photocred Wikimedia Commons

Joe Thornton has been in the news a lot recently, with talk that the Sharks are looking to move him as part of what they are labeling a rebuild.

Thornton is elite. In the past five seasons, he’s tied with Corey Perry as the sixth best point accumulator in the league. The Sharks signed Thornton, who turns 35 within two weeks, to a three year contract with a cap-hit of $6.75 million back in January.

I’m sure any potential acquiring team is intensely interested in how Thornton may depreciate over the life of his contract.

Career-progression charts show that it’s rare for players to produce at elite level in their mid-to-late thirties, but what about among players who are elite in their early thirties? Thornton is one of 78 forwards who averaged more than 0.8 Pts/GP in 150 games or more in their age 32 to age 34 seasons (excluding active players yet to player their age 37 season). I took a look at how that group of 38 players depreciated from age 35 to 37 — the span that Thornton’s contract covers. I compared the total games played by the group in each season compared to their age 34 season, and the weighted averaged points-per-game, again relative to their age 34 season. (Click on any image to enlarge)

The points-per-game rate didn’t budge too much but their games played total dropped more than 31% in the three seasons.

I think it can be useful to separate the population in half based on their characteristics. You can see that players that played their age 34 season in or after 1994-95 saw a very different distribution than players from earlier generations.

In earlier days, it was much less common for players to play at advanced ages, so you see that the pre-94/95 group was dragging down the games played mark in the total population. They did maintain their points per game exceptionally, likely due in part to the less successful among them retiring earlier.

The sample divided on PIM/GP from ages 32-34 also shows a divergence.

The games played trendlines for both samples are similar, but the low PIM group, of which Thornton is a part, held their points-per-game significantly better.  To the extent that one accepts penalty minutes as a proxy for physical play, that would be expected.

Now it wouldn’t make much sense to separate the sample in two based on points-per-game, as I purposefully picked a sample where Thornton is around the median in points per game. Instead, I split the population into three samples, with Thornton in the middle group.

As one would expect, the highest scoring group, which features Gretzky, Dionne, Richard etc., succumbed to attrition much slower than the rest. The other two groups were fairly similar, but it’s clear the that the elite’s elite is buoying the entire population some.

I think the data presented would lead one to feel good about Joe Thornton’s future play throughout his contract. Any acquiring team should feel good about Jumbo Joe. So should the Sharks — why do they want to trade him again?