What Are They Worth - Ryan Callahan


This is the week of reckoning for the contract of Brandon Dubinsky, with his arbitration hearing scheduled for this Thursday, July 21st. There is an equally important arbitration hearing for the Rangers the following week, as Ryan Callahan's hearing comes the following Thursday. I already looked at the expected value of Brandon Dubinsky, and determined that a reasonable deal for him would be $4.28m for 4 years. Using similar methods, we'll determine what can be expected as Callahan moves through the prime of his career.

Just like last time, I ran a search on hockey-reference.com, searching for the combined seasons of all players who played at least 200 games, post lockout. This time the age range is from 23 to 25, representing Ryan's last three seasons. The list of 74 players is once again headed by Alex Ovechkin (304 points), down through Darrell Powe of Philadelphia. Then I narrowed the list to all those within 0.1 points of Callahan's point per game mark over that stretch (0.57).

The list of those players is after the jump.

Player From To Tm GP G A PTS GC P/M PIM EV PP SH S S% TOI PPG
Jarret Stoll 2006 2008 EDM 214 49 94 143 51 -17 196 19 25 5 545 9.0 3888 0.67
Stephen Weiss 2007 2009 FLA 226 47 104 151 54 32 90 28 18 1 462 10.2 3957 0.67
Milan Michalek 2008 2010 TOT 222 69 77 146 57 18 117 47 19 3 575 12.0 4054 0.66
Joe Pavelski 2008 2010 SJS 229 69 81 150 58 7 100 45 19 5 701 9.8 3979 0.66
Tomas Plekanec 2006 2008 MTL 229 58 87 145 55 29 110 36 18 4 435 13.3 3647 0.63
Drew Stafford 2009 2011 BUF 212 65 66 131 52 20 98 40 25 0 543 12.0 3288 0.62
Andrew Ladd 2009 2011 TOT 245 61 85 146 56 18 134 50 9 2 538 11.3 3929 0.60
Scott Hartnell 2006 2008 TOT 225 71 59 130 54 29 356 38 30 3 537 13.2 3604 0.58
David Backes 2008 2010 STL 233 61 72 133 52 -18 370 45 14 2 500 12.2 3953 0.57
Ryan Callahan 2009 2011 NYR 218 64 61 125 51 -12 139 42 21 1 620 10.3 4070 0.57
Chris Higgins 2007 2009 MTL 200 61 52 113 47 -12 70 34 22 5 551 11.1 3533 0.57
Joffrey Lupul 2007 2009 TOT 216 61 63 124 51 -26 138 43 18 0 542 11.3 3522 0.57
Lee Stempniak 2007 2009 TOT 237 54 80 134 50 -14 106 40 14 0 499 10.8 3718 0.57
Andrei Kostitsyn 2009 2011 MTL 214 58 61 119 48 -3 118 41 17 0 501 11.6 3384 0.56
Clarke MacArthur 2009 2011 TOT 234 54 74 128 49 -23 142 38 15 1 391 13.8 3576 0.55
Tuomo Ruutu 2007 2009 TOT 227 53 71 124 48 8 265 38 15 0 405 13.1 3902 0.55
Tomas Fleischmann 2008 2010 WSH 217 52 66 118 46 -1 66 37 15 0 359 14.5 3154 0.54
Matt Stajan 2007 2009 TOR 240 41 86 127 45 -12 145 30 8 3 373 11.0 4161 0.53
Valtteri Filppula 2008 2010 DET 213 42 69 111 41 21 94 36 5 1 365 11.5 3614 0.52
Jiri Hudler 2007 2009 DET 239 51 73 124 47 34 78 39 12 0 393 13.0 2949 0.52
Jussi Jokinen 2007 2009 TOT 225 37 80 117 41 -16 64 23 14 0 353 10.5 3268 0.52
Alex Steen 2008 2010 TOT 225 47 70 117 44 -4 92 31 12 4 506 9.3 3805 0.52
Antoine Vermette 2006 2008 OTT 240 64 61 125 51 18 147 45 7 12 449 14.3 3665 0.52
Mason Raymond 2009 2011 VAN 224 51 64 115 44 10 82 36 14 1 559 9.1 3514 0.51
R.J. Umberger 2006 2008 PHI 228 49 67 116 45 -23 78 36 11 2 445 11.0 3465 0.51
Matthew Lombardi 2006 2008 CGY 218 40 68 108 40 3 163 24 8 8 429 9.3 3523 0.50
Petr Prucha 2006 2008 NYR 209 59 45 104 43 -1 84 33 26 0 355 16.6 2679 0.50

Since Dubinsky and Callahan are so often compared to each other, it makes sense that there's considerable overlap in comparable players, with names like Joe Pavelski, David Backes, and Drew Stafford appearing on both players' lists.  Where they don't overlap, there are a number of solid 2nd line scoring options, such as Tuomo Ruutu, Scott Hartnell, and Mason Raymond.  Plotting this group's Goals Versus Threshold yields the following graph (click to enlarge):

Rcgvt_medium

This graph looks a bit different than Dubinsky's, so let me explain why. In the Brad Richards and Brandon Dubinsky plots, I used the 82 game equivalents of their year to year GVT. That line is shown in purple here, and clearly would not work for the purposes of this study (well unless you think Callahan will retire the greatest player ever.). So to get a better projection, I instead used Callahan's actual GVT from season to season, shown in green. This gave a plot that more closely resembles his peer group (in red). The final line is an expansion of the peer group, including all players post 1999, which serves to give a glimpse of how they actually perform after age 30.

Going back to the Dubinsky chart, we saw a group that peaked at age 26 with about 12 GVT.  This group shows a peak at 28 of around 10 GVT, although the trendlines show the peak should be closer to the 26/27 age.  This past year, Callahan surpassed the play of this group, but injuries kept his numbers down to fit right alongside the group.

Assigning a dollar value to the data, the trend line says that Callahan will have a GVT of 34.39 over the next 4 years.  Using the same $367k per GVT rate we used for Dubinsky, that means that the 4 year contract for Callahan would be $12.62m, or a cap hit of $3.155m.  That's more than $1m less than Dubinsky, which seems out of line with public perception.  This is where the note below the graph comes in.  The comparison right now is not completely apples to apples.

Dubinsky's $4.28m was based on the assumption that he will play 82 games every year, while Callahan's is based on his current rate of games played, which has not been close to 82.  We can adjust for that two ways.  First, we could compare Dubinsky's actual GVT year to year, which gives an estimate of $15.03m, or a cap hit of $3.75m.  The alternative is to use Callahan's last season 82 game GVT (10.6), and repeat it.  That would also give a cap hit of $3.75m.

The end result in both cases says Dubinsky would be worth about $600k more than Callahan.  Given the current market, I'm more inclined to go with the higher values, where Dubinsky stays at $4.28m, and Callahan comes in at $3.75m, for a combined total of around $8.03m, perfect for the Rangers team.

Voice your disagreements with Callahan's expected value below.