Views
1269
Comments
23

 YODA was here - And during the 2008 BVOF Thanksgiving Feast Blogging Battle, he wrote a little about...

ABUSING STATS!

It's American Thanksgiving and I am pleased to thank an American member of FanNation for inspriring this blog.    Ironically, this person has certainly abused more than his share of stats over the past year - not the least of which is the statistical abuse that he has put on his own Throwdown record.  That's right, I am giving thanks to Saint Simons Island's finest son, Peachy.  

Recently Peachy started a TD titled, 'Make a new NFL Category or Stat'.   He came up with a good one, too.  I'll get to that a little later, but for now when it comes to stats, my mind semi-randomly wanders back to hockey and some particular definitions.  (Is this a case of the cerebral vortex that Bigalke speaks of?)

      

Hat Trick (hockey) When a player scores 3 goals in one game.  Fans acknowledge the feat by throwing their own hats from the stands onto the ice.

Gordie-Howe Hat Trick:  When an player scores a goal, an assist, AND gets into a fight in a single game. 

Earlier today, I joined a thread where somebody was insinuating that Wayne Gretzky is perhaps not on the same level as other elite NHL athletes such as Gordie Howe or Bobby Hull by stating, "I don't see Wayne scoring too many Gordie Howe Hat tricks for some reason."

Well, allow me to retort.  I don't see Gordie Howe scoring too many 'Gordie Howe Hat Tricks' for some reason.   The reason?   Because he DIDN'T.

      

In Gordie's 34 year pro career, he notched exactly TWO Gordie Howe Hat Tricks.    (Sidney Crosby already has one in his short 3 year NHL career.)   The term Gordie-Howe Hat Trick actually has NOTHING to do with Gordie's proficiency at achieving them.   Gordie hardly ever fought.  He only spent about 60 minutes a year in the penalty box.  The term is simply an honorific that is used to pay tribute to Gordie's ability to mix offensive production with tough agressive play.   Sure, it's a 'fun' statistic because of what it comprises (goal!  assist!  FIGHT!) but its rarity makes it almost wholly invalid in terms of actually analysing performance.  If a glam-stat like Gordie Howe Hat Tricks is used to gauge the best players, then Brenden Morrow and Shane Doan would be considered greater than Gordie himself.  

Obviously, the point is not whether Gordie is better than Wayne or vice-versa.  The point is (as BigAlke might say) that the spirit of debate demands more.  It humbly requests that one should be prepared to back up their statements with relevant information. 

Stats are routinely abused with respect to team sports.  This isn't unique to FanNation; the abuse runs rampant thoughout the sports media.    Here are just a few of the common pitfalls that we encounter when using sports statistics:

  • CASE 1: People assume stats apply to certain players without checking the facts.   
  • CASE 2: The wrong stat is given priority over a more valid one.
  • CASE 3: Stats aren't adjusted for variables such as era, venue, or league structure. 
  • CASE 4: The more meaningful stat has yet to be popularized. 
  • CASE 5: The 'right' stat has yet to be developed and thus isn't even tracked.

While 'CASE 5' isn't really a pitfall per se, it is the one that I find the most interesting.   Peachy's TD really got me thinking about what stats we do use and what stats we should use.  Finding a new way to analyze the game is an enticing endeavor.

 

 NFL

Peachy's idea for a new category is something he dubbed a 'Tip-Ception'.   If a QB throws a ball that is catchable, and it bounces off the receivers hands into the hands of the opposing team, the INT will no longer be credited against the QB.  Instead it will count against the receiver!   I can't stand the name 'Tip-Ception', but the idea makes perfect sense.   If a MLB shortstop makes an error, it isn't credited against the pitcher.   It properly penalizes the one who didn't do his job.  The same thing should apply in football.  A QB's TD-INT ratio and completion% would become a more accurate assessement of his performance. 

In the same thread, another member proposed that the NFL should start tracking 'Missed Tackles' - those tackles that should have been made but weren't followed though on properly.   Again, I wholeheartedly agree!  Tackling prowess should absolutely consider tackles made and tackles missed.  I'd much rather have the guy that has 78 tackles and 2 Missed-Tackles than the guy with 94 Tackles and 25 Missed- Tackles.  In MLB, they track saves and blown-saves.   Let's have the same thing in the NFL.  

   

 

 MLB

The science of statistical analysis has been applied to baseball more than any other sport.   As a result, newer and more accurate stats (in terms on judging overall performace) HAVE been developed.  And yet, the media continues to glorify the more tradtional and convenient ones.

When it comes to pitchers, I really don't care how many wins they have.  They have virtually NO control over how many runs their own team scores (in the American League, they have NONE).  I'd rather hear about their ERA, which they can control - but even then, one's ERA is misleading unless it's compared to the league average that year.  So give me *ERA+ and now I have something relevant that I can even compare across eras.  WHIP is a nice stat, too.  Wish they'd use it on TV. 

The same thing applies to hitting but perhaps to a slightly lesser extent.  RBIs, for example, are dependent on who is hitting in front of you!  *OPS+ is a nice stat that encompasses a mix of on-base % and power hitting to give an overall measure of productivity.   So again, why is this stat banned from the major networks? 

Are WHIP, *ERA+, and *OPS+ too difficult for the announcers to understand?  Is there a union issue here?   I really don't get it.  Even some of those in the position of voting on league awards still point to 'Wins' as the decisive reason they voted one way or the other.   Disgusting!

     

 

 NHL

Points are the most commonly referenced stat when ranking NHL players.  I, for one, have never liked the fact that a goal and an assist are given 'equal' value in terms of one's total 'points'.  Somebody that scores 22 goals and has 78 assists is hardly as valuable as someone that has 78 goals and 22 assists.  Goals are MUCH harder to get than assists.  Why?  Simple: Up to 2 assists are awarded per goal (An average of 1.6 assists are actually awarded per goal scored).   Therefore, for every 80 assists one notches, it only results in 50 team goals.  

Furthermore, a large proportion of the assists awarded are 'second-assists' (sometimes from weak dump passes) that happened to lead to two other people actually together the goal-scoring play.  Yet, if I make a perfect pass to a teammate and he makes a perfect pass back to me and I proceed to put the puck in the net, I am not awarded a 'second-assist' on my own goal.   It's bollocks!   In the NBA a 'second assist' is worth nothing and nobody complains.  In the NHL, a 'second-assist' is worth as much as a goal - a FULL point!   Give me a break.  My proposal is a happy medium.  Make 'second assists' worth only 0.5 points.   This isn't perfect but it's better than the current system.  Going forward, a Cam Neely-style sniper would be more properly recognized for his contribution to the team.

On another note, 'blocked shots' ARE now calculated in the NHL, but for some reason the stat just isn't given the attention it deserves.  Throwing your body in front of a puck that's travelling at 95 mph is not for the timid.  This is one of the most selfless and ballsy moves in all of sports!  There is an art to it that very few NHLers have mastered.  In the NBA, blocks are coveted and respected.   But it's the NHL shot-blockers that deserve the real praise!  They are putting themselves in harm's way and risking their bodies.  Let's get this stat moved up the statistical depth chart, pronto! 

 NBA

The main problem with the popular NBA stats is that they don't accurately measure one's defensive contribution.    The key role in the defensive end is to cover your man and prevent your opponents from scoring.   However, quantifying 'coverage' statistically is extremely difficult.   Rebounds, blocks, and steals are easy to measure but don't even come close to telling the full story about who's really doing the most to prevent opposition points. 

Over the past few days, I played with a number of mathematical scenarios in an attempt to quantify one's contribution to 'point-prevention'.   I went off on several theoretical tangents - but I got stuck when I realized that it would probably involve looking at every single player matchup during every minute of every game.    In other words, I have no idea where to start - and I'll thankfully leave this challenge for somebody else to work on.    There is an answer out there.

    

     

A Final Word on Stats...for now

I suppose we shouldn't be surprised that statistics are abused on a daily basis.  Especially when 85% of people rank themselves as above average in intelligence.  (Think about it)   The effective use of stats by sports fans at large is perhaps an ideal rather than an achievable goal.  Nevertheless, we can play a role in furthering the cause by endeavoring to use stats wisely in blogs, TDs, and the discussions we have in the various threads.  If we are willing to share and to listen, we can definitely learn more from each other on FanNation than from the national media. 

So on this day of giving thanks for one's blessings such as freedom, family, and friends, I'll also extend my thanks in the direction of another 'F' - to my fellow fans on FanNation.   This is our forum to rant and ramble about sports - and perhaps more importantly, a place where inspiration often comes at the most unexpected times from the most unexpected sources. 

So thank-you!  And I sincerely hope everyone had a 'Peachy' Thanksgiving.  

  

____

YODA is a NOT a freelance journalist, but has been a member of FanNation since May 2007. YODA is the founder of the Throwdown Zingers group which can be found here.  Like all Canadians, he is a die-hard NHL fan, however he has chosen to leave his homeland and is currently a personal trainer in San Diego.  Got something to say to YODA -- questions, comments, suggestions, derision to sling, vengeance to exact, commendations to render, accusations of byline plagiarism (BigAlke only), or contracts to offer?  You can reach YODA through FanMail, the comments box below, or in a galaxy far far away.

November 28, 2008  02:04 AM ET

nice blog...i agree most with the "tip-ception"

Since in the Super Bowl Eli threw an "interception" even though it bounced off the receiver's hands, and the defender caught it.

November 28, 2008  02:38 AM ET

This blog = awesomeness.

I effing hate it when people give more creedence to crap stats like RBI and runs when they're team-dependent and far less representative of a player's value than OBP, Slugging %, etc. Basically, there are 2 kinds of baseball people:

A: People that think Albert Pujols should have won the MVP award, and Howard was nowhere close.

B: People that think Howard should have won the award or finished second.

Group A gets it. Group B is full of idiots and/or ignoramuses.


As for the NFL, I wish you would have talked more about interceptions and cornerbacks. The people that don't get it think interceptions is the best barometer for judging the talent of a CB. These are the kind of people that have never even heard of Nnamdi Asomugha and think DeAngelo Hall is GREAT!


Regardless, great blog YODA!

November 28, 2008  02:53 AM ET

Thanks a lot Pitt. Cornerbacks were on my original list as well as NHL goalies. I was just afraid that if I included many more categories that people may get turned off by the length of the blog.
Some of the very best corners have relatively few interceptions.
1) They COVER their guys so well that QBs won't throw in that direction very often.
2) They choose to break up the play rather than going for the INT. 'Going for the INT' will more often than not result in the WR catching the ball on what could have been a preventable catch if approached correctly.

November 28, 2008  03:28 AM ET

Great blog Yoda personally I use many of those baseball stats you mentioned devising my strategy and draft order for fantasy baseball. Which I think is where the problem for the mainstream lies. Even though Fantasy sports are getting more and more popular the purists that still control baseball, the networks, and the media just don't get fantasy sports. Meanwhile the numbers crunchers who do play have created these stats that are far more accurate on a players performance than traditional stats, and they feel like outsiders are trying to invade their territory. Thats just my belief though.

November 28, 2008  03:37 AM ET
QUOTE(#4):

Thanks a lot Pitt. Cornerbacks were on my original list as well as NHL goalies. I was just afraid that if I included many more categories that people may get turned off by the length of the blog.Some of the very best corners have relatively few interceptions. 1) They COVER their guys so well that QBs won't throw in that direction very often. 2) They choose to break up the play rather than going for the INT. 'Going for the INT' will more often than not result in the WR catching the ball on what could have been a preventable catch if approached correctly.

The longer the better! :)

November 28, 2008  08:05 AM ET

Excellent! I enjoyed 98% of the blog. Would have been 100% if there was a picture of the Pens instead of Brodeur. ;)

November 28, 2008  08:12 AM ET

Excellent blog. It has always been mystifying to me why blocked shots don't get nearly enough publicity in the NHL. I mean, everyone glorifies them when they actually occur, but can people honestly name the league leaders without looking it up? For something so integral to the game, people sure ignore it.

My favourite stat there is the missed tackles one. As of now, there are very few NFL stats to highlight stupidity, and none of them are on the defensive end. This could end up becoming a vital stat, especially when deciding whether to give a multi-million dollar contract to a flashy, but fundamentally lazy, player.

November 28, 2008  09:54 AM ET

Excellent Excellent Blog my little green friend!

November 28, 2008  10:28 AM ET

Great work here, YODA... I must mention, though, that you blew your deadline... wonder what's going to come of that? ;-)

November 28, 2008  10:30 AM ET

Ya know why you had difficulty measuring the defensive effort of NBA players right? Because they don't play it! Hard to get enough evidence. Good blog. Way to point out the lack of recognition shot blocking in the NHL gets. A very important part of the game. Again good job.

November 28, 2008  11:03 AM ET

... and I see you're still loving on MY endline tag... THESE are the stats about which I'm most interested... times YODA cadges a blog styling of mine... times YODA defeats me with a post despite getting 'em in late... ;-)

November 28, 2008  11:04 AM ET

In all honesty, though, this really was an interesting read. I have long revered statistics, manipulating the numbers of any and every athletic contest... and I agree that far too often we take those statistics on face value and fail to delve deeper to realize their underlying meaning. Thanks for shedding some light on this issue, and may the next generation of statisticians be inspired by your words to take action in implementing the tracking of new stat categories in sports across the board...

November 28, 2008  05:38 PM ET
QUOTE(#9):

Great work here, YODA... I must mention, though, that you blew your deadline... wonder what's going to come of that? ;-)

I'm fighting against East Coast bias, Bigalke. I expect your unwavering support on this time-zone issue.

November 28, 2008  06:31 PM ET

Hat's off to you, YODA. As a professional statistician, I am constantly made aware of the misuse of statistics in all walks of life. Some don't know any better, others seem to use statistics as weapons rather than to understand.

I like many of the proposals for new statistical measures in sports. To be complete, there should be some way of comparing different measures of the same statistical measures (for example, which is more productive: a 20 goal and 10 assist forward or a 10 goad and 20 assist defensive player).

But you have said something important here which really needed to be said.

November 28, 2008  07:17 PM ET
QUOTE(#13):

I'm fighting against East Coast bias, Bigalke. I expect your unwavering support on this time-zone issue.

Hey... I've got no hard feelings... you, unlike the other two "competitors", actually got something in before midnight set on your part of the world... I'd rather have something to compete against than simply win by default...

November 28, 2008  11:17 PM ET

This is good stuff. Your Blog is good too Yoda. Ha! Love whip and also batting average against (for pitchers) and obs for hitters.

November 29, 2008  12:59 AM ET

Great blog Yoda. Your right, stats are...yeah.

November 29, 2008  03:10 AM ET
QUOTE(#14):

To be complete, there should be some way of comparing different measures of the same statistical measures (for example, which is more productive: a 20 goal and 10 assist forward or a 10 goad and 20 assist defensive player).

Thanks for your feedback, Appleseed.

There are some ways to do partially accomplish what you are proposing.

Like the *ERA+ stat in baseball, you could create an index (set at 100) of all forwards against all other forwards in the league and do the same for defensemen.
Offensive productivity could then measured as your % above or below the average person that plays your position rather than as an absolute figure. Having a Pts* score of 125 would mean that you are 25% more productive than your peers playing the same position.
A defenseman with 80 pts might have a Pts* score of 180 where a forward with 105 pts might only have a Pts* score of 155 so the comparison might be more valid.

November 29, 2008  03:13 AM ET
QUOTE(#4):

Even though Fantasy sports are getting more and more popular the purists that still control baseball, the networks, and the media just don't get fantasy sports. Meanwhile the numbers crunchers who do play have created these stats that are far more accurate on a players performance than traditional stats, and they feel like outsiders are trying to invade their territory. Thats just my belief though.

That's a VERY interesting thought, wildbill.
I was putting 100% of the blame on the purists not wanting to update/change. I hadn't really considered that perhaps those that DO understand the better stats like to keep their 'edge' to themselves to some degree rather than have all the good stuff go mainstream.

 
November 29, 2008  10:00 PM ET

Great blog, Yoda. I've questioned the accuracy of many stats in the past and welcome a more accurate version.Players should be asked what their opinions are to better the stats being kept. They stand to benefit the most from any changes.

Comment

Remember to keep your posts clean. Profanity will get filtered, and offensive comments will be removed.


Coming soon: Log in with your Facebook account, send comments and Throwdowns to Facebook and more.

Start Your Own Blog

Start Now

The Si.com Cover Hub Go to the Cover Hub

Stub Hub

The 2009 schedule has been released. Search for tickets!

Truth & Rumors

MOST POPULAR

  1. 1
    Sources: Big Ben having second thoughts
    Views
    6473
    Comments
    907
  2. 2
    Report: Halladay would welcome Yankees trade
    Views
    23582
    Comments
    465
  3. 3
    Have Irish already offered Kelly job?
    Views
    30526
    Comments
    70
  4. 4
    Stoops to take Notre Dame job?
    Views
    3370
    Comments
    55
  5. 5
    Report: Tiger's injuries inflicted by wife
    Views
    1094
    Comments
    54

Most Active Users

Comments + Blog Posts + Throwdowns

  1. 1
    Chico 2.0
  2. 2
    BBK - Now it's all football
  3. 3
    rstowe
  4. 4
    gigi_iv
  5. 5
    Never NDing Struggle

Message Boards

  1. NCAAF > General NCAAF

    Your team sucks
    Views
    777
    Replies
    35
  2. NFL > Pittsburgh Steelers

    STEELERS VS RAVENS …
    Views
    270
    Replies
    60
  3. NCAAF > General NCAAF

    CFB Game Day 11/28/2009
    Views
    207
    Replies
    39

Blogs

SI.com

Swimsuit

SI Photos