Progression to the mean is fun.

Brad Richards is struggling this season. Everyone has seen it, everyone has complained about it, and everyone is waiting for him to rebound. What generally goes unnoticed is that Ricahrds’ shooting percentage is an astonishing low 4.5%. That’s less than most defensemen. Richards has a career average of about 9%, so his shooting percentage right now is half of what it should be.

We spoke about shooting percentages at the end of January and highlighted Carl Hagelin (at that point goalless) and Taylor Pyatt (at that point shooting at 43%). Both have since progressed and regressed to their career averages, and the same theory is going to apply to Richards here. His career worst shooting percentage was in 2002-2003 when he shot at 6%, which is still above his current pace.

Richards has had a career long enough that we can assume progression back to his mean of 9%. Since we are roughly 42% through the season, that means Richards is due for a hot streak of shooting around 13%, which would get him back up to 9% over time. It’s not an exact science, as there’s some flexibility between his career average and what he could finish with this season.

Looking at his numbers for this season, Richards has 2 goals and 44 SOG. To progress back to the 9% career average (assuming the same ratio of shots per game), Richards will take about 60 more SOG this season. If he finishes at the 13% clip mentioned above, he will score about eight more goals this season. That puts Richards at 10 goals on the season, which is 17 for a full 82 game season.

Richards certainly is struggling, but he won’t continue shooting at 4.5%, which seems to be the bigger issue with his offensive numbers. Regression and progression to the mean is something we discussed with Hagelin and Pyatt, and we have seen Hagelin rebound and Pyatt cool down. There’s no reason to believe the same won’t happen to Richards. He’s too good to sit with two goals.

*Related*

“Since we are roughly 42% through the season, that means Richards is due for a hot streak of shooting around 13%, which would get him back up to 9% over time.”

Technically not true. If his true talent is 9%, we would expect him to shoot about 9% for the rest of the season, not %13 to end at 9%. Regression means that they return to their true talent level, not play above it for a while to get to the same numbers at the end of the season.

What your saying is technically true Mike, but I think Dave was trying to illustrate the peaks and valleys of the statistical curve. Especially with the marathon of an NHL season, mean divination will be fairly substantial. In this shortened season he may stay well short of his career average, but over an extended period, there would need to be peaks of 13% or more to progress back to that mean.

Yup. What Justin said.

doesnt this sometimes happen to a player when he is given a star player to play with?

I see some of this comming from him or players like him trying to hard to pass the puck to players like NASH. When in fact players like him should become evenmore dangerous if they just shot the puck more. And at the same time they would make that star player even more dangerous. Because he wouldnt be drawing all of the coverage.

I agree. Nash’s presence on the ice shouldn’t always lead players to try to get the puck to him. Sometimes, his mere presence should open up ice for other players to get quality shots as oppposing players cheat over to Nash’s side. Richards and Gaborik are both guilty of giving up shooting opportunities in an attempt to feed Nash the puck. Doesn’t always work.

This could be a factor, but right now I’m focusing on Richards shots only. He is taking 2 shots per game and converting at 4.5%, which is well below his average. That won’t last.

I agree with Dave on this one. This post didn’t seem to be about the amount of shots he’s taking, but more about the low shooting percentage and an expected progression to the mean.

Absolutely, giving up shooting opportunities doesn’t relate necessarily to lower shot percentage. This was more of a corollary observation about Richards not shooting enough, in addition to his lower shooting percentage numbers.

Surely one has a better chance of scooring when he shoots the puck tho NO???

Ricky,

One could surmise that there would be more goals with more shots, but there is no relation between number of shots and shooting percentages. We’d like Richards to shoot more, and with will come more goals. But, regression/progression to the mean isn’t really related to number of shots, it’s related to a player coming back to his natural numbers.