When seeking advice on fantasy hockey trades, most people will tell you to go with proven talent over unproven potential stars. There is a limited amount of space at the top of the NHL pyramid and those who are at that level have shown that they can maintain a high level of play over a number of years. Meanwhile, the unproven talent may or may not achieve a high level of success at some point in the future.

Unfortunately, the stars of the NHL come at a high price tag. This is a logical idea –players are mostly paid relative to their level of production. The chart below shows Dobber’s projected point totals for each forward (no names attached) and the associated cap hit for that player. Clearly, there is a relation between production and how much the player costs on your payroll.


Now that this relation is established, we can use this information to help make decisions in building a team in a league that has a salary cap. Evidently, managing the team budget is not as simple as adding the best players available. You have to find the best combination of talent that fits under the cap ceiling. Your best players are the most important to your team’s success, so we will start there.

We like owning star players because of the massive production they provide from a single roster spot. But their NHL contract can be influenced by a number of other factors including marketability and intangibles. It is worth investigating whether or not the NHL’s elite forwards are worth the money they are paid in fantasy hockey. For this we are going to look at each player’s production efficiency in points per million dollars of salary:



The results show that the star players are worth the money they are paid, even in fantasy hockey.  There are obviously exceptions but the trend line shows that production efficiency actually increases as you move up the NHL ranks.  It is important to know that your best players are not in fact hurting your team.

The chart also reveals a great tool that can be used to complement your star players.  Without question, efficiency is key in cap leagues and there are many great bargains in the 40-70 point range.  The more efficiently you spend your cap dollars in that point range, the more freedom you have to own and keep superstars at the top of your roster.  Furthermore, they can be helpful in accommodating other contract renewals on your roster, for example when Claude Giroux’s current contract with a $3.75-million cap hit expires at the end of the 2013-14 season.

Unfortunately, in past Capped articles we learned that long-term bargain contracts are extremely rare.  Most of the cap bargains are very good players who are nearing the end of their existing contract and are clearly due for a big payday.  In a one-year league, this is not an issue.  However, in a keeper league you will not be able to rely on them as super bargains long-term.  You will have to decide if they are part of your core and either accommodate the new bigger contract or move on and find alternative options.  In a future article we will look at methods to obtain good cheap labor to round out your roster.

Using this information you should have the tools necessary to allocate cap dollars more effectively in your keeper league.  Do not be afraid of Steven Stamkos’ $7.5-million contract or even Nicklas Backstrom’s $6.7-million cap hit.  Those are inflated salaries, but their production justifies the money.  But most importantly, you must retain that a player’s cost is a variable that greatly impacts his fantasy value and that each player on your roster affects the amount of flexibility you have to improve your team.  If you can manage your team’s talent and budget properly, you may be able to avoid a plateau and break through to win the championship in your league.


Recent posts from Eric Daoust:


Write comment
Comments (5)add comment

ericdaoust said:

... A few points to answer:

1) Yes that is Scott Gomez in the upper-left corner.

2) Coming out of the original draft the talent is going to be spread out and teams will have a star player or two. But as time goes on that is probably not the case (bad trades, failed rebuild projects, etc.), especially in a larger league which is becoming more and more common these days. You can have a team full of efficient players with tons of cap space but not enough points to win. Someone has to do the heavy lifting. The main point of this article was to see if there was a major drop-off in points vs cap hit as you move towards the elite.

3) I sampled all players because of the increasing number of huge leagues. I am in a 24-team league where only a couple dozen of NHL regulars are not owned. I can only imagine what a 30-team league would look like. If I had limited the sample to players projected to get over 50 points, then according to this data the trend would indicate that the stars are not worth their money. But in that case your waiver wire would be so rich in talent that you would be able to find cheap alternatives while sacrificing very little (or nothing at all) in production. As Ryan pointed out, finding cap bargains is huge and in this setup it would be easy to make it all work. So while the trend would have been different, I do not believe that the conclusion could be made that the elite are overpaid.

4) Yes I did look into this. The slope is significant within a 95% confidence interval. Thanks for pointing this out, because the slope is very small which is pretty close to actually being flat.

I am at my limit for the evening (illness) but if you guys want to discuss more you can reply here or PM me and I'll get back to you by the end of the weekend. I am having fun doing these articles, but I am also looking to get better so every little bit helps.
October 12, 2012
Votes: +0

Ryan Ma said:

... Solid Article Eric,

Only thing that I'd like to add is that "high priced' forwards aren't going to win you the championship...

Every team, if they've drafted properly, is going to have a Crosby/Malkin/Ovechkin/Stamkos/Backstrom type... so every team probably is going to have their high priced forwards anyway. And generally speaking most will pretty much score around the same range anyways, so there probably isn't going to be a large discrepancy, where a high priced forward is going to be leaps and bounds better than another option.

What is going to win you championships is being able to identify those more "efficient" options year after year. They ones that will tally 60+ points, but cost you $1.5 mil... Generally speaking it's probably your young kids on ELCs...

So to answer your question, are star forwards worth their money? Well the easy answer is of course they are. But will they be the ones that will garner you the championship, probably not...
October 11, 2012
Votes: +0

lesouder said:

$7.5M Even without names attached my gut tells me the ~$7.5M ~25 point player MUST be Scott Gomez haha.
October 11, 2012
Votes: +0

Hey_Robbie said:

Hey Robbie
... Oops: "including them greatly affects the results.
October 11, 2012
Votes: +0

Hey_Robbie said:

Hey Robbie
Do these data really show that top forwards are worth it? Thanks for your work and presentation, Eric.

I have two questions regarding your statement, "The results show that the star players are worth the money they are paid, even in fantasy hockey. There are obviously exceptions but the trend line shows that production efficiency actually increases as you move up the NHL ranks. It is important to know that your best players are not in fact hurting your team."

First, looking at the small slope of the trend line on the cap hit vs. projected points relative to the spread of the data, is the slope significant, i.e. is a slope of zero within a 95% confidence interval? 90%? 80% even? A linear regression of any data set is unlikely to find a slope of exactly zero, but that doesn't mean the trend isn't actually zero. I have no idea whether you actually looked at this, and my sincere apologies if you did and I am calling it into question. I am just pointing out that not all regressions mean there is a trend that is any different than you are likely to see due to random error.

Second, and much more importantly, I think, is the fact that your data set includes all forwards with projections. This seems to throw the phrase "even in fantasy hockey" especially into doubt. It is a vanishingly small percentage of leagues in which the forwards at the far left of the plot are fantasy-relevant*, and including them greatly effects the results. If you look at the relationship of only players projected to get more than, say, 25 points (very roughly it looks like the same would be true for 30 or 20 as well), you actually might find the exact opposite to be true, i.e. that points efficiency decreasesas you go up the ranks. At a minimum, the trend would be even less likely to be significant.

*N.B. I'm not saying that none of these players is fantasy relevant, but if they are it is probably for multi-cat attributes, in which case points efficiency would not be a measure of whether these players are bargains or hurting your team.
October 11, 2012
Votes: +0
You must be logged in to post a comment. Please register if you do not have an account yet.