06 May 2012

ESPN Sucks: Faux-Moneyball Analysis

Friday's Puck Headlines at Puck Daddy contained an inexplicably stupid nugget from ESPN:

NHL fans lost an entire season due to hard-line owners seeking a tight salary cap. So it’s shocking that since the empty 2004-05 season, payroll is linked even more with winning. Before the lockout, a 10% increase in spending was worth about 5.8 team points (roughly three wins) over a season. Since the lockout, that number has ballooned to 9.2 points. The Wharton researchers theorize that this counterintuitive trend is a result of the CBA’s producing a tighter range of spending between teams. “Each dollar became that much more valuable,” they concluded.  
Hurricanes GM Jim Rutherford agrees: “Despite not having as big a gap, $16 million between the cap and the floor, teams that consistently spend at the top will still have an advantage in getting top players.” He also says low-payroll teams can succeed only for a short period. Nashville earned the Western Conference’s fourth seed this year with a slightly below-average payroll. But the team is winning on borrowed time. “The Predators are at that point where they’re either going to spend toward the cap or risk losing top young players,” Rutherford says. “So from a consistency basis, you can see the advantage for the teams that are able to routinely spend at the upper end.” 

The portion of the piece discussing hockey is quoted above, and there are several things that stick out as being blatantly and stupidly wrong just on the first read.  I have no idea why Wyshynski would link to such an awful piece, but hey, I guess it gives me something to talk about.

First of all, it's not shocking that payroll should be linked even more with winning (if it even is - I'll get there in a minute) - part of the idea of cost certainty was that you had limited dollars to give out, so the way you used those dollars becomes more important.  Furthermore, you couldn't make mistakes and pay your way out of them (which notably ended up not being the case - ask Wade Redden!).  The idea was that across the board, no one would have to spend as much to try to compete, but cost certainty undeniably makes the money that you do spend more important.

It's also worth pointing out that frequently, the floor or near-floor teams are not just victims of circumstance - they're quite often in the poor house because they build shitty teams, and then bemoan the fact that no one comes to their games.  Consequently, that more money tends to mean more wins should strike no one as a surprise.

The second thing that immediately jumps out: Nashville has succeeded for a lot longer than a "short period".  Nashville has been one of the 5 most successful teams in the regular season since the lockout, along with New Jersey, Vancouver, San Jose, and Detroit.  You'll notice that Nashville has also done so on a much smaller budget than those other four teams.  Rutherford's Hurricanes are 17th, though I suppose you can point out that they lucked into a Stanley Cup in the strangest year ever.

In any case, Nashville has clearly outperformed at least 80% of the teams in the league, and have done it on a shoestring budget.  They may never have finished #1 in the conference or something, but they've been remarkably consistent about remaining in playoff contention, and that's more than you can say for the 'Canes.

The third point is the fact that the comparison from the article of "increasing payroll by 10%" is highly unequal between the two time periods as well.  If you ran a cap-floor team in 2012, you had to spend around $48,000,000, and the cap was around $64,000,000.  You can only increase your payroll by 10% three times, so the entire difference between the top and the bottom of the league is condensed into only three such payroll jumps (and magnified further by the increased number of points in the league - getting there in just a second).  Prior to the lockout, team spending ranged anywhere from under $20,000,000 to over $70,000,000.  A payroll increase of 10% for a team at the bottom of the pay scale still left you in the bottom of the pay scale, and you could increase your payroll by 10% many more times.  The method of comparison is incredibly disingenuous and misleading.

The fourth and final point is the reason why this piece should've never been written, and betrays the fact that those doing the research for the piece didn't know anything about the NHL.  The reason you get more points per million dollars spent in today's NHL compared to the pre-lockout NHL is because everyone gets more points, no matter how much they spend!  The advent of the charity point for losing in overtime/shootout means that everyone gets a bunch of extra points each year.  Any measure in which you compare points earned under the current CBA, to the pre-2005 CBA, will always show that there are more points up for grabs under the current CBA, because there are so many more points awarded to NHL teams nowadays.

A quick check of hockey-reference.com shows that in 2004, the league average for standings points was 87, whereas in 2012, it was 92.  That's about 150 extra standings points in the system, that weren't originally there.  And remember, there were charity points from 2000-2004 for overtime losses, though not nearly as many as we currently have.  If ESPN was including numbers in the pre-lockout points/payroll figure from pre-2000, that would make the difference look even bigger, because in 1998-99, the league average for standings points was just 82 points.  Is it any wonder that you should see an increase in standings points earned by nearly any measure?

Anyways, the important thing to remember here is that ESPN sucks.



No comments:

Post a Comment