Breaking Down Existing Power Play Data By Shot Type

One of the things I was curious about at the outset of this project was the shot type distribution over the years. I wondered whether power plays have trended away from slap shots and towards wrist shots in particular, as teams have gone away from the overload formation with two defenders at the point, and towards more aggressive setups. So I pulled the data to take a look.

What is difficult with data like this is to separate trends from recording bias. It seems as though wrist shots have slightly increased at the expense of slap shots, and also that we have seen slightly more deflected shots, but we don’t know for sure that this isn’t the result of scorers adjusting their definitions of each kind of shot, for example.

The more interesting and actionable question, therefore, is has the success rate of each type of shot changed over the years?

If we ignore data from prior to 2008 — where it appeared as though there may have been some issues in recording accuracy —  it’s interesting to note some of these movements. One would expect that, considering the decline in scoring, high frequency shot types would see a slight declining trend. That is the case with wrist shot attempts, which last season hit the twine 7.08 percent of the time. For slap shots, however, the success rate has increased. In 2015, players scored on 3.93 percent of slap shots, the highest rate in a full season since 2008. While more variable, we have seen the same trend in snap shots. Perhaps there could be an inefficiency on wind-ups.

Tip-ins and deflections, obviously, are high percentage plays, though what we don’t know from play-by-play data are how many wrist shots were taken hoping for a tip that were unsuccessful. Backhands likely hit at a high rate because they are almost exclusively taken close to the goal line.

Neither of these graphs are particularly actionable, but these rates are something to keep in mind when you watch player shot selection and evaluate decision-making.