Sports Illustrated’s baseball preview is on newsstands, and I’d be remiss not to mention it. This is SI’s most stat-heavy preview that I can remember; it includes a good definition and use of Wins Above Replacement (WAR), and of course, you can’t go wrong with Joe Posnanski. Posnanski’s article is an especially important read for anyone who need to restrain themselves from getting carried away with Tyler Colvin excitment (who had zero walks in spring training). Let’s not call him “Felix Pie-proof” until he has a better OBP than Felix Pie.

But back to WAR. WAR, as SI explains it, is “the baseball version of physics’ unified theory.” The WAR formula takes everything into account — hitting, fielding, baserunning and pitching. Everything is adjusted to the park the player plays most of his games in, their opponents, etc.; each player’s WAR is expressed as the number of wins he adds over the most average (replacement level) player possible. In a mathematically perfect world, WAR would spit out the league’s MVP every season — the definition says as much.

But like SI points out, WAR isn’t perfect; however, it does point out which players may be undervalued or overvalued. At the very least, WAR can give a generally good indication of who’s the most productive player at each position at various pay grades. Well, at least, that’s how SI uses it. And for people new to sabermetrics or those trying to find the value of an all-encompassing number, it’s really well done.

In the article, SI designates seven salary categories: up to $1M, $1M to $2M, $2M to $3M, $3M to $4M, $4M to $5M, $5M to $7.5M, $7.5M to $10M and $10M and up. Then in each category, they complete a lineup with the top WAR at each position. There are names you would expect to see; some you wouldn’t Among the not surprising aspects to this article, there are zero Cubs among the 69 players listed.

Does this mean the Cubs are a bad team? No, not at all. The Twins have as many players listed (Joe Mauer in the “$10M and up” category) as the Pirates (Akinori Iwamura in the $4M to $5M category); still, the Twins are the better of the two teams. What is obvious is that the Cubs are not the most efficient operation around. No, really, I’m serious. The Cubs are not well run.

In one way, it’s kind of unfair to judge the Cubs by this list: They have eight players competing for a spot on the “$10M and up” list, where they are (or at least should be) going up against the best players in the game. Of the eight, Derrek Lee is the closest the Cubs comes to an expensive value.

Over the entire roster, Ryan Theriot is the Cubs’ best value, based on SI’s report. Theriot will make $2.6M in 2010 and he had a 2.8 WAR in 2009; this isn’t too far behind Angels shortstop Erick Aybar’s 3.8 WAR, who will make $2.05M.

A list like this does fairly demonstrate how poorly the Cubs have spent their money. The Yankees spend a lot of money, but, at least last season, these players produced — Alex Rodriguez, Derek Jeter and Mariano Rivera are all the best value among players making $10M and up.

And the Cubs aren’t finding value at other price levels, either. Theriot is as close as they get to the best value; meanwhile, the Angels, Rays, Braves, Dodgers and Red Sox have five or more players considered the best value at their salary and position.

Again, no surprises here. But it does suck to see in print.

In the past, Bill Simmons’ two-faced approach to sabermetrics has left some readers dissatisfied. When Michael Schur (Fire Joe Morgan) came on the B.S. Report, sabermetrics are a great way to learn more about baseball; when Jack-O is on, it’s annoying that stat-heads don’t think RBI should be a tool in determining a player’s value.

The Jack-O podcasts and columns with his dad in mind seemed were frustrating for people who want to see advanced stats more widely accepted and comprehended. One would assume someone, like Simmons, involved in the MIT Sloan Statistical Analysis Conferences (SSAC) would feel the same way. So it was disheartening that the most popular sports columnist of our time decided to just ignore advanced stats until mathematicians explained them better.

Especially because Simmons would be one of the ideal writers to bridge the gap — the gap between the 1000+ SSAC attendees and the casual fan who wants to know more.

In Friday’s column, Simmons declared an end to his close-mindedness. He jumped in and wrote a lot of words about advanced statistics in baseball, well, at least the ones he liked. To his credit, he joked about his old way of thinking, calling some his methods of thinking “moronic.”

I’m not a sabermetric expert; his column wasn’t aimed me or anyone else “saber-obsessed.” He kept me interested enough to read all 6000+ words, though. More importantly, I think his column intrigued a group that never cared before.

Jump to Saturday morning: I received a text from a friend of mine asking if a certain stat existed. (It didn’t). But my friend noticed advanced stats. He has heard me say Reed Johnson is a bad fielder because he has a horrible UZR/150, but I think he just tunes me out. Now, he’s asking me about the existence of advanced RBI stats, that account for specific situations.

Some people don’t care, and like Simmons wrote, “you can’t put a gun to their heads and force them to care about this stuff.” But some (like my friend) don’t have to be forced, and if they see well explained advanced stats on a regular basis, they’re in.

I have 20 readers a day, so even if I wrote the most relatable, comprehensive advanced statistical analysis in the world it wouldn’t matter; Simmons is incredibly popular, follows a team among the most influenced by what the numbers say, and he has a writing style that’s both interesting and conversational enough to reach someone who has never heard of FIPS and make it understandable.

It can’t just be Simmons that tries to bridge the gap, though. One thing is clear about sabermetrics, not everyone gets them. A few months ago, Baseball Prospectus’ Will Carroll called on baseball writers to do what Simmons did on Friday: try to relate advanced stats to the casual fans. Because, at least the way Carroll sees it, sabermetrics, even the ones that people assume are simple and easy to understand, just aren’t simple enough.

Remember, ESPN tried to roll out OPS last year as a “new statistic,” explaining it on virtually every broadcast of Sunday Night Baseball. Jon Miller would read a short script, give an example, and then Joe Morgan would say something about the Big Red Machine. All planned out, in great detail. Yet people I know at ESPN have told me that in their testing, their viewers rejected OPS as “too complicated.”

The only thing complicated about OPS is that people aren’t familiar with it. If beat writers started using it more often than BA, and regularly included a sentence-long description to setup the stat, there’s your familiarity. With time, even the most casual fans could drop Alex Rios’ OPS into casual conversation.

The problem with step-by-step explanations is that the writer may come off as an ass to people who already know this stuff. After Simmons’ column, blogger Dan Levy joked on Twitter, “… Still cute to read stuff like “An excellent website called fangraphs.com…” and he’s right. Here in Chicago, we get this every so often from the Daily Herald’s Bruce Miles, who is known to use a sabermetric or two.

To anyone who knows just where to go to find someone’s GB%, it reads awkwardly. However, it’s necessary to mention because a lot of people don’t actively search for Geovany Soto’s GB%, and wouldn’t know where to look if they wanted to search for it.

Which is why Simmons deserves some credit. Assuming Simmons has heard of FanGraphs before (he had to have, right?), there’s a chance he knew mentioning it the way he did would make him sound like an idiot. Rightfully, it was included anyways.

If only more columnists and beat writers would write, “An excellent Web site called fangraphs.com …”

It’s been two days without baseball, and there is still one day to go. Hey, at least there are other things to watch on TV. Oh, wait.

Since it’s been so long, I though I’d remind everyone that there are four teams left in playoffs, two teams in the National League and two in the American League. And, with the NLCS less than 24 hours away, it seems appropriate to give you readers a couple predictions (otherwise known as guesses when it comes to me.)

Or, I can just let Nate Silver and his super-secret formula do it for me.

After Silver invented PECOTA, Baseball Prospectus’ system that projects a player’s performance based on comparisons with thousands of historical players-season, he moved on to slightly simpler projections, like predicting post-season results.

First introduced by Silver and Dayn Perry in Baseball Between the Numbers, the “Secret Sauce” looks at three statistics and team traits generally shared by teams successful in the post-season. These three traits are:

• A power pitching staff
• A good closer
• A good defense

According to the “Secret Sauce,” these three team characteristics were the only ones that related to teams winning post-season games and series, and Silver went out of his way in a post titled “Lies, Damned Lies” to dismiss a few other possible playoff success theories.

“Ending the year hot doesn’t make a whit of difference, for example, nor does having a veteran club, or a smallball offense.”

In the “Secret Sauce,” a power pitching staff is measured by normalized strikeout rate (EqK/9.) EqK/9 is a statistic designed to adjust a major league pitcher’s strikeouts per nine innings, while also taking into account the era in which he played. This allowed Baseball Prospectus to go back years, to further determine the legitimacy of the “Secret Sauce.”

The quality of the closer in the “Secret Sauce” is based on WXRL, a commonly used statistic for relievers. Here’s a fairly basic definition of WXRL, from an article by Derek Jacques of Baseball Prospectus:

“WXRL is a metric developed by Keith Woolner which is based on a Win Expectation framework. Win Expectation has its own statistical report (the Win Expectancy Matrix), and is a pretty complicated topic, so I’ll give you a bare-bones, no-math explanation. Win Expectation breaks down each game situation—inning, score, number of outs, number of runners on base, and which bases they’re on—that occurs in the major leagues, all to measure how the transition from one situation to another alters a team’s chance of winning the game.”

The third factor, a good defense, is measured in the “Secret Sauce” formula with fielding runs above average (FRAA.) FRAA measures the runs saved by a player beyond what a league-average player at that position might have saved.

Together, or separate really, this is a complicated formula with a lot of information going into it. The result at the end, at least this season, is pretty obvious though.

Steve Goldman at PinstripedBible.com rounded up the three parts of the formula and put a list together of this season’s eight playoff teams (and the Tigers and the Rangers) to see how the contenders stacked up against each other.

Ranked first according to Goldman’s calculations, the Yankees, ten spots ahead of the Angels. Right behind the Yankees in second place, the Dodgers. And, the Dodgers have a huge advantage over the Phillies (ranked 20th). ‘

Joe Torre going back to a new Yankee Stadium. Manny Ramirez going back to New York for another big series against the Yankees. The two biggest TV markets in the country in the World Series. I guess we all know FOX is hoping Silver is right.

]]> http://www.outsidetheclubhouse.com/2009/10/fox-is-rooting-for-the-secret-sauce/feed/ 0 http://www.outsidetheclubhouse.com/2009/10/some-people-really-hate-math/ http://www.outsidetheclubhouse.com/2009/10/some-people-really-hate-math/#comments Wed, 14 Oct 2009 10:08:20 +0000 OTC http://www.outsidetheclubhouse.com/?p=1550

Last Friday, ESPN’s Henry Abbott delved into the short career of Kevin Durant in a post titled “The Kevin Durant Conundrum.” Abbott skimmed the surface of the statistical movement in basketball, while pointing out that one of the most common new stats in basketball (+/- rating) say Durant is “one of the Thunder’s worst players.”

Now, if Abbott said that Durant will be one of the worst players on the 2012 Olympics team it would be ridiculous. Twenty-one-year-old, 6′9″ forward-guards with his ability to score (which is pretty much anyway he wants) are welcome on my team any time. I can find a way to get past the fact that his team was outscored by 501 points when he was on the floor last season.

Two years ago, I would have taken him over Greg Oden, and right now, there’s only a handful of people in the league I’d rather have. So, for this one statistic, especially one as unproven and simplistic as +/- rating, to say he’s one of the worst players on the Thunder, makes me more inclined to stop believing in math than it does for me to stop thinking Durant is going to be a superstar.

Plus/minus rating can’t change my opinion of Durant heading into this season. (And it seemingly can’t change the opinion of a lot of NBA GMs. Despite, Durant’s poor rating, he was voted the most likely to have a breakout season in the annual NBA GM survey. Still, I appreciate people a lot smarter than me looking to math to find more efficient ways to scout players, determine a player’s importance to his team and try to gain an edge for their team.

In some ways, front offices are getting closer to improving statistical analysis. “The No-Stats All-Star,” an article by Michael Lewis for The New York Times Magazine, further describes these advances and ways teams have improved finding value in players outside of points and assists. Despite improvements, basketball has yet to be fully Bill Jamesed; there is no VORP or WAR to appropriately values a player’s worth to his team, and who knows if there every will be.

Saying this, what I don’t understand is the harsh, vindictive reaction to Abbott and his article. I know you shouldn’t take everything you read on the Internet seriously, but reading some of the comments attached to the article makes you wonder about how these people can even figure out how to use a computer.

For example:

abbott i hope espn fires your dumb ####

worst blog topic ever, abbott u are a complete ####

In case you can’t figure it out, #### means some form of a curse word. When I skimmed the comments, I played Mad Libs with the ####.

And my personal favorite:

this article makes me want to come to your house and take a s-h-i-t on your front steps.

Those tricky ESPN commenters, using -’s to get through the censors.

It’s funny because when it comes to Durant, I’m on the same page with the guy who misspells the writer’s name and uses “i,” instead of “I.” I think Durant is going to have a great season, 30 points per game is a distinct possibility. But, if he scores 30 points per game, and the Thunder are still bad, I’d like to know why.

[True Hoop]
[The New York Times Magazine]

Using math to predict the future in football results is a tall task, though it doesn’t stop Football Outsiders from trying to do it.

I don’t necessarily buy into all of their predictions, because the formula for predicting football is so much more complex and unreliable than baseball. But on the Internet, where every football site out there feels it necessary to subjectively rank every team, every week, FO gives football fans an objective, statistically based look at the NFL.

FO uses a statistic they conceived called DVOA (Defense-adjusted Value Over Average) to determine their rankings every Tuesday. DVOA breaks down the entire season play-by-play, compares that team’s success to the league average based on down, distance, location on the field, current score gap, quarter, and opponent quality.

Basically, it’s crazy complicated, but I trust it more than I trust some expert overreacting to each crazy week in the NFL.

After Week 5, the Bears are ranked 16th. Right in the middle of the pile of very average, an NFL specialty. Sixteenth place is good enough for third in the NFC North and ninth in the NFC. Not a good sign for their playoff chances.

On the bright side, the Bears’ special teams ranked first in the NFL.

Bow to the man, the myth, the legend, Brad Maynard.

[Football Outsiders]