2016 Disabled List Information and the Future of Injury Analysis

The Dodgers lost a major league-record 2,418 man days to disabled list in 2016. (via Keith Allison)

The Dodgers lost a major league-record 2,418 man days to the disabled list in 2016. (via Keith Allison)

It’s that time of year when I get around to slogging through the past season’s disabled list data and investigate trends. The big change this season is that I don’t have any bonus coverage. After examining injury information for seven seasons, I have hit a road block with new topics. Instead of rehashing studies already done, I am going to suggest some ways injury analysis can expand.

2016

To put in simply, 2016 was the season to break all disabled list highs. Here are the major league totals for days missed and DL trips.

The new high for DL trips (571) broke the old record by 17. The number is a three percent increase over the previous high in 2013.

The real eye-opener is with the days lost. The total (36893) broke the previous high by over 6,000 days (21% increase). This total averages out to 200 extra days per team. In 2016, the days lost per trip got over 60 for the first time ever (average of 2002 to 2015 was 54.7).

As you can see the average days lost per trip hovered around the 55-day mark until this season. The jump may be attributed to Tommy John surgeries taking 14 months for a pitcher to return versus the historic 12 months. Otherwise, nothing yet sticks out as being abnormal.

Now onto the individual team totals. First, here are the one, three, five, and 15-year averages for each team.

While the White Sox still dominate the 15-year total, three different teams lead the one (Astros), three (Indians) and five (Twins) year time frames.

The most interesting piece of information is that the Dodgers, Athletics and Braves each broke the all-time single-season high in days missed, a record previously held by the 2014 Rangers. The Dodgers didn’t mess around and set the all-time high with over 2,400 days lost.

MOST DAYS LOST TO DL IN A SEASON
Team Season Days Lost
Dodgers 2016 2418
Athletics 2016 2308
Braves 2016 2205
Rangers 2014 2116
Diamondbacks 2004 2017
Padres 2012 1973
Royals 2007 1857
Rangers 2015 1846
Rangers 2004 1811
Mets 2015 1790

The Dodgers also tied the 2012 Red Sox with 33 DL trips. The 2016 Braves and Athletics also make the top 10.

MOST DL TRIPS IN A SEASON
Team Season DL Trips
Dodgers 2016 33
Red Sox 2012 33
Twins 2011 31
Braves 2016 30
Rangers 2004 30
Nationals 2008 30
Red Sox 2010 30
Yankees 2013 29
Padres 2002 28
Athletics 2016 27

Now, one more all-time high set by the Dodgers. Here are the DL days and trips by team divided up by pitchers and hitters.

Here are the all-times from the two graphs:

  • Pitchers (total days): Dodgers (1st), Braves (2nd), Athletics (3rd), Rockies (9th)
  • Hitters (total days): Athletics (5th)
  • Pitchers (total trips): Dodgers (1st), Braves (3rd), Rockies (6th), Athletics (6th)
  • Hitters (total trips): none

This season, it was the pitchers who missed the most days by far. Here is a comparison of 2016 days missed by pitchers to the overall total a few seasons back.

The number of total days lost to the DL in 2006 was 23,833. The days lost only to pitchers this past year was 23,121. If the pitcher-related increase in DL days continues, at some point in the near future, more days will be lost to just pitchers than to the combined total of hitters and pitchers.

Finally, here is a customizable table showing the number of total days per team lost to the DL over time (thanks to Bill Petti for updating the information).

Now that the new yearly data have been examined, here are some ways injury information can be expanded to help with future research.

1. More and better information

The biggest issue with the injury information is the lack of it. When I started examining this stuff years back, I was amazed there was none. Even though everything was clearly available on MLB.com’s website, not one person was tracking and processing it. Baseball stats are available going back to before 1900, but current injury information might as well have been nonexistent.

I was first able to have a working injury database when FanGraphs paid me to create one from 2002 to 2009. Since 2010, I have volunteered my time every year to go through the MLB’s suspect transaction data. With this information, I have been able to get the preceding information, but the disabled list is a limited resource and mainly is used as a roster management tool.

The first area in which more information could be obtained is collecting older disabled list information. The information dating back to 2001 is available at MLB.com and even more information going back to 1964 is available at Pro Sports Transactions. More information, if used correctly, will lead to richer information. There is no simple way to scrape the data. I tried. The biggest issue is matching up the days on which a player enters and leaves the DL, especially the further back a person goes. Some of the transactions are missing or doubled up or the wording is inconsistent. The data are a mess and it takes time to comb through by hand. From experience, it takes about three to four watching-paint-dry boring hours to complete. If a few brave souls sacrificed one evening, this project could be done quickly, but since it hasn’t been done in 15 years, I expect it never will.

Another area in which the dataset can be expanded is more detailed information. This past year, three instances existed where an injury was listed just as a back injury while, on the other hand, Kevan Smith had sacroiliac joint dysfunction (i.e. a back injury). The first bit of information is completely useless and the second gives a detailed account of the information. Most of the time the information is stated as a side then a location then an injury such as “left forearm strain.” The information on the extent and how the forearm was strained is missing. The extent (grade) of the strain isn’t stated in the transaction, but the data may be available from media reports.

The increase in details could help to get a better picture of an injury once it happens to another player.

On the same front, it would be interesting to have a repository of initially stated return times to get an idea of how much teams oversell or undersell the extent of an injury.

I feel I’m where researchers were with Pitchf/x a couple of seasons back. While the information we were receiving was still useful, it had been extensively studied. Better batted ball information was available and teams were using it but researchers in the general public were left in the cold until StatCast began releasing data two seasons ago. With MLB’s new injury database, people will now be left to speculate on how the information is being used. I understand the players don’t want their detailed medical information to go public, but the data can be censored while still allowing increased understanding of information like return times and various trends. I expect some of the final information to eventually get into the public’s hands, but the process will likely be slow and limited.

This is by far the most important obstacle to understanding how injuries affect the game. I am sure there are a few ideas I’ve missed and will touch on but not much more is to be learned with available data.

2. Trends vs. noise

With the amount of disabled list-related injury data currently created, we are finally getting to the point where we can begin to find out if a change is noise or a yearly trend. The past season, Tommy John surgeries were way down from the previous season. Instead of guessing if this drop is from normal variation or systematic change, we can now look back to see how often these drops occur. With basic statistics, we can look back over 100 years to find various changes. Right now, we are just getting into that frontier with injuries analysis.

3. Predictive vs. Reactive

The real unexplored frontier in injury analysis starts with getting away from reactive injury analysis and becoming more predictive with it. A big step forward happened this season with the addition of spin rate to the public Statcast data. We were able to notice changes in velocity and spin for pitchers and see which ones were changing. This type of analysis help to pinpoint Wade Davis’ changing fastball and the eventual DL trip.

Additionally, we can start to determine if a hitter isn’t hitting with as much power and therefore an injury may be nagging him. The work is just starting in this frontier and the results are a little slow in coming. It will probably take a few seasons to finish optimizing and back checking but having an idea if a slump is a from bad luck or a drop is production is the next step in our understanding of the game.

4. The New 10-Day Disabled List

Finally, the new minimum disabled list time frame of 10 days will mark a definite point in change for data collecting. When I examine the 2017 data next year, I expect there to be all-time highs in days missed and trips to the DL, led by pitchers. The 10-day DL is a perfect vehicle to bring up an extra bullpen pitcher whenever a starting pitcher will miss a start. With the team having no plans to use the fifth starter anyway, he can go on the DL and then come back to make his next start. Even with all-time high numbers posted this past year for time lost and some regression expected, I expect all the all-time highs to fall by the end of next season.

Conclusion

While I feel injury analysis just began, I also feel I have already hit a roadblock without any possible new and better information. I am not going to hold my breath on anything coming out anytime soon. Please let me know if you have ideas for future analysis. I will dabble in a study here and there but don’t expect any major breakthroughs without the amount of available data increasing.


Jeff, one of the authors of the fantasy baseball guide,The Process, writes for RotoGraphs, The Hardball Times, Rotowire, Baseball America, and BaseballHQ. He has been nominated for two SABR Analytics Research Award for Contemporary Analysis and won it in 2013 in tandem with Bill Petti. He has won four FSWA Awards including on for his Mining the News series. He's won Tout Wars three times, LABR twice, and got his first NFBC Main Event win in 2021. Follow him on Twitter @jeffwzimmerman.
14 Comments
Oldest
Newest Most Voted
Inline Feedbacks
View all comments
evo34
7 years ago

How much are concussions driving this?

David
7 years ago

I find the bar graph split by pitchers and hitters difficult to gauge the hitters, when they’re not all starting from the same vertical line.

Perhaps if you could weight the disabled days by projected WAR (then divide by 182 = days in a season?) as an estimate of the impact on W-L for the season?

Definitely more break-outs on type of injury, especially for pitchers.

But in my line of work, I certainly feel your pain of everyone suggesting grandiose ideas of how to improve analysis, when they don’t realize the tedium of sifting through the data, limitations of data, millions of micro-decisions to be made, etc. Keep up the good work!

obsessivegiantscompulsive
7 years ago

OT: Enjoyed the article, but wanted to see if anyone knows of any good study of TJS survivors of two TJS surgeries, or more specifically, how long has players been able to go after their first TJS until their second TJS?

Jonathan Davis
7 years ago

I appreciate that you started the conversation around Reactive v Predictive but I don’t think that just looking at velocity and spin rate on StatCast is touching the surface. In fact, that’s still reactive.

Are you able to comment on what Teams are doing to actually predict injury or look at the likelihood of injury for certain players based on arm angle or stride length or height for pitchers? Or any number of behaviors for hitters? I’d love to hear what they are doing.

Greg Pope
7 years ago

If you’re looking for people to help with combing through data, head on over to Baseball Think Factory where there’s already a link to this article (http://www.baseballthinkfactory.org/newsstand/discussion/2016_disabled_list_information_and_the_future_of_injury_analysis_the_hardba). I’d bet that you’ll get plenty of volunteers. Especially if there’s no specific deadline for it.

MCT
7 years ago

If you are eventually able to track information back historically, I think you’ll find that the DL was used less and less the further back you go, for various reasons. Some of them:

–Changes in what I’ll call “baseball culture” (e.g., players wanting to tough it out and not admit they were hurt, players being expected to play hurt, both teams and players having less to lose financially from players playing hurt)

–Changes over time in the way the DL has worked (at times in the past, there have been limits on the number of players who could be on the DL, different minimum stays, etc.)

–Changes to various other transaction rules that, at least under certain circumstances, made it less simple to call up a replacement from the minors to replace an injured player than it is today

–Changes in roster management (historically, when teams didn’t use as many different pitchers as today and benches were deeper, it may not have really been seen as necessary to bring in a short-term replacement for an injured player)

–Changes in medical science and contractual arrangements casing injuries that may have been career-ending in earlier times (prompting the player to quickly retire or be released) now resulting in a lengthy stay on the DL

I suspect you would only have to go back as far as the ’70s, maybe even just to the ’80s, to start finding teams that went an entire season without using the DL at all.

Cliff Blau’s web site has a history of the disabled list; this was obviously written before the introduction of the 7-day DL for concussions in 2011:

http://cliffblau.ucoz.com/dl.htm

Based on that, except for the creation of the 7-day DL, the basic rule structure that was in place up through 2016 was created in 1990.

Gary McCoy
7 years ago

One factor is neglected in the report: Roster Average Age. Look at a teams average roster age, and with further granularity- previous injury. This is the biggest predictor of injury. Great summary though. In Asia- the trend is soft tissue injury reduction. Open your eyes MLB- the answers lay outside of your scope of current thinking.

Jonathan Davis
7 years ago
Reply to  Gary McCoy

Can you comment more on this soft tissue injury reduction Gary? I’m intrigued.

Wilma
7 years ago

This is way more helpful than anyihtng else I’ve looked at.

sofortkredit selbststaendig ohne schufa sofort löschen
7 years ago

, November 14, 2012 at 12:22 pm Hello…..Thanks for sharing superb informations. Your web-site is so cool. I am impressed by the details that you’ve on this blog. It reveals how nicely you understand this subject. Bookmarked this website page, will come back for more articles. You, my pal, R…

http://www./
7 years ago

Hi Shane, If it’s on a good site, It will help your SEO. Whether it helps your traffic as much depends on a variety of factors. If it’s an old post that still gets comments, it’s probably getting traffic. And some of that traffic will make it back to your site. But if it’s a post that isn’t getting traffic, there won’t be a lot of traffic coming through to your site.

http://www.shopmiscere.com/
6 years ago

Coolest part about that commercial is that almost none of it is computer generated. They actually snuck him onto a little rail car to slide him off the boat onto the horse.

site web
6 years ago

Super site! I am Loving it!! Will return once more, Im taking your food additionally, Thanks.

Mark Vellutini
6 years ago

Why in the world isn’t the GM held more accountable, in one fashion or another, for excessive use and time lost on the DL. The GM has to sign off on its utilization, the GM hires the strength and conditioning coaches, trainers etc. I’m entirely aware that a savvy GM can use the DL to a Clubs advantage. But, when the Oakland A’s are continually in the top 5 for DL visits and time lost, in addition to losing nearly100 games and finishing at least 20 games out of first with 0 reason to believe that it will soon get better, why on Earth do they keep Billy Beane. They managed to somehow sever the relationship with Lou Wolff, maybe the worst Managing Gen Partner in modern history along with being a huge proponent of Beane. Why on Earth there isn’t more outcry by the Oaklands A’sfan base, is beyond me. I wish that Oakland A’s fans had 10 % of the passion that the Raider fans have, Billy Beane would be working for an entirely different organization, maybe league. The Commissioner should have investigated every single transaction that occurred in the last 3 yrs. under the watch of Beane and Wolff. The list of players that have gone through the doors of the Oakland A’s Clubhouse in the last 5 years could easily put a World Series LineUp together. They made a Wild Card appearance and have virtually nothing in return for all the trades that were made. Including trading Donaldson with 2 years remaining in the A’s possession for virtually nothing in return. And in the middle of an MVP Year. All because Donaldson asked him why he was fleecing the Club. I sure wish that Josh would have hit him with a left hook, instead.