PBD Points By Defensemen

[Flames Draft Watcher]

Originally Posted by KootenayFlamesFan

Yeah I think that's part of it too, Brodie was confident with Gio and they meshed so well for a while Brodie wasn't really overthinking, he was just playing.

And I wasn't really just thinking Brodie offensively (even though that's the point of the thread) but defensively lately he's made some really strange and poor decisions. I don't know what's up with him but if we could get Brodie from 2/3 years ago it would be massive.

Yep. I think moving Brodie to the left side made him uncomfortable. It seems like he's never regained his comfort. Would moving him to the right side solve it? It's hard to say as confidence is a weird thing. But moving him to the right side doesn't really seem like an option unless you move Hamilton or Stone. So the whole thing is awkward now as he really hasn't looked like a fit on the left side in the top 4 but you're happy with everybody else where they are.

I think one of the biggest decisions Treliving has to make this summer is in regards to Brodie. Personally I'm thinking it might be best just to trade him. I think top 4 defensemen with mobility are highly valued league wide and his value will be high despite the lacklustre last couple years. There's always GMs ready to think a player of his age could still regain lost form in a new situation.

I think Brodie makes the most sense to move because Kulak frankly looked as good as him last year and may be able to take over that spot immediately. And then you've got Valimaki on the left side depth chart and he's probably going to be a top 4 defenseman within 2-3 years so long term you've got Brodie's replacement there as well.

I'm sure management and our scouts have thoughts on whether they think Brodie can rebound. If they have doubts then I think he's the guy you move in order to find another offensive threat. If they have a strong belief that Brodie may rebound for whatever reason then maybe you keep him. I'm leaning towards doubt.

[heep223]

I really like Brodie but if you can trade him for a bonafide top six RW, I think you have to do it. With the D depth the organization currently has, including prospects, it’s kind of a no brainer.

I would have the sads though. Gio and Brodie were phenomenal when they played together.

[tkflames]

Great work on this thread!! I really appreciate the homework that went into this.

Through the last week of the season and the first week of the playoffs, I have been watching some of the "faster" teams with respect to how they move the puck and/or retain possession. From this I have the following observations:

1. Even Pittsburgh and Washington hold up and reverse the puck on the breakout if nothing is doing. Washington and Tampa power plays also regularly uses the dreaded drop pass. However, they can also play a fast transition via quick puck movement from their defensemen.

Hypothesis: The best teams read what the defense is giving them and play the long game. Much like football, a team that just runs the ball (GG Flames) will get shut down when the defense catches on. A team that just throws the ball (Hartley flames) opens up the pick option. The best teams seem to play both approaches and seem to use the stretch pass on occasion "intentionally unsuccessfully" to buy more space on future rushes.

2. While many teams use the 4th forward, the flames could certainly increase their defensive production by going with a 3 and 2 line up. Gio/Hamilton on the first unit, Brodie/Stone on the second. You have 3 guys that regularly score from the point and one that is perfectly capable of carrying it into the zone. In my idea world the Flames pickup a "power" top 6 winger to help the first power play unit with Gaudreau and Mony (maybe that is Tkatchuck already) and a high skill top 6/top 9 forward who's only job it is to gain the line on the second power play unit. That alone should boost defensive points by 15-25. If the Flames can score an extra 20-30 PP points next year, I would have to assume that at least one defenseman (using a 3/2 line-up) would be in on half or more of those.

[ricardodw]

sorry to throw a wrench into the works but your PBD is not a solid stat.

A significant number of teams use 4 forwards on the PP with one rotating into the point position.

This depresses their PBD by a lot.

[GranteedEV]

PBD is no RGI

[Calgary4LIfe]

IMO, the defence have slowed down their production because the way the Flames have been playing is slower. They play at a much slower pace and don't attack the net like they used to in prior years.

I do think that a higher number of goals are being scored with the assists going to the forwards. Defencemen push the puck up to the forwards right now, and then the forwards are generating offence. Defence seem to be responsible for playing defence (duh) and then once they get the puck, are passing it off to a forward.

I would like to see defencemen like Brodie and Giordano actually take the game into their hands and drive play. They simply don't drive the play throught he neutral zone. It is all Gaudreau, Backlund, Jankowski/Bennett and Stajan. I think the Flames have 5 good options with the 6th being able to blast hard shots at the net (though he needs to work on his accuracy).

In playing like they do, they DO 'play it safe' by having two defencemen back at all times. Under Hartley, I thought they rotated pretty well (just like Enoch mentioned how the Winnipeg forwards rotate to cover the defencemen) under Hartley - and when they didn't, they would get an earful on the bench. Now, would you prefer having 2 defencemen on an odd-man rush when the puck got turned over, or 1 defencemen and 1 forward? Of course you would say 2, right? I think that is the issue the Flames are having. Odd man rushes STILL happen, but they don't generate enough odd-man rushes the other way.

These odd-man rushes directly correlate to the Deluxe Moustache's thread about shots. Odd man rushes create a lot more cross-ice passes that force the goalie to move and lead to higher scoring rates.

So, not only are the Flames not generating enough QUALITY HIGH SCORING shots, but the defence are merely 'pushing' the puck up the ice, rather than actually driving offence.

It is what I have been talking about when I say: "Gulutzan has thoroughbreds but he is using them to plow the fields rather than racing them at the track."

Now, I will not bash on Gulutzan. There are obvious advantages to playing this way. First, you SHOULD be playing better defensively. Having 2 defencemen - even of the high offensive skill-set - should lead to better defensive play than having a forward on the back-end. Also, the proof is in the pudding - the Flames DO generate a tonne of shots AND they do manage to get the puck into home-plate - which in itself gives them a better chance to score.

However, they do it slowly - and just like the other thread mentioned above - the goalie is set for the shot for too many of these shots.

By using the D more, it will cause MORE movement. Allow the D to join the cycle. Cycle you say?

"What the heck is a stupid cycle designed for anyway? I hate that word. It is a Brent Sutter word! Cycles are for teams that don't have talented scorers and playmakers, and are designed for the over-sized slow and plodding types."

Hold on.. not so fast. A cycle is a FANTASTIC way to create offence. It should NEVER be the ONLY way to generate offence, but it is incredibly effective. It pulls the other team to the boards in an attempt to get the puck. The puck comes loose, and ideally a play is made to an open player with a cross-ice pass or (if the other team really screws up like the Edmonton Oilers swarm) a mini-breakaway.

Flames - IMO - excel on the cycle. Starting from the top line. Gaudreau gets pinned, but he seems able to do something with the puck still. Monahan is getting better and better at it, and Ferland is so damn good along the boards. Jankowski and Bennett? Fricken the best on the team (especially Bennett). Tkachuk? Gets my 2nd best on the team vote. Backlund and Frolik - while not very big - are like the Sedins in a way that they manage to cycle very well even though they aren't burly.

So if the Flames are so good on the cycle, then why aren't they scoring?

Just like Enoch posted - the defencemen aren't joining the cycle. Nobody is being pulled away from the opposing team. They are not creating much out of the cycle except lower percentage shots and zone time. Just watch how Winnipeg does it - when a defencemen joins, watch what the other team does. Often it draws in another person from the opposing team, or it allows a forward to slip behind coverage. Puck doesn't always get there, but it does sometimes. It creates 'chaos' for the opposition defencemen and the opposing forwards. "Cheat and try to get a breakaway, or end up losing coverage on a guy and having him score a goal against".

The reason I say that this team should not have any trouble scoring and that it had a lot to do with structure and system being employed is that this team can legitimately score well off the rush, and it can legitimately cycle well.

If the Defence aren't a part of BOTH, then you get limited quality shots on net, even when those shots are coming from home plate. You will always manage to create glorious shots - there are always defensive breakdowns after all - but a strong team can create more defensive breakdowns.

Having defencemen that can join the rush and join the cycle can greatly help in creating more confusion and higher quality chances.

Enoch's thread and Deluxe Moustaches' thread perfectly point out why I think this team CAN score at a much higher clip and why I don't necessarily think Treliving HAS to acquire a RW'er (though it would help!).

The only trade-off is possibly having a forward getting caught on an odd-man rush. Worst case scenario is a clean break-away, but I think that under Hartley they rotated fairly well and probably were well in the positives of the 'breakaways for and against' (and definitely in the positives for odd-man rushes for and against). I think this is a smart enough team to play that system competently.

It still can be a very possession-oriented system - aggressive forecheck to get the puck back if lost, pass back rather than throw it away, and (please!!!) aggressive defensive system (like the Wild used to play where you are NOT giving time and space to anyone).

[SebC]

I would guess that points by defensemen correlates pretty well to points in general. How does this analysis look if it's does as a percentage of team points?

[Enoch Root]

Originally Posted by ricardodw

sorry to throw a wrench into the works but your PBD is not a solid stat.

A significant number of teams use 4 forwards on the PP with one rotating into the point position.

This depresses their PBD by a lot.

Yes, some teams employ 4 forwards on the PP. The question though, is which teams? If they are the same teams that are in the bottom half for PBD, then you have a point.

However, based on the fact that high PBD equates to a virtual lock for getting into the playoffs, I don't think your concern has any merit. But why don't you look into it and get back to us.

[Enoch Root]

Originally Posted by Cali Panthers Fan

Great thread Enoch,

I have one question. When determining the correlation numbers, did you just look at last season or have you looked back at multiple seasons over the last few years? The only reason I ask is that you might find the data varies a fair amount from year to year.

First of all, in order to have any validity, yes you have to look at lots of seasons - as many as you can. A few points:

1) I made it clear in my post that the data was only last year because that was all I had time for

2) I HAVE looked at this for several different years in the past. This has been something I have thought about - and looked at - for a long time. And every season that I have looked at except for one, has had at least 14 of the top 16 PBD teams make the playoffs. My unofficial, and sporadic, testing over the years has all generated very similar results

3) While I would like to actually do the research, the problem is that I simply don't have time. Also, I haven't found a site where I can extract the data easily. I literally go to NHL.com and add up each team manually. So getting 10 years of data would take hours, and I simply don't have the time.

One of the things that I am hoping from this thread was to see if other people are interested in this stat. And if so, if it is worth discussing, maybe people know where we can get the data more efficiently so that we can actually back-test it.

[Enoch Root]

Originally Posted by Cali Panthers Fan

That being said, the teams with high end defensemen usually have success. Think of the teams that missed. How many really good defensemen do they have? How many guys that you could conceivably put on an all-star roster?

Buffalo: Ristolainen
Ottawa: Karlsson
Arizona: OEL
Montreal: Weber? (might be a stretch at this point in his career)
Detroit: ---
Vancouver: ---
Chicago: Keith (mostly defensively though)
Rangers: Shattenkirk supposedly, but he's pretty underwhelming now
Edmonton: LOL
Islanders: ---
Carolina: --- (plenty of talent depth, but no stars yet)
Calgary: Hamilton, Giordano
Dallas: Klingberg
St. Louis: Pietrangelo
Florida: Ekblad, Yandle (missed by 1 point)


What's weird is that when I look at that list, I see a lot of very talented offensive defensemen. I know that overall points by defensemen is the argument, and you need depth for that, but shouldn't some of those heavy hitters make up for it a little. Is PBD a product of good teams that score a lot of goals so the defensemen get more secondary assists etc., or is it a direct cause and effect?

At any rate, I agree in principle that the Flames have the talent both offensively and defensively from the d-group to be on the same level as Nashville. All it takes is a strategy that allows them to play to their strengths. Gio and Dougie both have great wristers and good offensive instincts. Brodie is a smooth skater and great passer. Stone has a cannon of a shot (needs to work on accuracy). Hamonic is a much better offensive player than he showed last year because he had to make up a lot for Brodie's questionable defensive play. Even Kulak showed that he has some offensive abilities at times later in the year. That's not taking into account some of the prospects like Andersson, Kylington, and Valimaki who all have displayed strong offensive games. The development system in the AHL needs to exploit this angle just as much as the NHL level needs to. I hope that a change is made in that regard to get as much offense as possible out of those kids, while still teaching them responsible defensive play.

To the bold, I have thought about that as well. But the more I look at this, the more I think it isn't about the great offensive defensemen, it is very much a team stat.

OTT, CHI, DAL, CAL all in the bottom half. VEG, TOR, WAS all in the top half.

Off the top of your head, name me two other defensemen on Vegas, other than Engelland.

Obviously, personnel matters. Better players are going to get more points. But the more I look at t, the more I conclude that this is a team stat. And what really matters is the style of play.

And the correlation with making the playoffs is really high.

Again, I go back to the Flames. I would argue that the talent level improved every year for the stats I posted. Yet the points fell off a cliff.

[Enoch Root]

Originally Posted by Flames Draft Watcher

IMO a big part of the reason why the defensemen scoring is down is because we didn't have two good scoring lines. As the teams scores more goals, as our GF goes up our defensemen are typically going to get more points as a result.

13/14: 163 Pts 209 GF 15.7% PP
14/15: 195 Pts 241 GF 18.8% PP
15/16: 203 Pts 231 GF 17% PP
16/17: 176 Pts 226 GF 20.2% PP
17/18: 146 Pts 218 GF 16% PP

So there is some correlation between GF and Points by Defensemen. Doesn't explain it all. Not a strong correlation between PP% and Points by Defensemen but it has to have an effect as well.

Seems clear that the defensemen got a slightly higher percentage of points under Hartley than Gulutzan. Obviously Hartley's stretch pass, quick break offense may have contributed.

Part of the weak year last year in terms of points from the backend can be explained by the fact that GG didn't play two defensemen on the points for half the year on the 1st unit PP. Only Brodie had the opportunity to get points for half the year with the 1st unit and personally I'm not even sure Brodie should be on the PP at all, I don't think thats a strength of his. Were Hamilton and Giordano to have been on the 1st unit PP all year surely their points would've been higher.

Will points by defensemen change with Peters? Hard to say. But if Treliving gets a bit more offensive depth and Peters has more success than GG as a result the GF will go up and the PP% will go up and so should the Points by Defensemen as they are all somewhat related.

Overall I think the biggest thing last year was that we only had 1.5 scoring lines. So for over half the game we were a black hole offensively. If the roster can be set up so that we have 2.5 scoring lines we'll be in a much, much better position to be scoring.

I disagree with the bold (at least within the confines of looking at the Flames from year to year). I do not think this team had less talent than the lineups from 2 and 3 years earlier. I think it was much more talented.

I think the lack of scoring from the 2nd through the 4th lines was as much style of play as it was lack of talent.

I would also turn the argument around and say that maybe the lack of scoring from the bottom 9 was because of the lack of support from the D not joining the offense enough.

Also, you tried to argue that scoring was down, which would lead to PBD being down. However, that isn't really true. From 15/16 to 17/18, team scoring dropped 13 goals, or 5.6%. But PBD dropped 28.1% over the same period.

I look at the precipitous drop in PBD and for, that goes a long way in explaining why the forwards aren't scoring as much, other than Johnny's line, which I attribute to him and Monahan continuing to get better. And to which I add: how many points would Johnny be getting if the D were contributing to the offense the way they should?

[Enoch Root]

Calgary4Life: I agree with everything you said - and everyone who didn't read it because it was long, you should. Seriously, go back and read it.

But I just wanted to point out that it wasn't me that was discussing WPG's cycle, it was Drewtastic. It was a great point and IMO, WPG is one of the best teams at rotating their D into the cycle (and pinching) and having the forward rotate high to support. BOS is really good at it too.

To me, that is what attacking as a 5-man unit is supposed to look like

[Enoch Root]

Originally Posted by SebC

I would guess that points by defensemen correlates pretty well to points in general. How does this analysis look if it's does as a percentage of team points?

I don't have that data and don't have time to try and get it, but I can tell you that PBD correlated to total goals at .76

In other words, PBD has a higher correlation to team standings than it does to team scoring (which I think is interesting).

(again: all my correlations in this thread refer to this season only. I would love to back test further at some point, but haven't had the time)

[Enoch Root]

VEG is also really good at rotating their D into the forecheck.

One of the things that pissed me off the most about Gulutzan's style of play was how reluctant the D was to pinch, and how quickly they backed out of the zone.

One thing I want to see from Peters is much more pinching and forechecking from the D (ala WPG and VEG)

[Jeff Lebowski]

Originally Posted by Enoch Root

VEG is also really good at rotating their D into the forecheck.

One of the things that pissed me off the most about Gulutzan's style of play was how reluctant the D was to pinch, and how quickly they backed out of the zone.

One thing I want to see from Peters is much more pinching and forechecking from the D (ala WPG and VEG)

VGK are committed to playing aggressive with their D especially in the offensive zone. In GM 4 against LA they were up 1-0 late in the third and Engelland is pinching hard down the wall to create a turnover by the King's winger.

When the D is that aggressive in the offensive zone you get turnovers so close to the opposition net and basically get chances for tap in one timers or chances of the highest quality.

In my mind, Peters has to implement this type of uptempo aggressive style to be successful. If that's how d are utilized and you get high% shots and good corsi numbers I'm all for it (rather than shooting a high volume of low% shots).

[ricardodw]

Originally Posted by heep223

All I know is that anytime he’s played the left side in the last 5 years, he’s looked horrible. Maybe not causation but certainly correlation and seems super obvious to at least try him back where he feels more comfortable.

Or maybe Gio just elevates whoever he plays with.

Except for the most successful 30 game stretch in Flames modern history (2015 make the playoffs run and 11 playoff games) where with Gio hurt he was the best player on the team out of the LD spot playing 25+ minutes a game with Engelland on his Right side.

Right after that the Flames traded their future for Hamilton and team defense became a tire fire.

[Cali Panthers Fan]

Originally Posted by Enoch Root

To the bold, I have thought about that as well. But the more I look at this, the more I think it isn't about the great offensive defensemen, it is very much a team stat.

OTT, CHI, DAL, CAL all in the bottom half. VEG, TOR, WAS all in the top half.

Off the top of your head, name me two other defensemen on Vegas, other than Engelland.

Obviously, personnel matters. Better players are going to get more points. But the more I look at t, the more I conclude that this is a team stat. And what really matters is the style of play.

And the correlation with making the playoffs is really high.

Again, I go back to the Flames. I would argue that the talent level improved every year for the stats I posted. Yet the points fell off a cliff.

I think we are in agreement here. I meant to include something in the bolded part: The prevailing wisdom says that the teams with high end defensemen have the most success. My point in posting that was that there seems to be no link between the talent level and success on the back end. It does seem to be a result of the style of play, which leads to more offense overall, which leads to more points accrued by defensemen over the course of a season.

So I guess what we will be doing is looking at a couple of things instead of pure corsi events this season (although that should still be included as a PART of the analysis): shots created that cross the royal road, and ranking of points by defensemen (as a product of strong team play? I guess TBD). These two threads have given me something interesting to look at over the course of the next season.

[Bingo]

Good work, Enoch! Love it, very interesting.

Trying to help you with the data ...

Here is the top ten points by defensemen by team in the last 5 years with I caveat that I'll summarize at the end.

NSH 764
CGY 706
SJS 704
NYI 657
MIN 655
STL 654
WSH 653
LAK 648
TBL 644
PIT 642

Top ten teams in terms of rank in the NHL Standings (average) in those 5 years

Anaheim Ducks 5.4
Washington Capitals 7.2
Pittsburgh Penguins 7.4
St. Louis Blues 8
Tampa Bay Lightning 8.8
Chicago Blackhawks 9.4
Boston Bruins 10
Minnesota Wild 10.4
New York Rangers 11
Nashville Predators 11.2

Only 5 of the top ten total defense production teams are in the top ten in standings over that time period.

Caveat was traded defenseman. The site that would pull the data all at once assigned traded defenseman to the team "Total", instead of their breakdown. So I painstakingly went back and assigned the vast majority of them to the team where they produced the most in those seasons.

The summary above is only one way to look at it, and there may be better ways that are more revealing.

[Kipper_3434]

Didn't read all the replies. An improved power play should help. Not having bride as the only first unit d-man and scoring 20 pp goals should put us middle of the pack. Also put stone with Brodie, hes put up points before, hopefully this season is an outlier. Go flams

[Philly06Cup]

Interesting hypothesis. Love that you put work into this, but seems shaky.

how does points by forwards correlation compare to PBD? That number should be somewhere and lower than PBD if you're going to make a claim that PBD should be stressed (over PBF).


Either way, the Flames have heavily invested in the D, so regardless of correlation, it makes to discuss how to get the most out of them.