Does Moneyball Work in the NHL?

[hbk]

And of what use is that? The only thing that matters is how those picks turn out, not where experts (on either side) have them ranked. Wheeler turned out just fine, as an example. Turris too.

The only thing such a measurement would be useful in calculating is if your GM is getting relative value at the picks. You certainly wouldn't want DM taking the consensus 30th guy at 13, even if that player turns out to be great. He should have traded down in that scenario.

Beyond that, everyone is stabbing in the dark. Some are better at it than others, but nobody is consistently money at the draft. Where teams put distance on others is really in development - it's something that's controllable. The Coyotes have historically sucked ass at this and it has only improved marginally since Maloney took over.

It tells you if our scouting staff sucks. If we drafted 12th and we fnished 15th in the ranking I would likely go yeah average. but when you draft on average 12th (and this is since 2010) and we finish at or near the bottom third I'm saying yeah we can make the same conclusion that Maloney did when he elected to remove Keith Gretzky from the organization. This board didn't need that metric though. We knew by out lack of 2nd rounders making the squad and lack of homegrown talent in general for a long stretch.

Another metric I've seen tossed around is number of players drafted each year that played more than 200 games at the NHL level. Again very general and way too much of a Monday morning quarterback stat but it essentially equates out that you need to average between 2-3 players from each year to be considered "elite".

 

[rt]

Wait. I think we've lost the plot. The point of this thread was that the coyotes are doing as well as can be expected for a poor team, right? I disagree.

 

[XX]

It tells you if our scouting staff sucks.

I don't agree. You could be drafting wildly out of position but if you're picking up all-stars, who cares? Draft "experts" are not actually experts. If they were, there'd be no misses in the draft. Everyone would be neatly aligned by maximum potential and their position.

The draft is the furthest thing from that. They're experts in consensus opinions, nothing more.

 

[Tom Polakis]

Hello. Long time lurker here. I felt like I had to interject. There is not a clear line in this data. The simple fit line is a bit misleading. Most of the data points are clumped together...likely due to the salary cap and floor. You should test for correlation which can be done using free excel add ins.

For a moneyball type analysis, I think you need to look at how specific player stats correlate to wins or points in the standings. I think salary is too general for this to be effective. You can figure out player value once you find the strongest correlations.

Agreed that it's not a tight curve fit, particularly for the Eastern Conference plot. The Western Conference data shows a clear trend, however, whether or not you want to agree with the slope of the regression. My point of including the simplistic line was only to see what the outliers looked like, and how the Coyotes fit in. At least for last season, we were in the middle of the pack for spending, and not grossly over- or under-achieving.

I think the intent of MIG's original post in the thread was not to ask about Moneyball correlations with specific stats. Rather, he was asking about whether cash correlates with winning, and if so, if the Coyotes can be taken seriously as a contender. The data shows that spending does loosely correlate with winning, even in a sport with a narrow salary range.

 

[hbk]

I don't agree. You could be drafting wildly out of position but if you're picking up all-stars, who cares? Draft "experts" are not actually experts. If they were, there'd be no misses in the draft. Everyone would be neatly aligned by maximum potential and their position.

The draft is the furthest thing from that. They're experts in consensus opinions, nothing more.

Is that what we did though? What you are saying is there is special cause variation which can explain the statistic. That I agree with.

 

[MIGs Dog]

Wait. I think we've lost the plot. The point of this thread was that the coyotes are doing as well as can be expected for a poor team, right? I disagree.

Sort of. My point was that if the owner says the goal is to win the Cup, then you must spend the money required to put a contender on the ice. Doesn't mean you need to spend the most, but if you are bottom 1/3 in salary, can you really expect top 1/3 results?

 

[DomiToDuclair]

Sort of. My point was that if the owner says the goal is to win the Cup, then you must spend the money required to put a contender on the ice. Doesn't mean you need to spend the most, but if you are bottom 1/3 in salary, can you really expect top 1/3 results?

I think that it's sort of silly to say we should be spending more - what would we spend it on? Free agents, right? There's certainly some people I would be interested in this year (so far at least) but it's quite likely we lose some of those options due to re-signing, and I don't really think we're in a position to spend to the cap right now either.

Maybe we can revisit this when some of our prospects come into the league, and our current contracts expire.

 

[MIGs Dog]

I think that it's sort of silly to say we should be spending more - what would we spend it on? Free agents, right?

Exhibit A would be Verby. How much more did we need to resign him, $2 mil?

I know it's not something you can create instantly, but if you consistently spend in the bottom third of the league, what is the message you are sending to your fans and players?

How much did they spend on stadium renovations? Do more people attend because they can now get Tim Hortons? I think people are more likley to attend if you win.

 

[DomiToDuclair]

Exhibit A would be Verby. How much more did we need to resign him, $2 mil?

I know it's not something you can create instantly, but if you consistently spend in the bottom third of the league, what is the message you are sending to your fans and players?

How much did they spend on stadium renovations? Do more people attend because they can now get Tim Hortons? I think people are more likley to attend if you win.

True, and I do wish we had signed him, and would be ok with his current deal on our team, and I think the majority of fans would be too.

But there's a large difference between 2m, and 10m.

And again...I don't see any point in spending to the cap when it probably doesn't really fit our team right now anyways.

 

[rt]

True, and I do wish we had signed him, and would be ok with his current deal on our team, and I think the majority of fans would be too.

But there's a large difference between 2m, and 10m.

And again...I don't see any point in spending to the cap when it probably doesn't really fit our team right now anyways.

Oh I don't know. I think money doesn't hurt. I'd have been thrilled to add Grabovksi, Kulemin, Leddy, and Boychuk, for example. Just like the Isles did.

 

[DomiToDuclair]

Oh I don't know. I think money doesn't hurt. I'd have been thrilled to add Grabovksi, Kulemin, Leddy, and Boychuk, for example. Just like the Isles did.

Yeah sure, but that's more than we have room for no?

And even if we did we wouldn't be cup contenders by any means.

 

[jacobhockey13]

I'd love to have enough money where we can keep our two-way guys up and send our one-way scrubs down

Edit: Yes, I intentionally spelt "skrubs" like that

 

[Guest]

I don't think Moneyball works in hockey because hockey requires players to have several key skills/characteristics. It's not often that these skills go unnoticed to rot in the minors or underutilized on other teams to be able to take advantage of that. I may be wrong, but isn't Moneyball based on taking players that may have been overlooked for specific skills and throwing those individuals together to make a sum larger as a whole? In hockey as an example, that would be like taking someone who is really good at faceoffs, someone who is really fast skating, someone who is really hard hitting, someone who is really good as passing, etc. The sum of those parts don't work for hockey because the skills have to mix together due to the level of team interaction.

 

[rt]

Yeah sure, but that's more than we have room for no?

Sure we would. We could stash deadweight in the minors because we have money in this scenario. Anyway, I wouldn't actually want all of those guys for the prices paid. But if we had the cash to keep Vrbata on that two year deal, and pick up Boychuk via trade, I think we'd be in a lot better shape.

 

[XX]

I may be wrong, but isn't Moneyball based on taking players that may have been overlooked for specific skills and throwing those individuals together to make a sum larger as a whole?

It wasn't about specialization at all. Just the opposite. It looked for a baseline of production from guys, regardless of how old they are, how they got on base, their hitting style etc... They also shifted their approach to things like stealing bases and how bunts are fielded. "No more stealing bases. I pay you to get on first, not get thrown out at second" is a good quote from the movie.

It basically was a mathematical approach to playing the averages. Assemble enough misfit toys that are strong mathematical players for whatever reason and the wins should come. It's important to note that the early success was largely due in part to their being a lot of guys shunned by traditional baseball out there that were still productive players. The same thing doesn't occur in modern day sports, as nearly everyone is using analytics.

The Kings have roughly taken this approach by emphasizing Corsi. Hockey is a different beast from a statistical standpoint. Each at bat in baseball is a single game or event in statistics, with almost everything measurable. Hockey flows constantly, so analysts have taken to stats that roughly measure how much of that flow is being directed at the other team by players. It's far from a perfect science, or even the applicability of baseball.