Rumor: Leafs significantly expanding their analytics department

[Bluelines]

Yes, that's how they practiced today.

Kadri on the wing? F off... I've been saying that they should do this for a few months now.... that is awesome.

 

[WTFMAN99]

I would have been right there with you a couple weeks ago, but I think I disagree now. Compared to last year, we probably have more points if Sparks plays like even a decent backup goalie (instead of one of the worst). Last year, we won a lot of our 1 goal games, while this year we've lost a lot of them. Now you might think this says something about the teams "mental strength" but realistically, a lot of it just comes down to luck as one goal games aren't really a repeatable thing. We're a better possession team, and have a higher xGF% this year than last year. Our special teams are worse (though our PK actually does a better job a limiting shots against this year, really it's just the goal tending isn't the same as last). We've also dealt with more injuries this year than last. Is it enough to beat Boston? Idk, but I do think this team has improved this year over last.

On paper adding Tavares, Muzzin and the internal improvements of Dermott, Johnsson Kapanen and even having Kadri for the full series should offset Komarov, Polak, Martin(?), Bozak and JVR.

Our defensive zone coverage and break outs haven't been all that good though, and it wasn't good last year either.

 

[Mess]

Not my line but I had saw the tweet earlier and it was perfect IMO.

The team has had a ton of individual great success stories this season but as a team, they've looked disjointed, they've been soft, and just haven't seemed to gel as a team. The success is less than what you would expect given the sum of the parts.

Problem being +$50 mil on extra analytics can't accurately predict goals behind the icing line by Sparks nor Stamkos scoring SH on a breakaway etc etc that result in team losses.

Teams results not directly connected to individual player stats in determining success.

Case in point: 3 of the top 5 NHL scorers all with 100 points or more in McDavid #2 116 and Draisaitl #4 104 points have combined for 90 goals and Oilers are bottom 5 team. Patty Kane is #3 OA with 107 points, with Hawks golfing next week. Check out their individual advanced stats and analytics, compared to their team performances.

Marner is a young poor man's Patty Kane, and Matthews will be chasing McDavid in points for almost the same Cap hit and JT is #3OA with 47 goals just behind Draisaitl with 49. Analytics will suggest great individual results all around, hopefully team results don't face the same bitter ending when Leafs 3 X forwards = Kane, McDavid and Draisaitl combined and NON playoffs.

Player analytics might be able to predict top individual scoring leaders but that doesn't prevent bottom 10 NHL team finishes. As seen with 3 of the top 5 NHL scorers there is almost an inversely proportional team result. If the goal of using advanced stats and analytics is to increase team winning% than teams must be using them wrong because its clearly not working that way.

 

[RLF]

Sure it has.

Teams realize that having a guy play 3 minutes a night while, small, speedy, skilled players like Moore sit in the minors is a stupid strategy.

The league is so fast because teams realized through analytics, speed and skill trump size and physicality. There is a reason why the league is getting smaller, faster and more skilled because analytics tells us its the best route to take.

No, teams don't want to get an instigator penalty for having a goon start a fight and now have to kill off a penalty. This was a deterrent to fight, exactly what the NHL wanted. If you take a look at the reduction in fighting, it started with the rule changes. The same with the progression to speed and skill. This is also what the NHL wanted to move toward. The end of the goon and the move to speed and skill were pushed by the NHL. They even said this is the direction they wanted to go. To say these changes are due to analytics is a massive reach.

What do you think would happen if the League got rid of the instigator fighting rules and started to allow the clutch and grab again and announced they wanted more physicality in the game? Think teams would still build quick, skilled teams with smaller players who don't want to play physical?

 

[Menzinger]

They don't use analytics?
What other teams don't?

The Leafs 2004-2014.

 

[Menzinger]

I don’t know why they have to expand the analytics department.

If HF Boards has taught us anything it’s that when someone uses publicly available stats to defend or critique a player.... that’s that. Definitive.

If you question the interpretation or the validity of the stat in question, you simple “don’t get the new NHL.”

So... why the need to expand? Just get a list of blogs, look at them and then pretend to be an irrefutable expert. That’s what HF Boards does.

Public stats gives you better insight than when you don’t use them.

Public stats + proprietary stats will lead to better analysis than either of those other options (providing your proprietary ones actually mean anything)

 

[WTFMAN99]

Problem being +$50 mil on extra analytics can't accurately predict goals behind the icing line by Sparks nor Stamkos scoring SH on a breakaway etc etc that result in team losses.

Teams results not directly connected to individual player stats in determining success.

Case in point: 3 of the top 5 NHL scorers all with 100 points or more in McDavid #2 116 and Draisaitl #4 104 points have combined for 90 goals and Oilers are bottom 5 team. Patty Kane is #3 OA with 107 points, with Hawks golfing next week. Check out their individual advanced stats and analytics, compared to their team performances.

Marner is a young poor man's Patty Kane, and Matthews will be chasing McDavid in points for almost the same Cap hit and JT is #3OA with 47 goals just behind Draisaitl with 49. Analytics will suggest great individual results all around, hopefully team results don't face the same bitter ending when Leafs 3 X forwards = Kane, McDavid and Draisaitl combined and NON playoffs.

Player analytics might be able to predict top individual scoring leaders but that doesn't prevent bottom 10 NHL team finishes. As seen with 3 of the top 5 NHL scorers there is almost an inversely proportional team result. If the goal of using advanced stats and analytics is to increase team winning% than teams must be using them wrong because its clearly not working that way.

This is all accurate, I think Dubas has tried to replicate Tampa to a lesser degree with having skill etc throughout the line up but we seem to lack....the system? tenacity? ability? grit? that Tampa Bay possesses.

I won't claim to have the answers here but I see a lot of good things with the team but there is still a lot I would want to change.

 

[rumman]

The interesting thing about ADV stats is that there are typically 5 guys IIRC at each game compiling the stats and using their judgement via the eye test on what to record. Each game is recorded by different people at each arena with a different opinion on what counts and what doesn't, using their own judgement and interpretation of what does and doesn't fit the criteria. So in every arena, the results are different based on the people recording the data.

The irony is, all those who claim the eye test is not reliable and ADV stats are more reliable - are using data recorded by an eye test.

never thought of it that way, but you bring up a very good point. Human error and passive bias would be at the heart of all this data.

 

[rumman]

Does anyone think it's a bad idea?

maybe not a bad idea, but $50 mil is a lot of cash, not my money, but wow, just wow.

 

[rumman]

The other talk around town is that gambling will destroy the integrity of the league. That the league will rely too much on the gambling component for it's finances to the point where if laws are changed, it could throw the league into financial chaos.

I am almost expecting more terribly officiated games. Maybe some Pete Rose action

I don't think the gambling angle will pan out for the NHL as a revenue source. By and large most Americans couldn't care less about hockey, why would they suddenly be interested in gambling on a sport they neither like or understand? Not sure about the Europeans being that interested either? That would make the revenue generated meager at best.

 

[WTFMAN99]

I don't think the gambling angle will pan out for the NHL as a revenue source. By and large most Americans couldn't care less about hockey, why would they suddenly be interested in gambling on a sport they neither like or understand? Not sure about the Europeans being that interested either? That would make the revenue generated meager at best.

Some revenue will come but it is hard as hell to bet, the most "lucky" or "random" sport there is

 

[Joey Hoser]

Only in Toronto could we have fans that are not happy with a very significant investment in something that doesn't count against the cap and that can only help with the team's performance (for reference, Ottawa doesn't even really have a front office staff aside from Melnyk/Dorion...)

It's got nothing to do with Toronto.

Anytime you take a stubborn person who thinks they know hockey, and speak to them about something they are not familiar with, they are likely going to be offended by the notion that something they are unaware of, matters or is useful.

"I know hockey real good, and I've never heard of Corsi, therefor Corsi can't be important".

 

[Mess]

This is all accurate, I think Dubas has tried to replicate Tampa to a lesser degree with having skill etc throughout the line up but we seem to lack....the system? tenacity? ability? grit? that Tampa Bay possesses.

I won't claim to have the answers here but I see a lot of good things with the team but there is still a lot I would want to change.

I can tell you with a strong degree of certainty that Cap hit % and usage >>> advanced analytics.

Dubas is trying to replicate TB stylistically but failing miserably capitalistically.

Kucherov has 126 points and leads the NHL in scoring and will start his $9.5 mil X 8 year deal next season while Stamkos #8 in NHL scoring is on his 8 X $8,5 mil deal.

Which one of Leafs top 3 scorers in MM, JT and AM will be replicating the Lightning under Dubas via CAP % and AAV?

Is Dubas planning on using that +$50 mil in analytics to figure out how to duplicate the TB success model while spending far more AAV to get less production on top players and then get more depth production with far less cap space to do so on the remaining players?

Good luck with that !!

PS.. I know you and I are aligned on salaries so I'm preaching to the choir here.

 

[WTFMAN99]

I can tell you with a strong degree of certainty that Cap hit % and usage >>> advanced analytics.

Dubas is trying to replicate TB stylistically but failing miserably capitalistically.

Kucherov has 126 points and leads the NHL in scoring and will start his $9.5 mil X 8 year deal next season while Stamkos #8 in NHL scoring is on his 8 X $8,5 mil deal.

Which one of Leafs top 3 scorers in MM, JT and AM will be replicating the Lightning under Dubas via CAP % and AAV?

Is Dubas planning on using that +$50 mil in analytics to figure out how to duplicate the TB success model while spending far more AAV to get less production on top players and then get more depth production with far less cap space to do so on the remaining players?

I wouldn't have minded paying another $1M to Matthews if it meant an 8 year term, the cap hit he got at the term he got didn't make sense to me mathemetically.

I do think the NHL has a problem though and needs to address the cap imbalance when you look at what Toronto has to spend vs Nashville, Tampa etc because of tax implications. Should we count only the net cap hit?

 

[rumman]

Some revenue will come but it is hard as hell to bet, the most "lucky" or "random" sport there is

another reason gambler's will stay away, smart gamblers know when to stay away from elements that favor the "house." I used to bet the ponies at Woodbine religiously on rainy days until they brought in the polytrack, used to be there was money to be had when it rained, they realized that and closed the loop.

 

[WTFMAN99]

another reason gambler's will stay away, smart gamblers know when to stay away from elements that favor the "house." I used to bet the ponies at Woodbine religiously on rainy days until they brought in the polytrack, used to be there was money to be had when it rained, they realized that and closed the loop.

I made money for a while gambling on the NHL but I had to bet a certain amount to cash out and started to just bet big money on UFC fights to hit the quota. I did alright and the last fight I had to gamble on was Anderson Silva vs Chris Wideman.

I watched on as Silva got cocky and started dancing around. He got tagged on the chin and got knocked out. Watched all the profit and my initial investment go down the drain. I got gun shy after that, plus adult expenses etc now 5 years later lol.

 

[Funk21]

Now focus that analytics team with a new parameter RHD that are above 6"2

 

[Mess]

I think the Leafs current analytics must be broken that is why they're investing $50 mil starting next year.

The Leafs playing at the cap ceiling needed to add a #1C for $11 mil in Tavares and dump $850k McElhinney to promote $750k Sparks to get 46 wins and 99 points.

Carolina playing at the Cap floor grab McBackup and now have 45 wins and 97 points, while the NYI lose their star franchise player also have a bottom 5 Cap and have 47 wins and 101 points.

One team loses a star #1 C and their results go UP and one team adds a star C and their team results go Down. One team tosses a backup goalie on waivers and their team results go Down and another team grabs that goalie off the scrap heap and their team results go UP.

I'd love to know what kind of advanced analytics can be produced that don't already exist today to help explain this situation above?

 

[zeke]

never thought of it that way, but you bring up a very good point. Human error and passive bias would be at the heart of all this data.

and a good chunk of that $50m would go towards hiring and training people to collect data properly....and not only on future games, but to go back and watch game tapes for the last 10yrs, re-collect the data more thoroughly and accurately, and re-do all the analysis ofbthat new data and its correlation with predictivity of winning.

and then to go out and start collecting our own database for every minor league as well.

 

[zeke]

I think the Leafs current analytics must be broken that is why they're investing $50 mil starting next year.

The Leafs playing at the cap ceiling needed to add a #1C for $11 mil in Tavares and dump $850k McElhinney to promote $750k Sparks to get 46 wins and 99 points.

Carolina playing at the Cap floor grab McBackup and now have 45 wins and 97 points, while the NYI lose their star franchise player also have a bottom 5 Cap and have 47 wins and 101 points.

One team loses a star #1 C and their results go UP and one team adds a star C and their team results go Down. One team tosses a backup goalie on waivers and their team results go Down and another team grabs that goalie off the scrap heap and their team results go UP.

I'd love to know what kind of advanced analytics can be produced that don't already exist today to help explain this situation above?

analytics have shown for years now that the canes are actually a good team.

analytics aren't surprised by the canes at all.

 

[Mess]

"There are three kinds of lies: lies, damned lies, and statistics."

When puck possession metrics and OZone entries become more important than actual goal and assist and team wins and loss results, you know you've gone tooooooo far in the wrong direction.

You simply can't break down a boxscore post game analytically, to tell you the full story as actually watching and assessessing a game,

Myself as a 30+ year IT professional knows the game of hockey is played on the ice and not in the office behind a computer monitor, no matter how many fancy spreadsheets and data mining happens behind the scenes.

Analytics should be just another tool in the toolbox, but not the one you rely on the most. IMO

 

[Prominence]

Some thoughts...

Leafs have numerous problems besides that finding that one RHD whose 6'2. Weaknesses: Winning board battles, shot suppression, average drafting, etc. Analytics is about addressing areas of weakness and finding market inefficiencies to solve those problems.

Leafs recently started investing in analytics and since it is so recent, they do not have a competitive advantage in regards to using analytics. By spending 50 million (hopefully appropriately), they are catching up to the teams that have used analytics (e.g. Tampa).

There are numerous questions, grey areas and things we don't know such as goalie projections (high danger scoring chances...i know about that but even then you still cannot project goalies). By investing in analytics, leafs are creating their own library and research to better understand the game.

Analytics is a tool needed to reduce biases. Also, it can be used across many facets of the game. Cap management is important, but there is probably analytics involved with that.

Most importantly, the 50 million does not count towards the cap. The leafs are rich. IF (and I do not subscribe to this) the analytics experiment fails, there is really little downside. It does not hurt.

EDIT: I'm not suggesting that analytics is the end-all either. It is a tool/library and there is always room for the eye-test. I do however support the decision and I'm pretty sure dubas combines the use of analytics and scouting as he has said in some video.

 

[Commander Clueless]

Neat.


I mean, we all know hockey is controlled by the whims of the hockey gods, but whatever makes Dubas happy. At least they are trying, right?

 

[ULF_55]

That number seems very high at first glance, though I guess the future is ambiguous wording.

It does make sense that they would look to expand with the avalanche of tracking data that they'll have access to soon.

I guess that's cheaper than Watson.

 

[RLF]

analytics have shown for years now that the canes are actually a good team.

analytics aren't surprised by the canes at all.

lol. Yet, this is the first year they made the playoffs.
In any other world being right once in 5 years would be considered unreliable. Only in the sports analytics world was it predictive and unsurprising even if it took 5 years of saying the same thing before it was right.