The Contract Year Is Worth 1.8 Points

You have heard it in every draft room and every trade thread. He is playing for a contract. It is usually said as though it settles something: a reason to buy a player this year, and a reason to be nervous about him the year after he signs.

Two claims live inside that sentence. The first is that players raise their game in the final year of a deal. The second, which is the one people actually act on, is that they drop back once the money is guaranteed. I have contract history going back two decades sitting next to complete season stat lines, so rather than argue about it I joined the two and counted.

Only one of them survives the count, and it survives smaller than anyone would guess.

What actually happens

The average skater across the three seasons that matter (the year before the walk year, the walk year itself, and the season after signing the next deal):

The rise into the contract year is real. It is also +0.022 points per game, which over a full eighty-two-game season comes to about one and three-quarter points.

And it is close to a coin flip whether any given player gets it at all: 54.8% of them improved, 45.2% did not. The median improvement is +0.015, smaller than the average, which tells you the average is being pulled up by a handful of players having genuinely big years rather than by everybody nudging upward together.

So the first claim is true in the narrowest possible sense. Players do score more in a contract year. They score about two points more, and one in every two and a bit of them doesn't.

The half that is simply false

Now the second claim, which is the one that changes behaviour. Everybody is wary of the player who just signed. The theory is that he has his money and the urgency is gone.

The season after signing: +0.002 points per game. The median is −0.001. Half of players declined and half improved (50.6% and 49.4%), which is what a coin looks like.

There is no post-payday decline. The year after a player signs looks exactly like the walk year that earned him the contract. Whatever a player showed you while he was auditioning, he mostly kept showing you afterwards.

The belief has a real cost. Fading a player because he just got paid is a decision made against evidence that does not exist.

There is one clean exception:

Forwards keep climbing after they sign. Defencemen give a little back. If you want to hold a version of the post-contract worry, hold it about defencemen only, and hold it gently: a hundredth of a point per game is under a point a season.

It was never about effort

Look again at the second row of that first table. Ice time rises across the three seasons in almost exactly the same shape as scoring: 16:55, 17:18, 17:22.

That is the whole effect, and it points away from motivation. If players were bearing down in a walk year and relaxing afterwards, you would expect scoring to spike and fall while their minutes stayed flat. Instead the minutes move first and the points follow.

That suggests a test. If the bump is ice time, it should appear only in players who have ice time available to gain, and vanish in players who are already playing as much as anyone can play.

A player under fourteen minutes gains a full minute a night and about five points a season. A player already over twenty minutes gains nothing: his ice time actually falls slightly, his scoring is flat, and he is less likely than a coin flip to improve.

This is the useful form of the finding. The contract year does not do anything to a player who is already playing twenty-two minutes on the top power play. There is no lever left to pull. When somebody tells you a star is motivated this season because his deal is up, the honest answer is that the data has no idea what they are talking about.

Mostly, it is a birthday

The next result undercuts even the modest version.

The contract-year bump and being young are largely the same players. Twenty-two year olds improve by +0.053 and thirty-two year olds get worse by −0.018. That is the age curve, which does the same thing to players who signed eight-year deals as toddlers.

So I held age still and looked only at players aged 26 to 31, where the curve is roughly flat, and asked the ice-time question again inside that band:

It mostly collapses. Take age out and what is left is small and messy.

I want to be careful here, because that last row does not fit. Players over twenty minutes in their prime came in at +0.012, which breaks the tidy gradient, and every cell in that table is under 120 cases. I am not going to tell you the contract-year effect is exactly zero. I am telling you it is small, that most of what looks like a contract effect is a player getting older and playing more, and that the confident version of the belief is not supported by nineteen years of evidence.

Which brings us to this season

Forty ranked players had contracts expiring next July when I pulled the board. I was going to hand you the list. Then I started checking it against the signings and the list kept shrinking, and it shrank in a way that makes the point better than the list would have.

Cale Makar signed, for $20.4M. Macklin Celebrini signed. Ivan Demidov signed for $9M, Bowen Byram for $12M, Zeev Buium and Dan Vladar too.

Look at who those are. Makar and Celebrini aside, they are young players with room to grow, precisely the profile everything above identifies as the only group a walk year means anything for. Teams are not waiting to find out. They extend the players with headroom, because that is where the value is, and they do it before the walk year happens.

Six gone from forty, and the pool is down to thirty-four.

What that leaves in the walk-year pool is mostly players whose role is already fixed, the group the data says gets nothing.

So the list is short, and every name on it needs the same question asked: is there a minute available to him that he is not already playing?

Five who still qualify

Matvei Michkov, Philadelphia, 22, ADP 167. Fifty-one points in eighty-one games on 14:50 a night, the lowest ice time of any ranked forward left in the cohort, at a rate of 2.55 points per sixty. He is the profile in its purest form: young, entry-level deal expiring, and more unplayed minutes than anyone else on the board. He is also nearly free.

Will Smith, San Jose, 21, ADP 94. Fifty-nine points in sixty-nine games at 2.82 per sixty on 18:11. The rate is already top-tier and the minutes are still climbing. The most expensive of the five and the one whose case rests least on projection.

Luke Evangelista, now New Jersey, 24, ADP 173. The trade is the story. A new team is the mechanism this whole article is about, the thing that changes a player's minutes. He scored 2.50 per sixty on 16:37 in Nashville, and he arrives somewhere new in the final year of a $3.0M deal, which is the one contract here that represents a real negotiation rather than an entry-level formality.

Kirill Marchenko, Columbus, 26, ADP 76. Sixty-seven points in seventy-six games, 2.79 per sixty on 18:57. Right at the top of the prime-age band where the effect is weakest, which is the honest caveat on him, but he has genuine rate and there is still a minute or two on the table.

Jimmy Snuggerud, St. Louis, 22, ADP 119. 2.62 per sixty on 16:42. I recommended him on Monday on his scoring rate alone and did not notice he is also in a walk year. That is not a second argument so much as the same argument arriving twice, but it is the shape this piece is looking for, and I would rather point out that I missed it than quietly leave him off.

And one the method throws out

Quinn Hughes is the biggest name in the walk-year pool and the clearest example of it meaning nothing.

He plays 27:44 a night. There is no version of a contract year that finds him another minute; there is barely another minute in the game. He is 27, past the age band where the bump lives. And he is a defenceman, the only group that measurably declines after signing.

Every filter that predicted the effect rejects him. He is going to be excellent this season because he is Quinn Hughes, not because his contract is up.

The same goes, with less ceremony, for Nikita Kucherov (20:20), Jason Robertson (20:15), Erik Karlsson (23:36), Nico Hischier (20:43), Vince Dunn (21:40) and Ryan O'Reilly (20:24): all in walk years, none with a minute to gain. And for Sidney Crosby, Alex Ovechkin, Evgeni Malkin and Mark Stone, who are in the 32-plus band where the measured change is negative.

Kirill Marchenko aside, the mid-career names with some room left sit in the age range where the effect is fading: real, but small. They are Drake Batherson (28, 17:21, and the best rate left in the pool at 3.11), Pavel Zacha (29, 16:51) and Jared McCann (30, 16:26).

And a dozen goalies are in walk years too: Askarov, Wallstedt, Dobes, Hofer, Kochetkov, Gibson, Kuemper, Lyon, Andersen, Hart and Wedgewood among them. My analysis is skaters only. I have no measured claim about goalie contract years, and I am not going to imply one by putting them in a table.

What would make me wrong

• The filter removes the players it would matter most to. Requiring forty games in three consecutive seasons throws out everyone who was hurt, buried or scratched, which is exactly the population whose ice time could move most in a walk year. This measures established players, and it may understate the effect for everyone else.
• My player universe skews good. These are the players in my database, which is built around rostered fantasy-relevant talent, not a random sample of NHL contracts. Fringe players are underrepresented.
• The prime-age cells are small. Eighty-five to 111 cases each, and one of them contradicts the pattern. I would not bet heavily on any single row of that last table.
• I measured what happened, not why. A player who improves in a walk year and keeps improving afterwards might have a new coach, a new linemate or a healed wrist. The contract is a label I attached to a season, not a cause I isolated.
• Ages are computed from birth year, so they can be a year out either direction near a birthday.
• The list will keep shrinking. Six of these players signed between my last contract sync and the day I wrote this, and 15 September is the last day a player can sign an eight-year extension before the maximum term drops to seven. Expect more of the interesting names to be gone by the time you draft.

The short version

Players score about 1.8 more points in a contract year, a bit over half of them improve at all, and the effect is mostly ice time and age rather than effort. There is no drop-off after they sign: none for forwards, a fraction of a point for defencemen. A player already over twenty minutes a night gets nothing from a walk year, because there is nothing left to give him.

The only players it plausibly matters for are young ones with minutes still available. And their teams have already worked that out, which is why Makar, Celebrini, Demidov, Byram, Buium and Vladar are no longer on this list.

What is left is Michkov at 167 with fourteen minutes and change, and a lot of players whose contract status is the least interesting thing about them.