WILL CAM TALBOT BE AN OILER BY SUNDOWN? (DARMOK)

From the still fantastic Gare Joyce classic Future Greats and Heartbreaks, page 68:

  • Picks three and four shuffle back and forth. Toews—a flag goes up when the scouts see two concussions marked beside his name on the combine physical. Backstrom—talent, but questions about size, and no interview. Five and six are even more contentious. Kessel edges Brassard on the board, but there’s no consensus. McMaster has Brassard No. 2 overall on his list. “Love this kid. Goes into corners and comes out with the puck like there’s glue on his stick.” Wayne Smith is pro-Brassard. “I know more about this kid than he does about himself.”

Scouts (hockey and baseball, those are the sports I’ve followed closely in my life) use their own metaphorical language—much like the alien captain in the Star Trek episode ‘Darmok’ which has been mentioned on this blog a few times over the years. Scouts have a colorful language (hockey men use the f-bomb like you and I use oxygen) but get their point across effectively. As an example, can you think of a retort to Wayne Smith in the above paragraph? You’re either going to challenge him or back down and as it turned out Columbus drafted Brassard, who ended up being the key asset sent away in the ill-fated Marian Gaborik deal with Glen Sather (who keeps showing up this week in conversation).

It’s that anecdotal tidbit, framed around an inspired or determined monologue, that gets people drafted out of order. It has to be, otherwise there would be no Niinimaki, no Thomas Hickey, no Hugh Jessiman. I imagine the math people involved in making the Oilers final list feeling much like the crew of the Enterprise when they made first contact with the Tamarians. I also imagine the opposite to be true, veteran scouts rolling their eyes at the new value given to math (or at least we hope there’s value given, there is a lot of projection here). It’s vitally important both sides be heard and we’ll know what happened next weekend. The potential for a generational draft is there, with Edmonton possessing four picks in the top 60 overall.

IS THE ANSWER LAWSON CROUSE?

crouse capture

This is going to be a fascinating draft for the Edmonton Oilers. As we’ve discussed, Edmonton could walk away with substantial talent at No. 1, No. 16, No. 33 and No. 57—FOUR pieces of the future who could turn into something special as they grow up together as Oilers. Bob Green will be charged with getting the list right and he’ll have help from damn near everyone—Stu MacGregor, Scott Howson, his scouts, Craig MacTavish and Peter Chiarelli has a scouting past, too. What will Edmonton’s list look like? For starters, the top 16 will NOT look like my list. Here’s the LT top 16 followed by a guesstimate of the Oil top 16.

LOWETIDE TOP 16

  1. C Connor McDavid, Erie Otters (OHL)
  2. C Jack Eichel, Boston U (NCAA) 
  3. R Mitch Marner, London Knights (OHL)
  4. D Noah Hanifin, BC (NCAA)
  5. C Dylan Strome, Erie Otters (OHL)
  6. D Ivan Provorov, Brandon Wheat Kings (WHL)
  7. R Mikko Rantanen, TPS Turku (SML)
  8. C Mathew Barzal, Seattle Thunderbirds (WHL)
  9. D Zach Werenski, Michigan (NCAA)
  10. R Evgeni Svechnikov, Cape Breton (QMJHL) 
  11. R Timo Meier, Halifax Mooseheads (QMJHL) 
  12. D Jeremy Roy, Sherbrooke Phoenix (QMJHL) 
  13. C Travis Konecny, Ottawa 67’s (OHL) 
  14. R Nick Merkley, Kelowna Rockets (WHL) 
  15. C Anthony Beauvillier, Shawinigan Cataractes (QMJHL)
  16. C Pavel Zacha, Sarnia Sting (OHL)

OILERS TOP 16

  1. C Connor McDavid, Erie Otters (OHL)
  2. C Jack Eichel, Boston U (NCAA) 
  3. C Dylan Strome, Erie Otters (OHL)
  4. D Noah Hanifin, BC (NCAA)
  5. C Pavel Zacha, Sarnia Sting (OHL)
  6. D Ivan Provorov, Brandon Wheat Kings (WHL)
  7. R Mitch Marner, London Knights (OHL)
  8. L Lawson Crouse, Kingston Frontenacs (OHL)
  9. C Mathew Barzal, Seattle Thunderbirds (WHL)
  10. R Mikko Rantanen, TPS Turku (SML)
  11. R Evgeni Svechnikov, Cape Breton (QMJHL) 
  12. D Zach Werenski, Michigan (NCAA)
  13. R Timo Meier, Halifax Mooseheads (QMJHL) 
  14. D Jakub Zboril, Saint John Sea Dogs (QMJHL)
  15. C Joel Eriksson Ek, Farjestad (SHL)
  16. D Noah Juulsen, Everett Silvertips (WHL)

 IS THE ANSWER CAM TALBOT?

One of the things that could impact Edmonton’s ‘four horsemen’ draft in the top 60 this year is a trade. I think we’re at a point now where it’s reasonable to suggest the target (Cam Talbot) and the most logical return (second-round pick this year, draft pick next season) based on the NY Rangers lacking picks this and next year.

  • Darren Dreger: “Cam Talbot is generating a ton of interest and in fact I would be a little surprised if he wasn’t the first guy to go and I know that the Edmonton Oilers have considerable interest in Cam Talbot.” Source
  • Darren Dreger: ” I know Peter Chiarelli is working pretty hard to nail down a goalie and he’s got Cam Talbot of the New York Rangers in his crosshairs. Source
  • Bruce Garrioch: Yes, Rangers GM Glen Sather can be tough to deal with, but he knows he has a valuable asset in Talbot and it’s believed the Oilers have had serious discussions with the Rangers about the backup to Henrik Lundqvist, who did an admirable job when he was out for an extended period during the regular season. Source
  • Jim Matheson: Talbot, an unrestricted free-agent next July, is a favourite of Mark Messier, who is an Oilers’ advisor. Source

That’s my guess. Talbot to NYR for No. 33 overall and Edmonton’s second-round pick next year. I know the first reaction will be ‘that’s too much!’ and I agree with you completely. However, Talbot (like McLellan) appears to be best available and it’s likely the Rangers are entertaining more than one significant offer. Plus, Sather. I’ll suggest No. 33 this year and next year’s second, and if it happens consider myself lucky Nail Yakupov wasn’t included in the deal. Finally, I would argue that Edmonton has a history of getting the demo with 4,000 clicks on it, the showroom sofa with some wear and a little rip at the back. If Cam Talbot is the best available, and I think he is (we don’t really know what’s out there), then get the man. Enough with the compromise, let’s turn the hell north.

  • Dreger: “But beyond that, I don’t hear any bigger names from a goaltending perspective. But even those players – the New York Rangers, based on the interest they’re getting in Cam Talbot, they’re just giggling in the background. They know he’s a good goalie. They know that. But they’re going to hit a home run on rate of return.”

We wait. I suspect not for much longer, though.

written by

The author didn‘t add any Information to his profile yet.
Related Posts

244 Responses to "WILL CAM TALBOT BE AN OILER BY SUNDOWN? (DARMOK)"

« Older Comments
  1. frjohnk says:

    kooler:
    C Schneider was worth a 9th draft pick (Bo Horvat). A proven #1 in Van with Luongo. Where would Bo Horvat sit in this year draft and how does Talbot compare to Schneider?

    If Bo is a 16th pick this year and Talbot is not a proven #1, he can’t be worth more than our # 33.

    I talked about this at the top

    frjohnk: Value of the goalies
    Schneider till 2013 was 86 games started .927, 67% quality starts
    Talbot now is 53 games started .931, 62% quality starts
    Schneider might have a slighter edge because of more games started but it is quite close.

    Value of the draft picks
    The 16th overall pick this year because of the deep draft may have close to similar value now as the value of the 9th pick in 2013 at that time ( though, I do like Horvath, just comparing value of the picks at their respective times) The 9th has a slight edge here.

    So one could say that if everything else is relative
    Talbot for the 16th overall this year is comparable to Schneider for the 9th overall in 2013

    Corey Schneider was 2 years from free agency, Cam Talbot is 1 year from free agency, and cant be extended till after Jan 1.

    This lessens Talbots value

    In 2013, it was a goalie sellers market. There was actually diddly squat for UFA goalies. The goalies that got paid the most, are not even in the league anymore!

    In 2015, it is a goalie buyers market. 6 goalies who could be starters or who have given near league average save %. These goalies only cost money, no other asset. Plus teams like Vancouver, Chicago, Ottawa with 3 NHL goalies.

    This also lessens Talbots value

    I have no problem going after Talbot, but the problem I have is that the cost should not be anywhere near the first round. And if we do get him through a trade, we still can not say we have solidified our goaltending situation, not because he does not have good numbers, but because of his contract. I don’t like that contract. Too much uncertainty as the player holds all the cards.

    Way different if he was RFA next summer.

    I would be more willing to trade the 33rd plus or maybe the 16th, if there were no other goalies on the UFA market, and Talbot was RFA next year. A goalie sellers market would force us to overpay, but because he is RFA in this case, I would do it.

    If it was a goalie buyers market and Talbot was RFA next summer, I would do the 33rd.

    Offer the 57th and a conditional pick from next year. If Sather does not want to trade, then move on.

  2. Connorrhea says:

    Canucks sign another goalie:

    http://canucks.nhl.com/club/news.htm?id=771405

  3. wheatnoil says:

    Factotum:
    wheatnoil,

    Thank you for clarifying that.So the work was intended to be descriptive rather than inferential – personally, I don’t think anyone should have a problem with that and I find the observations quite interesting.Admitting that this is not my area of expertise (and not speaking for anyone else), I think the thing that was bugging me was the thought of using statistical inference to draw conclusions about a population of *one* – i.e. Talbot – based on the properties of the sample, which I realize was not your intent.

    As an aside, if I had my choice between Bernier and Talbot, I’d choose Bernier, too.

    Well, the next study that G Money is working on will be more focused on trying to project where Talbot will end up. That study looks at every goalie that has started their career since 2005/06 (in the last 10 years) and has had at least Talbot’s sample size. The results, I think, are more interesting than the initial exploratory study I did and that G Money posted today.

    No one can predict with 100% accuracy what will happen to Talbot or any other N of 1. However, as you’ll soon see, Talbot keeps some pretty lofty company with how well he’s started his career. Based on the results of that data to come, I’ve come to the conclusion that Talbot is a pretty good bet. How much that bet is worth is up for debate and it’s not a sure thing. However, all goalies are a gamble and Talbot looks like a good one.

    Edit: For the record, I choose Talbot over Bernier all things being equal, but as others have pointed out, cost of acquisition and contract status have an impact on the decision-making process. I’m less concerned about contract status because I think he’s signable. I’m more concerned about acquisition cost, but I’d be more willing to pay a premium for Talbot than I would on Lack or almost any other goalie on the market.

  4. Bag of Pucks says:

    It wouldn’t surprise me if Chiarelli and Sather are talking a bigger overall deal possibly involving Rick Nash along with Talbot.

    That’s two disappointing postseasons in a row for the power forward and given the maturity of its core, NYR hockey management is under a lot of optics pressure to change their mix sufficiently to get them over the top whilst they still have one of the premier stoppers in the game.

    Given Nash’s age and his massive cap hit that continues for eons, this is precisely the type of player Chiarelli should avoid imo.

    But I could definitely see Sather selling a Nash/Talbot for Eberle & good prospect type deal under the guise of “you need size and goaltending without giving up goal production and this deal fits the bill.”

    Beware of trades with Slats. He’ll hypnotize you with Jimmy Carson while he plunders your Joe Murphy and Adam Graves. Just when it seems the prison rape couldn’t get any worse, you find yourself throwing in Petr Klima too!

  5. raventalon40 says:

    frjohnk: I talked about this at the top

    Corey Schneider was 2 years from free agency, Cam Talbot is 1 year from free agency, and cant be extended till after Jan 1.

    This lessens Talbots value

    In 2013, it was a goalie sellers market.There was actually diddly squat for UFA goalies.The goalies that got paid the most, are not even in the league anymore!

    In 2015, it is a goalie buyers market.6 goalies who could be starters or who have given near league average save %. These goalies only cost money, no other asset. Plus teams like Vancouver, Chicago, Ottawa with 3 NHL goalies.

    This also lessens Talbots value

    I have no problem going after Talbot, but the problem I have is that the cost should not be anywhere near the first round.And if we do get him through a trade, we still can not say we have solidified our goaltending situation, not because he does not have good numbers, but because of his contract. I don’t like that contract.Too much uncertainty as the player holds all the cards.

    Way different if he was RFA next summer.

    I would be more willing to trade the 33rd plus or maybe the 16th, if there were no other goalies on the UFA market, and Talbot was RFA next year.A goalie sellers market would force us to overpay, but because he is RFA in this case, I would do it.

    If it was a goalie buyers market and Talbot was RFA next summer, I would do the 33rd.

    Offer the 57th and a conditional pick from next year.If Sather does not want to trade, then move on.

    The time til free agency doesn’t affect market price… the price of competing bids DOES. Look at the trade deadline every year. This is just my opinion of course. If you have evidence contrary to this claim I’d be happy to see it.

  6. godot10 says:

    Ca$h-McMoney!: I’m not so sure. If we took Schultz to arbitration, would that indicate to the world that we value him highly?

    I think generally taking a player to arbitration means you question their value to an extent.My understanding is also that arbitration can’t award contracts longer than 2 years (and usually results in 1 year deals), if they were as high on him as the average Lowetide poster they’d be locking him up for at least 4.

    Or I could totally off base.

    The Leafs took Bernier to arbitration because they want a 100% guarantee that he will be under contract for training camp. And they wanted to guarantee they would be able to trade him if they get a good offer.

    Newport Sports is willing to do some funky things. Bernier might not have opted for player arbitration, which might have meant a holdout. Which would have complicated the Leafs summer plans and made it more difficult to trade Bernier, if that is what they want to do.

  7. Dashingsilverfox says:

    Connorrhea:
    Canucks sign another goalie:

    http://canucks.nhl.com/club/news.htm?id=771405

    Vancouver also has Thatcher Demko (2nd round pick 2014 – .925 at Boston College) in their system as well as Joacim Ericsson in Utica.

    They’re loaded with G prospects which I imagine is why they can afford to give one up,

  8. Ca$h-McMoney! says:

    godot10,

    I hadn’t considered that.

    It would be a very ballsy move for an unestablished goalie to hold out, especially when his team is relatively comfortable with his backup, who is perceived to be more of a 1b than a #2.

    Still, if I were the leafs I’d want to lock him down. I guess that’s one way of doing it.

    I’d be interested to see what the guy’s market value would be if he was sitting on the sidelines through mid November.

  9. Bag of Pucks says:

    raventalon40: The time til free agency doesn’t affect market price… the price of competing bids DOES. Look at the trade deadline every year. This is just my opinion of course. If you have evidence contrary to this claim I’d be happy to see it.

    The market conditions that exist at the trade deadline aren’t the same conditions that exist entering the draft. Teams out of the hunt are paring expiring contracts whilst those looking to win are mortgaging their future.

    In the more rational market that exists now, contract status impacts asset value which should impact the market price for the shrewder GMs.

    The X Factor is the Ed Snider type personalities that want what they want at any cost. If that type of personality gets Talbot in their sights, it drives up the bidding price for everyone.

    That’s why I’m not convinced that waiting for the first chip to fall to set the market is always the best strategy. Sometimes you’re better getting out with a reasonably priced acquisition early before the competition heats up and egos start to prevail.

    Kind of like what Chia did in avoiding the Babcock zoo and locking down McLellan.

  10. AsiaOil says:

    Wheatnoil

    Look I review research for a living – and critiquing a piece of research does not require me to re-do the research of every one of my students or colleagues. So your suggestion that I cannot outline the faults in your analysis unless I do it myself is specious and attempts to stop the conversation. If you put stuff up – then man up and accept the critique instead of trying deflect it.

    You have a tiny pre-selected sample of 17 which you attempt to draw conclusions from – and as you note – all you can say is that “since 1998, goalies that got a foothold in the NHL (50 games) but were unable to establish themselves as long-term options had poor even-strength save percentages at the 50 game mark the vast majority of the time”. Translation: the tiny number of goalies that you pre-selected as not very good – did indeed prove themselves to be not very good. Well except for Maarkanen – that’s 1/17 – and 1/17 is a lot in a pre-selected sample set up to prove your point.

    My points remains – given the sample size, survivorship and selection issues – this research is not really “proving” anything other than goalies pre-defined as “bad” are indeed bad. Well except for Maarkanen – who was a bad goalie who appeared to be a good goalie early on – but then in fact proved to be a bad goalie 🙂

    I don’t know if Talbot will be good or bad – no one does only based on his SP after only 50 games in a highly unusual NHL career that only started at 26. He’s an outlier, a wild card and a therefore a huge risk to spend expensive assets on to obtain even with out the UFA issue in 12 months. Life would be really easy if you could predict goalie future performance with something as simplistic as 50 game SP. Evaluating netminders is still an area where trained eyes and demonstrated performance over time is critical due to the huge mental load carried by a starting goaltending – a few simplistic stats cannot replace this. You finish off the article by stating that “my conclusion out of this is pretty simple: if you’re an NHL GM looking to hire a young goalie, I would suggest you STAY AWAY FROM ANY GOALIE WITH A LOUSY SV% AT 50 GAMES.

    Doubt Minnesota agrees with you – eg. Devan Dubnyk – 54 game SP = .906

    Actually all you can say based on your analysis is: STAY AWAY FROM ANY GOALIE WITH A LOUSY SV% AT 50 GAMES WHO IS ACTUALLY OUT OF THE LEAGUE BY 200 GAMES.

    I agree wholeheartedly 🙂

    wheatnoil,

  11. Mtl-oiler says:

    raventalon40,

    Exactly why we shouldn’t be in any hurry to over pay for a goalie. As stated before it’s a buyer’s market there’s no rush. We’re not winning the cup next year. Wait for the dust to settle and pick up a very capable G for pennies on the dollar.

    Another good example is the past trade dead line and all the D that were traded. Who was one of the last teams to make a trade for a D? Mtl. Who paid the least? Mtl. Who got the best D? Mtl. (I would say at least a tie with Sekera IMO)

  12. godot10 says:

    I think Buffalo is in on Talbot, because they are also in on Nash.

    Murray and Bylsma are thinking Kane, Eichel, Nash as a #1 line. Nash, as a veteran mentor to both of them.

    i.e. I think the Rangers are considering addressing their cap problems by trading Nash, rather than one of the top 4 D.

  13. Lowetide says:

    Dashingsilverfox: Vancouver also has Thatcher Demko (2nd round pick 2014 – .925 at Boston College) in their system as well as Joacim Ericsson in Utica.

    They’re loaded with G prospects which I imagine is why they can afford to give one up,

    The Canucks will have a helluva goalie group for sure, I’d suggested holding on to the best ones though. That’s been a problem of late.

  14. Ca$h-McMoney! says:

    godot10,

    That would make sense. Owner isn’t risk averse, he locked up Ville Leino long term after all.

  15. Bag of Pucks says:

    Asiaoil’s post raises a great point and it’s one we should be intimately familiar with as Oiler fans when assessing Talbot’s potential.

    Both Scrivens and Dubnyk and to a lesser extent Fasth appeared to be reasonable bets when assessing their performance and numbers as backups. They ALL appeared to be viable options as future starters.

    But you just don’t know until you set the pressure of that mantle squarely upon their shoulders as starters. That’s when you truly find out what you have with that player. And even then you still dont know how they’ll perform under the added pressure of being a playoff starter.

    This is one of the reasons I really think we undervalue wins as a crucial stat throughout a goalie’s development, yes, it’s a team stat and thus can be unfair to the goalie. But have you ever noticed how often the true greats seem to win wherever they go?

    In simpler terms, did the Devils make Brodeur or did Brodeur make the Devils?

  16. RexLibris says:

    G Money: Man, given how outstanding a player we can expect Jankowski to be (best player in his draft year, as I recall), the only way we could balance that would be to send the Rangers McDavid!

    I don’t know. Some of what I’ve heard from Flames fans suggests that McDavid will have a disappointing year because Bennett will outscore him as a result of playing on a far better team.

    Might want to cut bait now, before the disappointment.

    Yeah, when Feaster said that I argued on FN that the best thing for Jankowksi would be to disappear into the NCAA for four years and just quietly develop. I have him pegged as a 3rd line C/W if/when he makes the NHL. Maybe he becomes more, maybe less, but he kind of ran in place last year offensively and was said to have developed defensively. I have Anton Lander as the bar against which I’ll measure him, maybe that changes once he turns pro. But we’re heading into draft +4 year now and it doesn’t look good. Especially when you consider the names they passed on to take him. If you ever want to passively troll a Flames fan casually mention how good Teuvo Teravainen has looked this past year.

    Poor kid. It isn’t his fault he was drafted by a man who maintains his health by sleeping in a hyperbolic chamber.

  17. G Money says:

    AsiaOil,

    Spoken like a true academic!

    In the real world, there isn’t the luxury of not making a decision because you can’t couch it in terms of a formal Fisher hypothesis, or waiting years to build up the requisite sample sizes and p < 0.05.

    Walk into a boardroom, or a hockey GMs office anywhere, and I doubt you will find a decision has *ever* been made on that basis. Except maybe Tambellini, given his dithering results, I could believe that to be the case. He's still waiting for his sample size maybe.

    The 17 sample size was NOT chosen to predefine the results. It is NOT A SAMPLE. It is the whole population. The whole thing. In 20 years, 17 goalies have gotten to 50 games and not made it past 200.

    The point of the analysis is that you now have a well defined population, small though it is, and, as it turns out, that population has very well defined characteristics. E.g. mean sv% at 50 games is .907, a narrow spread on that mean, and only one member of that population that did much better than that and failed out.

    You may consider the 1 in 17 and not the 16 in 17 to be the dominant decision making factor, but that's an academic luxury – I do not thing you will find many who actually have to make decisions for a living will agree.

    The question, the practical one we have at hand and not the academic one that lets us wait until our students have built up a 217 sample size (which would take approximately 250 years at current rates), is whether or not Cam Talbot belongs in that population or not.

    It is a Bayesian question, not a Fisher question.

    Here is a simple answer for you: no. He does not. He is NOT part of that population.

    If Cam Talbot puts up a .931 after 50+ games and fails out before 200, he will be the first one to do so in over 20 years.

    The second question at hand is: based on the entire population of goalies who’ve started their careers in the last decade in the NHL and played at least 50 games, what can we conclude about the chances of Talbot being an elite goaltender in the NHL?

    Answer: if he turns out to be not elite, he’ll very much be an outlier in that population. Not a sample – the entire damn thing.

  18. Bag of Pucks says:

    Imagine a scenario whereby Chiarelli discounts analytics and goes ‘ultimate small sample size’ in trading for Scott Darling.

    The Oilogosphere would lose its collective shit.

  19. AsiaOil says:

    Actually I would bet on Anderson SHORT-TERM because he looked real good in the playoffs – but clearly he’s at the point where the cliff is getting near. So you don’t pay much for him because of this, he needs a strong backup at his age, and I assume other better options are not available. I’d rank the top 5 likely “available” guys:

    1 Bernier
    2 Niemi
    3 Anderson
    4a Talbot
    4b Lack

    But it’s always a balance of risk, reward, contract, age etc etc. Bernier’s the only guy who checks off most of the boxes. Talbot is a massive boom-bust and could cost the team another season which it cannot afford.

    Ca$h-McMoney!: I think most people would be very happy with Bernier, especially if the price is right.

    You suggested Anderson was a good bet and Talbot was a bad bet.You said Talbot was a bad bet because of sample size:you are not presenting evidence that he will be bad, but you are saying that you don’t trust the evidence that he will be good.I’m fine with that position.

    What I’m pointing out is that you are showing a willingness to ignore evidence with respect to Anderson.There is strong empirical evidence suggesting that the odds are good that Anderson declines at a pretty rapid rate at some point due to age, and that it’s a reasonable bet that the decline will start right away.

    I’d say that both Talbot and Anderson present a certain element of risk.Because of that, I’m happy with either.I think if you are trying to argue that the Oilers need to be risk averse, that Anderson is the wrong backup plan.He’s got plenty of risk, and plenty of reward, associated with his acquisition.

    Bernier on the other hand is low(er) risk, you’re right about that.He’s my first choice if the acquisition cost isn’t prohibitive.

  20. Ducey says:

    LT, you may want to look into the timeout thing. My last post took me about 2 min to type and then was erased by a “Timeout” when I submitted it.

  21. Jesse says:

    I’d like to affirm the discussion going on between Asiaoil/Wheatnoil/G Money. It’s a fascinating one with good points on both sides, and does seem to be leading somewhere rather than running in circles. The very antithesis of regular internet discourse.

    Well done lads.

  22. speeds says:

    I could be misunderstanding, but one difference between Lack and Talbot is that Lack is in the last year of a 2 year deal so he can sign an extension this summer, while Talbot’s on a one year deal so not until Jan 1 for him.

  23. AsiaOil says:

    I’ve spent more time in industry than academe so be careful with the assumptions – and your smarmy little remark about “people who actually have to make all living” is actually pretty juvenile. I’ve got at least as much expertise than you and quite likely a hell of a lot more in all three sectors. But that”s OK – weak analysis always resorts to name calling in the end.

    But you get to the crux of the matter – yes you have a population – but you have self-selected it to fit your hypothesis. Big bloody surprise that it turns out as predicted – well not really – Jussi even throws that into a shambles. Rigor is not the sole prerogative of academe – sloppy analysis doesn’t cut it anywhere.

    ….and yes Talbot could indeed turn out to be an outlier in the entire population. I actually expect that’s the case and would not be surprised. How many other guys come from literally nowhere, have almost no history at all playing significant games in their career, and then get their first NHL start at 26 in a prime backup slot. Population of 1 most likely as I can’t think of any recent examples – maybe Thomas.

    G Money:
    AsiaOil,

    Spoken like a true academic!

    In the real world, there isn’t the luxury of not making a decision because you can’t couch it in terms of a formal Fisher hypothesis, or waiting years to build up the requisite sample sizes and p < 0.05.

    Walk into a boardroom, or a hockey GMs office anywhere, and I doubt you will find a decision has *ever* been made on that basis.Except maybe Tambellini, given his dithering results, I could believe that to be the case.He’s still waiting for his sample size maybe.

    The 17 sample size was NOT chosen to predefine the results.It is NOT A SAMPLE.It is the whole population.The whole thing.In 20 years, 17 goalies have gotten to 50 games and not made it past 200.

    The point of the analysis is that you now have a well defined population, small though it is, and, as it turns out, that population has very well defined characteristics.E.g. mean sv% at 50 games is .907, a narrow spread on that mean, and only one member of that population that did much better than that and failed out.

    You may consider the 1 in 17 and not the 16 in 17 to be the dominant decision making factor, but that’s an academic luxury – I do not thing you will find many who actually have to make decisions for a living will agree.

    The question, the practical one we have at hand and not the academic one that lets us wait until our students have built up a 217 sample size (which would take approximately 250 years at current rates), is whether or not Cam Talbot belongs in that population or not.

    It is a Bayesian question, not a Fisher question.

    Here is a simple answer for you:no.He does not.He is NOT part of that population.

    If Cam Talbot puts up a .931 after 50+ games and fails out before 200, he will be the first one to do so in over 20 years.

    The second question at hand is: based on the entire population of goalies who’ve started their careers in the last decade in the NHL and played at least 50 games, what can we conclude about the chances of Talbot being an elite goaltender in the NHL?

    Answer: if he turns out to be not elite, he’ll very much be an outlier in that population.Not a sample – the entire damn thing.

  24. G Money says:

    AsiaOil: Doubt Minnesota agrees with you – eg. Devan Dubnyk – 54 game SP = .906

    ?

    Dubnyk sv% @ 1,300 shots = .915

    Dubnyk adjusted sv% @ 1,300 shots = .929

  25. TheGreatMutato says:

    Jesse:
    I’d like to affirm the discussion going on between Asiaoil/Wheatnoil/G Money. It’s a fascinating one with good points on both sides, and does seem to be leading somewhere rather than running in circles. The very antithesis of regular internet discourse.

    Well done lads.

    Indeed – it’s been fun and informative following along 😀

  26. G Money says:

    AsiaOil: Jussi even throws that into a shambles.

    Your definition of shambles is laughable. Sorry. It is.

    One guy in 17, the only guy, failed out of the league (could have been for any out of an entire universe of reasons – injury, personal, whatever) after looking OK fifty games in. The only one. In twenty years.

    Given the infinitude of possible ways in which failure could occur, its actually astonishing that it only happened to one.

    Astonishing.

  27. striatic says:

    i could see something like the 33rd and either Davidson or Hunt for Talbot. 2 seconds is too much thoug ouch.

  28. Doug McLachlan says:

    TheGreatMutato: Indeed – it’s been fun and informative following along

    Fight!

    Fight!

    Fight!

  29. G Money says:

    Some other interesting stats:

    # of goalies who’ve started in the league in the last decade with at least a .915 sv% after 1,300 shots: 32

    # of goalies who subsequently failed out of the league within say 2 seasons (or really at all): 0

    Average change in sv% from 1,300 mark to current or end of career: -.0006 (e.g. .925 becomes .9244)

    Amazing how much 1,300 shots (the Talbot criteria) tells you about a goalie. Far more than I ever thought would be the case.

  30. v4ance says:

    Joe Yerdon ‏@JoeYerdon
    Based on what Tim Murray said today, if Ilya Samsonov is available when they pick at 21, Buffalo is taking him. #NotedCleric

  31. Bag of Pucks says:

    RexLibris: I don’t know. Some of what I’ve heard from Flames fans suggests that McDavid will have a disappointing year because Bennett will outscore him as a result of playing on a far better team.

    Flames fans suggesting our generational talent will underperform their latest overhyped saviour?

    Shocker.

    This is the same fanbase that gave the GOAT the affectionate nickname of ‘Whiner.’

    Jealousy is a natural human emotion. It is also the Manifest Destiny of Flames fans.

  32. Bag of Pucks says:

    v4ance:
    Joe Yerdon ‏@JoeYerdon
    Based on what Tim Murray said today, if Ilya Samsonov is available when they pick at 21, Buffalo is taking him. #NotedCleric

    Given Murray’s recent history, Samsonov will likely refuse to report.

    Guy gets passed over more often than a mute at the debate team tryouts.

  33. flyfish1168 says:

    G Money,

    Very interesting topic today from LT on goalies. Is there a way we can rank or compare goalies to the quality of the team defense or d-men on their respected team ? It would be interesting especially goalies that has played for two or more teams, you can then see if they defense or the goalie that carried the team

  34. spoiler says:

    G Money: One guy in 17, the only guy, failed out of the league (could have been for any out of an entire universe of reasons – injury, personal, whatever) after looking OK fifty games in. The only one. In twenty years.

    You are actually understating this, no?

    Only one guy in 43 (or whatever the total population was) has failed out after looking good at the 50 game mark.

    Or out of all the guys that failed out, 17, only one of them failed after looking good.

  35. wheatnoil says:

    AsiaOil:
    If you put stuff up – then man up and accept the critique instead of trying deflect it.

    I responded to your statement by trying to explain the purpose of what I (and G Money) did and acknowledge the limitations of it. I also pointed out the lack of feasibility of one of your suggestions (i.e. the 500 goaltenders) and also responded with my reasoning for why Anderson (one of your alternate suggestions) may also be a risky choice, linking to evidence as to why I made that statement. I ended by attempting to invite you into the endeavor because I genuinely welcome it.

    As for your statement that Talbot is an outlier due to the age at which he started his NHL career, I’m not sure that’s the case. As you mentioned, Tim Thomas started his career in his later 20s. Niklas Backstrom, Eddie Lack, Antti Niemi and Jonas Hiller all started at 26 or later. There are likely more, but I’d have to go back to the data to sort it out.

    That said, the age issue is a valid point. This study, nor the one upcoming, corrected for age.There are some studies out there that suggest that goalies are the best when they are quite young (early 20s) and then decline from there on, suggesting the decline can be as early as age 26. Based on this, some have proposed promoting goalies as soon as they do well in the AHL and basically keeping a stable of young, unproven goalies as they all inevitably decline with age quite young. That never made a lot of sense to me but I’d have to look into it closer to get a better grasp on the argument. However, if we go by that data, we should be looking only at acquiring goalies in their early 20s and selling high by the time they hit their late 20s. In other words, promote Broissoit and acquire someone else who looks really good in the AHL and is in their early 20s and roll the dice with them. Interesting and entertaining thought.

  36. spoiler says:

    flyfish1168:
    G Money,

    Very interesting topic today from LT on goalies. Is there a way we can rank or compare goalies to the quality of the team defense or d-men on their respected team ? It would be interesting especially goalies that has played fortwo or more teams,you can then see if theydefense or the goalie that carried the team

    Adjusted save percentage.

  37. Lowetide says:

    Fantastic thread folks. I would ask for respect for your fellow man or woman, but beyond that splendid and educational. Thanks!

  38. Doug McLachlan says:

    Reading up on the Bernier situation I came across this nugget from Labrun:

    Pierre LeBrun
    ✔ ‎@Real_ESPNLeBrun
    Should add that the Leafs were the only team before yesterday’s club arbitration deadline to file on a player

    11:05 AM – 18 Jun 2015

    That would mean, contrary to one of the possible plans (can’t recall if it was McCurdy, Gregor or Willis – sorry) with respect to Schultz was for the club to take him to arbitration and seek to have his salary cut (by a small amount). Does this mean that Chia thinks he’s worth his contract and maybe a slight raise? Intending on trading him? Preparing to let him drop off completely?

  39. G Money says:

    flyfish1168,

    Hi Fly,

    That’s what the idea is behind war-on-ice’s adjusted sv%.

    The adjustment is made by looking at how many chances, especially dangerous ones, and weighting them accordingly. This is their definition: “Adjusted save percentage; this adjusts for the fact that some teams give up more high-quality shots, while others give up more low-quality shots. This is the weighted-average of SvPctHigh, SvPctMed, and SvPctLow, where the weights correspond to the league-wide percentage of shots from each of those areas. In other words, this is a goalies save percentage if they faced a league average proportion of shots from each of the three shooting zones (high, medium, and low probability of success).

    If interested, you can read the painful details here: http://blog.war-on-ice.com/adjusted-save-percentage-taking-into-account-high-medium-and-low-probability-shots/

  40. G Money says:

    spoiler: You are actually understating this, no?

    Only one guy in 43 (or whatever the total population was) has failed out after looking good at the 50 game mark.

    Or out of all the guys that failed out, 17, only one of them failed after looking good.

    Sort of. We can say definitively that out of the goalies that made the (50,200) bracket, all of them looked lousy at 50, but one. The one ‘good’ one, ol’ Jussi Rebound, that dropped out … well we have no idea why it happened, but given the infinite number of ways to drop out (injury being the least of them), it’s amazing that only one did. Anyone recall what sent Jussi into a tailspin and then home, add some colour to the commentary?

    Out of the bigger sample, the 43 guys that started their careers and then played at least 50 games since 2005, yes it’s technically true that you could argue that just two truly failed (Justin Peters and JDD).

    The rest all played last season, or played several seasons past their 50 game mark, so its hard to call that failure.

    But I’m loathe to use that as any sort of benchmark, because of course out of that group:

    – several could pull the pin this year. Fasth as an example! I’d expect he’s not going to be back in the league this year or ever. His 1,300 game sv% of .909 of course puts him squarely in the ‘more than likely to fail’ bucket.

    – there are others that played longer than you’d expect given all started under .910 and, as expected, stayed there: Curtis McElhinney, Joey MacDonald, Peter Budaj. In those cases, you kind of look at their records and go WTF?! They probably should have been punted but managed to stick around, primarily as backups. Gets a little subjective at that point, so they just get rolled into the ‘success’ column because they stuck around.

  41. spoiler says:

    AsiaOil: Look I review research for a living – and critiquing a piece of research does not require me to re-do the research of every one of my students or colleagues. So your suggestion that I cannot outline the faults in your analysis unless I do it myself is specious and attempts to stop the conversation. If you put stuff up – then man up and accept the critique instead of trying deflect it.

    This is a straw man, no one has asked you to do this. And appeals to authority are only going to invite ridicule.

    AsiaOil: I’ve spent more time in industry than academe so be careful with the assumptions – and your smarmy little remark about “people who actually have to make all living” is actually pretty juvenile. I’ve got at least as much expertise than you and quite likely a hell of a lot more in all three sectors. But that”s OK – weak analysis always resorts to name calling in the end.

    OK… so you set up yourself as an authority and GMon gets “smarmy” to make fun of such a weak argument (which was based on a straw man put up by you in the first place), and now you’re complaining about it and re-entrenching that very same appeal to authority?

    This says way more about your analysis than his.

  42. v4ance says:

    I remember we were hot for Bishop when he was in Ott and he was eventually acquired for a 3rd and Cory Conacher,

    From 2008-13, he played for the Blues and the Sens for a total of 36 games and had a 0.911 save % after 922 shots. The one good trend was that his save % gradually got better each year up to a high of 0.922 when he was acquired by the Bolts. Since 2013, his stats are 0.920 over 134 games.

    Even though the sample size was very small, everything Bishop accomplished before Tampa traded for him showed that he was trending to being a good NHL starter. Talbot has had save % of 0.941 and 0.926 over two seasons in 57 games for a overall save % of 0.931. Based on his performance relative to Bishop, he is JUST as good or even better at the barest minimum

  43. flyfish1168 says:

    G Money:
    flyfish1168,

    Hi Fly,

    That’s what the idea is behind war-on-ice’s adjusted sv%.

    The adjustment is made by looking at how many chances, especially dangerous ones, and weighting them accordingly.This is their definition:“Adjusted save percentage; this adjusts for the fact that some teams give up more high-quality shots, while others give up more low-quality shots. This is the weighted-average of SvPctHigh, SvPctMed, and SvPctLow, where the weights correspond to the league-wide percentage of shots from each of those areas. In other words, this is a goalies save percentage if they faced a league average proportion of shots from each of the three shooting zones (high, medium, and low probability of success).

    If interested, you can read the painful details here: http://blog.war-on-ice.com/adjusted-save-percentage-taking-into-account-high-medium-and-low-probability-shots/

    Thx didn’t know it already existed

  44. Hockey News: Bernier to Arbitration; Cam Talbot to Edmonton? says:

    […] Cam Talbot destined to be an Oiler? [Lowe Tide] and [Copper N Blue]. The Rangers are asking a lot for him. [Today’s Slap Shot]. And what […]

« Older Comments

Leave a Reply

Want to join the discussion?
Feel free to contribute!
© Copyright - Lowetide.ca