Last week I established that one of the things raising James Reimer’s numbers to an elite level since last-year’s lockout is his sky-high save percentage on rebounds. As the Jonathan Bernier-Reimer competition emerged this season, an early talking point became Bernier’s superior rebound control. The eye test overwhelmingly identifies Bernier as better controlling the puck, as his hands are more active, allowing his stick to direct pucks to the corners instead of using his pads. He also corrals and catches pucks that Reimer sometimes cannot contain.
Considering how much emphasis is placed on rebound control by goalie coaches, it is an easy target when contrasting Reimer and Bernier. The conventional wisdom suggests that if a goaltender can place rebounds in the corners or hold them for an offensive-zone faceoff, he has increased his team’s chance for defensive success.
It is a logical assumption, but what if it isn’t that important?
One of my initial beliefs was based around rebound control being crucial to a goalie’s success, so I made sure when I began this study that I tracked the location point on shot origins as well as the placement location of any rebounds kept in play. Any shot that was frozen by the goaltender or directed over the glass was designated a non-rebound.
Below I have compiled the average NHL rebound results from my current data sets and charted them on a heat map. The map on the left indicates the rebound placement an average NHL goaltender produces. The map on the right indicates the amount of rebound shots a goaltender faces in an average game. An average NHL goaltender produces a rebound on 61 percent of the shots he faces, 27 percent of which end up in the dangerous, home-plate area. Of those, only 17 percent result in a shot on net.
Looking back at the Shot Quality Project introduction, we can confirm that the home plate is a danger area for uncontrolled rebounds.
With this in mind I isolated the home-plate area in regards to the Maple Leafs goaltenders. The results confirmed the wisdom of the eye test. Jonathan Bernier produces fewer rebounds, controls them and places them outside of the most dangerous area more efficiently than Reimer.
The only question that remained was whether or not that’s a significant factor. I projected both of their percentages on to a 1,000-shot sample to see if the results would lead to a substantial goal differential. Reimer’s numbers are in line with the league average, but Bernier, in his small sample, has placed himself in the upper echelon in regards to rebound control.
To create an expected goal total, I took the slot rebounds and used the league average return rate of 17 percent to determine how many would result in shots. Although 41 extra slot rebounds looks significant, it only results in 16 more high-quality opportunities against Reimer. Considering that his biggest strength the past 37 games has been rebound save-percentage, the impact of these extra shots is negligible. Even when we use the league average save percentage of .760 we are talking about a difference of three goals over 1,000 shots—a .002 impact on overall save percentage.
At this point the data seems to indicate that rebound control is statistically insignificant when it comes to the Reimer-Bernier duel. If you are a member of Team Bernier, my condolences, you will have to find a new talking point. At this point any defense of rebound control I make would rely on anecdotal evidence, but fortunately there are more layers of this data onion to peel before I come to the ultimate conclusion that rebound control is insignificant.