Jordan Eberle Comparables, Part II

Yesterday, the Oilers re-signed Jordan Eberle to a six-year, $36 million contract extension. If you were wondering what that shrieking noise around mid-afternoon yesterday was, now you know (and you also know that its epicenter was Wanye Manor).

Earlier this month, I looked at some comparables for Jordan Eberle. It’s a pretty good looking list. Today, I’m going to try something similar but with some modified criteria.

The problem with my first list – a list that included Jason Allison, Patrice Bergeron, Martin Havlat, Ales Hemsky, Marian Hossa, Vincent Lecavalier, Patrick Kane, Anze Kopitar, Rick Nash, Brad Richards, Mike Richards, Sergei Samsonov, Alex Semin, Paul Stastny,Petr Sykora, Alex Tanguay, John Tavares, Joe Thornton and Jonathan Toews – was that it was based solely on points. The criteria on that list was to look at all forwards from 1997-98 on to have a season with a points/game total between 0.90 and 1.05 at the age of between 20 and 22 (Eberle, with a 0.97 points/game season at age 21, fell midway between those points).

Doing it that way doesn’t take into account Eberle’s unreal shooting percentage. At 19.8 percent, Eberle had a higher shot percentage than can reasonably be expected to continue (the league leader since the lockout, the phenomenally selective shooter Alex Tanguay, has an 18.0 percent conversion rate).

So, while the list of names above is nice and rosy, it’s not a list I particularly trust.

The Pessimist’s Method

Since shooting percentage is fickle from year to year – and history shows, barely short of certainty, that Eberle’s true accuracy level is lower than he managed this season – I decided to ignore goal totals and instead focus on two other things: assists/game and shots/game. Then I sorted every player since 1997-98 who came close to Eberle’s assist totals (within 10 over an 82-game season, basically) and shot totals (plus or minus 20% of Eberle’s shots per game this season) and came up with a list of 43 seasons between the age of 20 and 22. So, I narrowed it down some more. I removed all of the centers from the list, leaving only wingers, created an adjusted point column (basically assists plus shots multiplied by the group average shooting percentage) to negate shooting percentage differences and removed everybody who was no longer within 10 points of Eberle.

I haven’t compensated for individual points percentage (Eberle’s has been on the high side over his first two NHL seasons) and I also haven’t really allowed for Eberle’s high on-ice shooting percentage – at least not directly. Eberle’s 5-on-5 on-ice shooting percentage was driven in large part by his personal shooting percentage; 36.4% of 5-on-5 goals with him on the ice came off his stick (as per behindthenet.ca). By negating Eberle’s personal shooting percentage, we negate a good portion of his insanely high on-ice shooting percentage as well.

That’s a lot of explaining, but basically what we’re looking at here is a cynic’s list of comparable players. It’s the kind of list designed to handle the objections of a guy like me. Here’s the list:

It’s a pretty strong list, overall. Aside from Heatley those names lack star power to some degree, but when we look at their average 82 game performance over the next seven seasons (as long as Eberle’s under contract in other words) we get this:

The “S-1” column asks whether the player had a comparable season to Eberle the year before the big breakthrough; in a surprising number of cases the answer was yes. The players that didn’t – Horton and Friesen – are both below average on this list. The “Data” column indicates whether there were seven full seasons to grab information from after. Meanwhile, the “Drafted” column shows overall draft position in an attempt to add the context of these players’ pre-NHL career.

Looking at the data, I don’t really think Eberle’s going to crash and burn the way Friesen did, and I don’t really believe he’s going to set the NHL a-flame the way Heatley did, either. I think we’re probably looking at a pretty good player – at worst, a Milan Michalek fringe-first line type, and with the potential to be a Patrik Elias-style winger. Elias might sound like a disappointing comparison, but he is not: he had a 96-point season and was a top line power-vs.-power winger on a very good Devils team for years.

The group average is in Petr Sykora/Bobby Ryan country: a 30-35 guy during a healthy season. For those who believe Eberle’s a significantly above average NHL shooter (the group average here is 12.8 percent – an excellent number and not far from where I’d put Eberle), feel free to bump his goal totals a little higher.

I tend to be more cynical than the norm on Eberle. This list reflects my thinking on what constitutes a reasonable baseline – not necessarily the best-case scenario, but a reasonable baseline. Looking at it now, I’m not wild about the Oilers’ long-term offer to Eberle – I’d like a shorter term, or some patience getting him under contract – but I’m also not terribly worried it’s going to be an ugly overpay either.

Bottom line: if you’re a true believer that Eberle’s a franchise player and that his shooting percentage was for real, stick to that first list, the one that looked at him strictly by points. If, on the other hand, you’re more pessimistic on him (as I am), I think these comparables send the message that at worst – even with shooting percentage regression – he’s going to be a pretty good player for the next seven years.

  • I understand why advanced stats are important, but I think there are cases where the stats just don’t so a good job at projecting the amount of points a player is capable of getting. I think instead of saying Eberle lucked out with a high shooting percentage that he’s likely not able to keep up, why aren’t we saying Eberle’s shooting percentage this year is a good indicator of how Eberle takes smart shots that have a high percentage of finding twine?

    I have read article after article breaking down a ‘reasonable’ expectation for Eberle, using all manner of stats; however not one single article has been able to explain why his percentage was so high this year. The obvious answer would be Nuge and the passes he was supplying, but Eberle didn’t always play with the center.

    So maybe it was all powerplay. But of his 76 points, only 20 came on the pp 10 and 10 respectively. So maybe it was the soft competition, but I do remember reading somewhere that Ebs and Nuge were often up against top pairing defensemen this year.

    I guess my point is that stats used for expectations and point projections from year to year are pretty boring and unproductive. Every player is capable of having an off year, and likely it has more to do with the playstyle of the coach or other factors like injury and line mates as oppose to what can be extrapolated from stats. Is there any particular stat explaining why Ovechkin doesn’t score 60 goals every year? How Malkin has been able to play at such a consistently high level? Etc.

    I’m just happy we have what has appeared to be over the last two seasons an incredibly good player, in addition to all the other incredibly good players we’ve acquired recently.

  • Is that with Eberle on the ice?

    Yes.

    If hi IPP was around 85 then out are talking about a 2.8% bump on only 15% of points scored while he was on the ice. How many points is that?

    You can dock him big for IPP or you can dock big for ON% but I don’t see how you can dock him so heavily for both because they are related.

    No they’re aren’t related. Every shot your team takes could go in (100% shooting percentage) and you could get zero points.

    • Yeah my mistake – I misunderstood something in my math. I shouldn’t comment between clients when I have no time to think about what I’m saying.

      Interestingly though if that 10.8 is corrected to 8 and his rate of assisting stays the same his IPP goes up.

    • Yes and no.

      Say Player X has an on-ice shooting percentage 20% above his line’s true talent level and an IPP 10% above his true talent level.

      Without that high on-ice shooting percentage, he’d be scoring 110% of what we’d expect in a normal year. With that high on-ice shooting percentage, he’s now scoring 132%.

      The one inflates the other, no?

  • By negating Eberle’s personal shooting percentage, we negate a good portion of his insanely high on-ice shooting percentage as well.

    Not really. The Oilers shot 10.8% at 5v5 on shots Eberle didn’t take last year, which is nuts.

    • Is that with Eberle on the ice?

      If hi IPP was around 85 then out are talking about a 2.8% bump on only 15% of points scored while he was on the ice. How many points is that?

      You can dock him big for IPP or you can dock big for ON% but I don’t see how you can dock him so heavily for both because they are related.

      I’m on board with the idea that he is going to regress, but you’re double dipping a bit when you come up with your numbers.