Jump to content
The Official Site of the Vancouver Canucks
Canucks Community

Horvat +/- stats


wai_lai416

Recommended Posts

1 hour ago, wai_lai416 said:

Sorry for starting a new thread didn't really look thru the others to see if this has been discussed.. but can someone explain to me how Horvat is leading the league in -?? even with his current hot streak with 12 points in his last 10 games.. he's somehow still -6.. with 4 of those points coming from the powerplay.. which means his line contributed 8 goals on even strength but given up 14 in the last 10 games? tbh i haven't been paying attention to WD line matching.. but is he tossing Horvat's line vs other teams top scoring line? and if that's the case.. perhaps that line matching with other teams top line ain't working? with 8 even strength points in his last 10 games.. i'd expect at the very least +/- 0.. not -6.. in a way that's counter productive.. for every point his line scores.. they give up almost 2 points back the other way =/

The posts in this thread have identified several relevant factors. When I think of plus/minus I think the most important considerations are as follows.

1. Plus/Minus is a very "noisy" stat. It has a low "signal to noise ratio". A guy can play well but another player on the team might make a mistake that leads to a goal. Or a guy might do nothing while his team scores because other guys make good plays. This "noise" can explain a lot of what we see in small samples.

2. This noise implies that we need large samples for +/- to be meaningful. But in large samples it is meaningful. The top all-time +/- players are Larry Robinson, +730, Bobby Orr, +597, Raymond Bourque, +528, Wayne Gretzky, +518 and Bobby Clarke, +506. This is probably exactly right. Gretzky is the greatest offensive player of all time, and the other four have a pretty strong claim to being the best combined two-way players of all time. And on the Canucks, the top all-time +/- players are of course the Sedins, which does capture something real. In large samples +/- is highly meaningful. It is not correct to say that +/- is "meaningless" in large samples.

3. Even if we have a large enough sample so that noise is not a big problem there are biases to consider. Obviously the quality of a player's teammates and the quality of a player's opponents are important. Other small biases are created by zone starts and time on ice when a goalie is pulled.

4. In evaluating two-way play it is useful to look at +/- corrected for these biases and other two-way metrics can also be used (like adjusted Corsi).

5. So what about Bo? My read is that all these things are relevant.

6. First, Horvat has had below average luck (PDO) so far this season. So we would expect this +/- to improve as the sample size increases. Thus, part of his low +/- is due to noise.

7. Second, Bo has a lot defensive zone starts, often against strong opponents and is playing with wingers who are not particularly good defensively. So quality of competition, quality of teammates, and zone starts all work against Bo as well.

8., Third, even after you correct for everything, Bo still has not had a good year defensively. Not terrible, not even bad, but not good either. Most people on CDC love Bo and don't want to see a word of criticism. He a great guy, is an excellent young player, and appears to have a great future. But, for most most of the season, he has been over-matched in the role he has been asked to fill -- as the primary shutdown center who takes key defensive zone faceoffs, plays PK, is expected to score and is playing with wingers who are not strong defensively. (Vrbata has been particularly weak defensively, putting a lot of pressure on Bo in his own zone.)

9. Ideally, Bo would have been playing 3C this year with more protected minutes, in which case his +/- numbers would have been much better.

10. Bottom line: At this stage in his career Bo is not as good as average CDC opinion would suggest. However, he is a very good young player. Next year, with Sutter taking on the shutdown role, with a better defensive RW to replace Vrby, and with Bo continuing to improve his own game, I am sure his +/- will be a lot better.

Link to comment
Share on other sites

Horvat is playing some of, if not the hardest minutes on this team.

Strong quality of competition, high defensive zone starts.

Keep that in context and his +/- is not anywhere near as significant as some people may perceive.

Were those things not the case - ie high ozone starts, weak competition and his +/- were still negative - with sheltered minutes - and you might have cause for concern and want to look closer.

Link to comment
Share on other sites

1 hour ago, JamesB said:

The posts in this thread have identified several relevant factors. When I think of plus/minus I think the most important considerations are as follows.

1. Plus/Minus is a very "noisy" stat. It has a low "signal to noise ratio". A guy can play well but another player on the team might make a mistake that leads to a goal. Or a guy might do nothing while his team scores because other guys make good plays. This "noise" can explain a lot of what we see in small samples.

2. This noise implies that we need large samples for +/- to be meaningful. But in large samples it is meaningful. The top all-time +/- players are Larry Robinson, +730, Bobby Orr, +597, Raymond Bourque, +528, Wayne Gretzky, +518 and Bobby Clarke, +506. This is probably exactly right. Gretzky is the greatest offensive player of all time, and the other four have a pretty strong claim to being the best combined two-way players of all time. And on the Canucks, the top all-time +/- players are of course the Sedins, which does capture something real. In large samples +/- is highly meaningful. It is not correct to say that +/- is "meaningless" in large samples.

3. Even if we have a large enough sample so that noise is not a big problem there are biases to consider. Obviously the quality of a player's teammates and the quality of a player's opponents are important. Other small biases are created by zone starts and time on ice when a goalie is pulled.

4. In evaluating two-way play it is useful to look at +/- corrected for these biases and other two-way metrics can also be used (like adjusted Corsi).

5. So what about Bo? My read is that all these things are relevant.

6. First, Horvat has had below average luck (PDO) so far this season. So we would expect this +/- to improve as the sample size increases. Thus, part of his low +/- is due to noise.

7. Second, Bo has a lot defensive zone starts, often against strong opponents and is playing with wingers who are not particularly good defensively. So quality of competition, quality of teammates, and zone starts all work against Bo as well.

8., Third, even after you correct for everything, Bo still has not had a good year defensively. Not terrible, not even bad, but not good either. Most people on CDC love Bo and don't want to see a word of criticism. He a great guy, is an excellent young player, and appears to have a great future. But, for most most of the season, he has been over-matched in the role he has been asked to fill -- as the primary shutdown center who takes key defensive zone faceoffs, plays PK, is expected to score and is playing with wingers who are not strong defensively. (Vrbata has been particularly weak defensively, putting a lot of pressure on Bo in his own zone.)

9. Ideally, Bo would have been playing 3C this year with more protected minutes, in which case his +/- numbers would have been much better.

10. Bottom line: At this stage in his career Bo is not as good as average CDC opinion would suggest. However, he is a very good young player. Next year, with Sutter taking on the shutdown role, with a better defensive RW to replace Vrby, and with Bo continuing to improve his own game, I am sure his +/- will be a lot better.

Do it dude!  Drop the Mic and walk.. Do it!

Link to comment
Share on other sites

3 hours ago, baumerman77 said:

I think one of the biggest misconception of +/- is that it accurately measures two-way play. Most people think the better the plus/minus a player has on his team the better, but this isn't the case. Take for example a bad team, where would you expect their top defenceman to be on their teams plus/minus? Near the top? No, in all likelihood the teams top defender would be near the middle perhaps a little lower.

That is because of two things: ice time bias and the fact that plus/minus is a team stat. Because the best defender is likely to be on the ice for the most minutes and with a team that isn't good we could expect his plus/minus to be very poor. Now the fact that he is the best defender will slightly bring up his plus/minus but not to the top of the team. Anyway this is often what happens (although there are exceptions). My point is plus/minus doesn't tell us much about anything, and a lot of our preconceived notions about it are wrong. There are just better ways to measure what is claims to measure.

http://edmontonjournal.com/sports/hockey/nhl/cult-of-hockey/just-how-horse-$&!#-is-the-nhls-official-plus-minus-stat

Any actually the same for things with turnovers and other numerical stats. If someone is getting huge ice time minutes against the top lines, they are likely to have more turnovers. It would be better to express it as a percent (% passes intercepted or something) but that still only tells a very incomplete picture.

Intelligent observation of a player still wins over most of these type of stats IMO.

 

 

Link to comment
Share on other sites

Personally, I think any concern over his stats is unfair at this point. 

Due to circumstances not under his or managements control, Bo was, almost forced into the NHL. Fear of stalling his development by sending him back to his Junior team and his ineligibility to go to the AHL created this. Unfortunately, the injury to Sutter has put a lot of pressure on him to perform at a higher level than he's ready for. I think he's done a great job, all things considered. It's easy to forget he is barely out of his teens and still very raw as an NHL player. I see a bright future for him here. He is listening to his coaches and is always a professional. His stats may not be top notch right now but those will improve with time. I, for one, give him a definite thumbs up.

Link to comment
Share on other sites

Huge learning curve = slow start = tough numbers.  He's succeeded at every tough assignment he's been given and has improved his game in every way.

What we are now seeing is the making of a tremendous player who will have a huge impact on this team.  Imo, he is going to be a better player than what was projected.

Link to comment
Share on other sites

Analyzed his last 6 games:

3 goals and 3 assists, but a net -3 in the +/- column.

Took 63 defensive zone face-offs (winning 28) out of 150. (Vey took 47, winning 21.... McCann 10, winning 4..... with the remaining 30 split among Hank, Cracknell and Zalewski, who collectively won 13)

So our overall d-zone face-off success was 66 won out of 150 taken :o

Link to comment
Share on other sites

29 minutes ago, N7Nucks said:

I think it's clear we should trade him for Drouin before his +/- gets worse and no one wants him. Drouin is an elite talent and Horvat is a second liner at best. 

Yeah that's why Yzerman is jumping at the chance of a exchange of Horvat and Drouin...  :rolleyes:

Link to comment
Share on other sites

Archived

This topic is now archived and is closed to further replies.

  • Recently Browsing   0 members

    • No registered users viewing this page.
×
×
  • Create New...