, the next place teams will be looking to “save money” (i.e., quality for low cost) is in the bullpen, particularly non-closers.
The point being made is agood one, not that OBP was not important until now, but just that teams weren’t paying market price for OBP the way they were for BA. Just like now, where teams don’t pay market price for defense, even though a “run saved is a run earned”.
I would say that this is ONE area where sabermetrics is going to make a drastic impact on player evaluation. Soon, teams won’t be over-paying for the mammoths that drive in 90 runs, but allow half that to be scored because of their poor defense. It’ll be kinda like hockey where guys have a +/- type of score (runs scored / runs allowed), IF the defensive metrics ever get completely accepted and fine-tuned. I suppose this is where WAR comes in and allows us to get a numerical representation of how much more Mark Reynolds’ big bat is worth than his bad defense.
A +/- value might be more “casual fan friendly” than WAR is. Reynolds’ 09 season might be a “+17″ rather than a “3.9 WAR”. I’m not sure the casual fan will ever get into the aspect of “positional adjustments” and the like. However, a +/- number has the potential to end up on scoreboards.
Adam Dunn = -1
Nyjer Morgan = +22
Most casual fans would throw a poop fit if Morgan replaced Dunn on their team.
I think Bill James probably likes that the natural order of things is coming back to baseball.
I always found it weird how he disparaged the 50′s and 90′s for being mostly slugging eras, and praised decades like the 60′s and 80′s, even though the 60′s and 80′s were marred by overmanaging, bad baserunning, and other forms of stupid baseball.
But now “baseball will look like baseball again”. Fortunately, it will be without the excessive hit and run, the SB rates in the low-60%’s, etc.
I think what James disparaged about the 1950s was how one-dimensional the baseball was. He’d agree as much as anyone that the three-run-homer style wins ballgames, but the point is that when everyone plays that way the sport becomes dull. I don’t remember him praising the 60s so much; the 80s, yes. Who wouldn’t? I recall how much the 1985 season got me back into baseball after being down on it after the strike (especially down on the owners).
I think we seem to be missing a central point about moneyball, no matter what conclusions it stated or tone it reflected, it invigorated a fan base to an even greater extent then any of Bill James earlier works. Just talking about the controversies therein were enough fuel to pique our curiosity. Even if Micheal Lewis intentions were far from that how many of us would be here arguing any of these metrics? How much ‘demand’ would there be for writers like Dave here?
His impact on the MLB front offices will likely not be fully known but I know what he’s done for the fans, and we’re all the better off for it.
How many fans, though?
Go to most average baseball message boards, people will still be “analyzing” with batting average, errors, strikeouts, and whatever they can wrap their minds around aesthetically. Imagine what a cow people will still throw if you try to argue that Scott Rolen or *gasp* Adam Everett were the best defensive players of the decade. Hell, I once wrote an article arguing against the criticism Mark Reynolds was receiving for his high strikeout totals using things like facts (here’s a fact: Babe Ruth led the AL in strikeouts 5 times).
Was I insightful? More like a nerd who doesn’t watch the games. I didn’t “get” that Reynolds’ strikeouts were killing the team when he could’ve been getting singles or at least moving the runner over (forget the fact that he was 2nd on the D-backs with a 26.37 RE24, which was 2.63 runs less than his wRAA, what horror for $450,000, a 3.6 adjusted WAR player). Also, apparently strikeouts are not a product of eras, so using Babe Ruth was “stupid”. Apparently strikeouts increased because hitters suck now, not because pitchers have established out pitches more as the league has progressed.
I will say that I write on Bleacher Report, which is an “average message board” of sorts, and I’m the #1 MLB writer on the site right now. Most of my writing deals with stats from Fangraphs. So obviously my writing, nerdy or not, is working.
Do I run into the sort of people you’re describing? Hell yeah. Here’s a hint: NEVER tell the Giants fan base that Matt Cain didn’t improve from 2008 to 2009. EVER. Just don’t do it.
But just as often there are people who will actually quote UZR, WAR, and all these other advanced metrics right back at me, and sometimes even go further into the numbers than I do.
And then there’s the most common response I get, which is “Wow. Great numbers. I’d never seen that before. It’s really interesting.”
We’re always inclined to remember our negative experiences, and sure, there are plenty of people will have the importance of pitching wins taken from their cold, dead hands…
But the landscape is changing. I myself got into all this stuff when I read Moneyball five and a half years ago. Do I disagree with a lot of the stuff in it now? Yeah. But it started me on a journey through BP and here, and into a whole lot of really interesting stuff.
I think what will happen is proving conventional baseball wisdom.
I think as more analysis (whether improved stats, advances in scouting, maybe even some sort of biomechanics, whatever) leads to more accurate evaluations, what we’ll see is some conventional wisdom being proved (as with “Hey, defense actually is important after all”) and some conventional wisdom will be disproved (as with sacrifice bunts losing much of their luster). Fact is, conventional wisdom evolves. So does scouting. The “old fat guys” out there scouting while Lewis wrote Moneyball were not scouting in exactly the same way as the guys who signed Babe Ruth and Lefty Grove were.
Stats evolve too. xFIP is certainly better than ERA for evaluating pitchers, but ERA in it’s day was a lot better than W-L record. UZR is better than Fielding Pct. wOBA is better than OPS, etc. And one of the important things about using numbers for anything is doing a little common sense validation of the results. If, for example, some guy create a new valuation metric and his numbers routinely say the best starters on a team are not the guys the teams consider their #1 guys, well, maybe he’s a genius with a great new metric, and maybe his math is just wrong. It happens.
The shift to defense being the en vogue tool doesn’t invalidate the books central premise or its tone
The central premise, no, but the tone, yes, absolutely. The tone of the book was that the scouts were clueless, picking guys based on how good they looked in blue jeans, while the “smart guys” (usually with Ivy League degrees like Lewis) were quietly scarfing up the undervalued OBP guys. Turns out a team really only has room for one or two “old-player skills” guys who can’t cover much ground in the field. The scouts Lewis was mocking could have told him that – you put the big slow guy at 1B or DH. If you already have two big slow guys you like, don’t go get another one for your lineup.
So yeah, the premise of digging out market inefficiencies is still valid. The tone of mocking guys who didn’t use lots of complicated math in their evaluations hasn’t held up so well.
This article makes one of the really annoying mistakes about Moneyball. Thanks for perpetuating it.
Scouting is about evaluating talent. How good is the glove? How good is the bat?
Constructing a roster is about knowing which talents score and prevent runs and acquiring them.
Moneyball was about doing the latter. Scouts are completely clueless about that. They have no idea what a wOBA is. Many wouldn’t understand where the weights come from. Some who didn’t understand wouldn’t even care. What Lewis criticized was using scouts talent evaluations as guides to roster construction. The problem with doing that is that scouts have no idea how to weight gloves against home runs, doubles and strike outs.
Scouts might be able to tell you everything you could know about Brandon League and Brandon Morrow, but that still doesn’t tell you whether you should trade one for the other, because in the scouts mind one is a reliever and the other is a starter (maybe) and scouts don’t understand precisely the relative values of each.
Whoever said that analysts are eventually going to justify all of conventional wisdom is making so many mistakes it’s hard to count. First, there was a lot less agreement withing “convention” than this idea represents. Second, the way people constructed a roster in 1980 is conventional wisdom. If you gave them the necessary social tools to do it, most of the readers of this blog could run a team better than most of the people running team in 1980. Just knowing how to crunch a FIP, relative values of positional abilities and 2*OBP+SLG would do it.
In defense of sabermetricians, offensive statistics have been and continue to be for the most part of much better quality. That defense has come into focus the past year or so really speaks more to the fact that defensive data is finally becoming worthwhile, rather than a realization that defense matters.
It’s always nice when data fits perfectly into little buckets, and almost entirely a product of one’s ability.
Two guys, 600 PA, one K’s 100 times, walks 80 times, hits 30 HR, 4 triples, 40 doubles, 76 singles, that’s a .288/.383/.554 hitter. The other K’s 60 times, walks 40 times, hits 10 HR, 0 triples, 30 doubles, 150 singles, that’s a .339/.383/.446. Just bust out the linear weights and you know who’s better.
Player A: ~ 102.4 RC
Player B: ~ 84.1 RC
Simple.
Defense? Not so much.
“Really? One half-hearted attempt at using a closer-by-committee approach torpedoes that idea forever?”
Half-hearted attempt? Are you for real? They let their ace closer Ugie Urbina walk after the 2002 season and they intentionally began the 2003 season without a designated closer, let alone anyone who even resembled a legitimate closer. And we’re not talking about a crap team with nothing to lose, we are talking about a team with the 5th-highest payroll in baseball that fully intended to compete for a championship.
Closer by committee was an embarrassing blunder, all parties admitted it. Bill James was a moron for coming up with the idea, Theo Epstein was a moron for implementing it, and Beane/Ricciardi were morons for praising it.
If there was even the slightest chance that it could work out, why did JP never implement it himself? Why did he give BJ Ryan a recordbreaking $47 million contract to be the team’s designated closer? Why did Theo never try it again? Why did he give Foulke a $20 million contract to be the team’s designated closer?
Show me even ONE example where ANY team has intentionally used the closer by committee in a playoff season.
closer by committee is misleading. All we are advocating is for teams to use a “relief ace” in the highest leverage situations, and not necessarily bringing in a reliever so he can get his “save” with a 3 run lead. That is basically what the Yankees did with Rivera this last postseason.
Bill James didn’t come up with the “bullpen-by-committee” idea, so stop saying it.
“And, if you honestly think that (a) pitchers’ win totals are the single best metric for measuring pitching prowess; (b) runs batted in are the single best metric for measuring hitting prowess; and/or (c) a reasonable case couldn’t be made to vote Carpenter the 4th best pitcher in the NL, and/or a reasonable case couldn’t be made for voting Vazquez (or Haren) a top 3 pitcher in the NL, then…”
(a) Win totals have value, nobody said they have the most value.
(b) RBI have value, nobody said they have the most value.
(c) No, a reasonable case CANNOT be made to vote Carpenter the 4th best pitcher in the NL this season. Law tried, and what did he come up with? FIP, IP, and (in his own words) primarily WAR. And as we all know, his “reasons” have all been shot down.
“MASSIVE FAIL… *computer starts smoking* *screen blinks out*”
GIGO
“For the last time, the 2003 Red Sox “bullpen by committee” was NOT the model proposed by Bill James.”
I have no idea what you mean by “bullpen by committee” as EVERY team has a “bullpen by committee”. I can’t think of even one team that has a one-man bullpen.
Thanks troll for correcting the semantic error instead of discussing further.
“But instead, we go back to one opening day game vs. Tampa Bay in 2003 and crucify years of Bill James’ research as a result of Alan Embree and Chad Fox combining to suck one inning.”
Again, not sure what you are talking about here. Bill James convinced Theo Epstein that the baseball world had been overrating the abilities and the value of designated closers. That is PRECISELY why Theo let his ace closer Ugie Urbina walk after the 2002 season and did not replace him with someone who had the credentials to be a designated closer.
Have you ever actually read that article from the abstract? Bill James advocated a relief ace, not a committee closer, model, in that essay. He advocated bringing in your best available relief pitcher not for a “save”, but in the most important situation. That could be the 7th. 8th. 9th. Maybe even the 6th. The crux of the argument was that it was stupid to hold back your best pitcher if the 3-4-5 is coming up in the 8th just so he can pitch the 9th.
What the Red Sox did, whether James advocated the move or not, was NOT that.
And then the Rangers signed Urbina for a lot of money, and the Red Sox traded away a dead weight and brought in BH Kim. You’re bitching about a lateral move.
“He advocated bringing in your best available relief pitcher not for a “save”, but in the most important situation. That could be the 7th. 8th. 9th.”
That is an entirely different thesis of his, and one that I support in certain situations. However like with most everything, there are exceptions to the rule.
All I’m saying is that the 2003 Sox bullpen was a halfassed attempt at his theory.
His theory is good. The Red Sox execution of it was bad.
Wait, Michael Lewis wrote Moneyball?
I thought it was Billy Beane’s computer, in his mom’s basement, with the bag of cheetos, that was guilty of that crime against all that is just and right in the world of baseball.
Many have pointed out that Moneyball was about inefficiency, which i think Dave addresses in his post. My quibble is with the statement that “The teams that use statistical analysis the most are doing what their scouts have been recommending for years.”
No, they aren’t. What scouts have been recommending for years is to rely solely on the judgments of scouts. Analytical approaches, which perhaps aggregate the observations of different scouts over time and wring out the subjectivity, have led to beter evaluation of defense, but the current trend doesn’t validate the attitude of the “fat scouts” any more than a successful stock portfolio selected by a monkey throwing darts at the stock pages validates that approach to stock picking.
Dave, have you even *read* MoneyBall? If you did, did you pay attention to it or think about it at all. IT WAS NOT ABOUT OBP!!!! There’s an entire community of people out there who have been explaining what this book was about ever since it was written, and STILL people assume it was about OBP. Lewis even explained it himself: the book was simply about a business approach to baseball focused on identifying undervalued skills, and buying them at a discount. I think he even states in the book (he certainly did in interviews in the years following publication) that as advanced metrics were developed, defense would be The Next Big Thing. Now we’re reaching the point, or will soon enough, where defense is no longer undervalued, and it will get tricky once again. But please, repeat after me: the book was not about OBP. Say it 10x fast ;)
I cant find your RSS feed i would like to subscribe to your content.
One of my friends sent me link to this post saying I must definitely have a look at it, cause I am interested in such topics. And he was definitely true! Currently I am only preparing my small site bezecke pasy