Tuesday, January 15, 2008

When is the end for a goal scorer?
A study of 50-goal men throughout history

The Ovechkin email continues to come:
I don’t think the risk factor is quite as bad as you portray. Alex has proven to be remarkably durable and remarkably willing to play through pain. Bure is a decent comparison (for lots of reasons), but even if he has to hang ‘em up at 31, we get nine more years, which beats losing him as a UFA in four. Another thing ... the overwhelming majority of guys play well into their 30s, with the rare exception (Orr, Kluzak, Bure). Hockey has been lucky in terms of career-ending injuries (knock on wood).
— A friend from Washington
Such is the problem with evaluating a 13-year deal that lasts until a player is 35 years old: It's anyone's guess just how long said star remains happy, healthy and productive. How, exactly, do you measure longevity, and when does the average hockey star "hang 'em up?"

If you're expecting numbers and line charts, you get a gold star ...

Even though he's just 22, Alex Ovechkin belongs to some pretty exclusive clubs. Only two modern players, for instance, sit ahead of his .632 career goals-per-game average (Mike Bossy and Mario Lemieux). And the fact he's notched 252 points in his first 207 NHL games, as of last night, puts him in heady territory.

But one of the most exclusive groups Ovechkin's a part of is one he joined in his first season when he broke the 50-goal barrier.

Only 87 players in NHL history have scored 50 goals in a season, and that's a group that includes the majority of the game's top-producing players of all time. Of the 87, Ovechkin is one of 18 active players.

But for the purposes of this post, it's the other 69 fellows I'm interested in.

How long did they play for? How productive were they? And at what age does a former 50-goal scorer's ability to produce decline?

What are the chances, in other words, that Ovechkin's still filling the net 14 years down the road?

Quite a few of these stars retired early, with 15 (or about 22 per cent) playing their last 60-plus game season before age 31. More than half were gone before turning 34, but 22 managed to play at least 60 games in a season after age 35 (32 per cent).

Five former 50-goal scorers have played into their 40s: John Bucyk, Mark Messier, Steve Yzerman, Luc Robitaille and Dave Andreychuk.

Here are the ages when retired 50-goal men played their last 60-plus game season, plotted on a graph:

There's the big bulge, between 30 and 39, when the game's stars move on. As I mentioned above, 22 per cent are gone by 30, 35 per cent are gone by 31 and Mark Messier's the freak of nature at the end.

What are the chances Ovechkin's playing at 35? Decent enough, it would seem — or a little less than 40 per cent anyway if you trust historical averages — but the next question is just how productive will he be? When does a player's goal-scoring ability leave him?

Here's a look at what age these 50-goal scorers had their last 40-goal season:

Forty goals, during any era, was still a star turn, and an awful lot of players were pulling that off after age 30. But you certainly can see a definitive peak here, from 27 to 30, and it's from that point onward the goals began to fade.

Fifty-seven per cent fit into that range of longevity. Seventy-eight per cent had their last 40-goal season before turning 31.

On the freak end of the scale: Phil Esposito and John Bucyk beat Father Time by putting up 40 at age 37. Joe Mullen and Messier were the only other two who managed it after 33. (And Teemu Selanne joins these four when he officially calls it quits.)

But a goal scorer doesn't necessarily need to pot 40 every year to be of use. What if we take it down a notch, to 30 goals, and see how the numbers shift. Do more superstars produce in that range late in their career?

It's sad to say, but 30's the magic number. Fifty-seven per cent had their last 30-goal season before turning 31.

On the flip side, one-third of these stars were able to manage a 30-goal season at 32 or older. Luc Robitaille and Marcel Dionne were 36 in their last really productive season, Dino Ciccarelli and Mike Gartner were 37. Brett Hull managed one at 38, while Mr. Bucyk really deserves some credit for being the lone 40-year-old 30-goal man in the group.

Keep in mind that those fellows are the extremes. The average is somewhere near the graph's pointy edges, which means if I had to pick out players representative of the rise and fall of the average superstar career based on the numbers, they'd be Dale Hawerchuk and Michel Goulet.

Hawerchuk was an Ovechkin-like star right out of the gate, the first-overall pick in 1981, and he won the Calder the next summer after a 45-goal, 103-point campaign at age 18. He would have six 100-point campaigns in the next seven seasons, notching 70 per cent of his 518 goals by the time he was 25.

Hawerchuk was over the 80-point mark up until age 30, ran into all kinds of injury woes the following season and scored just 34 total goals over the last three years of his career. He retired with arthritis in his hip a few weeks after his 34th birthday in May, 1997.

Goulet was three years older, but hit his stride at about the same time. His first 40-goal season came in 1981-82 when he was 21, and he would score 50 four consecutive years after that. He fell off the 40- (and 30-) goal pace by age 28 and never bettered 65 points over his final six seasons.

In March, 1994, Goulet suffered a serious concussion at the Montreal Forum while playing for Chicago, and he retired at 33.

Both are in the Hall of Fame, and both are still among the top 40 goal scorers in NHL history. In terms of production and longevity when it comes to 50-goal men, they represent the average.

Average age they last played 60 games in a season: 33.3
Average age they last scored 40 goals in a season: 28.5
Average age they last scored 30 goals in a season: 30.5

For every Messier, there's a Cam Neely — and it's actually quite sad how many 50-goal scorers never even made it to age 30.

Eleven of the 69 former 50-goal men (16 per cent) played their last 60-game season before their 30th birthday.

Scoring by Age: An In-Depth Look [Dan Tolensky]
The Lucrative Long-Term Deal
[Dan Tolensky]

Labels: ,


At 2:48 a.m., January 15, 2008, Anonymous Anonymous said...

I see where you're coming from, but couldn't eras have something to say, here?

Did the high-scoring NHL survive for more than 15 years?

At 3:40 a.m., January 15, 2008, Anonymous David Johnson said...

Brendan Shanahan scored 40 goals in 2005-06 at age 37 so he'll be added to your list when he retires as well. Alfredsson should easily get 40 goals this year and could even get 50 at age 35. There are a number of players who are playing great hockey into their late 30's. Sakic, Alfredsson, and Sundin are still top 20 forwards late into their 30's and are probably better all round players than they were 10 years ago. Elite level players that can avoid chronic injuries generally age well. The Ovechkin contract is really a gamble on his health. I am just not sure the risk is worth the potential reward of having an top level player at a below fair value price 10-13 years from now.

At 4:26 a.m., January 15, 2008, Blogger Old said...

This deal depends upon the guys the Caps put around Ovechkin. If the Caps can't put a competitive team together, does any of it really matter? Isn't this just like the deal they gave Jagr writ larger and longer? They made the finals once with Jagr, and I expect the same with the Ovechkin.

At 6:53 a.m., January 15, 2008, Anonymous Geoff said...

Great work. You consider doing this with number of (100-point) seasons played instead of age?

I'd argue issue is not whether or not Ovechkin can have a 100-point season at 35 (plausible), but whether or not Ovechkin can have 16 straight 100-point seasons (less plausible, since this puts him in the elite HOF category). Because if he pulls a Hawerchuk in his last few years, there's no way he'll be a bargain, unless the cap is $100+ million. And if you're expecting to overpay him in the first year and the last few years of his deal, you better hope he's easily one of the league's cheapest superstars during his prime.

At 7:44 a.m., January 15, 2008, Anonymous Gerald said...

Good work, James.

You may recall that I posed a theory very much along the lines of what your data shows over on Tom Benjamin's site. My theory was posed in the context of the merits of the younger age for UFA's. I took a quick look at the more stellar scorers and point producers and, like you, found a number who flamed out early. It supported my contention at the time that the UFA age was likely accepted by the NHL very deliberately in the CBA negotiations, as it is almost universally true that elite level players have their best scoring seasons before age 26. As you observe, there are some outliers, but the exceptions prove the point.

You will find very much the same distriution if you consider 100 point seasons, you will find.

Probably, if someone did want to do an exhaustive analysis, they could assess elite players by comparing their scoring against league averages, in order to take out the effect of eras.

At 10:02 a.m., January 15, 2008, Blogger Oiler Mag said...

Hard to believe that 99 posted the last of his 40 goal seasons at the age of 30. Although he did have 12 in a row before that.

At 11:23 a.m., January 15, 2008, Anonymous sooooo good said...

hey Old, Jagr wasnt even close to being on the team when they went to the finals, they did have a 100 point season though if memory serves.
mirtle, its hard to compare with only players who are now retired because you are comparing eras. The year round fitness regimens of todays players are allowing for them to be more productive for longer(shanahan, selanne, lidstrom), and we are seeing more players hit their offensive 'stride' later in their careers (lecavalier, zetterberg, alfredsson). I understand that your analysis was being done strictly from a numbers point of view, and that you were just using probabilities and averages, fair enough. however, there is one other problem with doing that in this situation. nothing about alex ovechkin is average or normal. he 'breaks the mold' in every way imaginable in everything he does. he is AMAZING.
what is unfortunate is the feeling i get from you, that if he were to get injured and had to end his career prematurely or his production were to drop off dramatically, you would derive some sort of pleasure out of it just to prove your point. or is it just a canadian thing?

At 12:05 p.m., January 15, 2008, Blogger Doogie said...

That seems like kind of an unfair comment, SG. Actually, before I even came up with his countryman Pavel Bure, my first thought was Mike Bossy. True, AO "only" managed 46 last year, whereas Bossy had 50+ every year but his last, but Bossy also only lasted ten years before a back injury forced him out of the game. Given how Ovechkin plays, it's not out of the realm of possibility. See also Bure, Cam Neely, Peter Forsberg, and Bobby Orr, for just a small sample of Hall of Fame talents whose careers were shortened by injuries. That's the risk Leonsis is taking, that he might be Shanahan or Messier, but that he might also be Neely or Bossy.

At 12:08 p.m., January 15, 2008, Blogger James Mirtle said...

what is unfortunate is the feeling i get from you, that if he were to get injured and had to end his career prematurely or his production were to drop off dramatically, you would derive some sort of pleasure out of it just to prove your point.

I'm afraid that might be one of the dumbest comments I've ever had here.

The point is whether it's wise to sign a player to a 13-year contract or not, and I think the data shows there are long odds to beat in order to be productive to age 35. I'd hope every star could stay healthy and play well into his final years, but that's simply not the reality of the NHL.

Hopefully this shifts as we move into Ovechkin's era, but even then, there will be the Bures and LaFontaines who suffer injuries through no fault of their own.

At 12:35 p.m., January 15, 2008, Anonymous Anonymous said...

Good work, James.

Still, in determining whether Ovechkin's contract was wise - it would be interesting to plot the longevity of NHL'ers in general, not just 50-goal men.

Who knows, it might be revealed that a guy who scored 50 goals at age 22 is 300% more likely to still be playing by age 30, for instance. In terms of risk, betting on such a player might actually be one of the best risks out there.

But that would take a lot of work to graph out...

But I'm sure you're up to it! ;)

At 1:35 p.m., January 15, 2008, Anonymous Anonymous said...

Be careful with the numbers produced versus value argument as well. Many of your elite players take time to learn to sacrifice the numbers in order to achieve greater team success. For example, Yzerman's numbers went down as the winning went up.

The injury bug is still a valid argument and the reason I believe a long term injury to a forward is far more likely than to a goaltender. Again this make the DiPietro signing look like less and less of a risk. Supplemental insurance beyond the 7 year mark is the key. I would think that large market teams would have the advantage to absorb bought out or career ending contracts, but two small markets have set the precedence so far.

At 1:40 p.m., January 15, 2008, Blogger Art Vandelay said...

Nobody has mentioned monetary inflation, which accounts for most of the 5% increase in league revenues, year-over-year. At that rate, the cap in 13 years will be $95 million, even if the league continues to sell the same number of tickets, jerseys, beers and TV ads the entire time. OV's cap hit, over time, becomes a smaller percentage of team payroll. This contract is reminiscent of the long-term deal Teemu Selanne signed in Winnipeg which, by the end of it, he was making Lars-Erik Sjoberg money.
Brilliant deal for the Capitals. OV should fire his mom for being a horrible agent.

At 2:59 p.m., January 15, 2008, Anonymous Junior said...

Excellent and thoughtful work, Mr. Mirtle.

I share the view that a 13-year deal is too risky from the Caps' perspective in a contact,nay "collision" sport such as hockey. The obvious risks are exacerbated somewhat by AO's rambunctious style of play; so many guys who asked too much of their young bodies night in and night out never got a chance to finish out the string with prolonged periods of healthy production: Cam Neely, Wendel Clark, Eric Lindros, Owen Nolan (before he unretired - and again as soon as he retires again). A few commenters have mentioned improved contemporary conditioning and training regimes, suggesting that the players of today must inevitably therefore last longer than the players of yesterday; but don't forget that for every top conditioned scorer, there are now ten highly conditioned well-trained checkers using their improved muscle power to crash into guys like AO - when they can catch him.

It has to be remembered though that one can't evaluate the value or contribution of any player, even a goal-scoring virtuoso, purely on the numbers. He may not be scoring 35 at 35, but if he can still play, his mere presence may be one of those intangibles that are so important in a dressing room. The problem from the Caps' perspective is that he has to still be healthy to contribute even that... which brings us back to the notion of risk.

At the end of the day, Leonsis has taken an enormous gamble that the hockey gods smile with favour on his young phenom; for all hockey fans, I hope he's right. But that doesn't make the gamble any wiser.

At 12:41 p.m., January 16, 2008, Anonymous Anonymous said...

RE: "sacrifice the numbers in order to achieve greater team success. For example, Yzerman's numbers went down as the winning went up."

I luv Stevie Y, but... In 1988-89, Yzerman scored 155 pts (far & away the most on his team) but was criticized as “all-offense” & not a “leader” or “winner” because the Wings weren’t going far in the playoffs. By that time, Stevie Y had been in the league for 6 yrs & the rumours were that he was going to be traded so that the Wings could have a chance of re-loading & winning the Cup someday.

However, note that his +17 was 2nd best on the team that year.

In 1997-98 Yzerman only scored 69 pts in 75 games but was lauded as a “leader”, “2-way” player, “winner” etc. as the Wings won their 2nd straight cup.

However, his +3 was only 16th best on his team! He was actually below-avg in +/- compared to his teammates.

Yzerman was a far better & more dominant (at both even strength & in the p.p.) player in 1988-89 than in 1997-98. The difference in the team success was the improvement in his teammates.

Let’s see what A.O. can do with some good teammates.

At 6:38 p.m., January 16, 2008, Blogger The Falconer said...

I've crunched a lot of numbers and where I adjust for scoring environment and Mirtle's conclusions are fundamentally sound.

NHL players actually peak earlier than most fans (and perhaps GMs) realize and the career attrition rate starts to kick in earlier than many would suspect as well. Sure there are a small number of players that play (and play well) past age 33 but their numbers are very small and often limited to Hall of Fame level guys. We tend to remember the exceptions and forget the rules, but if I were an owner I'd pay attention to both.

Now Ovechkin certainly appears to be on a Hall of Fame career path but the injury of a career ending injury before this deal is finished is quite substantial--even modern medical technology can't do much for concussions or back pain.

At 4:23 a.m., January 22, 2008, Anonymous Anonymous said...

OK, everyone talks on risks.

But why do you think Ovi costs $10M per year ? This price comes from league's salary limit but not from market balance. I think his price is higher and 13years duration covers this difference.

So to my mind Leonis is ok about have Ovi playing for so many so he can - even if it will be only 5 years. I think he counts Ovechkin well deserves it. I think Leonis is ok to redeem this contract after some time.

On other hand for Ovi it is like insurance.

So both sides happy.


Post a Comment

<< Home


Free Page Rank Checker
eXTReMe Tracker