Confirmed with Link: PIT (Sheary & Rodrigues) | BUF (Kahun)

[ChaosAgent]

Is Kovalchuk a Sid player though? The clear intent behind this move is to take guys that weren't seen as clicking with the big dogs and exchange them for those who would. And given just how much we've benefited from this hard intent to move more in line with Sully's philosophy and what suits Sid (and Geno, but he seems a bit more sorted for guys right now), I really struggle to argue with that as a philosophy. Maybe we're pursuing a faded dream in thinking Sheary is that guy, but we are at least following a plan. A guy like Kovalchuk?

One downside of this approach though is it feels like we risk having a very homogeneous and uncreative bottom 6.

I'm very content for our bottom 6 to just play their matchups to a draw. In fact, I'd be elated.

For what it's worth, I agree with @pixiesfanyo on Tanev. Teddy isn't half bad either. Hornqvist is a lot of things, but homogenous isn't one of them.

Though I can't help but wonder if Sheary-Sid-Horny will be the Crosby line. I am growing enamored with Zucker-Geno-Rust for this year. It seems like Rust is in the midst of deciding whether he wants to be a skill first guy or a grit-grind first guy with skill. I think it's a contributing factor to his poor form as of late. Having Zucker as a skill-first guy might help clarify Rust's role on Geno's line a lot (and make Geno's line a true L1).

 

[NMK11]

I'm very content for our bottom 6 to just play their matchups to a draw. In fact, I'd be elated.

For what it's worth, I agree with @pixiesfanyo on Tanev. Teddy isn't half bad either. Hornqvist is a lot of things, but homogenous isn't one of them.

Though I can't help but wonder if Sheary-Sid-Horny will be the Crosby line. I am growing enamored with Zucker-Geno-Rust for this year. It seems like Rust is in the midst of deciding whether he wants to be a skill first guy or a grit-grind first guy with skill. I think it's a contributing factor to his poor form as of late. Having Zucker as a skill-first guy might help clarify Rust's role on Geno's line a lot (and make Geno's line a true L1).

Anyone suggesting a combo of Sid and Horny should be given a 1 hour temporary ban. It ain't happening. We've seen it play out way too many times the past few years. I'd personally love to see that line, but I've given up on wishing for Horny and Sid to work together.

Considering our only for sure, 100%, totally belong top 6 forwards are Zucker and Rust, they're probably going to stay split between Geno and Sid.

 

[ChaosAgent]

Anyone suggesting a combo of Sid and Horny should be given a 1 hour temporary ban. It ain't happening. We've seen it play out way too many times the past few years. I'd personally love to see that line, but I've given up on wishing for Horny and Sid to work together.

Considering our only for sure, 100%, totally belong top 6 forwards are Zucker and Rust, they're probably going to stay split between Geno and Sid.

As much as I should embrace that ban...

I just don't like Rust-Geno-Horny at all. I realize it's produced well in a small sample (due to Geno & Horny) but I don't dig Rust at LW. I'd just as soon move someone else in that spot and try Rust with Crosby again.

 

[rintinw]

I am surprised that so many of you are so tied to Kahun's performances here. I've been trying to avoid using WoWy stats recently because I don't think they paint as detailed a picture as we think they do. Example: there's 10 players on the ice, with one-line stats, we're looking at 3 of those 10 players. Who are they with? Who are they against? Crosby, Guentzel and Simon performs differently with Letang/Dumoulin than they would Johnson/Schultz, right? The WoWY data doesn't really provide context to that in any way.

As a result, my personal shift has been more towards stats that track individual player impacts or focus on data relative to teammates. It can help us discern the true underlying details about a players' ability to drive offense or mitigate quality defensively. By doing this, we can remove all the unknowns that exist in the WoWY data. The context is much richer in these individual evaluations and they often account for mostly everything outside an individual players control.

To me, this trade is more about the impact that depth plays over anything else, but to feed the fire of the debate as I am thoroughly enjoying it, here is some stuff to chew on. These are the individual isolate's for Kahun and Sheary over the last few year's from Micah's model and website (HockeyViz.com). In these, the higher the + the better on offense, the lower the - the better on defense.

Kahun:
18/19
O Threat: +8.7%
D Threat: -5.1%

19/20
O Threat: +2.6%
D Threat: -7.2%

Sheary:
17/18
O Threat: +5.9%
D Threat: -2.5%

18/19
O Threat: +14.4%
D Threat: -5.9%

19/20
O Threat: +12.9%
D Threat: -12.3%

There are few issues I have with this approach.

1) I do not like to use Corsi (which I understood to be parameter which is Micah tracking with those numbers - players impact to teams CF rate in O Threat and CA in D Threat) instead of Goals to grade/rate players past performance. I understand the ratio (less influence of chance, more data points, etc.) but I believe it excludes significant factors that contribute to scoring (or preventing of scoring for that matter). It's because it operates with assumption that every shot attempt is equal to each other (on average). This results in my opinion to favoring certain type of players.

2) I also understand that Micah (and others) recognize that fact and try to accommodate their models to that through various means (scoring chances, high danger scoring chances, expected goals, shot maps, ...). Nevertheless there are still things absent either due to simply not having data (for example quantification of individual shot - corsi quality: speed, quickness of release, accuracy, screens) or not having good enough quantification (how much these factors result in goals on average). Right now I believe we have average rates of shot going in from different areas of offensive zone - I am aware of no adjustment for type and quality of shot being made. Again, I understand that there are simply no data that could be used for such adjustments.

 

[Jesse]

There are few issues I have with this approach.

1) I do not like to use Corsi (which I understood to be parameter which is Micah tracking with those numbers - players impact to teams CF rate in O Threat and CA in D Threat) instead of Goals to grade/rate players past performance. I understand the ratio (less influence of chance, more data points, etc.) but I believe it excludes significant factors that contribute to scoring (or preventing of scoring for that matter). It's because it operates with assumption that every shot attempt is equal to each other (on average). This results in my opinion to favoring certain type of players.

2) I also understand that Micah (and others) recognize that fact and try to accommodate their models to that through various means (scoring chances, high danger scoring chances, expected goals, shot maps, ...). Nevertheless there are still things absent either due to simply not having data (for example quantification of individual shot - corsi quality: speed, quickness of release, accuracy, screens) or not having good enough quantification (how much these factors result in goals on average). Right now I believe we have average rates of shot going in from different areas of offensive zone - I am aware of no adjustment for type and quality of shot being made. Again, I understand that there are simply no data that could be used for such adjustments.

1. Corsi isn't involved in threat like you think it is. It certainly weights shot volume - after all, we're trying to account for a players' offensive impact and not looking at shots would make for a crappy evaluation. You can't score if you don't shoot. That being said, there's an expected-goal piece to this, too, and one that serves as the foundation of the model. Micah is not assuming that all shots are created equal, in fact, the model bases threat off of historical shot success from specific locations. Not player shot success, league shot success. We're attempting to remove the talent of the shooter and the talent of the goaltender to evaluate a true players impact based on the shots they've taken and the league historical data behind those shots. This is according to league average shooting percentages from given areas.

2. Despite all the shortcomings you mention which will likely exist until player tracking becomes public domain, expected-goals are still significantly more predictive of future scoring than past goals are. If we went into the playoffs head to head this year and bet every series - you taking goals and me taking expected-goals, I'd probably clean you up pretty good money-wise.

 

[Dipsy Doodle]

1. Corsi isn't involved in threat like you think it is. It certainly weights shot volume - after all, we're trying to account for a players' offensive impact and not looking at shots would make for a crappy evaluation. You can't score if you don't shoot. That being said, there's an expected-goal piece to this, too, and one that serves as the foundation of the model. Micah is not assuming that all shots are created equal, in fact, the model bases threat off of historical shot success from specific locations. Not player shot success, league shot success. We're attempting to remove the talent of the shooter and the talent of the goaltender to evaluate a true players impact based on the shots they've taken and the league historical data behind those shots. This is according to league average shooting percentages from given areas.

I have no background in analytics, but to the uninitiated, player shot success seems like a pretty important variable...not that I would know how to take that into account.

If you've got Ovechkin or ZAR taking a shot from the left half-boards, there's going to be a stark difference in their success rate.

 

[ChaosAgent]

For what it's worth, I though Sheary looked quite good yesterday. Rodrigues was also noticeable to me...the 4th line in general played well. Good early returns.

 

[rintinw]

1. Corsi isn't involved in threat like you think it is. It certainly weights shot volume - after all, we're trying to account for a players' offensive impact and not looking at shots would make for a crappy evaluation. You can't score if you don't shoot. That being said, there's an expected-goal piece to this, too, and one that serves as the foundation of the model. Micah is not assuming that all shots are created equal, in fact, the model bases threat off of historical shot success from specific locations. Not player shot success, league shot success.

Now you have me a little confused. Before my post I went to check for explanation (Isolated Impacts) what exactly is tracked by those 2 numbers specifically (for the very reason you mention) and found nothing that would suggest those 2 numbers measuring change in teams expected goals. I thought that for location based impact there was the shot map and threat number was just raw summary. But it's possible I missed it.

We're attempting to remove the talent of the shooter and the talent of the goaltender to evaluate a true players impact based on the shots they've taken and the league historical data behind those shots. This is according to league average shooting percentages from given areas.

This is exactly why I think it favors certain type of players (more specifically disfavors skill based players):
1) talent of the goaltender - ok
2) talent of the shooter (if shooter is linemate) - ok
3) talent of the shooter (if shooter is player in question) - not ok
4) not measuring individual impact of player to change in ratio of shots attempted (player vs. linemates) - not ok

2. Despite all the shortcomings you mention which will likely exist until player tracking becomes public domain, expected-goals are still significantly more predictive of future scoring than past goals are. If we went into the playoffs head to head this year and bet every series - you taking goals and me taking expected-goals, I'd probably clean you up pretty good money-wise.

You may have misunderstood me. I was talking about grading past results (that have already occurred), not about predicting future results. I simply believe that once result occurred the luck is integral part of that result. I am well aware of the shortcomings of goals stat when predicting future results though.

 

[Sidney the Kidney]

I have no background in analytics, but to the uninitiated, player shot success seems like a pretty important variable...not that I would know how to take that into account.

If you've got Ovechkin or ZAR taking a shot from the left half-boards, there's going to be a stark difference in their success rate.

This continues to be my biggest beef with these models. If the model assumes "the average player" is taking this shot, then it's not good at predicting anything but what average players will do in those situations.

Like you pointed out above, Ovechkin and ZAR getting the exact same opportunity from the exact same location will have the exact same xG number, but anyone who would then suggest we should be expecting ZAR to score just as many goals in those situations as Ovechkin would be questionable.

 

[Hell Yeah!!!]

Just want to weigh in that this trade was horrible.

thanks

 

[Empoleon8771]

The bad part about the Kahun trade is that he'd be helping out dramatically with their bottom-6 production. But the debate with this trade is whether it's nicer to have someone who meshes well with Crosby or someone who can drive offense from a bottom-6 role. To me, Sheary has played well for the Penguins overall, while a lot of the rest of the team has been utter dog shit. I would have loved to have added Sheary and kept Kahun, but there wasn't a realistic way that the Penguins were going to be able to do that (Buffalo wasn't interested in selling for futures at the deadline, according to JR). Sheary is not a problem on this team, and if his play so far continues, I hope JR will find a way to keep him for a reasonable price.

 

[Gurglesons]

The bad part about the Kahun trade is that he'd be helping out dramatically with their bottom-6 production. But the debate with this trade is whether it's nicer to have someone who meshes well with Crosby or someone who can drive offense from a bottom-6 role. To me, Sheary has played well for the Penguins overall, while a lot of the rest of the team has been utter dog shit. I would have loved to have added Sheary and kept Kahun, but there wasn't a realistic way that the Penguins were going to be able to do that (Buffalo wasn't interested in selling for futures at the deadline, according to JR). Sheary is not a problem on this team, and if his play so far continues, I hope JR will find a way to keep him for a reasonable price.

Let’s be honest about this trade Kahun would have probably been healthy scratched today for Lafferty just like Rodrigues.

Still curious what their rational was behind that one.

 

[SHOOTANDSCORE]

Much to my surprise Tanev seems like a very creative player with the puck. Rodrigues from my viewing is definitely one and then you have McCann who sucks at doing it with linemates but is creative individually. Also, Sheary on L1 should drop Simon into the bottom six.

Agreed, or at least relative to my expectations.

I got jumped on earlier in the season for saying I was surprised by his playmaking ability.

 

[Fatty McLardy]

Awful trade then still an awful trade now. Despite Sheary playing ok. Rodrigues is pretty much useless, basically a poor man's ZAR.

Kahun was an awesome bottow 6 player for us. Feelsbadman.

 

[vodeni]

Anyone suggesting a combo of Sid and Horny should be given a 1 hour temporary ban. It ain't happening. We've seen it play out way too many times the past few years. I'd personally love to see that line, but I've given up on wishing for Horny and Sid to work together.

Considering our only for sure, 100%, totally belong top 6 forwards are Zucker and Rust, they're probably going to stay split between Geno and Sid.

Never had we seen such a desperate situation where you have 4 good forwards and Horny...four good forwards are 60% Sid, tired Geno Zucks and Rust.....I never liked Horny up there, but they need to rejuvinate that top 6 cause you AIN'T getting anything from your bottom 6...those are meh player slumping worst combo....so whether you put him up with Sid or play him with Geno, Horny should be in TOP 6

 

[Pens x]

I guess we can add Kahun to the long list of players that weren’t liked by Sully, aka Dan Babcock.

The coaching ability of Bylsma, the personality of Babcock.

 

[vodeni]

Let’s be honest about this trade Kahun would have probably been healthy scratched today for Lafferty just like Rodrigues.

Still curious what their rational was behind that one.

people don't pay attention, Sully could not wait to seat Kahun for whatever reasons...when everyone healthy he would be scratched anyways

 

[BeatenLikeRentedMule]

Sully certainly doesn't make his preferences transparent. It's well reflected in his lineup choices. That's when a coach's effectiveness expires cause what's the point playing hard if you aren't rewarded? Whats the point playing hard if you're rewarded for doing nothing? This leads to the team we have now.

 

[Turin]

Kahun and his... 2 goals in 21 games before the trade definitely would fix this team.

 

[SouthGeorge]

We've seen this time and time again. Kahun was never playing on Top 2 lines here. Gets traded to worse team, gets 2nd line/more minutes and produces. It doesn't mean we lost the trade.

 

[metalan2]

We've seen this time and time again. Kahun was never playing on Top 2 lines here. Gets traded to worse team, gets 2nd line/more minutes and produces. It doesn't mean we lost the trade.

We lost the trade.

 

[Peat]

Kahun and his... 2 goals in 21 games before the trade definitely would fix this team.

Kahun's 2 goals and 2 assists in 5 games in Buffalo would be manna from heaven right now.

 

[Turin]

Kahun's 2 goals and 2 assists in 5 games in Buffalo would be manna from heaven right now.

Probably wouldn’t be happening on this team and Sheary has been pretty good himself.

 

[Peat]

Probably wouldn’t be happening on this team and Sheary has been pretty good himself.

Maybe it wouldn't but someone who even has the hope of that would be nice. And Sheary... well, decent enough, but I'm not sure he's been the top 6 upgrade that made sacrificing scoring depth make sense.

Tbh, I'm somewhat wondering if trying to bolster the top 6 at the cost of a third line that could score lots was the right call anyway. Time will tell.

 

[Shady Machine]

Probably wouldn’t be happening on this team and Sheary has been pretty good himself.

Why wouldn't it be happening? After a slow start, Kahun was good here and then got injured. He came back and was traded shortly after. In my view, he was the most talented player traded. I like Sheary just fine, but I'm really not sure why we needed to move Kahun to get Sheary and if that was the price, I'd have probably just kept it as is.

When the trade went down I knew nothing about Rodrigues so that was the wildcard. He's an okay player but is in the class of our bubble players.