October 24 2014 09:21AM
Normal distributions of team PDO over nine seasons at each score situation
The idea that PDO converges to a value of 1.000 (or 100.0 or 1000 depending where you want your decimal point) over long periods of time is a useful one for hockey analysts. However, it can also provide an additional lens on performance over short periods of time. Knowing that the sum of shooting percentage and save percentage will eventually converge to 1 can indicate whether the short term results are indicative of what we can expect over larger samples.
You can call it a measure of luck or randomness or chance or bounces or whatever. The point is that they all even out eventually.
Or do they?
Nation World HQ
October 24 2014 07:00AM
The Canucks adjusted plus/minus, T.J. Brodie signs on the cheap, Oilers are streaking, Yakupov, Draisaitl, Richard Peddie, the Twitter crowd and more in this weeks Roundup.
October 23 2014 01:57AM
Folks, eight games are in the books for the Calgary Flames, which means we're approximately 10% of the way through the regular-season schedule. In addition to realizing how quickly the season actually flies by, it's a chance to take a quick (and very preliminary) glance at the analytics behind the Flames successes and struggles thus far.
And remember the oft-repeated mantra of the analytics community: "sample size!" It's just 8 games - 6 of which were on the road. Take these numbers for a small sample of a long, long season and with the appropriate shaker-full of salt.
Percentage stats are all even-strength, except for zone entries (which is more of a tendencies measure than a possession stat). Only players who have played 5 or more games have been included here, so this excludes defenseman Raphael Diaz and forwards Josh Jooris, David Jones, Brian McGrattan and Devin Setoguchi.
October 20 2014 12:45PM
Alex Edler: actually pretty good at preventing quality shots
Most people in the stats community dislike plus-minus, but my feelings are much stronger than that. I loathe plus-minus. It’s personal. The draft for my keeper league hockey pool was about to start and I was considering taking a flyer on a young Erik Karlsson. He was coming off his 2nd season in the NHL, where he posted an impressive 45 points. However, at minus 30 I figured he must be in for a significant drop in playing time considering he was obviously such a huge defensive liability to his team. Karlsson ended up finding himself on the roster of my arch nemesis, scored an insane 78 points, and hilarity ensued. I hate plus-minus.
In theory, plus-minus should be an immensely useful stat. The team who scores the most goals in a game wins, so by extension it seems to make sense that better defensive players would be on the ice for more goals for than goals against. Unfortunately, because this statistic is so greatly affected by goaltending and team’s on-ice shooting percentage, amongst other factors, the usefulness of plus-minus is negligible and it has been largely abandoned by the stats community.
This got me thinking: while we know plus-minus itself is a useless statistic, is there a way we could more accurately calculate the impact a player has on their terms scoring differential? Or put more simply, can we create a better plus-minus? Read past the jump to find out.
Nation World HQ
October 20 2014 11:45AM
The Nation Network brings you the Nation Minute, a rapid fire look at stats and news around the network or NHL. On tap this week: The Canucks offense, Jonas Hiller, Ben Scrivens, Mark Fayne and more.