Compare and contrast (because I can’t leave well enough alone)

Let’s consider two players’ even-strength statistics together and separately. The two players are Vincent Lecavalier and Martin St. Louis. (GFON/GAON – goals for/against while on the ice, SFON/SAON – shots for/against while on the ice SAFON/SAAON – shots attempted for and against while on the ice, SH% ON – shooting percentage while on the ice, SV% On – save percentage while on the ice)

Vincent Lecavalier and Martin St. Louis

  • GFON: 30
  • GAON: 19
  • +/- ON: +11
  • SAFON: 497
  • SAAON: 491
  • SFON: 288
  • SAON: 276
  • SH% ON: 10.4%
  • SV% ON: .931%

Collectively, these guys have been heavily outscoring their opposition. Since both of them are tremendous offensive talents, that’s unsurprising. What is surprising is that they really aren’t out-shooting their opponents by much, especially considering how many goals they’re scoring. The reason for this is that while they’re on the ice at even-strength, their opposition goaltender has a .896 SV%, while their goaltender has a .931 SV%.

Vic Ferrari pretty much conclusively proved awhile back that forwards have a major impact on the opposition goaltender’s SV%, but very nearly no impact on their own goaltender’s SV%. Thus, while Lecavalier and St. Louis are likely responsible for scoring so much, the fact that Mike Smith has been exceptional behind them is Mike Smith’s doing.

Right now, this duo is benefiting from great goal tending; otherwise, the number of goals against would be much greater.

Martin St. Louis without Vincent Lecavalier

  • GFON: 5
  • GAON: 4
  • +/- ON: +1
  • SAFON: 102
  • SAAON: 106
  • SFON: 46
  • SAON: 66
  • SH% ON: 10.8%
  • SV% ON: .939%

Aside from modest changes to shooting and save percentage, the numbers look very similar. The biggest difference is that a much higher percentage of shot attempts are getting through, something unlikely to be replicated in a bigger sample (i.e. shots for and against are generally closely related to shot attempts for and against).

Vincent Lecavalier without Martin St. Louis

  • GFON: 5
  • GAON: 6
  • +/- ON: -1
  • SAFON: 94
  • SAAON: 121
  • SFON: 54
  • SAON: 69
  • SH% ON: 9.3%
  • SV% ON: .913%

Fairly significant changes to save and shooting percentage (which can likely be dismissed due to sample size) are present in this sample. Still, the general trend that the Lightning are hardly the dominant team (from a shots or shots attempted perspective) with Lecavalier on the ice continues.

Conclusion

With or without Martin St. Louis, the Lightning are not dominating the play with Vincent Lecavalier on the ice. I don’t understand people who insist he’s a top-ten NHL player. He may be a top-ten offensive talent, but he doesn’t carry the play against tough opponents. In point of fact he never has; the last time Lecavalier’s line seriously out-shot their opposition, Martin St. Louis won the Hart Trophy and Brad Richards’ line was facing the heavies. That was the year where Lecavalier was fourth on his team in playoff scoring and Brad Richards won the Conn Smythe.

He scores lots, and he plays a physical game. The simple fact of the matter though, is that for as much offense as he generates five-on-five, he surrenders a lot of chances the opposite way.

He’s an exceptional talent, albeit one who’s recorded more than a point per game only twice in ten seasons. He’s vastly overrated as a top-ten NHL player.

  • Hippy

    Tampa's defense is awful. They have difficulties gaining puck possession in their own zone and have poor puck movement out. I'd only compare Vinny's Corsi-stats to his other teammates as if they are all in the black and Vinny is negative then you can blame VL. To me, Corsi is very much like +/- … only useful in ranking players on the same team (while using QoC numbers).

    I'm glad we dodged the Brad Richards bullet last season. He was a terrible ES player on the Lightning and is still a terrible ES player on the Stars.

    At least Lecavalier and St. Louis can stay afloat at ES even though they aren't necessarily outshooting their opponents.

  • Hippy

    @ Dropping Deuces:
    I like Hemsky, but over Nash?!? I don't necessarily disagree, but I didn't put Hemmer on there bc I thought it might be looked at as a Homer pick. No doubt in my mind though that Ales will be on most people's top 10 within a year or two. (Is that guy popular right now, or what?)

    Doing a list like this is fun and a great waste of the workday, but it also reminded me of a couple of things:

    Forwards are over-rated in this league. All teams that excel have amazing D men.
    Goalies are also under-rated.
    There are a lot of great players out there.
    You can't win a cup with 22 "Top 10" type players. You need the muckers, pluggers, checkers, and grinders as well.
    Did I already mention it's a great waste of the work day?

  • Hippy

    Ok. At least you admit you have only watched 25 games of Vinnys 10 year career to come to a conclusion that he isnt good defensivly. Yet can score short handed goals. Must have a little defensive awarness.

    I have watched alot of Vinny. Maybe a little biased here because he is my favorite player. He used to lite me up in bantam the same way he can lite dmen up in the NHL.

    All i can say, is watch 40 games and watch him closely. I wouldnt be surprised if he becomes ur favorite player in that time Jonathon. He really is that good.

  • Hippy

    Great article and defense of it in the comments, Johnathan. While the conclusion you've come to has come under fire, the way you run the numbers is refreshing and the arguments you present make sense.

  • Hippy

    "There are three kinds of lies: lies, damned lies, and statistics."
    -Benjamin Disraeli.

    "He's (LeCavlier) vastly overrated as a top-ten NHL player."
    -Johnathan Willis Jan 21, 2009. (Using statistical Analysis)

    Statistical analysis is 100% reliable 0% of the time. This dialogue has been interesting, but very inconclusive. Hockey players are people not robots. Could advanced statistics have predicted or accounted for Marleau's turnaround this season? Do numbers account for girlfriends, nagging injuries, or team chemistry? Putting TOO much faith in numbers is a big mistake.

  • Hippy

    While numbers do tend to show interesting trends, and data can be a useful TOOL, when it replaces all other measuring boards as the measure of choice, there are going to be real problems.
    Also for the "elite players to have to be great defensively" argument, im going to go out on a limb here and say that mario and gretzky were not exactly the first ones back blocking the shots.
    And if you try and say they were not elite, you may get strung up in a tree.

    Another flaw in your argument and reasoning in not including a definition of "elite Player", you just gave examples (crosby, malkin, ovechkin). And if they are the only players you consider "elite" then just come right out and say it.

    There is more to hockey that stats can tell (an echo of Chris's statement) and when flawed predictors are used, and the categories are ill defined, we run into major problems.

  • Hippy

    was vinny ever considered elite before the 06 07 season? Everyone knew he had offensive talent by the ton but he was considered inconsistent. Why does it mqtter that he was fourh in playoff scoring, he only had 60some points that year, what else would you expect. Although he put up good stats they were nothingto write home about. Very similar to hemsky actually, although he has good numbers just the other day Pierre mcguire labelled him inconsistent, and that won't change until he puts up the numbers to prove himself to the rest of the league. That season vinny put up over 100 points and he entered this'elite category. What has he done since then? 90 some points the year after and under a ppg this year. So you could easily come to the conclusion that his elite year was a fluke and that he doesn't deserve the label but the fact remains that he was on the worst team in the league last year and on a terrible team this year combined with injury problems. And as long as we know that, his stats over the past two years will always be taken with a grain of salt as the eye tells us that his 100 of year was no fluke. Due to the quality of his team, he hasn't even had a fair shake to put up the stats that you are looking for, and you can't use stats from earlier on in his career because you're right he probably wasn't elite then, but that's got nothing to do with if he's elite now

  • Hippy

    JW: the world is flat and the quicker you understand that, the less time you'll waste.

    Do something more productive like finding out the PK kill-rates when it comes to killing off aggressive penalties!!

    !!!

  • Hippy

    @ Dennis:
    Right.
    We just don't get it. We're not as smart as you. We're not as insightful as you. We're not willing to consider other ways of assessing players. We don't buy into every conclusion (some of them downright bizarre) you spit out based on numbers and that makes us (and people who have forgotten more about hockey than you'll ever know) ignorant. We think the world is flat. You arrogant jackass.

    Lost in the numbers, alright.

  • Hippy

    holy eff, Im not a longstanding member of the community, so I might be missing out on some of the history here, but it really seems like you guys are giving jw an overly hard time here.

    he made a claim, he supported it with statistics, and that's it. I'd give Lecav a little more credit than he does as well, but I certainly didn't find anything he posted offensive.

    can't we just go back to taking our frustrations out on mactavish, and not each other? good lord.

  • Hippy

    RobinB wrote:

    Somewhere, somehow, somebody could come up with a set of numbers showing that Bobby Orr wasn’t that great and Rocket Richard was overrated.

    Orr was +597 over his 596 GP (all except his rookie year for which I can't find +/- numbers). If anything, the stats guys would like him more than the others given his gaudy +/- numbers.

  • Hippy

    @ speeds:
    Most would, yes. The point is, it's possible to make any argument you want with statistics alone.

    I'll repeat my bottom line, in case you mistakenly think I have no use for advanced statistics.

    Never checking the stats and relying only on what you see to assess a player is flawed and lacks context. Checking only stats and not watching a player perform over a reasonable length of time in making an assessment also lacks context.
    For me, stats are a tool, a secondary way to back-up or disprove what I've seen.

  • Hippy

    What a fun thread

    Jonathan says “He’s vastly overrated as a top-ten NHL player.”

    A couple of guys pile on, they respond:
    “This truly proves he is not a top NHL talent.”

    “Who says a top NHL talent has to outshoot the opposition?”

    Maybe you guys ought to respond to Jonathan's argument, not a different argument!

    Maybe you could've asked "Who says a top NHL talent has to be a pervert?"

    Its called reading, gentlemen.

    but maybe his line keeps the opposition to the outside surrendering lots of shots but not a lot of quality scoring chances…

    That one should be easy; is Mike Smith’s Save % better or worse when The Vinny is on the ice?

  • Hippy

    RB: Actually, I think you should give them close to equal weight but I'd fall back on stats when the money's on the table.

    I appreciate the access you have so you get pin down a Lowe and ask him what the plans might be for Brule etc etc. But when the chips falls you look at the game one way: the same you've looked at it for the last 50 years.

    Other guys grew up with that but now they're embracing stats as well. And when stats pass the sniff test, you know you're on to something.

    Overall, you'll always have the adv because conceivably you could study up on your stats and then ask the people in charge about them. But until you do, you remain ignorant to the new big picture.

  • Hippy

    Shoot, I screwed up the first reply.

    Statistical analysis is 100% reliable 0% of the time.

    Be that as it may, are you saying that we should throw all of science out the window and go back to positing that medical problems are due to an imbalance of the humours? All the science we rely on today to keep us alive and using this here series of tubes is itself reliant on statistical analysis, which is usually only considered around 95% reliable.

    I also love how people are all too willing to plug their ears when the numbers say something they don't want to hear, but are all ears when it says something they do want to hear. I'm sure you'd all have the time of day for the numbers if they told you that Rob Schremp was actually awesome and not one of a dozen washed-up junior scorers in the American League. What do people have invested in Lecavalier, anyway? He's not an Oiler, so why do you care if the numbers show that maybe he's not the all-around star that someone in the upper echelon, like Crosby et. al.? Is it your pride on the line? Come on, people, think for a second here.

    Orr was +597 over his 596 GP (all except his rookie year for which I can’t find +/- numbers). If anything, the stats guys would like him more than the others given his gaudy +/- numbers.

    Recently, Matt Fenwick linked to an adjusted +/- study over the first 40 years from expansion on HF. Orr came in #2 in the raw numbers, and given that he played less than half as many games as the leader, Ray Bourque, and given that his first couple of seasons came before +/- was even recorded, let's just say he's #1 with a bullet and carry on.