skyking162

Sal Baxamusa recently wrote a great article on the hierarchy of baseball knowledge and regression (there can never be enough written about regression). It’s worth reading — and even has nice pictures — but the reason I mention it is because of a question left in the comments. I can’t figure out if the poster is being sarcastic or not, but either way, it’s hilarious. What do you think?

I have heard it said that all teams win sixty and lose sixty [and] it’s the remaining 42 that determine who goes to the playoffs. So all this number crunching over pitching, batting, defense etc is the tail wagging the dog. Figure out which games are the most important. Is a win or loss in April more or less important than one in September? Why risk injuring your best players on a meaningless loss or meaningless win [and] save them for the important 42?

Popularity: 81% [?]

I’m certainly not a big believer in repeatable clutch skill, but there are certainly players who end up performing better in important situations. If you enjoy perusing clutch stats, check out Baseball-Reference’s new clutch splits. For example, David Ortiz.

There are two sections to look at. The first is titled “Clutch Stats” and breaks down performance by score differential (tied and within 1, 2, 3, or 4 runs), plus “close and late” and 2-out RISP. The second is titled Leverage and breaks down performance into one of three leverage index categories, courtesy of Tom Tango’s work.

To me, the most useful column of these splits is tOPS+ which compares each player’s performance in each split to his overall performance. For example, David Ortiz’s tOPS+ in high-leverage situations throughout his career is 112, meaning he’s performed 12% better when the leverage index is 1.5 or greater. Interestingly, his low-leverage tOPS+ is 99 and his mid-leverage tOPS+ is 94. Evidently Big Papi just can’t focus when the game is reasonably competitive.

Also, take a look at Ortiz’s tOPS+ by score differential. Do you spy a pattern? Better yet, does it mean anything?

Diff    tOPS+
tied    119
w/in 1  108
w/in 2  107
w/in 3  106
w/in 4  102
 >4     87

And don’t forget about Fangraph’s WPA leaderboard.

Addendum: Here’s a better use of the new split data, one that doesn’t ask whether a player is clutch or not — Ichiro steal when it matters.

Popularity: 20% [?]

I’ll admit it, this is a mail-it-in preview. We had no Orioles’ bloggers interested in a round table and I had an eight-hour fantasy auction yesterday where I tried my best not to bid on any Orioles not named Luke Scott, Ramon Hernandez, or George Sherrill (Brian Roberts and Nick Markakis were keepers). And I think I overbid on Sherrill. Here are some quick-hit Oriole thoughts before I turn this into an Around The Web column:

• Luke Scott isn’t being talked about enough. Yes, he’s old (already 30), but he’s hit 28 homeruns in 663 major league at-bats and sports a .273/.366/.516 career line. That’s better than Nick Markakis‘ .296/.357/.469 line. I know, past performance doesn’t guarantee future results.
• The starting rotation is overrated. Daniel Cabrera might still have the potential to be an ace, but a 10% shot isn’t worth much. He is who he is. Jeremie Guthrie was lucky with his BABIP last year and should regress to a league-average ERA. Adam Loewen is all potential, and the rest of the options are not pretty: John Leicester, Matt Albers (isn’t that a Studio 60 character?), and Hayden Penn who seems to be moving backwards, not forwards.
• The bullpen looks solid with Jamie Walker, George Sherrill, and Greg Aquino, but playing high-scoring games is a great way to unleverage the impact of a good bullpen. Just Baltimore being Baltimore.
• I actually like this team a little better than most people, but there’s no way they’re better than the Blue Jays or Rays, who are both at least .500 teams. Given that the under/over is 65 wins, I think I’ll take the over.

Around The Web

Bashing on ignorant sports writers who try to bash sabermetrics isn’t as fun as it used to be, but with Dusty Baker, a local hack, and FireJoeMorgan teaming up, there’s bound to be some hilarity. Even if you think FJM has gotten old, you should read this post. Here’s another good take on the Cincinnati issue, in the form of an open letter to Dusty Baker on behalf of Joey Votto and Jay Bruce.

Joe Posnanski wrote a great article about the stupidity of the anti-stathead argument. Point one is that those folks often counter numbers-oriented arguments with… more numbers. They just stick with their own. Point two is that traditional numbers might seem simpler, but they’re actually anything but straight-forward. As Joe writes, “Sometimes, the anti-stathead crowd doesn’t realize something that’s worth realizing: The reason so many people keep working on new statistics is because the stats we grew up with are STUPID. And plenty confusing too.” Traditionalists are just choosing to stay inside their comfort zone.

The Brian Bannister statgeek story is getting a wider audience, this time over at Yahoo.

Brian, from my favorite NFL analysis website, offers an explanation as to why maximizing your score in betting pools doesn’t necessarily maximize your chances of winning. Think about that when filling out your March Madness brackets in a week.

Speaking of well-written explanations of sports-related statistics, check out Sal’s write-up of regression and why it’s important. (Hat tip: everyone on the web.)

The Cardinals are giving fans a chance to find their next big draft success. Power to the people!

Joe P. Sheehan has yet another great post using Pitch f/x data. He breaks the strike zone up into smaller pieces and found the linear weights value of the average pitch to each location, including balls and strikes.

Wow, there certainly is a theme to today’s links. Some teams continue to take the traditional route and think Dusty Baker is their savior. But some organizations have embraced changing the way things are done based on all the data now available.

Some fans look at the Rays and Pirates and think it’s the same-old plan (finish last, draft top players, repeat). But some fans notice the Rays’ fielding is much improved and smart contracts have been handed out to talented young players and cheap veterans, or that the Pirates are doing things differently.

Some sports writers continue to whine for the good old days and wonder why print media is going extinct. But some writers embrace the blog format and are reaching a wider audience than ever.

I apologize for the preaching, but I’d love for this level of ignorance not to be around in another ten years. I have no problem with your favorite player being David Eckstein. Just please realize he’s not that good.

Ok, how about some (mostly) non-baseball links…

From longtime reader and non-baseball fan Ari comes this story of the mathematical study of Joba Chamberlain’s arch-enemy: the midge.

This project might sound a bit anal, but it’s probably more productive than anything I’ve done since graduating from college.

Oh, and this article is amusing enough on its own, but reading the comments left by readers who couldn’t figure out that it’s satire is a whole afternoon’s worth of fun.

Your Turn

Popularity: 50% [?]

I have another guest post up about historical fielding, this time about the Yankees over at Steve Lombardi’s WasWatching blog. It’s a fun read, I think, as even a non-Yankee fan would recognize most of the names. Some of the more interesting tidbits include:

• Clete Boyer was a monster at third base, and deserved at least three Gold Glove awards.
• Don Mattingly and Bobby Richardson were extremely overrated at first base and second base, respectively. I’d love to hear some thoughts on why Mattingly doesn’t rate well.
• Wade Boggs did deserve his two Gold Gloves.
• Dave Winfield’s arm had to rival Sid Fynch’s to deserve any of his.
• Scott Brosius’ reputation as a godsend at third base in 1998 isn’t just legend talking. He was a black hole (in the good way).

If you have an interest in fielding numbers, either with today’s players or going back to 1956, it’s a lot easier than you think to find that information. This page has links to simple data tables on range, outfield throwing arms, catcher ratings, and first base scooping ability. I’m happy to answer any questions or point you in the right direction for any request.

Or, if you’re lazy, I’m happy to do the number-crunching for you. I’ll even right something up if you have a blog. Just send me an email at skyking162@gmail.com.

Popularity: 44% [?]

Wow, linkage twice in one week. Let’s get it on…

What are the best pitches in baseball? Joe P. Sheehan chimes in with a brilliant analysis. Using PITCH f/x data, Joe computed the change in expected runs every time a pitch was thrown. For balls put into in play, that’s easy. But for balls and strikes, Joe calculated the change in expected outcome for the plate appearance (hitters tend to do better with 2-1 counts than 1-2 counts). It’s an awesome read. Here’s a teaser: Kason Gabbard’s changeup to righties comes out on top.

VegasWatch tears apart CBS Sportsline for matching ethnicity in 21 out of 25 comparisons between rookies and veteran major leaguers.

Lookout Landing makes a brilliant point about bringing more people to the stathead point of view — we need more easily accessible primers and fewer black boxes.

Speaking of well-written primers, here’s one on replacement level.

How many projection systems include platoon split data? PECOTA does, and even adjusts for percentage of lefties and righties that each hitter is expected to face. Pretty cool.

On a football note, NFL Stats rates kickers on their ability to make field goals. Adam Vinatieri comes out third-worst, and Olindo Mare was dreadful. Remember, kick-off distance is almost half the value of kickers.

Play fantasy baseball? Here’s more proof that drafting closers on crappy teams isn’t a big deal.

Finally, here’s another small reason to pick the Cubs over the Brewers in the NL Central this year: they have an easier inter-league schedule. .054 points of winning percentage over 15 games is .8 wins — on the free agent market, that would cost about $3.5 million dollars.

Popularity: 27% [?]