Goals vs. Assists: If you had to weight them, how would you?

[Jurivan Demidovsky]

82 gp - 25 g - 0 a
82 gp - 0 g - 25 a

At the forward position only one guy would be considered valuable.

 

[Hockey Outsider]

I don't have exact ratios. I'm okay with valuing primary assists as much or close to as much as goals (I'm sure no one would say Zach Hyman was a lot better offensively last season than McDavid), but secondary assists definitely shouldn't be considered close to as valuable as either (generally speaking - yes, there are the rare instances where a player makes a brilliant solo effort and then another player shovels the puck toward the net and a third player scores on the rebound).

I believe @Hockey Outsider did a study in which he concluded a secondary assist has about two-thirds the value of a goal.

That's right - see this post. I found that goals and primary assists have almost the same predictive value. Secondary assists were (on average, of course) worth about one-third less.

 

[Leaf Fans]

The way they do it now is fine for me.

 

[kugelbahn]

That's right - see this post. I found that goals and primary assists have almost the same predictive value. Secondary assists were (on average, of course) worth about one-third less.

"predicitve value" != "relative value (weight)", though, whatever the latter means

 

[Plastic Joseph]

There are about 60-70% more assists than there are goals. Statistically that shows that goals are more valuable and hard to accumulate. This is further ilustrated from the qualitative understanding that the primary objective of hockey is to score goals.

Realistically, I'm on board with a primary assist = a goal, but secondary assists do not have the same value to me as goals do. The 2nd assist is arbitrary and if the NHL decided to never count them back in the day, nobody would really bat an eye - similar to how most soccer leagues don't count secondary assists, and similar to how we don't see people here advocating for tertiary assists.

I agree that sometimes the primary or secondary assist can be more fundamental to the goal. But the reality is that the further you go down the chain: Goal -> primary assist -> secondar assist -> tertiary assist -> 4th assist etc. the less impactful or meaningful that play or pass goes towards the final outcome which is the goal.

I will take a guy that is 80 goals, 20 assists over the 20 goal, 80 assist guy. Therefore also logical to say that I would still take the 80 goal (100 point) guy over the 20 goal (110 point) guy even if he had 10 more points.

Yep, goals are actually a rigid metric in that 1 goal scored = 1 goal for your team that counts towards winning.

Assists are a bit more subjective and I've seen plenty of iffy secondary assists be awarded , or not, just depending on interpretation of the officials.

Obviously some assists are more impressive than the goal. Sometimes, the secondary assist is actually the most impressive. Same goes for basketball, but only the primary is tracked, and only the person scoring gets points.

I would actually prefer to separate them entirely and have a trophy for assists champion. But since it's "the ways it's always been done " we're not gonna see the AR trophy go away.

At the end of the day though, I'm general goals are more valuable. But the thing is - historically most of the greatest players have been better playmakers.

So goals > assists, but also the best playmakers have historically been better than the best pure goal scorers, if that makes sense?

At the end of the day stats are just stats, it's far more nuanced than just looking at totals when we rank a player's impact on winning

 

[ijuka]

As a predictor for team wins, if a goal was 1, primary assist was like 0.85 and secondary assist like 0.6 or something of that nature, don't remember the precise numeric values but secondary assists were significantly below primary points.

 

[daver]

Let's go with some narratives:

First and foremost, generally speaking, the best playmaking forwards, as identified by having the most first assists, get the most second assists. The narrative that one can randomly "pile up" 2nd assists is false but the expection to this is Backstrom who clearly benefitted from being on OV's line the vast majority of the time.

There is no different weight when talking about Crosby vs. Ovechkin given other factors such as quality of linemates and deployment. The same can probably be said about McDavid, Jagr and other generational offensive point producers who produced regardless of the quality of their linemates.

There was no difference in weight when comparing the value of Jagr and Thornton in 05/06.

Though not a bit controversial, Sedin's assist to goals ratio was irrelevant in 09/10.

Wayne was as valuable when he scored 92 goals or over 150 assists.

The theme here is that, generally speaking, among superstar offensive talent, the weight becomes a lot more irrelevant than it may be among bottom 6 talent.

 

[Goose]

I didn't want to complicate things by getting into primary versus secondary assists with this, but it's clear that's the most important part.

 

[Scorevechkin]

1 goal =1 point
1st assist = 0.75 point
2nd assist = 0.50 point

 

[CashMash]

I don't know, and I don't think anybody has a good answer. Yes, yes... there are more assists than there are goals and we should adjust player values according to that. RIGHT?!

But...

I would NEVER take Ovechkin over Crosby in the same draft. Spare me the "B-b-b-but Crosby had more help!!!" excuses. Ovechkin had Backstrom glued to his hip (excellent passer, good defensively) for a very long time and won Presidents' trophies. Teams will never be equal, but it's not like he was playing with bums. Even early Semin was very good.

 

[banks]

It's nearly impossible to quantify the difference in "value" between a goal and an assist. They're pretty close to equal, since clearly neither would have happened without the other.

Goals are probably worth "more". But it's not enough to matter unless you start looking at individual plays to determine how difficult each part of the play was. Since no one can do that on a large scale, I'd say its close enough to equal "value" that it isn't much worth talking about the difference.