2013 MLB Team Health Reports: Introduction to Injury-Risk Series

Use your ← → (arrow) keys to browse more stories
2013 MLB Team Health Reports: Introduction to Injury-Risk Series

This is the 12th season of the MLB Team Health Reports, but the first for THR at Bleacher Report. Besides some internal changes to improve the system, the move to Bleacher Report gives us a new format. It think it is not only much better looking than in the past but also more functional. I hope returning readers think so as well.

AL East Baltimore Boston New York Tampa Bay Toronto
AL Cen Chicago Cleveland Detroit Kansas City Minnesota
AL West Houston L.A. Oakland Seattle Texas
NL East Atlanta Miami
New York Philadelphia Washington
NL Cen Chicago Cincinnati Milwaukee Pittsburgh St. Louis
NL West Arizona Colorado L.A. San Diego S.F.

 

Really, they shouldn't be called "Health" reports. If I'd thought about it more when these started, "risk"would have been a better term. We've elected not to change the name since tradition counts in baseball.

For those of you new to this process and this system, allow me to answer some of the more common questions and deal with some confusing issues:

What Is a Team Health Report?

These 30 reports, one for each MLB team, are injury-risk ratings for every expected starting position player, including DH in the American League, plus risk ratings for all five expected starters as well as two relievers. The Team Health Reports are based on a proprietary system that adjusts an actuarial baseline. This baseline, determined from over 20 years of MLB data, classifies players by age and position.

This baseline risk is then adjusted, positively and negatively, by 12 factors that include injury history, team history and expected workload. There are several subjective factors that are created with the assistance of Major League Baseball medical personnel, scouts and executives.

The rating is then put into three tiers of risk: green (lowest risk, less then 34 percent chance of going on the disabled list), yellow (medium risk, between 35 and 51 percent risk) and red (highest risk, 52 percent or greater). These are probabilities, not projections, and are based on large sample sizes.

My old colleague Nate Silver—Nate was a pretty good baseball writer before he got distracted by this politics stuff—does a great job explaining probabilities in his book The Signal And The Noise. We can discuss things like the chance of rain, but we cannot say definitively that it will rain until we're wet and wishing we brought an umbrella. 

The risk The System measures is that a player will be placed on the disabled list at some point during the season. Day-to-day injuries cannot yet be accurately and consistently tracked.

Leon Halip/Getty Images
No one could have predicted Victor Martinez's injury last year.

 

So Who's Going to Get Injured?

The single most misunderstood thing about these ratings is that they are based on probabilities of risk. They are not predictions. There are times when you will look at some of these ratings and wonder why a player might be risky. "He doesn't have a thick medical file!" you might think. "I disagree with that," some say. Good. Fine. I'm the exact same way. 

Washington Nationals pitcher Stephen Strasburg had no history of injuries up to the point that he snapped a ligament in his elbow. He was healthy up to that point, but most people understand that he was always risky.

Another confusing point for some is that the risk is actuarial. It is based on broad probabilities and designed to work over a large population. The problem is that baseball is not a large population and in some cohorts (groups of similar players), there are only a couple players. The System hated Jamie Moyer and year after year spit out that he was high risk—and year after year, Moyer kept pitching. 

Think about the group of starting pitchers over the age of 40. It's a very small group and often includes some oddballs like Tim Wakefield or Moyer. It also includes possible Hall of Famers like Nolan Ryan and Roger Clemens. The group is split nearly perfectly; half are very healthy, half are injured. At the baseline level, The System doesn't care whether it's looking at a soft-tossing lefty or a flame-throwing Texan. It's looking at risk. 

So someone like Andy Pettitte opens at a 47 percent risk rating due to his age and position. That makes it very difficult for him to have a low injury-risk rating even though he's had a largely healthy career, with only minor arm issues and a couple traumatic issues like last year's fractured leg. Once his baseline is adjusted, Pettitte isn't much more risky than most pitchers, but considering how many pitchers get injured, that's not saying much.

Which is why the Team Health Reports go beyond a mere rating. If all you needed was the color-coded rating, I could do that in a lot less time, but it would be much less instructive. By taking the time to read the player comment for any player you're interested in, you'll be at a significant advantage to those who just skim. 

You want a prediction of who's going to get injured or what part of his body is going to snap? Call Miss Cleo. You want to know who's risky? Read on.

 

How Good Is This Thing? 

The System has been tested several times. Internally, I am constantly backtesting to make any possible improvements. Externally, I've provided several people and organizations with ratings in order to test how good or how bad The System does. In every case, the ratings were found to be very accurate.

The Steamer Projections, widely considered to be the most accurate projection engine, took a look at the Team Health Reports in a detailed study back in 2010. Their findings were very positive and at that point, I was using a playing-time estimator that was very inaccurate. That estimator is no longer used, and I believe this year's ratings will be even more accurate.

For the record, here are last year's Team Health Reports.

 

What Changes Have Been Made?

I've mentioned some of the changes already, but beyond the new location and format, there are some changes under the hood as well. As always, there is a new baseline actuarial chart that adds another year of data to the models. This is very similar to the chart used to help set insurance rates for major-league players. 

There are also adjustments to the 12 factors. Innings increases for pitchers were lessened as an effect after the research performed by Russell Carleton at Baseball Prospectus refuting the Year-After Effect. I also struggled to model a league shift, as the Astros are switching the American League. The Brewers made the only similar switch and they had a tough season, but was that bad luck or a pattern? I couldn't find any consistent adjustments, so there are no changes made due to that move. 

I also made a significant change to the underlying playing time projections, using the Steamer projections you can find at Razzball.com

Ezra Shaw/Getty Images
Chris Carpenter's lost season cost Cards $22 million ... and ring?

 

Why Is the Baseline So Important?

If you could only know a few things about a player, what would you want to know? There have been a lot of statistics like WAR and VORP that have tried to put as much information about a player as possible into one easily comparable number. Some traditionalists stick to batting average and RBI.

For injuries, the things I want to know are age, position and injury history. Those are, by far, the most important factors. Nothing else can move the numbers the way those can. A player moving from shortstop to second base increases his risk by almost 10 percent. 

The baseline tends to be pretty common sense. Catchers are risky, both young and old. Left fielders get hurt more than right fielders. Designated hitters are all-or-nothing risks, who are either lumbering guys who don't tend to do more than swing and trot or guys who were breaking down so much that they had to move to DH. 

 

Why Not Day-to-Day or Minor-League Injuries?

Unfortunately there is no consistent record of day-to-day injuries or of minor-league injuries. Worse, there is no historical record of injuries. I could look up who pitched for the Cubs on a given date in 1902, but there's nothing about Doc Gooden's arm injuries until you dig into the archives of the New York Times and Sporting News

You'd think it would be better now, but it's not. Quick—how many days did your favorite team lose to the DL last season? MLB just began electronically collecting this information a few years ago and there's no accurate information available publicly before 2002. (There were internal reports called "Red Books" that were published by MLB's insurance carrier, but they have been discontinued.)

Even sources now tend to be collected. Something like the news feed from Rotowire is an amazing tool, but it's not searchable. MLB's feed is DL-only, and even then it has some staggering inaccuracies. A resource  called the eBIS feed is the only accurate system, and MLB doesn't share it publicly. I'm left doing it myself—or, rather, relying on my amazing interns. 

It's worse at the minor-league level. There is simply no consistent source of injuries. Given the inability to get consistent data on either of these, I have no choice but to ignore them in these ratings. 

 

Do You Ever Disagree with the System? Do the Ratings Fluctuate?

The System—yes, it's capitalized, but not yet self-aware—is blind in many ways. It doesn't let emotion get in the way of information, which is one reason it's often better at projecting things than I am. 

In the player comments for each team's slideshow, I seek to give additional information to help explain the ratings and help you understand why in some cases the risk that The System sees is over- or under-stated. My opinion never changes the ratings.

In addition, once the ratings are "locked down," they do not change. I completed these ratings on Feb. 20, so anything that occurred after that date is not factored into the ratings.

For instance, Curtis Granderson is a green risk, but as you know, he broke his arm after being hit by a pitch. Since he will start the season on the DL, he should be an automatic red risk. He's not. I feel like The System must stand or fall on its own merits. Plus, it tends to be smarter than me. Look back at these next year and you'll get some "aha" moments, the way I did going back through the last couple of seasons' ratings.

 

Why Should I Care About Health?

Over the past five seasons, MLB teams have lost $1.6 billion—with a B—dollars to the disabled list. That's a talented player, often a star, who's not playing. In any sport, the only way to win is to collect talent and keep it on the field. Injured players don't play or at least don't play as well.

Making the game incrementally healthier would make the game incrementally better. Every fan can think of that one "what if" player. What if Mark Fidrych had been healthy? What if Sandy Koufax could have been saved by Tommy John surgery? What if Bo Jackson had stuck to baseball?

It's that simple. Baseball could be a better game for everyone if we could make those kinds of changes. Just a fewer Tommy Johns, fewer fractured wrists and a bit more protection for pitchers and we're making quantum leaps.

Moreover, I think it's important to know who's in charge of that. Fans tend to know their general manager and field manager, but who is the guy in charge of keeping that team healthy? Most fans have no idea what an athletic trainer does, let alone what his name is. Those 20-hour days they average during the season deserve our respect and a lot more help.

 

What Are the Numbers Near the Beginning of Each Team's Report?

At the top are three key facts. They're pretty simple and offer a quick way to get an overall view of a team's general health.

The first is 2012 rank. Teams will rank from No. 1 (Cincinnati) to No. 30 (San Diego) based on DL days and dollars lost last season. More is worse. The midpoint is a good dividing line this year, as the injury stats were much more linear this season, with the outliers on the positive end.

The second number is what injury had the biggest impact on the team last year. Using a value-based number I call Injury Cost, which adjusts a players value based on his expected contribution rather than his actual salary, these are the most significant injuries. Usually they're the ones that make you go "ouch" in remembering them, but occasionally they will surprise you. Just check out the Rockies.

Finally, I list the name of the head athletic trainer for each team. As I said above, I think these men, along with their assistants and the team's doctors, deserve a lot more recognition. While some teams are shifting to a "medical director" model, I think the guy who's first to help the players on the field deserves this annual tip o' the cap.

Rich Pilling/Getty Images
Braun is durable. Does that make him Fantasy MVP again?

 

Can I Use These for Fantasy Draft Prep?

Absolutely. I suggest that you use these in concert with your favorite projections or rankings.

What I feel works best is to use these rankings as a guideline. I call it the "Maalox Test." If you like a player, but the risk of that player makes you reach for the antacids, move him down. Use the ratings as a check on your own views. Albert Pujols is risky, but he's also pretty good. The red rating should be noted, then used only to set a maximum value. Maybe you don't go that extra auction dollar for El Hombre this season. 

The same is true for a pitcher like Stephen Strasburg. He's incredibly risky, but could put up 20 wins and get a strikeout title. That's tough to find, but it depends on what else is available and your own risk tolerance. Alone, the ratings shouldn't be used for fantasy rankings since the context is key.

 

What If I Want a Cheat Sheet for My Fantasy Draft?

Here you go: This handy guide will tell you the ratings and only the ratings. Fair warning: if you haven't figured it out from what's above, there's a lot more to this process than just the ratings. Don't skim and don't skimp.

 

So What Do I Do Now?

Go read them!

Know that a lot of people put a lot of work into these. Editors and designers made them look good. My research assistants Dan Wade and Richard Ke did a stunning job helping me through the process, as well as maintaining the injury database. I was honored to have Jeff Stotts from Rotowire help with a couple of the teams, adding his vast knowledge to the mix. 

By reading these, you might be entertained or educated, but if you really dig in, maybe you'll become a bit of a medhead. You'll be surprised through the course of a season just how much health, perhaps more than any other single factor, defines a season and changes who wins and who loses.

AL East Baltimore Boston New York Tampa Bay Toronto
AL Cen Chicago Cleveland Detroit Kansas City Minnesota
AL West Houston L.A. Oakland Seattle Texas
NL East Atlanta Miami
New York Philadelphia Washington
NL Cen Chicago Cincinnati Milwaukee Pittsburgh St. Louis
NL West Arizona Colorado L.A. San Diego S.F.
Load More Stories

Follow B/R on Facebook

Out of Bounds

Michigan Wolverines Football

Subscribe Now

We will never share your email address

Thanks for signing up.