Gardiner vs Rielly: Who is Toronto’s Best Defenceman?

It’s well known that September is Jake Gardiner Appreciation Month.  But did you know that Toronto has other defencemen?  It’s true.  One of those defencemen is Morgan Rielly, the former 5th overall pick who recently played a major role on the exciting Young Guns team at the World Cup of Hockey.  In fact, some people even think Morgan Rielly is better than Jake Gardiner.  What a world!

I decided it was time to answer the question once and for all.  This is the post that will end all arguments on Twitter, forever, I think.  So I gathered a bunch of stats and put them in some charts and now you can see them.

OFFENCE

I think Jake and Morgan are both generally thought of as “puck moving” defencemen, which is a phrase that people usually use to mean “defencemen who are not bad”, but also to mean “defencemen who create a lot of offence”.  If you were to ask people who generates more offence, though, I think almost everyone would say Morgan Rielly.

To see if that’s true, I gathered some statistics from Corsica.  In particular, the stats are goals, primary assists, and individual Corsi For (ie. shot attempts taken by the player) at both 5v5 and 5v4.  I’m using statistics from the past two years for this comparison.  Generally, I like to use a longer window, like 3-4 years, but because Morgan Rielly is so young I thought it would be better to use more recent data so the results weren’t as likely to be biased against him simply for being inexperienced.


Gardiner Rielly
G/60 0.20 0.22
A1/60 0.25 0.29
iCF/60 9.90 12.43
5v4 G/60 0.00 0.51
5v4 A1/60 0.78 0.68
5v4 iCF/60 15.82 14.61

A quick glance at this chart suggests that the two players are quite similar.  Rielly does come out slightly ahead in all the even strength categories, but their primary points are only slightly different.  Rielly does have a clear advantage in terms of individual shot attempts at 5v5, but it doesn’t seem to be helping him generate points at a higher rate.

On the powerplay, Gardiner is ahead in assists and shot attempts but behind in goals.  He’s had some incredibly bad shooting luck on the powerplay over the past couple of seasons, and clearly, no player will score zero goals forever, so the gap there isn’t really as big as it appears to be.

If we take the numbers here in total, I think it’s fair to say that Morgan Rielly has the edge in terms of offensive ability, but it’s a very small edge – certainly much smaller than most people would likely guess.  Jake Gardiner’s ability to generate offence is close enough that the gap is not very noteworthy.  Nevertheless, if we have to give the edge to someone, it’s Rielly.

POSSESSION

Scoring points is great, but one of the most important ways that a defenceman can influence the game is by driving possession.  We use shot attempts as a proxy for that, both because evidence has shown that Corsi matches actual puck possession very closely and because Corsi does a pretty good job of predicting future goals, which means it’s an important, repeatable talent.  This chart once again looks at the last two years, and I’ve split it into Corsi For and Corsi Against so that offensive and defensive impacts can be seen more clearly.

Gardiner Rielly
CF/60 58.3 59.3
CA/60 54.7 61.4
CF% 51.6 49.1
5v4 CF/60 86.9 84.1
4v5 CA/60 55.7 78.4

The two players have pretty similar results in terms of their ability to drive possession for the Leafs, but there’s an enormous gap in their ability to prevent shots against the Leafs.  Jake Gardiner’s defensive ability, at least as measured by a propensity to reduce the number of shot attempts that are taken against the Leafs net, is vastly superior to Morgan Rielly’s.  I don’t think this should be too surprising, as I’ve already written about how Gardiner is very good, and very under-rated, defensively, but this helps demonstrate how big the gap is between the Leafs’ two most exciting defenders.

In addition to the even strength numbers, I’ve also included Corsi For on the powerplay, which shows that the players are pretty even in terms of shot attempts taken by the Leafs when each player is on with the man advantage.  However, Gardiner has considerably better results in terms of limiting shots on the penalty kill.

There is one caveat here, and that’s that Gardiner has not played a large number of minutes on the penalty kill over the past couple of seasons: about 71 to Rielly’s 276.  It is possible that Gardiner’s results are just a fluke of the small sample size and aren’t really indicative of a real ability to limit shots on the penalty kill.  It does line up with his ability at 5v5, but he just hasn’t played enough minutes to say with any certainty.

LINE COMBOS

One thing that might affect the results seen here is linemates.  For defencemen in particular, their primary defensive partner can have a big impact on what kind of results the pairing is able to get.  To take a look at what kind of effect that might be having on these players, I grabbed a list of all of the defencemen who both Gardiner and Rielly have played at least 40 minutes with over the past two seasons.  The results here are not weighted in any way, so there’s nothing compensating for the fact that, for example, Morgan Rielly has played way more minutes with Matt Hunwick than Gardiner has.  But this at least gives us a snapshot.

Gardiner Rielly
Carrick CF/60 62.0 48.7
CA/60 51.1 56.9
Corrado CF/60 58.0 60.2
CA/60 56.5 57.9
Franson CF/60 64.1 53.9
CA/60 48.7 86.6
Hunwick CF/60 68.1 56.1
CA/60 58.4 61.2
Marincin CF/60 57.8 56.8
CA/60 57.8 56.7
Phaneuf CF/60 60.9 67.2
CA/60 55.7 71.8
Polak CF/60 49.1 56.2
CA/60 56.6 60.1
Robidas CF/60 43.5 59.0
CA/60 46.1 64.7

The results in terms of offence are split evenly; Gardiner generates more Corsi For with four of these players, while Rielly generates more with the other four.  On defence, however, the results are much more stark; among all of these players, the only one who has a lower Corsi Against with Rielly than with Gardiner is Martin Marincin, and even then he’s only just barely better with Rielly.  All the others are better with Jake, and some of them are much better.

This backs up what we saw above, where both players are similar in terms of their ability to generate offence, but Jake is superior defensively.  This all seems to suggest that Gardiner is the better player.  But there’s one criticism I know many readers will have with what I’ve said so far: Rielly played top pair minutes against difficult competition last season, while Gardiner has never taken on that kind of duty.  Don’t we have to take that into account when evaluating these kinds of results?  That’s exactly what I’m going to take a look at in part two of this series next week.

  • Don Seitz

    Great article.

    I’ve always felt that Gardiner is very underrated except by informed observers. I’m wondering how much experience factors into his CA numbers. Has he always trended this way? Is there any likelihood for Rielly to improve his numbers as well?

    Maybe Babcock and his systems can continue improve upon their defensive results.

  • Glen Kappel

    Great article but pretty sure the names and stats as are reversed here. (Gardener should have the higher Corsi against numbers; check the point about Marancin being the only defender who is marginally worse when with Gardiner.) In the interest of being definitive and settling the matter you might wanna fix that.

    But overall, this is why we probably shouldn’t swap Gardiner for Trouba. The only part that makes me hesitate is the fact that our best defense prospects (Neilson, Dermott) are both lefties and (hopefully) potential top-four. The better way would be to trade one of N/D++ (a Leivo, a Leipsic, a Kapenen, a Brown; pick 2 or 3).

  • Dave C

    Too many qualifiers here that aren’t backed up by anything other than your opinion on perception (weird for an analytics-focused piece, isn’t it?) – the fact that you’ve written about Gardiner being so underrated might explain the bias a bit. You keep defending everything that looks negatively at Gardiner and qualifying all of his flaws with reasons for optimism that for some reason Rielly doesn’t get to enjoy, despite being younger and the chances of his numbers improving much higher than Gardiner’s for the remainder of their careers.

    “I think it’s fair to say that Morgan Rielly has the edge in terms of offensive ability, but it’s a very small edge – certainly much smaller than most people would likely guess.”

    How can you possibly evaluate what “most people would likely guess” about an offensive edge? Time to revisit the wheel and decide whether you’re going to be objective or fill this post with your own opinions….

    You also point to Rielly’s point production being low despite a high shot rate, but then in the next paragraph explain how Corsi is a good predictor of future goals. And somehow Jake Gardiner scoring 0 goals on the PP gets him a mulligan.

    • Tigon

      Jake will most likely be playing with Carrick or Zeitsev. Maybe Polak if he is used as a top 4. Hunwick is a LHD. It actually might be more beneficial to have Carrick on the third pair with Hunwick instead of a Polak/Hunwick/Corrado third.

      This is assuming they go 8 D because I don’t think Hunwick should, should being a keyword, be in top 7. I’d rather MM be paired with Polak/Carrick/Corrado.

  • LukeDaDrifter

    Let’s not forget we are comparing defencemen who played on the 30th overall team last year.

    Both players are good at moving the puck out of their own end if given a little space. They do have different styles of doing it. I like Rielly’s ability to see and make a good pass when crossing his own blueline. Gardiner is more inclined to carry it right in and take a shot on goal. Last year they were close to even. Rielly is still developing and I expect him to surpass Gardiner at some point this year.

  • Harte of a Lion

    As a Leaf fan it’s nice that we have 2 defencemen who are both good to possibly great talents and we can have this argument. Rielly seemed to turn a corner at the World Cup and hopefully he carries that over into the season. I look forward to seeing the team they assemble this year.

  • Rick

    Whoever wrote this needs to have his keyboard privileges taken away. Rielly shots against are higher because he played the tougher minutes, against the world’s best players while Gardiner was sheltered playing against 2nd/3rd line players. If you don’t even take that into consideration, then all your “facts” mean nothing. Yes Jake is good at recovering loose pucks and getting them out, but there is a LOT more to being a good defender. Corsi only tells part of the story.
    Then you add PK stats. HA! Gardiner averaged 21 seconds a game. Too small of a sample size to even consider. Then when his stats come up stale on the PP, you make excuses (bad luck). Jeezus. Maybe he just wasn’t that good
    Rielly is miles ahead of Gardiner as a d-man, and he is still just a kid, while Gardiner is a veteran. Stupid story. Don’t ever write again.