Article by Neil Seeman
May 22, 2003

The Canadian Statistical Assessment Service

CANSTATS

Body Checking at the CBC

By Neil Seeman

If you haven’t heard of Dr. Montelpare, you’re not a hockey Dad or Mum. Dr. Montelpare sits at the eye of a storm over body checking -- when the hip or shoulder is used to separate an opponent from the puck. His 1998-2001 research on atom-aged players (nine- and 10-year-olds) for the Ontario Hockey Federation and the Canadian Hockey Association found no statistically significant difference in injury rates between groups that allowed body checking and those that did not.

"It appears that body checking can be introduced at the atom-age level without significant deleterious effects," the study concluded. "Checking is a skill within the game of ice hockey that can be taught in such a way that it does not become a negative influence."

What the CBC Said

•  It’s counterintuitive. Disclosure felt the research "didn’t seem quite right." So CBC interviewed Dr. Montelpare and questioned his data. This is the program that seeks to "discover the truth, uncover the facts and expose hidden agendas."
   
• Under the CBC’s questioning, Dr. Montelpare admitted that some of his data were incomplete (the research was ongoing). Female and house league players (who are prohibited from body checking) had been jumbled in the data set. Rejigging the numbers, it appeared that injuries in the kids exposed to body checking were four times higher.
   
• The CBC has been hyping the story ever since. The Canadian Hockey Association’s new ruling -- it has banned body checking for players under 11 for the time being -- received top billing on the CBC’s National news, beating out coverage of a bloody string of weekend suicide bombings in Israel. In various reports, the CBC now says that it has "discredited" Dr. Montelpare’s work.

What the CBC did not Say

• What the CBC has not said is that independent investigation has confirmed the lack of a statistically significant increase in the rate of injuries at the atom level. Dr. Montelpare’s substantive conclusions were based on data involving insurance rate claims. The anomaly unearthed by Disclosure related to self-reported data, which are unreliable. Even before the numbers were recalculated, the self-reported data revealed the same higher rate of injury in the body-checking group.
   
• There are problems with self-reported data. One person’s interpretation of whether somebody received a body check, and was injured as a result, is different than another’s. There’s also a difference between incidental body contact and a body check. Disclosure showed lots of pictures of children falling to the ice. But most get up again, uninjured. Trainers in leagues where body checking is allowed will think that incidental body contact has led to more injuries. But only independent variables (like insurance claims or medical reports) can verify whether or not an injury occurred.
   
• None of this is to suggest that the study gave a green light to body checking. Nor did it promote body checking per se. It was a pilot study, and used a highly selective cohort of amateur players. More detailed research is forthcoming, and the final, complete study has yet to appear in final form. But the CBC has pilloried the research, offering its own re-analysis as definitive proof that body checking should be banned among youngsters.

Other Considerations

• It’s not a black-and-white issue. It’s even possible that lowering the age at which children can body check will reduce injuries. Teaching children to give and take body checks at a younger age might better prepare them for when contact is more integral to the game.
   
• It’s also potentially dangerous to suddenly hand kids the body-checking tool right after puberty. If kids are able to start checking earlier, it wouldn’t be such a big deal down the road. On the other hand, there is evidence that hockey injuries increase as children age and enter divisions where contact is more prevalent. Whether body checking or some other independent variable is responsible is unclear.
   
• Whatever risks may ensue from body checking, far more will ensue from bad coaching. It’s tricky to adjust the statistics for the quality of coaching, just as it’s tricky to externally validate self-reported injury data. One must also correct for differential skill development.

What CANSTATS Says

• The CBC is simply not playing fair. Disclosure depicts a young atom-aged player standing in his dressing room after a game. Off camera you hear the boy’s father ask what’s wrong. "What do you think?" the boy asks, trembling. "I’m almost in pieces." At another point in the program, a young boy slams into the boards, while an anguished parent’s voice cries out: "Get up Michael. Get up."
   
• This kind of fear-mongering is unwarranted. It’s no wonder registration in minor league hockey has been in a tailspin, notwithstanding the success of the men’s and women’s Olympic teams. Gotcha journalism deserves a bone-crushing hip check.