Wednesday, March 28, 2012

How Much Are The Teams With The Largest Salary Commitments Actually Spending? (Or State Of Contracts In The NHL, Part 2B)

Fancy stats guru Eric T. tweeted to DrivingPlay after my State Of NHL Contracts Part 2 was posted, asking thus:

"So it matters whether you look at % of spend or % of cap, then, I guess? You'll find same % of spend, but lower % of cap?"

Decoding this question out of Twitter pidgin, he essentially asked - are teams spending as much money (proportionally) as they did in my comparison year (08-09)? And the answer is: Yes. They're actually spending more, at least the top teams are. Sussing out this sort of information is time-consuming, so rather than looking at the entire NHL, I took the top 12 teams in terms of cap hits (according to from this year and from 2008-09. I then looked at the total salary they paid out, counting all one-way contracts¹, and counting all call-ups proportionally to the time they were called up. Most of these were estimates based on me wanting to get through this data collection as fast as possible, so these numbers could be off by a tiny amount.

I'll spare you the individual details, but it suffices to say that the top teams are spending more, proportionally to the cap, than they did in 2008-09, while their cap spending has gone down (relative to the cap). Here's an average of the top 12 teams' actual spending as well as their total cap hits.

YearTotal Paid Salary% Of CapTotal Cap Hits% Of Cap

It's clear, then, that among top teams, salaries have continued to rise even faster than the cap. We can see that cap hits and cap payments used to be pretty much equal, but now one has far outstripped the other. Part of that is a function of the fact that these severely front-loaded contracts are new to the NHL; I imagine there will be some settling. On the flip side, one also imagines that salaries among bottom teams would show a similar fall - as teams like the Islanders struggle to make a profit even with their team parked right at the salary floor, they are using players with significant bonuses and backloading contracts to get around paying the full $49M that is this year's salary floor.

It's also clear that something in this system has to change if the NHL wants to maintain that the salary cap increases parity, because as it is now, the top teams are all spending well over the cap in total salaries each year, and I don't see why this would change in the future.

¹ Data on one-way contracts was incomplete from 2008-09

The Driving Play Podcast - 2012 Playoff Race Edition

Good evening, friends. We're back again with another edition of the Driving Play Podcast, this time discussing the tight playoff races in the Eastern and Western Conferences. We are happy to be joined by Corey Sznajder of the Carolina Hurricanes blog Shutdown Line. Just as a production note, we recorded this podcast yesterday before Buffalo's win over Washington, the Pens losing to the Islanders and the Rangers beating the Wild. Small sample sizes and having confidence in Dale Hunter can be cruel mistresses.


Download (Right Click and choose "Save Link As...")

Saturday, March 24, 2012

The State Of Contracts In The NHL Today, Part 2

In Part 1, I examined how the top contracts in the league have essentially stagnated in the last 4 years, which means that relative to the salary cap, they've gone down. We know that entry level contracts have a cap on them which essentially limits how much they can pay out, and cursory examination of will show that the NHL essentially has a slotting system for entry-level deals.

However, one place that contracts could be exploding is for RFA forwards. Perhaps teams, wary of other teams offer sheeting their own players, or fearful of being taken to arbitration, are being more generous with their RFA players. A closer study shows that this isn't the case - once again, RFA salaries have stagnated despite the expanding cap.

To study this properly, I first sorted for RFA-age forwards who played more than 40 games with at least .4 Points Per Game (numbers courtesy of I then looked at all the eligible forwards who signed new contracts in the summer of 2008 versus the summer of 2011. However, one cannot just sort contracts willy-nilly - I examined whether or not an RFA player was eligible for arbitration, how many years of UFA eligibility the contract he signed gave up, and how long his contract was signed for. This didn't leave me with too many contracts to directly compare, but I think it makes my point just as well. If one looks at the entire list of contracts signed, they would see that there is simply not much difference between then and now despite the nearly 15% increase in the salary cap between the two summers.

Without further ado, here's our first table; we're comparing forwards who signed 2 year contracts, who were not arbitration eligible, and who gave up none of their UFA years:

YearsTotal ContractsPPG AverageAvg. ContractAvg. % Of Cap

(Salary numbers courtesy of and

If we remember that 2008 was a higher scoring year than 2011, we can basically call the points per game even, and yet the players are making less money, and far less money against the cap. Perhaps that's a fluke - we are after all only looking at 8 contracts, and points per game isn't a scientific way to examine a contract's value.

Here's players who sign 5 year contracts with 1 year of UFA included who aren't eligible for arbitration:

YearsTotal ContractsPPG AverageAvg. ContractAvg. % Of Cap

Again, we find a similar pattern. If we look at the percentage of the cap these contracts take up, it's higher in 2008.

Here we're looking at contracts that go for 2 years, don't have any UFA years, and are for players who are arbitration-eligible:

YearsTotal ContractsPPG AverageAvg. ContractAvg. % Of Cap

Once again, players who signed in 2011 got less money than players who signed in 2008. It's hard to imagine that agents aren't aware of these sorts of numbers - teams must be able to sell the fact that they don't have the same kind of money to sign these players.

In Part 3, I'll be examining possible reasons why contract amounts are stagnating and contracts are going down relative to the salary cap.

Tuesday, March 13, 2012

The State Of Contracts In The NHL Today, Part 1

It's been ripping good fun watching teams ink their best players for periods of time that I hadn't even considered possible. By the time Ilya Kovalchuk's contract ends, people will be riding on Hoverboards, but not for leisure, as they will be using them in the fight against the deadly invasion of space aliens. Remember that you read this on Driving Play first; perhaps that will be your last thought as you enter the terrible tentacled maw of an interstellar conqueror.

I seem to have gotten off track here, but it's going to be fascinating watching teams maneuver with these contracts on board over the next 5-8 years. Some of them are time bombs, waiting to explode in a team's face. We can't say which. Others will provide the foundation for a Stanley Cup championship. And while I expect this trend to be reversed under the next CBA - even as many teams have one of these contracts on their books, they no doubt realize the enormous advantage conveyed upon teams that have multiple ultra-long-term contracts, and are going to work to remove the possibility of such a contract under the next CBA - the cat is out of the bag. Let's look at when these big contracts end.

2018: 9
2019: 4
2020: 6
2021: 5
2022: 2
2023: 1
2024: 0
2025: 1

That's a total of 28 contracts that end more than 6 years from now, and that number will no doubt increase when Ryan Suter, Zach Parise, and Shea Weber are re-signed this offseason, among others. 17 NHL teams have a contract that ends in 2018 or later.

Most of these contracts are at 'discount prices.' For instance, Marian Hossa and Ilya Kovalchuk are signed to deals that have a 5.2M and 6.6M cap hit, respectively, and that end roughly when your life's course will be fully determined and there's no extricating yourself from your fate. Their cap hits on contracts both signed after the end of the lockout, when the salary cap was 39M? 6M and 6.3M, respectively. Their cap hits have barely moved even as the salary cap itself has nearly doubled, and this was for players who were RFA when they signed their post-lockout deals, and UFA when they signed their lifetime contracts.

I decided to look at the top 20 largest cap hits in the league in 2008-09 and the biggest cap hits in the league in 2011-12. The summer of 2008 was when teams seemingly decided that the cap would never go anything but outrageously upwards - the economy was robust, NHL fans had forgotten about the lockout, and Glen Sather knew that the surge of Wade Redden jersey purchases would pay for that contract by themselves.

YearTotal Top 20 SalarySalary Cap% Of Total Cap Space

The 'Total Top 20 Salary' column refers to the summed amount of the top 20 biggest cap hits in the league. As we can see, it's barely moved despite the fact that the cap has grown by more than $7M. The '% Of Total Cap Space' refers to the top 20 salaries' total being divided by the salary cap amount times 30, for the 30 teams in the NHL. We can see that the percentage has gone down by nearly a full percent. That doesn't sound like very much, but we're talking about nearly 1% of 1.7 billion dollars. That means that cap-hit wise, the top 20 contracts are down by nearly an average of $766K relative to the cap. In total dollars, as we can see, things have remained relatively stagnant.

The deals that are super-long term don't really fit into this mold either, as only 5 contracts in the current top 20 biggest cap hits end in 2018 or after. That means these players whose big contracts end earlier will almost certainly face pressure to sign for a lower cap hit - odds are they will be out of their prime when their current deal ends.

In Part 2, I will examine RFA forwards to see if this trend also holds true among lower paid players, and in Part 3, I will give some reasons why the NHL is evolving in this way.

Wednesday, March 7, 2012

Has Ilya Kovalchuk Changed For The Better?

It's a thing that gets repeated in the media all the time - Ilya Kovalchuk's making a Commitment to Defense. He Wants To Win and is now Listening To The Coaches. See here as Bobby Holik holds him up as a paragon for Alex Ovechkin to emulate. See here as Pierre Lebrun calls him a Hart Trophy candidate. One thing that's not getting talked about is how Kovalchuk's shooting percentage has gone to seed as a New Jersey Devil. Kovalchuk, as of March 8th, is shooting 11.1% as a Devil, scoring on exactly 1 out of every 9 shots. As an Atlanta Thrasher, he shot 14.95% - the difference may not sound like much, but that's scoring on more than one out of every 7 shots. For a player who shoots upwards of 3.5 times a game, it's a significant difference - assuming Kovalchuk's career shot rate and an 80 game season, it's a difference of 11 goals a year. That's going to be difficult to make up in other ways - the question is, can we account for this difference?

First, we must point out that shooting percentage for everyone should be down - league-wide goaltending has gotten slightly better over the course of Kovalchuk's career. According to, the average save percentage was .908 in Kovalchuk's first season, it is now .914. In addition, there were many more power plays league-wide early in Kovalchuk's career - power plays are at a 30 year low again this season. Kovalchuk had 27 power play goals in 2005-06, the leader this year is not even on pace for 20. So we should expect some sort of drop in total shooting percentage, but his percentages have dropped both at even strength and at 5 on 4.

Here's Kovalchuk's shooting percentages since 2007-08, split into 5v5 and 5v4 (EN goals not removed, source

YearGoals5v5Shots5v5S%Goals 5v4Shots 5v4S%

We don't have enough information to say that's a trend, but it doesn't look good. Still, a smart fan who's good at the eye test will tell me this: 'Sure, Kovalchuk's shooting percentage is down at even strength - it's because he's not cheating for breakaways like he used to. He's gotten better defensively.' And sure enough, an examination of save percentages yield some hope. I compared Kovalchuk's save percentage while on the ice 5 on 5 to when he was off (source:

Again, it's difficult to ascribe this to anything besides randomness, but at the very least, what appears to have been a trend is reversing. It's difficult to tell though, as Kovalchuk's teams have largely gotten below-average goaltending. However, at least this season, it's difficult to ascribe that to him or players playing like him on his line.

Here's a look at Goals For per 60 minutes at even strength and Goals Against per 60 minutes (source:


Again, what seems to have been a high goal player on both ends has become a player much closer to the NHL average in terms of goals for and against. So let's go to the real stats, the Corsi and so forth (source: and

YearCorsi RelCorsi ONZone StartES +/-
The Corsi Rel is slightly improving but it's difficult to say that's that great given the zone starts - New Jersey is slightly above 50% in total zone starts. Plus, Kovalchuk plays with Zach Parise, one of the game's best play drivers before this year. It's hard to find much of an indication here that Kovalchuk is becoming a substantially better player - we can't know for sure if his defense was responsible for his goalie's decreased save percentage, nor can we necessarily call his upswing in that department this year a 'trend', as last year he was substantially below par.
The one place where we can say for sure that Kovalchuk has 'improved' is a place where he's rarely played - the penalty kill. While Kovalchuk is getting better zone starts than just about anyone on the PK, his Corsi Rel ranks 2nd in the league among PKers with more than 1 minute per game of short-handed ice time, and he has 3 short handed goals and 2 assists. Kovalchuk has been on the ice for 5 goals for and 2 goals against while shorthanded 4 on 5, a remarkable feat in 70 minutes of PK time. As far as I can remember, New Jersey has not been shorthanded at the end of a game in which they trailed by 1 or 2, which might warp the Corsi results, as the Devils would try to pull their goalie and attempt lots of shots on the opposition's net. Kovalchuk has been a top penalty killer, and that's certainly a skill most wouldn't've thought he was capable of before this season.
In summary, while Kovalchuk's shooting percentage numbers have declined, his Corsi Rel has slightly improved. We also can't forget that Kovalchuk has moved to right wing for most of this season - perhaps that's part of why his shooting percentage has declined. Still, even though Kovalchuk has earned many plaudits for this season, Devils fans have to be concerned that the guy they got is not entirely the guy they paid for, and it remains to be seen whether that will turn out to be a positive development.

Tuesday, March 6, 2012

The Driving Play Podcast - Trade Deadline Recap

Good evening, friends. We're finally back with the final installment of our NHL trade deadline series of podcasts. This time, we're evaluating the winners, losers, and the catastrophe that was Scott Howson and the Columbus Blue Jackets.

We were also privileged to again be joined by Dirk Hoag of On the Forecheck to get his reaction to our take on Nashville's deadline dealings.

All four of us were able to get together to record this edition, so go ahead and give it a listen.

Download (Right Click and choose "Save Link As...")

Sunday, March 4, 2012

Yet Another Reason Why Carrying 2 Enforcers Is Stupid

Discussions that bash enforcers are the lowest hanging fruit in the hockey blogosphere. If you're reading this, you either think that A: having an enforcer on your team is dumb or B: I or anyone else is dumb for suggesting otherwise (psst, it's most likely A). Regardless, it's tough to not think either A or B if you expose yourself to any talk about hockey.

I've resigned myself to being a fan of a team that's going to carry at least one 'enforcer'. The players seem to like having guys like that around, and while I'm no longer a fan of the scheduled scrap, where the commentator talks about how these guys 'know their job' or whatever while both guys decide to whale on one another for what seems like no good reason, it does still give me a vicarious thrill on those very rare occasions that a fighter from my team straightens out a pest who's been taking liberties with our non-fighters. We don't see much frontier justice on the East Coast - it's more about lingering resentments and therapy bills out here. Still, I do not understand having two enforcers on the roster when the NHL roster limit is only 23 players - the roster is simply not large enough to carry around players who, when push isn't coming to shove, quite simply hurt the team any time they step on the ice.

Two enforcers are can be worked around for most of the season if a team carries a 22 man roster and always dresses at least one enforcer. This way, a GM always has an extra spot open in case a forward gets nicked up and is day to day - he can call up an extra forward who he actually considers a hockey player. He doesn't have to use emergency recall, which is allowed only if the team falls below the 12 forwards, 6 defensemen, and 2 goalies that the NHL considers a minimum roster. Judicious use of injured reserve means his coach is never forced to play both enforcers, though he might choose to in a rivalry game, or to make an example of a real hockey player, or possibly just to throw caution to the winds. As long as he keeps 22 men on the roster, the coach's hand is rarely forced unless he's on the other side of the country from his AHL team.

This is not the case after the trade deadline, however. After the deadline, while there is no longer a limit to the number of players a team can have on their roster, a team is only allowed 4 non-emergency recalls from their AHL team from now until the end of the playoffs. So let's say our example NHL team is going with a 22 man roster and a top 9 forward gets injured after the trade deadline. A player from the 4th line is promoted into the top 9 and either the coach has to dress both enforcers, dress a defenseman as a forward, or the GM has to waste a non-emergency recall to bring up a forward who can play hockey. If the team's out of the playoff race, using a recall is no big deal, but if the team figures on making the playoffs, it's going to want the maximum number of available bodies. This is precisely what happened to the New Jersey Devils, and yesterday they lost 1-0 on a goal that was scored while their 4th line, made up of 2 enforcers and a waiver pickup, was on the ice. Would it have made a difference to have a real hockey player out there? Who knows - perhaps not. It certainly couldn't hurt to try, though.