Sunday, April 06, 2008

Money can't buy you goals

I was working on something on this about a month ago, and never got around to posting it, but if you take a look at the impact unrestricted free agency has had on the scoring race this season, it'll surprise you. (Spector has a good look through the list of moves.)

Of the top 25 scorers in the NHL in 2007-08, not a single one was acquired as a UFA last summer. You've got to go all the way down to 28th and 37th to find Daniel Briere and Scott Gomez, bought up for ridiculous contracts that will pay them into old age, who both had underwhelming seasons in the 70-point range.

For all of the other big contracts given out last summer — Ryan Smyth, Paul Kariya, Chris Drury, Jason Blake, Michael Nylander, Sheldon Souray — very few translated to big numbers.

Looking at the top scorers this season, here's how it breaks down when it comes to how they were acquired:

Top 25 scorers
draft 68%, trade 24%, free agency 8%

Top 50 scorers
draft 58%, trade 28%, free agency 14%

In other words, only two players in the top 25 scorers this season, Martin St. Louis and Marc Savard, were acquired as unrestricted free agents. (And even then, St. Louis was signed for a song way back in 2000.)

And 17 of the top 25 scorers this season are still playing with the team that drafted them.

That's a group led by the likes of Alex Ovechkin and Evgeni Malkin, but there are players who have been around longer there, too, including Pavel Datsyuk, Daniel Alfredsson and Shane Doan.
The six traded players who sit among the leaders include Jarome Iginla, Joe Thornton, Alex Kovalev, Mike Ribeiro, Dany Heatley and Mats Sundin.

The lesson here is, if you're a team like Toronto or Vancouver that will be looking to add a scorer via free agency in the off-season, be prepared to (a) overpay and (b) receive dwindling production.

The UFA age may have come down considerably the past few years, but for the most part, the real stars are staying home to put up big numbers.



At 2:59 p.m., April 06, 2008, Blogger HockeyNutz said...

The analysis would get even more interesting if you fine tuned the 'traded' aspect of your analysis. For example you allocate Jarome Iginla into the Trade category, which is true, but he was traded before he was ever signed to an NHL contract.

Perhaps getting picky, but it further emphasizes the importance of developing prospects.

At 4:02 p.m., April 06, 2008, Blogger Dennis Prouse said...

Great post James. I had suspected that this was the case, and your smart research now confirms it. What you have seen in the last year or two is teams locking up their own key guys, thus preventing them from hitting the streets as either an RFA or a UFA. As a result, the only guys to hit free agency are, by and large, guys on the downside. If you look at the pending pool of free agents, there is precious little out there for the Canucks, no matter how much they want to spend. I guess they could go hard after Hossa, but so will about half a dozen other teams, and he's about it for legit snipers this summer.

At 5:10 p.m., April 06, 2008, Blogger Darren Barefoot said...

Way to kick us Canuck fans when we're down. What's left? I guess I have to hang my hopes on the narrow shoulders of Mason Raymond. Oy.

At 6:02 p.m., April 16, 2008, Blogger DAN THE STAT MAN said... have great blog.
However, this post shows the dangers of simple/quick analysis of statistics. 1) it is simply far too soon to judge the results of free agency. - with too small a sample number. considering contracts for the best players are usually 3+ years long There hasn't even been a cycle yet of free agents to fairly judge. There are many other plausible reasons for the descrepancies. 1.There has been a high number of young elite players
lately. Crosby Malkin, Kolvachuck etc. This is not always the case.
2.New players can take a year or two to fit in with a new team. i.e Briere, Arnott, Gomez Drury.
3. Why cut off at top 25 this is an arbartrary selection. (example Gomez is 5 points behind 25th player ? ARNOTT is 3 points so therefore they are not successful pick ups??
4. As we are clearly seeing in this years' playoffs there can be a big difference between reg. season scoring and playoff scoring
Briere for example leads all scores over the last three years in playoffs (currently 40points and counting!)
5. The Ducks last year had 5 free agents who made MAJOR contributions Selanne, Macdonald, Kuntz, Penner,Neidimeyer
5a.The Ducks last year had 2 players in top 30 and three in to 49 of scoring (all were signed as free agent)

Competent statistical analysis
is a high level skill that takes years of experience and patience (just like every other expertise( i, myself, have spent many hours learning this craft)
BEFORE one can draw any MEANINGFUL conclusions...
I suggest anyone interested in this topic pick up a university level statistics/philosophy of logic texts or explore some of the great statistical sites on the web...

Also, this is a classic case of having a subjective opinion
(i.e. free agent signings are
not helpful..THEN seeking out statistics to SUPPORT this position..IT's clear of the inherent bias in this common approach..The better way is to examine the statistics IN DEPTH
challanging positions along the way using a variety of tests and appropriate samples and then let the well-tested statistics themselves reveal the results.

TO quickly dismiss the value of siging free agents is simply a misinformed and ridiculious position

'quick analysis' like this post just reinforces the"Statistics are for losers" often quoted

At 6:08 p.m., April 16, 2008, Blogger James Mirtle said...

Dan, anyone who reads this site regularly knows I don't take a 'statistics are for losers' approach. This post was merely a quick look at the impact drafted players are having on the scoring race this season.

At 6:43 p.m., April 16, 2008, Blogger DAN THE STAT MAN said...

James...thats not my point.I really enjoy your blog and this is due in large part to you expertise - you are a great writer/reporter this is a skill that shows and you have crafted over time.
Statistical analysis is a similar expertise developed over time..My problem is not with your 'quick look' which is interesting but with the CONCLUSION you drew in your post"

"There's an awful lot of franchises and fans that are hoping to build through free agency, but it simply doesn't work."

Now this is a VERY strong conclusion to draw from a 'quick look' and I feel it is potentially flawed at the very least unsupported and I gave a few valid reasons why i think this is so.

Also, this is precisely the common mistake intelligent well meaning observers make time and time again ..(believe me i've made the same mistake repeatedly before
i understood statistical analysis

from my expertise I simply was offering a warning of the perils of making such strong all-encompassing CONCLUSIONS like the one in your post from 'quick looks' like the one you did..



Post a Comment

<< Home

Links to this post:

Create a Link


Free Page Rank Checker
eXTReMe Tracker