jarred smithson and such

I think we gave up quite early in the search for the worst forward in the National Hockey League. mc79hockey has a post up this week talking about sub-40 players:

If you keep your eye on the Corsi numbers, you notice that Lennart Petrell is posting a spectacularly bad number – his Corsi% is currently 32%. Last year, he was at 39.6%. This got me interested in something: what happens to sub-40% Corsi guys? Looking at seasons of 300 minutes or more, there are 29 of these guys in the BTN era, which started in 2007-08.

17 of these guys are no longer active in the NHL. For eight of them (Jeff Cowan, Jim Dowd, Jeff Giuliano, Byron Ritchie, Scott Thornton, Dan Hinote, Todd Marchant and Jon Sim), a sub 40% Corsi season was the end of the line.

Anyhow, I don’t want to quote the whole post, but Tyler has some interesting facts about what happened to those sub-40 players like Dowd and Hinote, and our favourite, Ryan Johnson, who played a grand total of 48 regular season and playoff games after Don Cherry assured us Johnson was a player that every coach would want on his team.

I had my own little formula, trying to determine the minimum offensive output a player needs to produce to stick around in the NHL. I found that most of the time, players who registered 10 or fewer points in 58 games, and had a plus/minus rating of -3 or less didn’t last in the NHL if they weren’t also fighters. I called it the Ryan Johnson Number.

Ryan Johnson fell under that expectation in 2010, and lasted just 34 games the next season. A year ago, the careers of Troy Bodie, Adam Mair and Todd Marchant were claimed. However the 15 players who were below the minimum expectations in 2011-2012 all seem to be in the NHL with the exception of Brad Winchester and Warren Peters. Even Jay Pandolfo, Matt Ellis and Colton Gillies have managed to stick around despite not bringing much of consequence to their squads. Ryan Carter, who scored 4 goals in 72 games last season, appears to have carved out a niche on a much more used New Jersey fourth line and has already beaten last season’s point production.

However, I’m upset that my model needs to be rethought, not because the players who fell below the Ryan Johnson Minimum Expectation are any good. Oh no, with players like Lennart Petrell and Scott Nichol, I’m quite convinced that the players in this range don’t have an awful lot of value. I don’t think that the minus-3 matters much, because I think plus/minus should only be used in situations where you’re trying to showcase how lousy of a statistic it is. There is no denying that some NHL teams and coaches use plus/minus to evaluate their own players and decide which ones need replacing.

One name on the list from last season is Jerred Smithson, formerly of the Nashville Predators and now with the Florida Panthers. He doesn’t actually fall below Tyler’s 40% rule in any of the seasons he’s played. In fact, Smithson’s puck-possession rate this season is 41.0%. Last year in Florida it was 42.1%. In Nashville it was 45.0%, and back in 2010 it was 48.9%. It seems that Smithson, originally an undrafted Los Angeles King out of the Calgary Hitmen organization, has aged less like cheese and more like yogurt—his career high in points is 16, and he has played 569 NHL games without doing much at the proper end of the hockey rink. Here is his only goal on the season:

Let’s look at Smithson this season. Behindthenet.ca, an excellent resource for statistics, has a number of filters. I wanted to find a list of players who had failed miserably given the easiest of circumstances, so I looked for players with a Corsi Rel QoC (quality of competition) rate of below NHL average, who started more than half their shifts in the offensive zone, and played fewer than 9 minutes a night. The list produced these names:

  • Richard Clune
  • Patrick Bordeleau
  • Chuck Kobasew
  • John Scott
  • George Parros
  • Gregory Campbell
  • Jared Boll
  • Jerred Smithson

Those players are ranked in descending order by Corsi. It’s worth noting that Clune is the highest on the list, with a Corsi ON of minus-3.91 per 60 minutes of play. Even in these easy minutes, not a single one of these players was able to positively influence puck-possession. That’s remarkable to me. It seems that in those easy situations, you could have a rookie on the farm who could give you some time in the offensive zone. No dice to NHL teams, who like to roll with thugs like Scott, Parros, Boll and Clune in the ripest minutes for exploitation.

Jarred Smithson, though, is at the bottom of the list. His minus-19.38 Corsi per 60 minutes is the ugliest number I’ve seen so far in the NHL, especially given how easy the minutes he plays. He doesn’t do too much in Florida. Given the way he’s aged, from the outside it’s almost as if he’s given a roster spot based off nothing but reputation and size. But maybe there’s a specialist role he plays, other than offensive anchor.

I looked up Florida’s time on ice stats and saw that Smithson plays 10:11 of ice-time a night—nearly 2 minutes of time on the penalty kill. I wanted to see how the Panthers did when he was killing penalties. It’s quite shocking, actually: the number of shots, and the number of goals he’s on the ice for per 60 minutes compares very well with the rest of Panthers penalty killers:

NAME Goals against/60 minutes Shots against/60 minutes
Shawn Matthias 0.0 57.1
Tomas Kopecky 17.2 55.6
Marcel Goc 9.3 52.8
Jerred Smithson 3.7 45.7
Jack Skille 4.5 31.3

(These numbers were last updated March 11—indeed, it is nuts that Matthias, who plays 1:08 shorthanded a game, has not been on the ice for a single goal against the worst PK in the entire NHL. He has been on the ice for a huge number of shots against, however)

His Corsi, though, remains hilariously bad. Here’s how bad he is on offence: Scott Reynolds during the lockout for NHL numbers looked at something called “individual points percentage”. It’s basically attempting to be a measure of offensive involvement, by looking at the number of points a player got per goal he was on the ice. Crosby between 2008-2012, for instance, led the league with 234 points on 280 goals at even strength, or an 83.6% IPP. Malkin, 264 points on 316 goals, also had 83.6% and Jordan Eberle was third at 85 points on 102 goals, or 83.4%.

Generally, forwards were north of 70%. When a goal is scored, typically the forwards are the ones who factor in. Gregory Campbell was 67.4%, and that’s one of the checking guys on the list. The lowest on the list of players who had been on the ice 100 goals for was Jarred Smithson, and it wasn’t even close.

Smithson registered a point on just 51 of 112 on-ice goals at even strength, for an IPP% of 45.5%. The second lowest, Josh Bailey, was at 55.6%. That’s, uh, better than all defencemen, but shows a player hilariously inept at moving the puck forward or getting involved in the offence.

It’s funny, because it’s not like Smithson blocks a whole heck of a lot of shots or prevents a tonne. He’s a remarkably effective penalty killer, but by almost any measurement you can find shows a player who isn’t good at all at even strength, yet continues to find a spot in the NHL year after year after year.

I don’t think he’s the worst overall forward in the NHL, but he’s almost certainly the worst from an offensive standpoint. It’s absurd that the lowest-ranked penalty kill in the NHL would have a specialist deployed like Smithson. Against Boston Thursday night, he played minutes primarily against their fourth line and had a minus-7 Corsi number as the Panthers were out-shot 8-2 when he was on the ice. He played :52 seconds on the penalty-kill, during which the Panthers out-shot the Bruins 1-0. Unreal.