Last week, Zig Ziegler became a household name in Portland. In a post on his blog, Zigsports, Ziegler related how he was hired by Trail Blazers assistant GM Tom Penn to evaluate the movement of Greg Oden and later other Portland players to assess their health and risk of injury. Ziegler found that Oden, Joel Przybilla and Brandon Roy all had issues, and ultimately all three suffered major knee injuries.
When I first read Ziegler’s account of his prophetic predictions to the Blazers, something struck me as off before Benjamin Golliverrelated his sketchy backstory on Blazersedge: They were too accurate. When you project the future for a living, the first thing you learn is that nobody is right all the time. That’s not the goal. The idea is to be enough better than conventional wisdom to provide value. If Ziegler had said he found four players at risk of injury, and three of them ended up getting hurt, I might have found his statements more credible. But three for three? That strains the bounds of predictability.
Yet it’s clear that Ziegler’s account of his involvement with the Portland front office has struck a chord with Blazers fans. For one thing, it reinforces growing skepticism about the Blazers’ medical staff. More than that, though, I think Ziegler’s core belief that catastrophic injuries are often due to overcompensation for other, more minor maladies speaks to our natural human desire to find explanations for outcomes, something I wrote about in my discussion of Thinking, Fast and Slow. When a player does get hurt, it’s compelling to believe that they were destined to suffer the injury.
In part, I think this stems from our difficulty thinking of injuries probabilistically. Instead, people are more likely to think of them in binary opposition: Either a player is hurt, or he’s healthy; either he’s prone to injury or he’s not. This simply isn’t the way sports work. It’s not the way life works.
We see this in the way teams evaluate players medically. Players “fail” physicals, which is kind of odd, when you think about it. The bar between passing and failing is set subjectively. Draft prospects are “red flagged,” which is an indication that they should be taken off draft boards entirely, but the possibility of injury should really be incorporated into the player’s overall risk assessment alongside the chances of them failing to contribute on the floor.
So there is signal here, but as I noted in today’s column quantifying injuries, it’s buried under a lot of noise. That’s especially true when we try to evaluate athletic training staffs based on how often their players get injured. To the extent training staffs can actually make a preventative difference, we’re talking a small shift in the percentages. Over a single season, that’s overwhelmed by the random nature of injuries. If you were looking to predict a team’s games lost by using the numbers of games lost the previous season, 5/6 of the projection would be made up by league average and only 1/6 would reflect the observed differences between teams.
There’s also the question of whether these results tell us anything about teams’ training staffs, or simply reflect the players on hand. The Golden State Warriors are an interesting test case. After losing more games to injury than any team in the league in 2009-10, they revamped their training staff last summer, bringing in Chad Bergman from San Antonio to serve as head athletic trainer. Yet the Warriors still rank below average in games lost this year, and 25th in WARP lost largely because Stephen Curry continues to battle ankle injuries. Was the issue with the training staff or the players themselves?
I’d compare using these numbers to assess the work of medical staffs to trying to judge players’ ability to shoot free throws from two attempts. Odds are, the players who make both are better than those that miss them, but the sample is too small to make any strong statements about them. It’s only over more time that the noise evens out and the underlying signal filters through. Players shoot enough free throws to get to that point. I’m not sure there is enough time for injury data to build up to get there at all but the extremes.
The Blazers might qualify as an extreme. Over the course of the three years, no team has lost more games or WARP to injury. Portland was 29th in games lost in 2009-10 and 28th in 2010-11. There are some underlying explanations here. In the case of Roy, GM Kevin Pritchard and the rest of the Blazers front office knowingly took a risk, and were rewarded handsomely for it before Roy’s knees broke down. Because he has missed so much time, Oden’s injuries also skew the numbers. Yet what stands out as much is the sheer number of different players who have suffered major injuries, including guard Elliot Williams undergoing season-ending surgeries each of his two years in Portland.