Stacking the deck on draft day



I want to point you all in the direction of a fantastic bit of research written last week by CIS Blogger and math graduate Rob Pettapiece. Rob expands on research he did into the CHL draft and found that players born earlier in the year were more likely to get drafted than players born later in the year. He divided the players into four camps, ‘Q1’ through ‘Q4’ and concluded that, since there were 4.3 times more Q1 players drafted, there are teams that make decisions, particularly in the late rounds, based on player size and strength rather than observable skill.

Advertisement - Continue Reading Below

“If you’re down to “projectability” at that point, rather than stats and performance, then fine, pick the tallest kid if you have no other information about them.”

So where does this put us as we head into the NHL Draft on Friday? If younger players are more likely to break out, particularly in the later rounds, does that give a team an incentive to draft, say, Jonathan Huberdeau, who will be 18.1 years old as opposed to Gabriel Landeskog, who will be roughly 18.6 years old? As we delve into the later rounds, teams are better served if they looked to draft a player who may not be as well developed but has the benefit of growing into an NHL body. Players born in June and May who just turned 18 are prime candidates, as Rob mentions the successes of Michael Ryder and Brooks Laich, both late round picks, Ryder having been born on March 31st and Laich on June 23rd.

Recent examples of this effect are Sergei Kostitsyn and Jannik Hansen, who were drafted at 18.3 years of age and slow developers before becoming key forwards on playoff teams. As Rob notes, “we see behaviour from hockey executives supporting the hypothesis that younger junior hockey players are undervalued. But this time it’s the NHL teams, not the CHL teams, whose valuation systems are off.”

Advertisement - Continue Reading Below

This site uses Akismet to reduce spam. Learn how your comment data is processed.

    • SmellOfVictory

      According to the article, the potential of younger players in later rounds tends to outstrip the more established/”projectable” older players once they actually play in the NHL (or don’t). This is something that appears not to be a common belief.

      • O.C.

        But that is obvious isn’t it? The young have more maturing? There’s more of a gamble on e younger… They may get better… We all get that… But does anyone believe that the pro scouts don’t know there is more potential for reward where there is greater risk?

        It makes sense. Draft unproven in later rounds.

        • SmellOfVictory

          It may be a case of pro scouts drafting based on what they see as an individual who won’t fit the mold as opposed to a general philosophy, but basically the result is that yes, certain teams do seem to continue to draft older in later rounds, and with poor results.

  • It may seem obvious but i’m not sure if I have the gist of it but here is what I conclude:

    Drafting younger is better not because the lower the age the more magic the number is, but simply because the younger have more of a potential for growth and improvement not yet manifested in the individual at the time of the draft.
    Cam, I may have thought myself into circles by now, but given these results, do you see a further importance of the combine in judging or quantifying this obvious ‘potential’ in the younger draftees? By comparing the average results in the combine by various Q1-Q4 draftees to an invidual’s performance, would we not at least begin to get a foothold on quantifying this ‘potential’?

    EDIT: In other words, if the average guy can bench 195lbs, and my 19.1 (base 10 age lol) guy can bench 250… wouldn’t it be reasonable to assume that he isn’t going to improve a whole lot more and assign that a specific number.

    • Exactly.
      Thinking along somewhat same (but different) lines, there were some baseball mathematicians who recently did something sort of similar with comparing players from different eras, but looking at how they fared against similar overlapping competition. In that way they could rank players who played in the rabbit-ball era against players who played in the dead-ball era etc.
      Having stated that, I am still not surprised that big league execs/scouts are slow on the uptake in looking at important info like Pettapiece’s work etc. (The book ‘Moneyball’ illustrates this point very well with regards to how players with valuable skill sets get overlooked for the wrong reasons.)

    • I think you’re onto something here. I’m not a big fan of the combine in general, mainly because it’s all the top prospects, and also because there’s no on-ice component. I’m not sure how being able to bench 250 over 195 is supposed to make you a better hockey player than somebody, particularly when you’re a good six months older.

      The combine hardly separates one 18-year old fitness freak from another, and I think the whole thing is a masturbatory ruse by scouts to see players without their shirts (I ~loathe~ scouts).

      I think that teams could get more value if they spent their money in hiring projection analysts to soak up all these variables and allow teams to make more of an informed late round pick than “he’s big” which is probably as useful as saying “he would look good in our ECHL team’s sweater colours”.

    • SmellOfVictory

      I had the thought that perhaps those younger players who are drafted are, generally speaking, quite outstanding for their age (a few months making a potentially large difference in development at that age), resulting in being drafted. Thus, if more teams subscribed to the concept of drafting younger in the later rounds, perhaps we wouldn’t see that great an uptick in success compared to the current set of results, as those additional young players who ended up being drafted would simply water down the sample, so to speak. That said, I’m generally all for going younger if faced with a set of potential prospects who don’t have a great deal else to set them apart.

      • That’s an interesting point and one which causes a pulsing twinge of pain on the inner left side of my skull when I think about it too long.

        Maybe it’s not a matter of the younger players who are selected being exceptional, but rather the older players selected due to size are rather unexceptional except for their size.

        Maybe it’s a combination of both?

        Let’s say that for the next 6 years, after round 3 in each draft every NHL team agreed to draft the youngest players first. What would be the result? Would we see this pattern of Pettapiece’s disappear? Would we then see data suggesting that older players are being neglected in the later rounds of the draft, and that if you draft for older age in the later rounds you will have success?

        What prevents this from happening?

        • Matty Franchise Jr

          I think most posters are missing the primary point. Players born later in the year are the youngest players on EVERY team they have ever played for. Consequently, they fight for scraps of ice time during their formative years and are never relied upon for the PP and PK. The effects on development are profound. That is the real issue. It has nothing to do with how well they are playing at age 18 versus 18.5 or how long it will take to ‘mature’ in the NHL.
          This is the reason it was suggested minor hockey rotate the birthday cut off date by a quarter every year. The result would be every player lands in the ‘young’ group and ‘old’ group every 4 years.

  • Derzie

    Fantasy sports comes full circle. These advanced stats topics are exactly the fuel for the Flames hiring a Video Stats Tech McGyver guy this year. But like stock market analysis, the past is limited in value to predicting the future. Gut feel is still huge (ie. how many Bruins players were top draft picks?)

    • Actually, the Bruins help prove Cam’s point. You are correct that several of the top players on the Bruins were not top draft picks. However, a good percentage of them later rounders that helped them hoist the cup are March to June birthdays (Ryder, Lucic, Chara, Marchand, Thomas, Krejci, Paille, Ference, Kaberle, Rask and Khudobin – yes, I know that Paille and Rask were both R1, though they still went later at #20 and #21).

      Sure, like stock-picking, past performance does not guarantee future performance. However, there are methods of analysis that can be of great value in evaluating the future performance of a stock or athlete that are far more effective than gut feel.

      Sometimes our gut points us in the right direction, but often it can fool us. Math/statistical analysis can take guesswork out of the equation. Referencing ‘Moneyball’ again – because of the way that things had always been, there have always been some very valuable players that would get overlooked by other teams because they ‘didn’t fit the mould’ (ie: gut and experience). It’s only the teams that were smart enough to look at the numbers that benefitted from eventually having them on their rosters.

      The last and most obvious example of the value of numbers that I will throw down is that of Tim Thomas (Stanley/Conn Smythe winner, likely Vezina winner). Tim Thomas, due to his non-traditional goal-tending style, had a difficult time getting any coach or manager to take a serious look at him. Even when the obvious numbers (not the esoteric, geek stats) pointed to him as a keeper, coaches at several levels ignored the stats and refused give him a chance. Finally, later in his career, a few people noticed that the guy that didn’t look like much of a goalie, actually might be. Tim Thomas – drafted 9th round, 217th overall. Born April, 1974. Butterfly style points: very low. Save %: .940

  • Clyde Frog

    The article is very interesting.

    One question I would have is the total number of players in each category being examined.

    Are greater numbers of older kids being taken because they comprise a larger set of the talent pool?

    Another small issue would be, comparing Vancouver’s drafting record to several lottery teams seems a little unfair. Top 10 picks should far outperform Vancouver’s late round picks.

    I see the message of the article and agree with the idea of choosing talent over intangibles when you can with later picks.

  • Wax Man Riley

    So your astrological birth sign DOES matter?

    I read the link and there is data to support that drafting younger can turn out better, but the Canucks are a team with very established veterans and less room for the youngsters?

    Does that have any impact to this article? Or am I missing the point?

    There are more NHLers and CHLers born in the first part of the year. I have to side a bit with Gladwell (“Outliers” is a great book btw, I have read ever Gladwell book and they are all amazing) on this one, as the kids born younger in the year are given preferential treatment in their developing years.

  • VMR

    Micheal Ryder? Brooks Laich? Jannik Hansen? Segei Kostisyn? If those are the best examples out there I’m not sure I’d put too much faith in it. I mean sure they are serviceable players picked up later but they’re no Zetterberg or Chara or Weber.

  • @ j

    Isn’t that more Gladwell’s point and not Pettapiece’s?

    So….how does this translate to the draft? We should draft younger players in later rounds because their formative years have been stunted and their growth retarded due to a coaching age-bias? Younger players suck because they have never been given a chance, umm….so draft them please.
    *points to the chart*

    I’m no longer convinced! I hereby officially resign as Head Scouting Director of Me™.

    *throw his #4 pencil down in disgust and stomps away*

    • The draft is simply a manifestation of the on-ice facts – the older kids (those with a birth date in the first 2 quarters of any year)have a better chance of being drafted highly (and making the league) because they are better players i.e. have benefited from greater responsibility/developmental efforts by the coaches. There are always going to be exceptional talents that can overcome these odds therefore you will always have some 3rd and 4th quarter kids who draft highly but the stats suggest these are the exceptions.

    • The draft is simply a manifestation of the on-ice facts – the older kids (those with a birth date in the first 2 quarters of any year)have a better chance of being drafted highly (and making the league) because they are better players i.e. have benefited from greater responsibility/developmental efforts by the coaches. There are always going to be exceptional talents that can overcome these odds therefore you will always have some 3rd and 4th quarter kids who draft highly but the stats suggest these are the exceptions.

  • Matty Franchise Jr

    I think the younger guys development is not stunted, it is delayed partially until after the draft, so that on draft day they appear to be late rounders but may eventually develop into a player who would have deserved to be drafted higher.

    Edit: Early rounders are proven commodities who deserve to be drafted (or taken a chance on) in the early rounds. Once you get to the late rounds, older guys are less likely to develop much more and are deservedly late rounders. Younger guys whose development was delayed due to ageism may still have undeveloped potential that the older guys do not.

  • further to this, as outlined in Federal Association based Athletic Development programs is that as these kids go through minor sports programs, the older kids show more talent that the kids born later in the year.
    Then as a result of this are subject to better coaching, and many times, better programs.
    The stats will bear out if you look at AA or AAA programs. Even major Junior.

    So that they’re younger doesn’t necessarily mean they are going to develop further (though it does to a certain extent), but it can also mean that the older kids have had better coaching/training/opportunities going through minor sports and thus have developed more thoroughly throw the years.