Toggle sidebar

The Advanced Stats Thread Episode VII: An Ode to the Sanity of Silverfish

  • Xenforo Cloud has upgraded us to version 2.3.6. Please report any issues you experience.
Status
Not open for further replies.
The only issue with taking a left handed defenseman is they loaded up on them at the trade deadline.

However I guess you could trade for right-handed guys if the left-handed D become a surplus

Hughes is my third favorite D in this draft after Dahlin and Boqvist so I’m all good with him
 
Machinehead said:
I still have a gut feeling about Bouchard but I can absolutely get on board with Hughes.
Click to expand...
I’d be happy with any of the 4 of them.

After the past decade plus, I really want this team to focus on building a defense like the Predators or the Hurricanes have goin on, take high upside forwards too, but I think it’s easier to trade sub-par defensemen for good forwards
 
With all my expertise of draft eligible prospects, Hughes would be my choice if the Rangers go defenseman. Assuming Boqvist is off the table by the time we pick. But I think we will trade up, especially if Zadina falls to #5, which is a possibility if Montreal take Tkachuk and Ottawa wants Boqvist to be a Karlsson replacement.
 
  • Like
Reactions: jas
Zadina is the only guy i’m trading up for. Montreal would be foolish to pass on him

But Beregvin gonna do his thing
 
Man, look at the difference in standard deviation for qual teammate and qual comp...

a5X36qc.png


The sample is, by season, d-men who have appeared in 25 or more games. It's 2331 player seasons.
 
But weren't his stats always going to forecast terribly based on his usage and performance at the end of the season? I mean thank you so much for putting in the time and effort, but is that sample size really indicative of what we should expect? I honestly don't know what to expect until I see who our new coach is and can gather a semblance of an idea of what style we'll play.

Sandelin might be a good option that's going under the radar. Robbing Boston's AHL farm team may be a good idea too.
 
  • Like
Reactions: Bozle
Mac n Gs said:
But weren't his stats always going to forecast terribly based on his usage and performance at the end of the season? I mean thank you so much for putting in the time and effort, but is that sample size really indicative of what we should expect? I honestly don't know what to expect until I see who our new coach is and can gather a semblance of an idea of what style we'll play.
Click to expand...

I am doing everything in my power to account for this. Context/usage are in the model. All relative metrics are RelT based.

In no way, shape, or form, am I calling this gospel. It's for fun. I think if someone smarter than me was doing this, it could have "real-world" implications, but I do not believe that person is me.

Full methodology and blog probably coming next weekend or tomorrow, pending how lazy I am the rest of the weekend.
 
silverfish said:
I am doing everything in my power to account for this. Context/usage are in the model. All relative metrics are RelT based.

In no way, shape, or form, am I calling this gospel. It's for fun. I think if someone smarter than me was doing this, it could have "real-world" implications, but I do not believe that person is me.

Full methodology and blog probably coming next weekend or tomorrow, pending how lazy I am the rest of the weekend.
Click to expand...
No, dude, this is definitely awesome work. I'm just throwing other ideas out there about what else we may see next season. It's not your fault AV played him in heavy first pairing minutes.

I mean unless you have a time machine and can go back in 3 years and whisper into Dolan's ear to fire AV before this whole downturn even begins.
 
Mac n Gs said:
No, dude, this is definitely awesome work. I'm just throwing other ideas out there about what else we may see next season. It's not your fault AV played him in heavy first pairing minutes.

I mean unless you have a time machine and can go back in 3 years and whisper into Dolan's ear to fire AV before this whole downturn even begins.
Click to expand...
Yeah, so this comes back to the whole usage argument, and how much it matters, and if we even have a basis of understanding. I think the standard deviation differences between xGF QoT and xGF QoC is really compelling. I think we tend to forget that if Pionk goes from first-pairing to third-pairing, as he probably should, he's also going to see way less minutes with Kreider and Zibanejad. So. Who knows. I don't.

Dolan doesn't need a time machine, he just needs an HF account ;)
 
silverfish said:
Yeah, so this comes back to the whole usage argument, and how much it matters, and if we even have a basis of understanding. I think the standard deviation differences between xGF QoT and xGF QoC is really compelling. I think we tend to forget that if Pionk goes from first-pairing to third-pairing, as he probably should, he's also going to see way less minutes with Kreider and Zibanejad. So. Who knows. I don't.

Dolan doesn't need a time machine, he just needs an HF account ;)
Click to expand...
See that's an interesting point because I wonder if that usage shift would result in Pionk driving possession. We know he has a good offensive skillset, and if he's getting more sheltered minutes, he should in theory have better CF% numbers.

I literally have no idea what to expect right now, and it's exciting/terrifying.
 
Mac n Gs said:
See that's an interesting point because I wonder if that usage shift would result in Pionk driving possession. We know he has a good offensive skillset, and if he's getting more sheltered minutes, he should in theory have better CF% numbers.

I literally have no idea what to expect right now, and it's exciting/terrifying.
Click to expand...
On the brightside, he really can't have a worse relCF60 next year than he did this year.
 
Not sure how much data will be readily available, but is there any way to check fancy stats for the WC?

More of a curiosity than anything.
 
Hey curious, do you guys know where I can find like a top 10 or 20 for the entire league in CF Rel% for the playoffs? I go to these sites but trying to figure what is relative or correct.
 
Bodi Finds the Fire said:
Hey curious, do you guys know where I can find like a top 10 or 20 for the entire league in CF Rel% for the playoffs? I go to these sites but trying to figure what is relative or correct.
Click to expand...
Go to naturalstattrick.com --> under the players tab, select on-ice --> change counts to relative & click submit.

I'd recommend adding a filter for GP because it's a little wonky with one game sample sizes.

silverfish said:
On the brightside, he really can't have a worse relCF60 next year than he did this year.
Click to expand...

Only place to go is up from here! Worst CF% team in the league from Pionk's first game through the end of the season. What a time February through April was.

On a second note, Sacco and Scott Gordon make me nervous as potential coaching options. Unless they've completely changed coaching philosophies, we may be in for a bumpy ride.
 
Last edited:
Hi was wondering if anyone here had seen this guys attempt to recreate Dawson's GAR. This is his write up:

goals ABOVE REPLACEMENT

And the full rankings are available on Sean Tierney's profile.

Tableau Public

The player rankings seem more consistent with what I expected.
 
  • Like
Reactions: Mac n Gs
Found this interesting from that article regarding how points are viewed among dmen by DTM:

This basically represents the weights for C-BPM. The longer bars show the more important inputs for each position. Some things of note here. For forwards, primary points are king. For defencemen all points are relatively equal, with slightly more focus on assists. And for either position, shot quality (ixFSh%) is slightly more important than shot quantity (iFF/60)
Click to expand...

@Machinehead, here's your Kredier explanation to everyone when you say he's elite:

I mentioned above that point production (CBPM) is only half of the equation. That's because there's another key input to players offensive output, their ability to drive play. Some players (especially defencemen) can be elite offensive players without massive point totals. They achieve this by driving shots and scoring chances towards the other teams net. Making sure their team generates goals even though they may not be directly picking up points. To account for players ability to drive play offensively I use C-XPM (again based on Dawson Sprigings WAR, and his XPM metric). This uses 2 key metrics, Relative to teammate Corsi for, and relative to teammate expected goals for.
Click to expand...
 
  • Like
Reactions: Dack
Since it’s coming up in the spec thread, who would be good FA targets to add to the bottom-6
 
Status
Not open for further replies.

Latest posts

Ad

Upcoming events

Ad

Top Bottom