January 31, 2008
Team Health Reports
It's that time again. Each year since coming to BP, I've done a pre-season health report. Last season, to switch things up a bit, I went from doing it on a team-by-team basis to doing it on a positional basis. We learned some interesting things by comparing players side by side, but we're going back to the Team Health Report because I believe that team health is an organic process. Injuries seem to feed on injuries, we learn about a team's strengths and weaknesses, and we can learn which teams might be held back by their health and which might surge.
So what's new this season and what's changed about the system? The biggest thing you'll notice is that I've had some help. The team of writers from SportsBlog Nation have assisted me in compiling the projected lineups, and each asked "The Big Question" about their team's upcoming season. We've tweaked the formatting a bit, but that's just visual. Behind the scenes, there are a few small changes to the simple system that underlies the ratings. The most important is that I've been able to get more accurate Body Mass Index figures on some players, which is taken into account this season for the first time. The downside here is that I wasn't able to get them for all the players. I've decided to include it where possible, and judge next year whether or not it helped the projections.You'll also note that I've scrapped the concept of the "blue" player. The blue rating was one saved for those that were well outside the norm for their position. The problem is, it simply didn't work. Almost all of those affected players--a near 50 percent injury rate--had some sort of traumatic injury, one that simply can't be predicted. In the red/yellow/green system, the larger numbers help smooth the data some. It serves as a reminder that these ratings are only probabilities. Just as you or I could walk outside the house and get hit by a bus, the greenest of green ratings isn't going to help if Prince Fielder crashes into you on the basepath.
With more data behind the scenes--five years of injury data plus the continual improvements to PECOTA--the system is getting more accurate. It will never be perfect and I'll never stop tweaking it, but we are rapidly approaching the point where the gains will be smaller. In fact, the one part of the system that might see the biggest change is the underlying actuarial data. Teams are slowly getting better at keeping their players healthy and players are getting generally healthier. The numbers might not look like it, but that's because we've essentially seen the end of the career-ending injury. Players that were once losing years of their careers, sometimes from a very young age, are now losing months. Terrible knee injuries are fixed by routine reconstructions. Blown elbows are now fixed with routine Tommy John surgery. Shoulders remain a bit of a mystery, but surgeons are having better success, and people like Dr. Ben Kibler are starting to find the root causes.
This is the sixth year of the team health reports, and if time flies when you're having fun, I must be having a ton of fun. My time chronicling the changes the game of baseball has seen on the medical side has coordinated with one of the most exciting times for the profession since Dr. Frank Jobe looked inside of Tommy John's damaged elbow. I can't imagine the next six or sixteen years will see the pace of change slow, which is good, because we're seeing better baseball because of it.
Beyond the THRs, you'll also see some changes in Under The Knife. The column has existed in essentially its current form since 2002, and it's time to freshen it up a bit. The biggest change you'll see is on Friday. Instead of the normal UTK, we'll start the UTK Wrap. It will be both a week-in-review column, an injury of the week focus piece on the most important injury in the game, and we'll use some new data that I've been working on, including the first public use of the Injury Database. I'm proud to say that the UTK Wrap will be featured at SI.com as a part of our continuing partnership with our friends at Sports Illustrated. You'll see that starting tomorrow as I take a look around the league at what rehabbing players do in the offseason.
One of the most noticeable changes will be the use of "Injury Cost," a metric that's not perfect, but will help to show the true costs of injuries. By using Nate Silver's MORP as a daily value, we'll calculate the lost MORP pushed out over the time a player is expected to be out. It's not a complex formula--MORP divided by 180 times days expected out. There are a couple of tricks here, though. First, I'll be estimating the time each player is expected to be out from the time I first report on them. This is a fluid number, so you're likely to see the IC change over the course of the injury's duration. For instance, an injury that is thought to put a player out for a week can often lead to a 15-day DL stint. That instantly increases the IC, though the injury doesn't actually change.
So all in all, it's an evolution, not a revolution. The UTK you've come to rely upon is still here and will still be focused on getting you the best information possible, whether you're a fantasy owner, or just a fan who wants to know how long his team's #2 starter will be sidelined. I'll keep calling my sources and advisors to make sure that you get that information on a near-daily basis. I'm always open to suggestions and as always, I'll always respond to your emails. It's just a bit until pitchers and catchers report to Spring Training, but I'm reporting early.