One of the first things that was offered to us from the NHLís real-time scoring system was shot distance. No longer were we forced to contend with gross shot totals. Now, we could finally determine the exact distance from which players took their shots, and consequently the level of danger of those shots. With this development, the notion of Shot Quality entered the hockey analysis world.
The trailblazers of the Shot Quality world were Alan Ryder and Ken Krzywicki, who first established the odds of any shot going in from a particular distance. Aggregating this, they established the total number of expected goals and therefore created a better tool for judging shooting percentage for forwards and, especially, save percentage for goaltenders.
The two biggest factors that influence Shot Quality are the distance at which the shot is taken and the game state. At 5-on-5, the most dangerous shots, taken from less than 10 feet away from the net, have about a 20% chance of going in, while the least threatening shots, which are taken from more than 60 feet away, will go in 2.2% of the time, making them 9 times less dangerous. Shots on the power-play are 1.6 times more likely to go in than 5-on-5 shots (12.8% last year vs. 7.9%), but this is not because they are taken closer to the net; last season, the average distance for a shot during 5-on-5 play was 35 feet, while on the power-play it was 34 feet. However, at all distances, power-play shots were more likely to go in, especially those from at least 30 feet or whose success rate nearly doubled.
Below is a graph listing the average shooting percentage of shots as a function of distance, depending on whether play was 5-on-5, 4-on-4 (as it is in overtime) or 5-on-4. I considered the sample size for 5-on-3 play to be too small to be included here, however, suffice it to say that they are yet again more dangerous. While the average distance only marginally decreased to 32 feet, the average success rate was a whopping 20.6%, and the success rate even for shots taken from 40 feet out and further was roughly 15%. 6-on-5 power-plays, when one team pulls its goaltender, are only marginally less dangerous than 5-on-4s, with an average shooting rate of 11.9%.
In 2007, Alan Ryder published a cautionary note on Shot Quality. It turns out that there are significant differences from arena to arena in terms of how scorers judge shot distance. The most notable example is Madison Square Garden in New York, where shot distances are routinely 80% of what they are in the rest of the league. Snipers from all across the NHL, when up against the Rangers, are judged as incompetent, while Henrik Lundqvist is deemed the greatest goaltender in the history of the world. Itís simply not true: Lundqvist is good, but heís not facing shots that are immeasurably tougher than any other goaltender in the league. The scorers just make it seem that way. To remedy this problem, it is important to use a normalizing factor for each teamís home games, based on the typical shot quality observed on the road.
In recent years, with the parity that has overtaken the league, there has been a challenge to the notion of shot quality. While nobody objects to the idea that closer shots are more dangerous than distant shots and more likely to go in, many hockey analysts believe that it all evens out and we can therefore eliminate the added layer of complexity involved in analyzing shot quality. I personally disagree with this notion. Given that different shots at different distances have different success rates, and that the aggregate distances allowed by different teams vary, there is every reason to believe that there might be shot quality differences in the aggregate as well. Last season, the average shot distance for home teams was 34.9 feet, while for road teams it was 36.2 feet. While this may not seem like much, it is enough to justify part of the home ice advantage that home teams enjoy.
More importantly, on a team-by-team basis, shot quality does correlate both with shooting percentage and inversely with save percentage, as it should. On the road last year, the team that allowed the most dangerous shots was (unsurprisingly) the Toronto Maple Leafs, while the team that took the most dangerous shots was (also unsurprisingly) the Pittsburgh Penguins. Both teamsí goals totals (the Leafs Goals Against, the Penguins Goals For) vastly exceeded what the raw shot totals would have predicted. My colleague Gabriel Desjardins has just shown how the Minnesota Wild manage to clamp down on the quality of shots against.
There may be other factors that affect shot quality other than distance and situation. In my next article Iíll delve into one of them.
Tom Awad is an author of Hockey Prospectus.
You can contact Tom by clicking here or click here to see Tom's other articles.