With Or Without You Analysis: Travis Zajac's Corsi Impact on the 2009-10 New Jersey Devils

Today, I've done a With Or Without You (WOWY) analysis for Travis Zajac.  Over at the Copper & Blue, Derek Zona revealed a few days ago that Zajac was one of the best forwards of the last two years based on his production while facing stiff competition.  Derek focused solely on even strength scoring, which makes his findings even more incredible.  In fact, he even used the word "Datsyukian" (as in Pavel Datsyuk, who has been the best forward against top competition three seasons running) to refer to Zajac's outscoring rate.   Therefore, I figured it would be appropriate to see how he has performed with and without his teammates on the New Jersey Devils.

This will be the third one done for a New Jersey Devil.  Thanks to Tyler Dellow's help and the new scripts at Vic Ferrari's Time on Ice, I was able to create a WOWY analysis for Zach Parise and then for Ilya Kovalchuk in the past week.  I went into more depth into what the analysis in the Parise's WOWY post, but here's a brief summary of what it is. A WOWY analysis breaks down a player's Corsi events at even strength, 5-on-5, no-empty net situations across a whole season, and determines the impact of each player's Corsi percentage with and without a particular player.   A higher Corsi percentage is definitely desirable, especially for top forwards, since it would suggest more puck possession and territorial dominance for New Jersey.  The WOWY analysis can be used to determine who has played the best with each other on the ice over a season, among other findings, based on Corsi.

The WOWY analysis on Zach Parise showed that he provides a positive impact on most his teammate's Corsi percentage.  It's additional evidence as to how effective he is from an offensive standpoint - more than just racking up the goals and assists.   In the WOWY analysis for Ilya Kovalchuk, despite the short time frame, I found that Kovalchuk worked best with 5 Devils and in conjunction with the Parise WOWY analysis, saw the reasoning for the makeup of Jacques Lemaire's top two lines going into the playoffs.

What follows after the jump are the findings from the WOWY analysis I did on Zajac.  From the Parise WOWY, I found that Zajac and Parise both contribute more from a Corsi standpoint than apart.  It may seem obvious to some fans, but it's good to have some hard numbers back up what we've seen and what the coaches have done in terms of lines.  Let's see how other Devils are impacted when playing with and without Zajac and determine who should and shouldn't be with Zajac on a regular basis.

Before continuing, please view this post in Wide before you continue after the jump. Please select it in the gray box next to the headline of this post to make it wider and be able to see the entire chart.

Travis Zajac's With Or Without You with the 2009-10 New Jersey Devils

Now, for even strength, 5-on-5, both goaltenders in net situations in the regular season, Zajac has had Corsi events with at least 25 skaters in 2009-10.  I will offer a criticism of my own analysis now. The true number is likely I didn't not split up every jersey number for every player called up, so the true number is closer to 35- however they have had such little ice time with Travis Zajac on the ice that it's not significant data.  They weren't going to reach 100 Corsi events so they would only be included for completion's sake.

The only player who may be undercounted is Vladimir Zharkov.  The base numbers available to determine Corsi% and such are all defined by jersey number. Well, with all of the injuries and called-up Devils in this past season, jersey numbers were used by multiple players.  For most of them, it's easy to figure out when they had a certain number since they didn't play so much.  Yet, I can't totally parse out when Zharkov was #28 (figuring out Martin Skoula was #28 was easy, though), when he was #18, and what other numbers he had.  Is this information out there? Something I can check out just to ensure what games Zharkov was what number?  I'd appreciate it and would use the information to correct the Parise, Kovalchuk, and this WOWY.

In any case, with Zharkov, I did the best I could and it's possible he did have over 100 Corsi events with Zajac despite largely being on the bottom six forwards; he actually could be wrongly among the grayed out players.The other player's numbers all match up though, so it is just one player whose numbers in the following chart should be taken with a grain of salt.  I would say that the overall conclusion from the Corsi% may hold up with definitive numbers.

That said, here is the team WOWY chart for Travis Zajac. It's too huge to embed, so please check the link to view the whole chart.

Among this group, Zajac has had at least 100 Corsi events with 15 players: 7 forwards and 8 defensemen.   The forwards consist of who has made up the top six forwards at different points of the season, no real surprises there. Zajac was notably rotated between centering Zach Parise's line and Ilya Kovalchuk's like in the second half of the season. Again, defensemen rotate differently and often play more minutes, so they've been present when Zajac was on the ice more often - hence, all of the regulars throughout the season racked up Corsi events.

Forwards and Zajac - 100+ Corsi Events Together

Again, I'm most interested in those forwards who had at least 100 or more Corsi events with Travis Zajac; forwards who have had significant time playing with the pivot.  Here's the chart of all of the forwards.

Immediately, it should be of no surprise that Parise tops the list.  We've learned that Parise and Zajac do well together and their Corsi% suffers quite a bit when they are apart from the Parise WOWY.  This just confirms it. Incidentally, the percentages are a little bit off between both posts because of rounding error.  Not that a change in a few hundreths of a percent changes any conclusions.

Zajac doesn't have the same level of positive effect on most forwards as Parise did, but for the most part the Devils forwards benefit with Zajac as their center.   Jamie Langenbrunner's Corsi% dropped significantly when he wasn't  playing with Zajac.  Zajac also suffered as well away from Langenbrunner, but not to the degree Langenbrunner did. Initially, one may look at this and conclude that Langenbrunner and Zajac should be kept together if only for Langenbrunner's sake.   Well, a WOWY for Langenbrunner would confirm that, but there's some validity to that statement.

Only, Zajac provided a bump in the Corsi percentages for Nicklas Bergfors, Brian Rolston, and Dainius Zubrus.  In the case of Rolston, it seems that Rolston clearly benefited more playing with Zajac than Zajac did with Rolston. That Zajac's Corsi% improving away from Rolston in this past season shows that to be true.  It suggests that the two may not be such a great matchup since the Corsi benefit appears to be one-sided for Rolston. As for Bergfors and Zubrus, the benefit in Corsi was a little more mutual.  Zajac at least saw a drop in Corsi% away from both players; but as with Langenbrunner, both Zubrus and Bergfors took even larger hits to their Corsi percentage. 

Incidentally, Langenbrunner, Parise, Bergfors, and Zubrus all had a Corsi percentage just over 55% with Zajac. A quick look at the forward chart in the Parise WOWY shows a similar Corsi percentage of just over 55% when Parise played with Zajac, Langenbrunner, and Zubrus.  Between the two WOWY analyses from he 2009-10 season, it's clear that the first line should be Parise-Zajac-Langenbrunner/Zubrus.  Let's hope the new coach understands that.

Now, Zajac did not mesh so well with Ilya Kovalchuk and Patrik Elias.  The Corsi percentage didn't change much for Kovalchuk with or without Zajac; but Zajac's Corsi percentage saw a boost away from Kovalchuk.  You could argue that it's not necessarily fair for all of Zajac's season to be seen without Kovalchuk.  Fine, but the Kovalchuk WOWY isolates Zajac's numbers as only time played with Kovalchuk as a Devil; and Zajac was found to be even better off away from Kovalchuk from a Corsi standpoint. Again, this and that should be of no surprise, since if Zajac wasn't playing next to Kovalchuk, then he was next to Parise.

The Elias findings are a bit more straight forward.  Given that Elias was moved to center for much of the season and when he was at wing, he was behind Parise anyway; the two haven't had much interaction with each other in games. But they had just enough to show that they really shouldn't be together. Maybe more Corsi events together would yield more favorable results, but both players saw their Corsi percentage significantly increase apart.

Defensemen and Zajac - 100+ Corsi Events Together

As with the forwards, here's the chart of defensemen's Corsi performance with and without Zajac.

Just like Parise and Kovalchuk, Zajac has a very large effect on most defenseman's Corsi percentage.  Unlike them, it's positive across the board and nearly with a significant effect.  With him, everyone is at 50% - often much higher than that.  Without him, only Paul Martin remains above 50% in Corsi. Among them all, Colin White was the most affected away from Zajac.  The effect isn't mutual, though. Zajac's own Corsi actually improves away from most of these same defensemen, most of all away from Mark Fraser.  Zajac's Corsi percentage only went down away from Johnny Oduya, Paul Martin, and Martin Skoula.

I don't know why Oduya and Zajac were so great with each other in terms of Corsi, and they both suffered so much, more than any other defenseman on this chart.  Yet, it led to little in terms of production - only 13 goals happened on the ice when they were together.  Maybe if Oduya stuck around all season and/or didn't get hurt, more could have happened.  At least it can be said that the Devils enjoyed positive territorial control when those two were together. 

Two other defensemen showed a similar effect on their Corsi, Martin Skoula and Paul Martin.   Skoula had the fewest amount of Corsi events with Zajac so perhaps the numbers aren't where they would be had Skoula been in New Jersey for 82 games.  However, the Corsi percentages reacted with and without Zajac similarly to Oduya and Martin, only with a smaller reduction on Zajac's end (Zajac without Skoula).  Unlike with Oduya and Martin, Skoula has never been a offensive force from the blueline so it really does just tell us about possession more than anything else. Which is good, I just wouldn't immediately say from this that Skoula-Martin absolutely has to be on the ice with Zajac whenever possible due to the defensemen's low Corsi event counts.  Just that it is a very favorable pairing.

Paul Martin was a very interesting player with respect to these WOWY analyses.  In the Parise and Kovalchuk WOWYs, Martin was found to have a higher Corsi% away from those players; much more away from Parise than Kovalchuk, incidentally.  However, Martin's Corsi% with Zajac was excellent at 59.4%; and both players suffer in terms of Corsi apart from each other.  I can't help but feel that the lower sample size may show a larger effect than what it really is, but I will say that the Martin-Zajac combination is a very powerful one indeed.  (I'd apply the same logic to Skoula, parenthetically.) I don't have an explanation as to why Martin did so well with Zajac but not Parise, given how much Zajac and Parise are together.  Do you?

Main Conclusions

Overall, 13 of 15 teammates who have had 100 or more Corsi events with Travis Zajac  yielded a higher Corsi percentage with Zajac than without him.   This was a similar effect that Parise had on his teammates but not as much.  Still, I would say this suggests that Zajac is an effective forward in terms of Corsi even if he's with Parise more often than not.

The only teammates who saw increases to their Corsi percentage away from Zajac were Ilya Kovalchuk, which was only a slight effect and didn't play the entire season in New Jersey; and Patrik Elias, who was often on a different line from Zajac and so didn't have many interactions to generate Corsi events with each other.

Based on the WOWY analysis, it appears that the best wingers from this past season or Zajac was Parise on his left and either Jamie Langenbrunner or Dainius Zubrus on right.  Langenbrunner has had more experience with Zajac and his Corsi percentage suffered more away from Zajac than Zubrus' did in 2009-10, so I'd give Langenbrunner the edge if Corsi is going to be a deciding factor. Niclas Bergfors would factor here, but he was traded and so can't be considered for next season.  For even strength situations, the Corsi percentage impact should be high enough to justify a combination where Zajac can continue centering effective lines against tough competition.

In terms of defense to be paired with, take your pick - all the defensemen other than Mark Fraser (slight positive effect) showed big gains to their Corsi percentage when playing with Zajac.  Johnny Oduya, Paul Martin, and Martin Skoula showed the largest positive effects as well as Zajac's own Corsi percentage suffered.  Again, Oduya isn't in New Jersey anymore, and Skoula's and Martin's futures are up in the air as they are free agents.  Were they to stay, I'd suggest that they should paired with Zajac out there.  Among the signed defensemen, White would be a priority as his own Corsi percentage increases drastically with Zajac and Zajac gets the smallest gain away from White among other signed defensemen. Overall, I can't help but think that since so many defensemen saw an increase in their own Corsi percentage playing with Zajac, most pairings would benefit.  Matching up pairings should only come into play in strategic situations anyway, like needing to possess the puck or get a late goal.

When you consider this analysis along with Derek Zona's finding of how well he does compared to other forwards who face stiff competition, Zajac's defensive prowess in this past season (I explained why here and his DGVT was the highest among forwards in 2009-10 at 6.2), his play exceeding my expectations set of him back in July 2009, and his own production of 25 goals and 42 assists off of 219 shots all sum up to Zajac having an excellent 2009-10 season.   The only blot on his record was, well, the playoffs. 

Still, I think he's earned his contract from last summer and then some already; and he's just turning 25.  I don't know about you, but I'm real excited for what Zajac can do in coming seasons.  Just continuing to play like he did in 09-10 would be fantastic on it's own; let's hope he can keep up the good work.

Now it's your turn.  What do you think about this WOWY analysis for Zajac?  Did it change anything you thought about Travis Zajac? Did it confirm what you already thought? If you saw these numbers and thought something different other than what I found (maybe I missed something or came to an incorrect conclusion), what do you take from them instead? How excited are you for Zajac hitting the prime of his career? Incidentally, I plan on doing this for Jamie Langenbrunner and Paul Martin in the near future so let me know what you think of these posts: Am I explaining too much? do I need to explain more? Thanks for reading and please leave your thoughts and answers about Zajac and the WOWY analysis in the comments.