Matt Henderson: New Report Suggest Oilers Being Conned

If Peter Chiarelli was told by some analytics firm that Niklas Kronwall and Kris Russell ranked 1 & 2 respectively in passing that lead to zone entries and they were ahead of Keith, Letang, Suter, Giordano, Hjalmarsson, and Karlsson then once again red flags should have gone off.

I have no words

I have no words



- Sabresfan-365

I'm glad I wasnt the only one floored

Thing is, and I'm not defending Russel per se, but the analytics followers also make conclusion leaps much like specters numbers do. They leave out too much nuance.

Analytics don't neccesarily say that talbots save % when russels on definitively has no connection. It very well may. And if the trend continues, long term coincidence doesn't really hold up to a ton of scrutiny either imo.


Add to that the use of regression to the mean or even luck in analytics. If russels passing is in in fact leading to high numbers of zone entries by their internal numbers , but not in a ton of scoring chances/shots, then by analytics standards that's just bad luck and the numbers will in fact start heading in the direction they're supposed to.
Not all that different from a player taking a ton of shots, but having a low shooting percentage.
Maybe one of those players that are often on the ice with him is experiencing this ?

Aside from any Russel positives or negatives, it Comes down to this, fancies don't tell the whole story. Or even close to it really considering how fluid the game is.

I'm certainly not advocating those insane rumours of a 4 year deal for big dollars, but imo, he makes our defence significantly deeper and does a lot of things fairly well. Important things in today's game. Even if he also is a ton of drawbacks

Thing is, and I'm not defending Russel per se, but the analytics also make conclusion leaps much like specters numbers do. They leave out too much nuance.

Analytics don't neccesarily say that talbots save % when russels on definitively has no connection. It very well may. And if the trend continues, long term coincidence doesn't really hold up to a ton of scrutiny either imo.


Add to that the use of regression to the mean or even luck in analytics. If russels passing is in in fact leading to high numbers of zone entries by their internal numbers , but not in a ton of scoring chances/shots, then by analytics standards that's just bad luck and the numbers will in fact start heading in the direction they're supposed to.
Not all that different from a player taking a ton of shots, but having a low shooting percentage.
Maybe one of those players that are often on the ice with him is experiencing this ?

Aside from any Russel positives or negatives, it Comes down to this, fancies don't tell the whole story. Or even close to it really considering how fluid the game is.

I'm certainly not advocating those insane rumours of a 4 year deal for big dollars, but imo, he makes our defence significantly deeper and does a lot of things fairly well. Important things in today's game. Even if he also is a ton of drawbacks
- HB77

Well said. Agree with this post completely.

Analytics of any kind are well and good but in the game of hockey are practically useless without context. Hockey is a team game played at a incredible speed with uncontrolled variables like ice quality, board flex, and as many small rounded surfaces as you can find. As well as a distinct psychological aspect that applies to most sports but is amplified by what seems like pure chance. Toss in a term like "chemistry" as well and watch a true analytics persons head rightly explode.

The "eye test" by an experienced hockey pundit is a far superior form of gathering information than replying on numbers without context.

It's not even close.

Kris Russel passes the eye test very well through his career. To the extent that the very popular rumor was that Russel was expecting to get a deal in the range of 5 mill per year for 5+ years.

He is clearly a top 4 defender even playing on his offside. Some of that may well be due to the unquantifiable aspect of "chemistry" with Sekera.

Guys like Nurse, Gyrba, Benning, Reinhart, and more often than not Fayne are clearly not.

Its actually great to mention Fayne since he was a analytics darling and is effectively being replaced by Russel who is anti analytics.

In a perfect world the Oilers would be able to find a right shooting top 4 defender with a decent shot via FA or minor trade. That just isn't happening without trading away McDavid apparently.

With the struggle to find a guy able to play top 4 minutes on the right side so obviously difficult signing Russel (as long as there are no, no movement/trade clauses) is a no brainer.

Edit: Signing Russel and leaving him exposed to the expansion draft is a very smart move cause if they take him we still have Davidson. If we don't sign Russel and leave Davidson exposed THAT would be very poor management IMO. Signing Russel and not exposing him is of course the safest route to go for defensive strength but not the way I would personally go. I would sign Russel and expose him to protect more of our forwards like Maroon and whichever of Davidson/Russel didn't get taken would be fine.

If Peter Chiarelli was told by some analytics firm that Niklas Kronwall and Kris Russell ranked 1 & 2 respectively in passing that lead to zone entries and they were ahead of Keith, Letang, Suter, Giordano, Hjalmarsson, and Karlsson then once again red flags should have gone off.



Why? HERO Charts place PA Parenteau as a top tier second line player bordering on first...and yet he's waiver wire candy and puts up second line points playing heavily gifted offensive minutes with top tier players like Tavares and Duchene.

That's an analytics firm telling me that PA Parenteau is better than (randomly search a top players name) Claude Giroux. But that's the "truth" and gospel.

great job on this article. clearly someone's being taken for a ride.

Analytics of any kind are well and good but in the game of hockey are practically useless without context. Hockey is a team game played at a incredible speed with uncontrolled variables like ice quality, board flex, and as many small rounded surfaces as you can find. As well as a distinct psychological aspect that applies to most sports but is amplified by what seems like pure chance. Toss in a term like "chemistry" as well and watch a true analytics persons head rightly explode.

The "eye test" by an experienced hockey pundit is a far superior form of gathering information than replying on numbers without context.

It's not even close.

Kris Russel passes the eye test very well through his career. To the extent that the very popular rumor was that Russel was expecting to get a deal in the range of 5 mill per year for 5+ years.

He is clearly a top 4 defender even playing on his offside. Some of that may well be due to the unquantifiable aspect of "chemistry" with Sekera.

Guys like Nurse, Gyrba, Benning, Reinhart, and more often than not Fayne are clearly not.

Its actually great to mention Fayne since he was a analytics darling and is effectively being replaced by Russel who is anti analytics.

In a perfect world the Oilers would be able to find a right shooting top 4 defender with a decent shot via FA or minor trade. That just isn't happening without trading away McDavid apparently.

With the struggle to find a guy able to play top 4 minutes on the right side so obviously difficult signing Russel (as long as there are no, no movement/trade clauses) is a no brainer.
- Aerchon

While im not completely onside with analytics quite yet, I think a lot of people would disagree with this statement. The eye test alone has a ton of flaws as well.

Missing over arching trends and inherent bias are undoubtedly problems with the eye test. Rather big ones. And nearly impossible to argue against as well. I see it every day on posters evaluations. Myself included.
The differences of opinion in eberle immediately comes to mind. I'm forever defending his value. But am I bias ? He's one of my fav players, so potentially yes. Are the haters that put too much stock (imo anyways) in his defensive deficiencies looking for every little negative thing to point out because of bias ?
Potentially yes as wel

If Peter Chiarelli was told by some analytics firm that Niklas Kronwall and Kris Russell ranked 1 & 2 respectively in passing that lead to zone entries and they were ahead of Keith, Letang, Suter, Giordano, Hjalmarsson, and Karlsson then once again red flags should have gone off.



Why? HERO Charts place PA Parenteau as a top tier second line player bordering on first...and yet he's waiver wire candy and puts up second line points playing heavily gifted offensive minutes with top tier players like Tavares and Duchene.

That's an analytics firm telling me that PA Parenteau is better than (randomly search a top players name) Claude Giroux. But that's the "truth" and gospel.
- AdamFrench

Hmmm. Some food for thought, danke.
Analytics are quite the mixed bag.

Great write up.

Why does Chia say these things? See my location ...

And the passing analytics back up what I see and what I've been saying.
Most specifically in regards to Larsson. He's not a good puck mover at all and dumps more pucks than anyone else on the team.
We've heard Chiarelli say these types of things about Russell, Larsson and Reinhart that just aren't true. If former oiler management said these things people would be publicly flogging them by now.

Good read.



McDavid and Russel ( seems like he's trouble ) for Rielly.

Kris Russell is not a top 4 dman. Im sorry maybe he fits that role w the Oil. He is a liability come playoff time. I get the analytics and those are nice and pretty but not all can be measured independently.

Oshie was probably great statistically during regular season but Blues still shipped him off because he did nothing in the playoffs. Playoff hockey is a different animal. And if your not building your team to excel at playoff hockey then you're just pissing in the wind.

Great write up.

Why does Chia say these things? See my location ...

And the passing analytics back up what I see and what I've been saying.
Most specifically in regards to Larsson. He's not a good puck mover at all and dumps more pucks than anyone else on the team.
We've heard Chiarelli say these types of things about Russell, Larsson and Reinhart that just aren't true. If former oiler management said these things people would be publicly flogging them by now.
- Iggysbff

Larsson has sucked?

While im not completely onside with analytics quite yet, I think a lot of people would disagree with this statement. The eye test alone has a ton of flaws as well.

Missing over arching trends and inherent bias are undoubtedly problems with the eye test. Rather big ones. And nearly impossible to argue against as well. I see it every day on posters evaluations. Myself included.
The differences of opinion in eberle immediately comes to mind. I'm forever defending his value. But am I bias ? He's one of my fav players, so potentially yes. Are the haters that put too much stock (imo anyways) in his defensive deficiencies looking for every little negative thing to point out because of bias ?
Potentially yes as wel
- HB77

Yeah I would beg to differ that the eye test is far superior. All aspects should play a part.
And none are definitively correct.

But the things Chia says continue to boggle my mind.

Larsson has sucked?
- Garnie

Not sucked. But he's not a top pair guy either. He's had some struggles but is a decent middle pair guy at best IMO.

well penned...some of these metrics just dont pass the smell test (forget about the eye test)

Not sucked. But he's not a top pair guy either. He's had some struggles but is a decent middle pair guy at best IMO.
- Iggysbff

Well I havent watched many Edmonton games...so I have no clues...that sucks though....I was hoping he'd be a homerun top pairing for you guys. Only 25 games in....lots of time for him to still get going.

I'm excited to see Russell and Wideman reunited in Las Vegas.

Larsson has sucked?
- Garnie

Only if he deems the efforts of his brethren as lacking, does he allow himself to appear as a mere mortal player. Larsson does not reward those who do not deserve his gifts.

Thing is, and I'm not defending Russel per se, but the analytics followers also make conclusion leaps much like specters numbers do. They leave out too much nuance.

Analytics don't neccesarily say that talbots save % when russels on definitively has no connection. It very well may. And if the trend continues, long term coincidence doesn't really hold up to a ton of scrutiny either imo.


Add to that the use of regression to the mean or even luck in analytics. If russels passing is in in fact leading to high numbers of zone entries by their internal numbers , but not in a ton of scoring chances/shots, then by analytics standards that's just bad luck and the numbers will in fact start heading in the direction they're supposed to.
Not all that different from a player taking a ton of shots, but having a low shooting percentage.
Maybe one of those players that are often on the ice with him is experiencing this ?

Aside from any Russel positives or negatives, it Comes down to this, fancies don't tell the whole story. Or even close to it really considering how fluid the game is.

I'm certainly not advocating those insane rumours of a 4 year deal for big dollars, but imo, he makes our defence significantly deeper and does a lot of things fairly well. Important things in today's game. Even if he also is a ton of drawbacks
- HB77

SUCK IT JERO!!!!!!

While I enjoyed the piece on accountable stats; wasn't there an Oilers game last night? Can anyone tell me how it ended?

SUCK IT JERO!!!!!!
- smellmyfinger

Wow finally someone said it. people have been dropping the ball with the first comment.

Thing is, and I'm not defending Russel per se, but the analytics followers also make conclusion leaps much like specters numbers do. They leave out too much nuance.

Analytics don't neccesarily say that talbots save % when russels on definitively has no connection. It very well may. And if the trend continues, long term coincidence doesn't really hold up to a ton of scrutiny either imo.


Add to that the use of regression to the mean or even luck in analytics. If russels passing is in in fact leading to high numbers of zone entries by their internal numbers , but not in a ton of scoring chances/shots, then by analytics standards that's just bad luck and the numbers will in fact start heading in the direction they're supposed to.
Not all that different from a player taking a ton of shots, but having a low shooting percentage.
Maybe one of those players that are often on the ice with him is experiencing this ?

Aside from any Russel positives or negatives, it Comes down to this, fancies don't tell the whole story. Or even close to it really considering how fluid the game is.

I'm certainly not advocating those insane rumours of a 4 year deal for big dollars, but imo, he makes our defence significantly deeper and does a lot of things fairly well. Important things in today's game. Even if he also is a ton of drawbacks
- HB77

Agree, especially with the luck statement. Luck is something that is too commonly used by armchair analytics people to explain discrepancies in data.

I think this is more likely a thing where Chiarelli misspoke or didnt fully understand the analytics given to him to be able explain it in an interview. I dont think thats a bad thing, it's not his job to fully understand all the analytics and explain it to someone, he just has to take all info given to him and make good choices. With the Oilers' record/stats as of now, I think he's done a decent job.