Confirmed Signing with Link: [TOR] Marner re-signs (6 years, $10.893M AAV) Part III
- Thread starter Lshap
- Start date
-
Work is still on-going to rebuild the site styling and features. Please report any issues you may experience so we can look into it. Click Here for Updates
- Status
Malkin scored 106 points and went to the finals in 07-08, he also finished second in Hart voting.
Matthews has a career high of 73 points
As I've already said, those were the years I had on hand. I didn't think everybody would have a fit over me using that year to show a point, when PPs were still heavily inflated during Malkin's ELC, and as I already said, the number I said actually under-represents the difference in PP opportunity between the two individual players.
uncleben
Global Moderator
Yes, they did. Though Malkin got more benefit than most, as Pittsburgh was near/at the top of the PP opportunities list both years. Not to mention getting to play with the best player in the game.
As much as Dekes and I disagree I’ll stick up at this, it’s not just about Malkin. The entire league averages more pp’s than Matthews elc years back then. I just don’t agree it’s the be all end all that he does
Tender Rip
Wears long pants
Dekes.... the reason your comparison is absurd, never mind how numbers are interpreted or abused, is that Malkin at the time of signing was a fair bet to become just as good as Sid and A.O - or better - and thus had a good shot at becoming the best player in the league. The very next season he arguably was and justified everything.
Matthews is a great player, but I don’t think there are anyone who have such thoughts for him, also because it is obvious that Connor is at least a cut above among his generation.
Matthews is a great player, but I don’t think there are anyone who have such thoughts for him, also because it is obvious that Connor is at least a cut above among his generation.
JoeThorntonsRooster
Don’t say eye test when you mean points
Since there is a lot of talk about what factors drive the market, I looked specifically at post-ELC contracts for players who signed on a 5-year term at over 7% of the cap at the time of signing, and looked at the following metrics: ELC P/GP, ELC G/GP, ELC ESP/GP, ELC PPP/60, Contract Year (Team Last 82) Points, Contract Year (Team Last 82) Goals, Contract Year (Team Last 82) P/GP, Contract Year (Team Last 82) G/GP, Contract Year (Year of Signing) P/GP, Contract Year (Year of Signing) G/GP.
If you're confused by the two different metrics for contract year, a few of these players signed in the middle of season. For "contract year goals/points", I used their team's last 82 games as their "contract year". For goals/points per game, I used their team's last 82 games and their numbers in just the season up to date before they signed their contracts.
Here are those players:
[TBODY]
[/TBODY]
Here are the correlations for cap hit % and each of these metrics:
[TBODY]
[/TBODY]As you can see, all the raw point based metrics vastly out-produce the goal based metrics, or the "context-based" stats that end up excluding more context than they provided.
Here is what these look like on a graph, for reference:
Despite what some posters tell you, market value is NOT primarily dictated by ELC ESP/60 or ELC PPP/60; it is primarily dictated by raw point totals. Whether intrinsic/actual value is dictated by those metrics, or whether or not market value should be dictated by those metrics is a whole other story. But these are NOT the factors that GMs have historically paid for.
There is an argument to be made that by paying for Marner, whose value is perhaps in this range or only slightly below his pay range based on ELC ESP/GP and ELC PPP/60, is fair based on his actual value to a team, if you believe that ELC ESP/GP and ELC PPP/60 are more descriptive of the value that a player provides than just points per game, either in the contract year or the full ELC.
I will look over 6-year contracts shortly to see if these factors drive pay any differently for players who sign for 6 years, but I strongly suspect that they will not.
I have tried to find some ways to combine ELC ESP/GP and ELC PPP/60, and when testing the correlation that each of these have with cap hit percentage, I've got R^2 values below 0.5, which pale in comparison to the R^2 value for any of the raw point based metrics and cap hit percentage.
If you're confused by the two different metrics for contract year, a few of these players signed in the middle of season. For "contract year goals/points", I used their team's last 82 games as their "contract year". For goals/points per game, I used their team's last 82 games and their numbers in just the season up to date before they signed their contracts.
Here are those players:
| PLAYER | Cap Hit Percentage | ELC P/GP | ELC G/GP | ELC ES P/GP | ELC PPP/60 | Contract Year (team last 82) Points | Contract year (team last 82) goals | Contract year (team last 82) P/GP | Contract Year (team last 82) G/GP | Contract year (this year) P/GP | Contract year (this year) G/GP |
| Sebastian Aho | 10.37% | 0.81 | 0.34 | 0.55 | 5.38 | 83 | 30 | 1.01 | 0.37 | 1.01 | 0.37 |
| Auston Matthews | 14.63% | 0.98 | 0.53 | 0.73 | 6.45 | 70 | 34 | 1.21 | 0.59 | 1.21 | 0.61 |
| Jake Guentzel | 7.55% | 0.72 | 0.33 | 0.60 | 4.51 | 60 | 25 | 0.73 | 0.30 | 0.89 | 0.41 |
| Dylan Larkin | 7.67% | 0.58 | 0.23 | 0.48 | 2.44 | 63 | 16 | 0.77 | 0.20 | 0.77 | 0.2 |
| Jamie Benn | 8.16% | 0.72 | 0.32 | 0.56 | 3.51 | 63 | 26 | 0.89 | 0.37 | 0.89 | 0.37 |
| Steven Stamkos | 11.64% | 0.95 | 0.49 | 0.56 | 5.75 | 91 | 45 | 1.11 | 0.55 | 1.11 | 0.55 |
| Bobby Ryan | 8.59% | 0.78 | 0.42 | 0.49 | 6.23 | 64 | 35 | 0.79 | 0.43 | 0.79 | 0.43 |
| Patrick Kane | 11.09% | 0.89 | 0.29 | 0.52 | 5.73 | 63 | 21 | 0.83 | 0.26 | 1 | 0.35 |
| Jonathan Toews | 11.09% | 0.84 | 0.39 | 0.52 | 4.85 | 63 | 32 | 0.79 | 0.42 | 0.8 | 0.3 |
| Phil Kessel | 9.51% | 0.57 | 0.30 | 0.45 | 3.2 | 60 | 36 | 0.86 | 0.51 | 0.86 | 0.51 |
| Paul Stastny | 11.64% | 1.00 | 0.34 | 0.62 | 5.55 | 67 | 20 | 1.02 | 0.30 | 0.94 | 0.19 |
| Evgeni Malkin | 15.34% | 1.19 | 0.5 | 0.69 | 5.57 | 106 | 47 | 1.29 | 0.57 | 1.29 | 0.57 |
| Corey Perry | 9.39% | 0.59 | 0.28 | 0.36 | 4.54 | 54 | 29 | 0.77 | 0.41 | 0.77 | 0.41 |
| Ryan Getzlaf | 10.59% | 0.73 | 0.28 | 0.32 | 6.50 | 61 | 22 | 0.76 | 0.28 | 1.15 | 0.31 |
| Dustin Penner | 8.45% | 0.51 | 0.33 | 0.34 | 4.01 | 45 | 29 | 0.55 | 0.35 | 0.55 | 0.35 |
| Sidney Crosby | 17.30% | 1.39 | 0.47 | 0.71 | 7.04 | 120 | 36 | 1.52 | 0.46 | 1.52 | 0.46 |
| Patrice Bergeron | 10.80% | 0.74 | 0.31 | 0.43 | 4.60 | 73 | 31 | 0.90 | 0.38 | 0.90 | 0.38 |
| Jonathan Cheechoo | 7.69% | 0.58 | 0.345 | 0.37 | 5.26 | 75 | 46 | 0.91 | 0.56 | 1 | 0.60 |
Here are the correlations for cap hit % and each of these metrics:
| Statistic | R^2 with Cap Hit % |
| ELC P/GP | 0.82 |
| Contract Year (Team last 82) P/GP | 0.75 |
| Contract Year (Season of signing) P/GP | 0.68 |
| Contract Year (Team last 82) Pts | 0.58 |
| ELC G/GP | 0.5 |
| ELC ESP/GP | 0.46 |
| ELC PPP/60 | 0.43 |
| Contract Year (Team last 82) G/GP | 0.18 |
| Contract Year (Team last 82) Goals | 0.12 |
| Contract Year (Season of signing) G/GP | 0.1 |
Here is what these look like on a graph, for reference:
Despite what some posters tell you, market value is NOT primarily dictated by ELC ESP/60 or ELC PPP/60; it is primarily dictated by raw point totals. Whether intrinsic/actual value is dictated by those metrics, or whether or not market value should be dictated by those metrics is a whole other story. But these are NOT the factors that GMs have historically paid for.
There is an argument to be made that by paying for Marner, whose value is perhaps in this range or only slightly below his pay range based on ELC ESP/GP and ELC PPP/60, is fair based on his actual value to a team, if you believe that ELC ESP/GP and ELC PPP/60 are more descriptive of the value that a player provides than just points per game, either in the contract year or the full ELC.
I will look over 6-year contracts shortly to see if these factors drive pay any differently for players who sign for 6 years, but I strongly suspect that they will not.
I have tried to find some ways to combine ELC ESP/GP and ELC PPP/60, and when testing the correlation that each of these have with cap hit percentage, I've got R^2 values below 0.5, which pale in comparison to the R^2 value for any of the raw point based metrics and cap hit percentage.
I don't agree that Malkin was a good bet to be as good as Crosby and Ovechkin at time of signing. He wasn't that far behind Crosby, but he was behind Crosby in pretty much every way, despite entering the league at an older age. He is closer to Ovechkin, but Ovechkin was so uncontested in goal-scoring until Matthews came along that it's hard not to put him ahead. While it's true that Matthews is likely a further bet at being best in the league, that is more a statement on McDavid, who is frankly currently underrated as well and has a really, really strong case that he'll be the best player ever drafted in the cap era so far. That said, Matthews has the best case of anybody for being 2nd best of this generation, and he's being paid less than 2nd or 3rd bests of previous generations who frankly, are worse in direct comparison even when giving them extremely generous adjustments. I know it's hard to admit when you've grown up putting so much stock into raw points and idolizing these players, but it's true.
Give Matthews the same quality of linemates and relative PP TOI/GP to the rest of the league that many of these past stars got, and he will also challenge for awards as soon as this year as he moves closer to his prime.
My favourite part was when you showed that Marner and Matthews both outperform the contract trendline for ES and PP production.
Tender Rip
Wears long pants
OK, so I was wrong. Someone is indeed thinking of Matthews in those terms 
.
For me it is a simple eye-test matter. Like Stamkos, Matthews is just not at that level, never mind that he will score tons of goals and probably challenge for the Ross at some point. But I assume we get no further. Time will tell.
.For me it is a simple eye-test matter. Like Stamkos, Matthews is just not at that level, never mind that he will score tons of goals and probably challenge for the Ross at some point. But I assume we get no further. Time will tell.
This just in: players who get lots of ice time on both 5v5 and PP on their team are very important to the team, thus get higher contracts.
But appealing to authority on ice time is just as useful as appealing to authority on AAV. So the best player makes the most, etc... The Leafs determined that Matthews and Marner were as important to their team as their AAV states, therefore they are set perfectly in the market.
You can't really believe finding r2 on only a small handful of players is going to be good enough to identify obvious outliers, such as how substantially lower the PP time is for the Leafs players compared to the others. It's just a massive elephant in the room when we sit here and talk about total points that you're not fooling anyone but yourself by trying to throw a blanket over it.
If "the market" will pay a player that scored total points 5v5/PP 20/40 the same as another player that scored 40/20 with a third of the PP time, I'm quite happy that we aren't paying the "market", because that will pay dividends in not overpaying replacement level players because their team structure allowed generous PP opportunity. Pretty plain and simple.
JoeThorntonsRooster
Don’t say eye test when you mean points
My favorite part is when you ignore the purpose of the post, and refuse to acknowledge that you were flat out wrong and that raw P/GP is a FAR higher driving factor of pay than ESP/GP and PPP/60.
You’ve been telling everybody in these threads, ever since you switched from ESP/60, that ESP/GP and PPP/60 are “the real way to judge production”, and that these factors are what really set the market. You have provided ZERO EVIDENCE for this claim at any point, going no further than saying “I looked at contracts, and it became clear to me that these factors are what drive contracts.”
Also, Marner isn’t even included in those charts, so I’m not sure what you’re looking at. I will post his along with other 6-year contracts tomorrow. The results are similar with contract year P/GP + ELC P/GP having the strongest correlation with cap hit percentage, but I believe Marner actually under performs even your cherry picked trend lines that don’t actually correlate with cap hit as well as raw P/GP.
This is incorrect. Beyond the flaws in your graphs, you are applying a causation to a correlation, and then pretending like any outliers are ridiculous overpayments, even when everybody knows exactly why they are outliers in that one stat.
Also... I noticed that "Contract Year (Team last 82) P/GP" has an R^2 that rivals your beloved P/GP, but interestingly, you make no mention of it despite Matthews' contract lining up pretty well with other contracts when going by it.
I've said that context is always necessary, but those stats give a much better view of production, which is true.
That's true. Matthews outperforms the contract trendline for ES and PP production. Marner outperforms the 5 year contract trendline for ES and PP production, while having a 6 year contract. Score!
JoeThorntonsRooster
Don’t say eye test when you mean points
There is no appeal to authority here. I am not saying that the market is correct for paying players the way that they do. Personally, if it were up to me to determine their actual value I would use something far more sophisticated than any of the models here. Think something that includes shot impacts, an annual age adjustment, etc, and places a heavy emphasis on 5-on-5 play driving in particular. But that is not what we are talking about here. This poster Dekes For Days has continuously told everybody that using ESP/GP and PPP/60 over their entire ELCs is the right way to judge production, and that it is thenway that the market has historically paid players. You once told me that P/60 was a bigger diving factor behind pay than raw points. When I looked at that, I looked at every contract signed by forwards in the 2018 summer and that was proven false as well.
The fact is, this is what the market is and this is how they pay players. This is not just one handful of players either. These are all of the forwards that I can find who signed >7% contracts on 5-year terms after their ELCs. Tomorrow I will post the same for 6-year terms and the results are similar. I am willing to do more in-depth research as I plan on building my own contract projection model at some point. All of my research has shown me that teams pay forwards mostly based on points.
I understand that contract negotiations do not just happen through using a points or points/game based regression and I understand that these players in particular had an argument to be outside of that trend line due to these factors. But the degree to which they were outside of that trend line is egregious. A lot of these factors also have multiple sides to them. 5-on-5 scoring and PP scoring rates per hour are both up considerably from what they were in the Malkin era. Toronto in particular is a 5-on-5 offensive juggernaut which is partly due to these players but the success of these players is partly due to that system as well. This half of the context gets ignored too often.
I understand the argument that you’d prefer to pay players based on their actual value than their market value, but the issue is that so far, it has only resulted in paying more. What happens if Tyson Barrie, who is 2nd in PPTOI/GP over the past two seasons and spends most of that time with MacKinnon and Rantanen, gets a market value extension based on his point totals? Then it is only going one way, and not the good way. It is not going to be easy to convince Tyson Barrie’s camp that his actual value is less than his market value, but by that you have used to try and convince me that Marner’s actual value is higher than his market value (as I have presented it), Barrie’s value is going to be lower than his market value. If Dubas can manage to get Barrie well below market value, I will give him credit where it is due, and respect that he is paying all of his players in a manner that contract value is closer to his perceived actual value than typical market value. At that point, that kind of pay scale would be a net zero relative to today’s NHL but it would be more fair to everybody if they were paid based on the value they actually brought. But right now, he has only done that with players who he has paid more than their market value. He paid Ceci more than double his actual value for example.
I assure you teams are as well, including Dubas & co.
I thought he mostly argued 5v5 p/60 PP p/60 but I don't really have the time to keep up with you or Dekes.
It's still just a handful though. A smart analyst knows when to concede there is not enough data to determine something. I've often thought it would be interesting to try and first determine the value of Non-Arb RFA year, Arb RFA year, UFA1, UFA2, UFA3, then use that distribution as an "adjustment" to estimate market value over a larger sample size.
I can understand that's why you would believe it to be so, but I have not seen any sign your research is built within the context of existing research. It's purpose built lazily for these discussions and comes off as disingenuous and makes me feel sad that you spend so much time when if you really were that interested, you could just do actual research, built on the shoulders of those smart people who have done research before you, and do something actually useful outside of the context and bias of these dumb discussions.
The comparison of their PP time, injury makes them egregious outliers. They are perfect examples of where some reasonable person who knows looking at trendlines can have mistakes in their outliers, then that reasonable person just makes a very small reasonable adjustment (adjust for TOI and PP), then they fall exactly right in line. While you want to sit there and average out the league and ignore outliers, these players are perfect examples of outliers for reasons that are completely out of their control.
Any time I've switched over to league rank for their 5v5 /60 and PP /60 stats they're still equal or better. The impact of the PP time far, far outweighs the raw /60 difference.
People only talked about the elite ones. Kapanen, Johnsson, Kerfoot and the swarm of league-min replacement level players.
You'll first have to not take what I said as saying that people who play a lot of PP time are inherently worse. But summing their PP time with 5v5 time is just, like, one of the worst things you can do when trying to find how impactful offensively a player is. I typically bow out of the discussion when it comes to valuing defensemen because I'm skeptical production is enough of the picture to get even a close estimate and I wouldn't pretend to know the more advanced team metrics teams will use to value them.
This ignores context around his QO, risk of arbitration, dumping Zaitsev, and needing a top-4 defensemen. Leafs are taking a bit of a risk, but it's only 1 year and a year where they already had the space available to sign mitch. So just give the guy a chance, why not? I would not be surprised if there is an evaluation efficiency on defensemen who play big roles on awful teams -- would you?
Malkin finished 9th and 3rd PPG before signing contract and was expected to be top 3 PPG producer going forward.
Matthews finised 27th and 21st in PPG and is expected to be TOP 10 in PPG by biggest defender of his contact on this forum
They were never comparable which was already proven even more during contract +1 year performance. Dubas did bad job if he was convinced by Matthews camp that Malkin contact should be used as comparable.
Matthews finised 27th and 21st in PPG and is expected to be TOP 10 in PPG by biggest defender of his contact on this forum
They were never comparable which was already proven even more during contract +1 year performance. Dubas did bad job if he was convinced by Matthews camp that Malkin contact should be used as comparable.
HoBOonFiRE
Registered User
Marner is overpaid compared to the rest of the league. If Matthews wasnt given that insane money @ 5 years this would of never happened.
When it comes to who is the leafs best player it's close between Matthews and Marner, they both do different things. Leafs management thought they could argue that Matthews was a center and scores at a ridiculous rate 5 on 5..but in the end they failed and paid Marner relative to Matthews instead of comparing him with the rest of the league. RFA deals are forever changed.
When it comes to who is the leafs best player it's close between Matthews and Marner, they both do different things. Leafs management thought they could argue that Matthews was a center and scores at a ridiculous rate 5 on 5..but in the end they failed and paid Marner relative to Matthews instead of comparing him with the rest of the league. RFA deals are forever changed.
JoeThorntonsRooster
Don’t say eye test when you mean points
How on earth is this information lazy or bad? This is looking at every 5-year post-ELC contract signed by a forward at over 7% of the cap in the cap era. All apples to apples, guys seem to get paid the most based on points. Here is the same stuff for 6-year contracts:
| PLAYER | Cap Hit % | ELC G/GP | ELC P/GP | Contract Year G/GP | Contract Year P/GP | Contract Year + ELC P/GP | ELC ESP/GP | ELC PPP/60 |
| Mitchell Marner | 13.37% | 0.28 | 0.93 | 0.32 | 1.15 | 1.04 | 0.63 | 7.19 |
| Johnny Gaudreau | 9.25% | 0.34 | 0.89 | 0.38 | 0.99 | 0.94 | 0.63 | 4.94 |
| David Pastrnak | 8.89% | 0.34 | 0.72 | 0.45 | 0.93 | 0.82 | 0.55 | 5.59 |
| William Nylander | 8.76% | 0.26 | 0.73 | 0.24 | 0.74 | 0.74 | 0.51 | 5.89 |
| Filip Forsberg | 8.22% | 0.33 | 0.73 | 0.40 | 0.78 | 0.76 | 0.47 | 5.83 |
| Jordan Eberle | 10% | 0.35 | 0.81 | 0.44 | 0.97 | 0.89 | 0.59 | 4.52 |
| Jonathan Drouin | 7.53% | 0.18 | 0.58 | 0.29 | 0.73 | 0.65 | 0.37 | 5.07 |
| Milan Michalek | 8.62% | 0.27 | 0.63 | 0.33 | 0.85 | 0.74 | 0.40 | 4.3 |
| Brandon Saad | 8.22% | 0.25 | 0.61 | 0.28 | 0.63 | 0.62 | 0.49 | 3.37 |
| Tyler Seguin | 9.58% | 0.26 | 0.57 | 0.36 | 0.83 | 0.70 | 0.46 | 3.60 |
| Bo Horvat | 7.33% | 0.21 | 0.51 | 0.25 | 0.64 | 0.57 | 0.40 | 4.10 |
| David Booth | 7.48% | 0.29 | 0.57 | 0.43 | 0.83 | 0.70 | 0.47 | 3.3 |
| Evander Kane | 8.75% | 0.30 | 0.59 | 0.41 | 0.77 | 0.68 | 0.51 | 2.88 |
| Aleksander Barkov | 8.08% | 0.26 | 0.61 | 0.31 | 0.69 | 0.65 | 0.42 | 2.93 |
| Statistic | R^2 with Cap Hit % |
| Contract Year + ELC P/GP | 0.69 |
| Contract Year P/GP | 0.68 |
| ELC P/GP | 0.57 |
| ELC ESP/GP | 0.52 |
| ELC PPP/60 | 0.33 |
What on earth is lazy about this?
This is incorrect. Beyond the flaws in your graphs, you are applying a causation to a correlation, and then pretending like any outliers are ridiculous overpayments, even when everybody knows exactly why they are outliers in that one stat.
Also... I noticed that "Contract Year (Team last 82) P/GP" has an R^2 that rivals your beloved P/GP, but interestingly, you make no mention of it despite Matthews' contract lining up pretty well with other contracts when going by it.
I've said that context is always necessary, but those stats give a much better view of production, which is true.
That's true. Matthews outperforms the contract trendline for ES and PP production. Marner outperforms the 5 year contract trendline for ES and PP production, while having a 6 year contract. Score!
Um, what? You think ELC P/GP is a metric that is beloved by me? I am the one who has previously told you that contract years have an extremely high emphasis placed on them, while you have told me it's pretty much all about the full ELC.
JoeThorntonsRooster
Don’t say eye test when you mean points
You in no way explained how it is lazy. You said it is “purpose built lazily for these discussions and comes off as disingenuous and makes me feel sad that you spend so much time when if you really were that interested, you could just do actual research, built on the shoulders of those smart people who have done research before you, and do something actually useful outside of the context and bias of these dumb discussions.”
That is just condescending drivel. You provided nothing to explain how that analysis is lazy or flawed in any way.
Tripod
I hate this team
Not so fast. Right now, Marner and Matthews are the outliers(only top 25 paid players NOT on 7 year deals or longer). Let's see what the other RFA forwards sign for. The Dmen deals did not really change.
- Status
Ad
Upcoming events
-
Australian Open - Women Madison Keys vs Danielle CollinsWagers: 6Staked: $1,760.00Event closes- Updated:
-
-
-
-