Is there a 'Bill James' of hockey?

[Jim Carey Price]

Help me out here, is there anyone that you would consider to be a Bill James in the hockey world?

If you are unaware of Bill James, he wrote Baseball Almanacs in the 1980s...he had a cult-like following of baseball fans after he was able to challenge many of the ways the MLB kept stats and backed up his ideas with better statistical analysis.

Can any of you think of someone out there who might fit this type of person, being a big time 'stats guy' to analyse the NHL?

Click to read more...

 

[Bear of Bad News]

Iain Fyffe might be the best out there at this point - he's done some very innovative work.

http://www.puckerings.com/index.html

 

[Resolute]

There is no Bill James of hockey. Baseball is a unique sport in that almost everything can be broken down into statistical components, and that stats can be a generally reliable means of indicating future performance (defensive stats notwithstanding). There is far too much subjectivity in most sports, especially hockey, that is impossible to quantify. Not to mention that some of what is quantified (ie: +/-) can have dubious value at the best of times.

 

[Jim Carey Price]

There is no Bill James of hockey. Baseball is a unique sport in that almost everything can be broken down into statistical components, and that stats can be a generally reliable means of indicating future performance (defensive stats notwithstanding). There is far too much subjectivity in most sports, especially hockey, that is impossible to quantify. Not to mention that some of what is quantified (ie: +/-) can have dubious value at the best of times.

What James did was challenge the subjectivity of certain baseball stats...then came up with some better ones. Sure baseball is a more 'stats' type sport, but the way James looked at things could be applied to hockey.

 

[joe_shannon_1983*]

I hear that Ron Wilson is big on using statistics and hockey-style sabermetrics.

A real number cruncher.

 

[Bear of Bad News]

There is no Bill James of hockey. Baseball is a unique sport in that almost everything can be broken down into statistical components, and that stats can be a generally reliable means of indicating future performance (defensive stats notwithstanding). There is far too much subjectivity in most sports, especially hockey, that is impossible to quantify. Not to mention that some of what is quantified (ie: +/-) can have dubious value at the best of times.

Actually, many people felt the same way about baseball until James and others challenged that line of reasoning.

If you consider American football to be a game which can't be quantified, you should read Palmer, Carroll and Thorn's "The Hidden Game of Football".

Many of us are currently doing the same thing with hockey.

 

[Tricolore#20]

I hear that Ron Wilson is big on using statistics and hockey-style sabermetrics.

A real number cruncher.

I've heard the same thing about Ron Wilson. I read somewhere too that the Sharks use some sort of state of the art technology, that was developed in Silicon Valley. I'm not sure if it was some statistical engine, or something completely different.

I like the fact that the Sharks did so well during the last NHL season, with so little. I kind of pull for them to succeed simply to encourage other teams to look at the game more critically and less through stereotyped visions as how the game should be.

The guys over at the Oilers board (igor, etc...) do a remarkable job in doing statistical analysis. Check it out, it is an enlightening read. I really do wonder if anybody other than Wilson applies the same philosophy.

 

[KH1]

Who cares. Bill James is an overrated moron--because he is so hung up on "sabermetric" stats his player ratings are at times just plain stupid (I'll try to dig up my copy of the Abstract and find some examples.)

 

[grapeshine]

Who cares. Bill James is an overrated moron--because he is so hung up on "sabermetric" stats his player ratings are at times just plain stupid (I'll try to dig up my copy of the Abstract and find some examples.)

What? He practically revolutionized the way stats are maintained and interpreted. Disagree with his methods all you like, but I don't know how you can justify calling him a moron.

 

[A Good Flying Bird*]

What? He practically revolutionized the way stats are maintained and interpreted. Disagree with his methods all you like, but I don't know how you can justify calling him a moron.

It's easy to call him a moron and easily justifiable.

It's fine to look at the stats and break them down and create new ones and all that.

But in the end, they aren't the be all and end all.

Not by a long shot. And sometimes, James got too hung up on his numbers.

 

[19nazzy]

I'd like to see someone come up with a valid indicator of goalie stats.
GAA/S% are too dependent on team play IMO

 

[habs9]

See Goalies World for stats

Goalies world magazine and site has some innovative stats on goalies and rankings. And of course the magazine breaks down exactly how a player plays which cannot be seen in stats.

Stats are good but seeing is believing. You could tell Crosby is an elite player now even if he had no points just by some of the things he does on the ice.

 

[Sens Rule]

The real question is who is more valuable statistically for his career -Ozzie Smith or Bob Gainey?

 

[Egil]

One quick stat adjustment that should be done is PK% and PP%. Currently, short handed goals have no impact on your PK% or PP%, which is dumb!

So I propose that PP% be calculated as follows (PPGF - SHGA)/(#PP oportunities). PK% should also be modified to (PPGA - SHGF)/(#PK oportunities).

Further adjustments could also be made for the actual time SH or on the PP. Why does the 5 second powerplay count the same as the 2:00 minute one, or the 5 minute one? That gets far more complicated though, unlike taking short handed goals into acount.

 

[stanley]

There is no Bill James of hockey. Baseball is a unique sport in that almost everything can be broken down into statistical components, and that stats can be a generally reliable means of indicating future performance (defensive stats notwithstanding). There is far too much subjectivity in most sports, especially hockey, that is impossible to quantify. Not to mention that some of what is quantified (ie: +/-) can have dubious value at the best of times.

To add to this, baseball has the benefit of having a high number of zero-sum situations and long, complete data record. Apart from the newly implemented shootout, hockey has practically no zero-sum situations and therefore, rather difficult to somewhere near impossible to accurately debit or credit individual players and subsequently, assign value.

There's really no debate to be had here. You don't get split peas when you open a can of chili, and you don't expect an objective data output from subjective input.

 

[Bear of Bad News]

It's easy to call him a moron and easily justifiable.

Too easy, apparently.

And, for what it's worth, James never claimed that his methods were the "be all and end all". If you want to call others morons for ascribing those claims to his work, then have at it.

 

[Injektilo]

I've heard roger neilson was really into statistical analysis long before anyone else was.


If anyone's interested in reading some stuff, some of the links i vist are :
The Hockey Project , Puckerings and Hockey Analysis.


You can also go to the "library" section of the Oilers board on here and look at some posts by either Igor or Mudcrutch, they used to post here alot more often and had some really interesting things to look it.




However, if you think stats are of the devil and that your pure, unbiased opinion is the be all and end all, then you probably shoudln't bother.

 

[Krusher55]

Stats are good but seeing is believing. You could tell Crosby is an elite player now even if he had no points just by some of the things he does on the ice.

Yes and no. If you just go by some of the things he does, Alexei Kovalev would be a superstar in the league. Ask most hockey people and they will say he has as much pure skill as almost anyone in the league. But he's just a 30 goal scorer. A good player, but not a superstar. Stats are what you get when you combine raw skill and work ethic, passion, dedication, endurance, and so many other factors.

David
HockeyAnalysis.com

 

[DownFromNJ]

Statistical analysis cannot work in hockey as it has in baseball. In baseball, every pitch is a perfectly recorded isolated event. We know everything about every pitch. Each player on the field only has one objective. For the hitter and runners on base, that objective is to score a run. For the pitcher and players on defense, that objective is to prevent runs from scoring. A single is always worth a single. A double is always worth a double. A home run is always worth a home run. Not only that, but the baseball season is so long that players compile a huge sample size of information: over 700 plate appearances.

It is not the same for hockey. If Grant Marshall hits Jagr, causing a turnover, and Scott Gomez grabs the puck, and passes it to Brian Gionta, who shoots and scores on Kevin Weekes, Gomez and Gionta get credit for the play but Marshall gets hardly any at all, and Weekes gets credit for surrendering a goal but Jagr hardly gets any at all.

When you are playing both offense and defense at the same time, and the defensive side of things is barely recorded, it is difficult if not impossible to do baseball-style statistical analysis.

And this is coming from one of the biggest baseball stat-nerds that you will ever meet.

 

[Bear of Bad News]

Agreed that it can never work as well in hockey as it can in baseball. But that doesn't mean that it's not worthwhile as a goal.

Bill James' goal for sabremetrics was to understand the sport of baseball better. His goal was not to have the televised baseball games entirely covered with statistics from one corner of the screen to the other.

There's no reason why we can't have the same goal for hockey - to understand the sport better. For instance, a lot of people still use goals-against average (or, heaven forbid, win-loss record) to compare one goaltender with another. In fact, even though saves and shots were recorded for every National Hockey League game, save percentage did not become an official NHL statistic until 1982-83.

And save percentage isn't perfect, either - unless you know the types of shots and the situations involved, you're going to get inaccuracies. That type of information is being tracked; we just have to figure out how to interpret it. The fact that it can't be as useful as it is in baseball shouldn't stop us.

 

[Jim Carey Price]

Statistical analysis cannot work in hockey as it has in baseball. A single is always worth a single. A double is always worth a double. A home run is always worth a home run. Not only that, but the baseball season is so long that players compile a huge sample size of information: over 700 plate appearances.

When you are playing both offense and defense at the same time, and the defensive side of things is barely recorded, it is difficult if not impossible to do baseball-style statistical analysis.

And this is coming from one of the biggest baseball stat-nerds that you will ever meet.

How is a double always a double? You are missing what Bill James has done. If a ball is always hit into "the gap" is it always a double? No...there are stats that can tell you that based on whereever a ball is hit it there is a x% chance it is a single, x% double, x% caught...etc.

The reason I was asking for this type of information is that I believe that in the hockey world, while yes, it is tougher to do than in the baseball world, are there stats that could explain more of the game? And some of the sites I have seen have shown that this is true, there are indeed people looking at this kind of stuff, which I think is great.

Baseball similarly has some of the same problems with the defensive side of the game being barely recorded or being too subjective...however some things have improved in that area since James as I understand...maybe someday hockey...there is more information contained with the game than we observe.

 

[Krusher55]

Statistical analysis cannot work in hockey as it has in baseball.

I do agree with this. Hockey is much more of a team game than baseball. But with that said, I do think that statistical analysis can provide some benefit.

It is not the same for hockey. If Grant Marshall hits Jagr, causing a turnover, and Scott Gomez grabs the puck, and passes it to Brian Gionta, who shoots and scores on Kevin Weekes, Gomez and Gionta get credit for the play but Marshall gets hardly any at all, and Weekes gets credit for surrendering a goal but Jagr hardly gets any at all.

Marshall gets a +1 and Jagr gets a -1. That is how you factor in their contribution to the outcome of that play. If Jagr can score 80 even strength points this year that should tell you he is a good offensive player. What if I also say that he is involed in 30% of his teams even strength goals and that as a team they score 15% more even strength goals than the average NHL team. Now we are starting to get an idea of how Jagr compares to his teammates, how his team compares to other teams, and thus with a bit more work how Jagr compares to other players on other teams (offensively). Now, what if he is also a -5. That will tell you a small bit about his defensive ability. But what if I also tell you that the average New York Ranger player is a +5. He produces a lot of offense but his "net" offense/defense rating is well below most of his teammates. Now we are really digging down into Jagr's effectiveness as a defensive player. At this point we should be able to combine his offensive ratings with his defensive ratings to get an overrall understanding of how good Jagr is as a complete player.

In a way hockey can potentially be more interesting to analyze statistically than baseball because so much more effort is required to factor out the "team" component. I believe it can be done, it just takes more work. And If I didn't have to work a real job to pay the bills, I'd dig into it much more. I'd really love to develop a player rating index of some sort.

(Note: all the numbers I just used are figments of my imagination and just being used to prove a point)

 

[Randall Graves*]

Who cares. Bill James is an overrated moron--because he is so hung up on "sabermetric" stats his player ratings are at times just plain stupid (I'll try to dig up my copy of the Abstract and find some examples.)

Yeah, lets keep evaluating guys based on batting average and errors.

 

[Bear of Bad News]

Who cares.

It should be pointed out that, in the future, you're allowed to skip threads that don't interest you. There won't be a test later.

 

[john g]

you can do anything you want with stats and make them out to be a result of any bizarre scenario you could imagine -- doesnt mean they do anything but if patterns are obvious then you are potentially on to something

i.e -- say you want to look at potential conditioning of players

from a wide view, you can statistically look at how players fair in the specific periods of the 2nd game, where the team plays back to back games.