OT: Hockey analytics-What gives?

[ozzzie19]

So to clarify my thoughts from another thread:

Eye test: Toews>Strome by a lot, it's not close

Numbers: Toews is just a bit better than Strome

Evolving-Hockey.com | Standard Skater Stats

How do you use hockey analytics, when it's very contradictory from the eye test?
Do you throw out the numbers and ignore them?

I'm kind of confused. Analytics are supposed to be an objective measure of the game, so when it's contradictory like this, which side do you lean on?

I tend to like things quantified, so if hockey cannot ever be completely quantified I'm a bit confused on how to use this stuff.

Which numbers are you looking at? EW’s non-Patreon #s don’t capture much. Naturalstattrick has the most for free. Looking at the last two years, Toews and Strome have identical deployment. And all the #s have Toews pretty decently better at even strength:
xGF% 46.03 vs 42.47
CF% 50.81 vs 48.03
HDCF% 45.55 vs 40.68
HDGF% 49.09 vs 43.48
ixG/60 .75 vs .52

 

[Kaners Bald Spot]

Which numbers are you looking at? EW’s non-Patreon #s don’t capture much. Naturalstattrick has the most for free. Looking at the last two years, Toews and Strome have identical deployment. And all the #s have Toews pretty decently better at even strength:
xGF% 46.03 vs 42.47
CF% 50.81 vs 48.03
HDCF% 45.55 vs 40.68
HDGF% 49.09 vs 43.48
ixG/60 .75 vs .52

It's way closer if you only use 2019-20.

 

[BobbyJet]

It's way closer if you only use 2019-20.

zzzzzzzz.... Wake me up when you figure it out.

 

[Kaners Bald Spot]

zzzzzzzz.... Wake me up when you figure it out.

I have the link from NST. Do you want me to list the on ice numbers for both? I'll do it.

 

[Styles]

Analytics are only used for nerds to win an internet argument when they only watched a player twice that year. Eye test trumps that

 

[RememberTheRoar]

Analytics are only used for nerds to win an internet argument when they only watched a player twice that year. Eye test trumps that

 

[Styles]

I have the link from NST. Do you want me to list the on ice numbers for both? I'll do it.

Are you really arguing Toews vs Strome?

 

[Styles]

“I don’t know anything about analytics”

“It’s closer than you think”

 

[RememberTheRoar]

Are you really arguing Toews vs Strome?

 

[Kaners Bald Spot]

Are you really arguing Toews vs Strome?

Not really.

However

This is all situations, not just 5-5.

Toews:
All Situations
52.76 CF%
51.52 FF%
51.53 xGF%
53.14 SCF%
52.26 HDCF%

Strome:
54.34 CF%
53.49 FF%
51.15 xGF%
53.92 SCF%
49.86 HDCF%

 

[Styles]

View attachment 373252

I just can’t. Go Rays baby!

 

[Styles]

Not really.

However

Toews:
52.76 CF%
51.52 FF%
51.53 xGF%
53.14 SCF%
52.26 HDCF%

Strome:
54.34 CF%
53.49 FF%
51.15 xGF%
53.92 SCF%
49.86 HDCF%

Yeah you kinda are. It’s actually hilarious at this point. Just stop bud.

 

[Kaners Bald Spot]

Yeah you kinda are. It’s actually hilarious at this point. Just stop bud.

No.....I'm wondering more about how the numbers couldn't be more wrong.

 

[Styles]

Oh shit Strome has better cf% game set match. Stop being dumb

 

[RememberTheRoar]

Not really.

However

Toews:
52.76 CF%
51.52 FF%
51.53 xGF%
53.14 SCF%
52.26 HDCF%

Strome:
54.34 CF%
53.49 FF%
51.15 xGF%
53.92 SCF%
49.86 HDCF%

“Not really”

*Proceeds to type out and post comparison*

 

[Styles]

No.....I'm wondering more about how the numbers couldn't be more wrong.

Because the numbers you are chasing don’t matter.

 

[Styles]

f*** this the asshole astros might tie this up.

 

[Kaners Bald Spot]

“Not really”

*Proceeds to type out and post comparison*

I'm doing this because I'm confused. I mean, the numbers couldn't be more wrong, according to board consensus.


The whole point of this exercise is that the entire board (myself included) thinks Toews is the better player by far, when the numbers say that it's close. I knew the numbers were close before I posted "Is Toews really that much better? Because this is what the numbers say"

So who is wrong? The nerds or the board? Are they both right in some way? I know a bit about analytics, but I don't know how to use them when they contradict popular opinion so greatly. That's the point of this thread and the previous discussion.

I'm not arguing one way or another. I'm discussing the validity of hockey analytics when the numbers and eye test tell two different stories.

Another one was Rundblad on the Hawks. Everyone thought he sucked but his numbers were pretty good.
David Rundblad - Summary - Natural Stat Trick

 

[MileHighHawksFan]

I wonder if anyone has sat down with Toews and talked about how he isnt as effective as he once was? Maybe that would light a fire under his ass and get him going, or maybe the flames just not burning as bright as it once did? All I know is I really miss when our Captain was our best player. He used to piss guys off with how good he was at shutting them down and taking play the other way, now not so much.

 

[Kaners Bald Spot]

To clarify: I'm using this Toews vs Strome comparison as the basis for a discussion about hockey analytics.

In the other thread I should have been more clear.

It's more about board consensus/eye test vs numbers and which matters more.

 

[MileHighHawksFan]

To clarify: I'm using this Toews vs Strome comparison as the basis for a discussion about hockey analytics.

In the other thread I should have been more clear.

It's more about board consensus/eye test vs numbers and which matters more.

I see you KBS I think this just circles back to what gms usually say regarding analytics which is something to the effect of “Its a tool we use but certainly not the only tool or the tool that is used the most.” I also believe it was chris snow on spittin chiclets who said a lot of the analytics they use or put more weight on aren’t the analytics you see being brought up the most. Sounded like they don’t put too much stock in corsi, xgf, etc but maybe I misunderstood.

 

[RememberTheRoar]

To clarify: I'm using this Toews vs Strome comparison as the basis for a discussion about hockey analytics.

In the other thread I should have been more clear.

It's more about board consensus/eye test vs numbers and which matters more.

“All sentimentality aside, is Toews really any better than Strome at this point in his career? He's a much better skater but besides that?”

 

[BobbyJet]

I'm not arguing one way or another. I'm discussing the validity of hockey analytics when the numbers and eye test tell two different stories.

Another one was Rundblad on the Hawks. Everyone thought he sucked but his numbers were pretty good.
David Rundblad - Summary - Natural Stat Trick

That should clinch it then.

 

[Kaners Bald Spot]

“All sentimentality aside, is Toews really any better than Strome at this point in his career? He's a much better skater but besides that?”

View attachment 373255

Again, I had already looked up the numbers before I posted that and threw it out there for the purposes of this discussion.

 

[Marotte Marauder]

It's actually pretty simple really. Do good thing happen when Player X is on the ice?
Example of good things:
Scores goal
Sets up a goal
Makes a good breakout pass
Chips it out of his zone when that is the right play
Covers the point when Dman goes down low
Chips it in to make a line change
Is in the passing lane to break up a play
Knocks an opponent on his ass
Is on the right side of the puck along the boards

I have no idea how one would quantify all of these things but they are all important. A player who does these things more often than the opponents is a positive player.

It is similar to the concept of a player who clearly has lesser skills or lesser speed but is an effective player. Why? Because he knows how to the play the game, period. Measure that? Good luck.