By The NumbersHockey Analytics... the Final Frontier. Explore strange new worlds, to seek out new algorithms, to boldly go where no one has gone before.
Red Sox are not an example of Moneyball. Yes, Epstein used a similar approach at times, but he had a huge payroll to work with. Neither are the Mavericks. Burke is right a lot here. Stats are pretty useless in hockey.
To be fair, Burke led a team to a Cup not 20 years ago, but 5 years ago. He's justified in believing in his methods as they had success for him not all that long ago.
In his second quote, maybe he's trying to say that nobody ever won a championship purely with statistics, and that a measure of eyeballing and having an eye for talent is involved as well? I think hearing the entire discussion would provide some context, instead of just yanking a sentence out of what was probably a larger idea.
Burke knows his hockey. He's great at getting parts in place, even though he'll sometimes over trade to do it, but seems like he'll tell an owner to stuff it if he has to.
The proof is in the pudding. Look at the teams he's built. He's usually not around for the rewards but he's gotten major parts of some winning teams.
Do people realize that 'moneyball' doesn't mean winning without spending? It means using advanced statistical analysis to find players that excel in traditionally (or currently) undervalued areas.
Statistics will never be as effective in hockey as they are in baseball. Burke's right. There are too many variables for them to be anything near an exact science.
Statistics will never be as effective in hockey as they are in baseball. Burke's right. There are too many variables for them to be anything near an exact science.
Burke's not right, as most organizations use statistical analysis in their player evaluations.
What does this even mean? He built the team that is doing well.
Yeah. I'm not a big fan of Burke, but you can't credit the Leafs current success to his dismissal. This is still the team he built. There's been no significant additions made to the Leafs through trade or free agency since Burke was canned.
Statistical analysis plays a role in proper player evaluations, but it is possible to attach too much importance to it.
Let's be honest - Any Red Sox World Series victory over the past decade or two has more to do with them having a bigger payroll than 95% of the rest of MLB than it has to do with anything else. Moneyball might help, but it's not the key reason why the Red Sox have enjoyed whatever recent success they have.
Show me a few teams with average payrolls that turned into championship teams through the Moneyball philosophy, and then it'll be impressive.
Burke's not right, as most organizations use statistical analysis in their player evaluations.
Well, no organization wants to be left behind in case statistics can eventually provide some fruition. But look at hockey - every metric has so many caveats and every variable has ten variables. It's almost impossible. It moves too fast.
It's not completely useless, but it's not ever going to be reliably accurate to a significant degree.
Yeah. I'm not a big fan of Burke, but you can't credit the Leafs current success to his dismissal. This is still the team he built. There's been no significant additions made to the Leafs through trade or free agency since Burke was canned.
Statistical analysis plays a role in proper player evaluations, but it is possible to attach too much importance to it.
Let's be honest - Any Red Sox World Series victory over the past decade or two has more to do with them having a bigger payroll than 95% of the rest of MLB than it has to do with anything else. Moneyball might help, but it's not the key reason why the Red Sox have enjoyed whatever recent success they have.
Show me a few teams with average payrolls that turned into championship teams through the Moneyball philosophy, and then it'll be impressive.
As far as examples go. The 2003 Florida Marlins, 2002 Anaheim Angels finished just above middle of the pack in terms of payroll. The New England Patriots won 3 out of 4 Super Bowls from 2002-2006, all with a payroll in the lower half of the NFL.
The only difference between the Red Sox and a true Moneyball team is the fact that the Sox have the luxury of retaining the talent that they obtain via a Moneyball approach to building a team. Same thing with the Dallas Mavericks. They are still a Moneyball team, just because they have the luxury of a large payroll doesn't change that...
Last edited by tigermask48: 03-01-2013 at 01:21 PM.
Show me a few teams with average payrolls that turned into championship teams through the Moneyball philosophy, and then it'll be impressive.
New England Patriots
The point the A's made with Moneyball or the Tampa Rays with the extra 2% was that there were and still are pretty big market inefficiencies. Some players are vastly undervalued by the old guard. Some strategies aren't used because some coaches are way too afraid of going against the grain and going through the criticism in case of a backlash. For instance, going for it more often on 4th downs. We all know what happened with Belichick when that happened against the Colts.
The idea of moneyball isn't that stats will make any team win every year. The idea is objective analysis. You may expand the concept to the use of sports psychology, biomechanics, financial analysis, topology, computer models the list goes on and on.
Saying that stats don't work because the A's never won a championship is like saying, the old fashioned way doesn't work because Milbury tried it with the Islanders and failed miserably.
That said, even if you built a terrific team with advanced stats like the A's did, the odds of winning a championship are still way low. A .600 team playing a .540 team in the playoffs with homefield advantage would win roughly 61% of the time. Then in the 2nd round, the odds of a .600 team beating a .580 team with home-field advantage are 57.1%. Then the odds of a .600 team beating another .580 team with home-field advantage are again 57.1%.
Through probabilities, we know that the odds of running the table are the odds of beating the 1st team times the odds of beating the 2nd team times the odds of beating the third team. In our example, that would be .61 * .571 * .571 = 19.9%
So, odds are really stacked against you to win a title in baseball to begin with.
What the A's did was building the best team they could with the budget they had.
The idea of moneyball isn't that stats will make any team win every year. The idea is objective analysis. You may expand the concept to the use of sports psychology, biomechanics, financial analysis, topology, computer models the list goes on and on.
Right. It's less about payroll and more about using the concepts above to identify a career .200 avg. hitter that hits .375 in smaller ballparks, obtaining said player at below market value and putting him in a smaller ballpark for 81 games a year and hopefully reaping the rewards when he becomes .300 hitter over a season. (extreme example I know...)
It just happened to be largely developed by a lower payroll team out of necessity to survive in a league with teams with payrolls 4-5 times greater. It's more about mentality and less aobut payroll.
Well, no organization wants to be left behind in case statistics can eventually provide some fruition. But look at hockey - every metric has so many caveats and every variable has ten variables. It's almost impossible. It moves too fast.
It's not completely useless, but it's not ever going to be reliably accurate to a significant degree.
Hockey probably won't ever reach the point of baseball, however when the NHL begins to track puck touches, possession, passes completed, etc., we'll see a big revolution. That stuff will explain a lot of the game.
