Comparing Nuge’s Oranges and Draisaitl’s Oranges

Since Draisaitl’s contract signing, one thing I have heard a lot of people say is that this spells the end of Nuge [probable] in Edmonton, especially because he’s been something of a disappointment [FALSE! SLANDEROUS!]

It’s easy to see why this latter point is argued though – as far as genos and apples go, it hasn’t been a great year for Toddler Nuge (note: he’s no longer a baby). The evil snake/monkey that sat on Eberle’s back and kept biting 14’s hands all season spent plenty of time making Nugey look like a hunchback too.

snake monkey guariba-boia small

 

Now, most people recognize that Nuge is the coach’s “hard minutes” guy: via PuckIQ.com, Nuge plays 42% of his time vs Elite competition, easily the highest percentage for any of the regular players, and so we cut him some minor slack for that as a result.

Is “minor slack” fair, though?

The McDavid Effect

While acknowledging QoC is a start, it’s not enough.

See, there’s a big challenge when it comes to assessing any Oiler not named Connor McDavid. And that challenge’s name starts with C and ends with OnnorMcDavid.

When Connor McDavid is on the ice, he has an outlandishly positive effect. For the Oilers he does … for the other team not so much. Unsurprisingly then, McDavid also has that same outlandishly positive effect on his teammates.

Generally speaking, the more time you spend with Connor McDavid, the better you will do. More goals, more shots, more corsis, more benjamins. (I won’t bother to present the hard stats-based evidence for this … seriously, does anyone want to argue the point?)

So I’ll argue that the implication of this outlandish effect is that, if you want to get a sense of how good a player actually is, you need to look at his on-ice results when he’s without The Franchise.

Nuge < Fayne???

Here’s an interesting thing though – Nuge spends basically no time with McDavid. He spent less time with McDavid this season than Mark Fayne – who played all of half an hour the entire season – did!

This means that Nuge’s numbers get absolutely no McDavid rocket sauce. If you’re going to compare Nuge with someone (like Draisaitl) who gets a sh*t ton of time with McDavid, you have to account for this effect. Have to.

Otherwise you’re comparing the lap times of an NSX with a Civic, and blaming the slower Civic lap times on the driver.

The Russell Effect

But wait! That’s not all.

See, there’s a second effect that takes place on the Oilers, around what has to easily be the most polarizing third pair defenseman in the entire g d league. Y’all know who I’m talking about.

I’m NOT going to spend any time analyzing Russell, a topic that has been covered to Hades and back by many, but simply point out the actual effect he has.

One of the reasons many fans (and presumably his teammates and coach) love Russell is that he’s the hockey equivalent of a soldier who throws himself on a grenade to save his squad. Russell’s like that. He will sacrifice everything and anything to prevent a goal, and you have to admire him for that.

Unfortunately, one of the things that Russell also sacrifices while he’s doing this is his own team’s ability to score a goal. Pretty much every Oiler scores less with Russell on the ice than when he’s off the ice – that’s just plain ol’ facts.

This is true even for McDavid, whose lines score like they’ve been touched by the Hand of God without Russell, but merely at garden variety supernatural levels with him.

Ryis Nugent-Russell

One weird thing though – Russell’s anti-offense effect is particularly acute when he’s on the ice with Nuge. Like, so acute it makes Ariana Grande look frumpy.

Ariana-Grande-2-500x375c

How bad is it? Take a look:

  • RNH 5v5 CF% and GF% with Kris Russell: 44.8% and 27.8%.
  • RNH 5v5 CF% and GF% w/o Kris Russell: 49.4% and 58.3%.

Check that line again: <28% vs >58%. This change in goal scoring rates is so dramatic it doesn’t even seem real.

I’m always super-wary of using GF% as a statistic because of its low sample size and resultant noise/volatility, but even at these sample sizes (pulled from ~326 mins of “with” TOI and ~557 mins of “without” TOI), the Bayesian estimators I use to assess uncertainty indicate that those numbers represent a real difference. (see details at end if you’re a power stats geek)

With Russell, Nuge’s lines score under a goal per hour (0.92), which is about on par with any fourth liner you care to name.

Without Russell, RNH’s lines (still without McDavid!) pop above 3 goals per hour, which is getting into that garden variety supernatural level I mentioned earlier. It is an excellent 5v5 scoring rate, worthy of a first line. (Have I used enough ‘super’s yet?)

Monkey Poo

I noticed this rather extraordinary effect in mid-season and tweeted about it. Which of course brought out the usual array of knowledge-free hacks who assumed this was an attack on Russell and immediately insulted my game viewing habits, my lack of NHL playing experience, and of course, stats in general. (I guarantee someone is going to attack this article as if it’s somehow a criticism of Draisaitl, which it most assuredly is not)

monkey_poo_for_you_pt6_by_jays_doodles

Fortunately, amidst the monkey poo being flung my way, I also had a brief conversation about the topic with ultra-smart Tyler Dellow, who wondered if the effect was because both guys play such a conservative defense-oriented game. He noted that RNH is the Oilers C that comes back deeper and more often than any other (to the surprise of no one).

Add that to Russell’s conservative style and that would leave the team with a lot of guys deep any time they recovered the puck, which means an awful long way to go to get back on the attack.

Seems like a credible explanation. Could be that. Maybe something else. That’s something video study by someone with time and an NHL Centre Ice subscription might be able to suss out (hint hint).

Either way, I cannot speak to why this occurs. Only that it does occur.

And if we’re going to assess a player’s true impact on the game, I would argue that it needs to be accounted for.

So let’s do that.

Backing Out the McDavid and Russell Effect

With David Johson’s invaluable puckalytics.com site gone, doing this type of multi-player WOWY is difficult. Fortunately, I have the enormous database I’ve built that feeds the PuckIQ site at my fingertips … I knew that would come in handy one of these days! [this also means any data errors you find are mine and mine alone … unless they’re the NHL’s]

Here’s what Ryan Nugent-Hopkin’s and Draisaitl’s lines accomplished in 2016-2017 when on the ice without either Connor McDavid or Kris Russell (note the axis does not start at zero to allow for more detail to be seen):

nuge drai without mcdavid or russell.png

Pretty damn similar, right? If anything, Nuge has numbers a little better overall, though most are within error bars. This data is from ~554 mins for Nuge and ~240 mins for Draisaitl – enough for this data to be reasonably stable and representative.

Of course, a key point here is that Drai is younger, so he’s still got some room to improve. RNH on the other hand likely generated these numbers facing tougher competition. Things to keep in mind either way.

In case you’re wondering, and you’re pretty sharp so you probably are, they only spent ~53 mins together with neither Russell nor McDavid on the ice. So these are pretty fair representations of their ability to independently drive lines.

Bottom line: the actual separation between the two when without the rocket sauce of McDavid and without … whatever you want to call Russell’s sauce … is miniscule.

Ignuzzling the Russizzle

For completeness, I should also add that mixing Russell into the picture may strike you as biasing this analysis by selectively post facto taking away a huge negative impact on Nuge (and that’s a very fair comment. See, I told you you were sharp).

But the broad conclusion – that there isn’t much difference between the two players as far as driving their own line goes – still stands if you just look at time away from McDavid alone: RNH CF/GF% = 48.7%, 46.8%, Draisaitl = 47.5%, 46.7%.

I added Russell because I felt it made the comparison better rather than worse, both because of symmetry of treating the best/worst possession players on the Oilers both as outliers, and of course that massive effect on goal differential.

But you may (rightly) disagree on the topic of Russell, in which case you might want to fall back on the McDavid-only numbers.

Farewell, Sweet Nuge?

At the end of the day, I think TMac is aware that a lot of what I’ve shown above is happening, and it’s a big part of why he gives Nuge the tough minutes. For that reason, I’m guessing that when the time comes, he’ll probably fight to keep him.

But … the first dawn of Oiler cap hell starts next year with McDavid’s new contract kicking in and Russell and Lucic with NMCs. So it’s likely Nuge who will be the one waving goodbye.

crying 200_s.gif

I just don’t want anyone to not be aware of how good a player it is that will be walking out the door.

Ya got that? GIVE MY NUGEY HIS DUE!

 


POST SCRIPT: For Stats Power Geeks

Using Bayesian statistics to suss out whether a GF% differential is meaningful


I mentioned earlier that I am loath to use GF% as a stat because of its low sample size. It’s common also to run into the issue of small sample size even with higher sample statistics like Corsi.

Rather than ignore data (the paucity of GF% data is offset by its crucial importance to the game, no?), or just claim ‘small sample warning!’ and go ahead and run the analysis anyway, it would be good if there was some way to characterize the uncertainty of the data.

Broadly speaking, we are starting to recognize that traditional hypothesis testing and the way in which p values are used therein is flawed, or at least is a limited approach.

And I would argue that in a situation like hockey, it’s particularly inapplicable (hence rarely used) because we typically don’t want to reject or accept a hypothesis, we want to know to what extent a hypothesis / a number / a set of numbers is accurate or reliable or likely to be true.

Well, there is a way, and it’s called Bayesian statistics. (I am of the opinion that the future of hockey statistics is Bayesian. I’m far from an expert but trust me, working hard on it!)

In the case of the GF% situation between Nuge with/without Russell, I modeled it this way:

1. To model GF%, I used a beta distribution, which is the ‘right’ distribution for proportions but also has some really nice properties when calculating posterior distributions.

2. I started with what I call a “modestly informative” 50-goal prior centred at 50% (equivalent to perhaps 1/3rd to 1/2 of a season), which is strong enough to provide stability in the face of volatile small sample data, but weak enough to “let the data speak” as they say.

3. I then calculated posteriors for Nuge’s GF% with and without Russell.

4. So how then do we determine statistically if the two posterior distributions are meaningfully different? If we call the Nuge with distribution μ1 and Nuge without μ2, one way is to compute P(μ1 < μ2 | data). That is to say, what is the probability that the true value of the lower distribution μ1 is smaller than larger distribution μ2 given the data we observed?

5. If the number is ~50%, it’s a coin flip and we shouldn’t read anything into the difference. If it’s ~99%, confidence should be very high. Make sense? (this technique and description is from Bayesian Methods for Hackers by Cameron Davidson-Pilon).

6. We can implement this by drawing samples from the posteriors (sounds gross, eh) – the probability P(μ1 < μ2 | data) is then simply the fraction of the count of samples from μ1 that were less than samples from μ2.

7. To do this, I drew multiple 10,000 count random samples from each of the two posterior distributions. The proportion for which samples from μ1 were less than samples from μ2 was consistently around 89.7%.

8. So despite the small sample sizes, we can be pretty damn (that’s a formal statistics term) confident that the observed GF% difference of 28% vs 58% is meaningful.

9. A chart showing the prior, the two posteriors, the mean and 90% credible intervals, and a histogram of the samples from each posterior is shown below.

10. For clarity, the label ‘w Russell [0.344,0.441,0.541]’ for example shows that the mean of the ‘with Russell’ μ1 posterior distribution is 44.1%, and the 90% credible interval is (34.4%, 54.1%). Note the effect of the prior as well – even though the calculated GF% are 28% and 58%, this analysis pegs the expected value of each distribution at a much less wide-ranging 44.1% and 54.1% respectively. The ‘wo Russell’ μ2 distribution is closer to the calculated value and also narrower because its larger sample size means lower uncertainty.

11. One way you can be comfortable in asserting that the difference is meaningful is based on where the means and credible intervals sit.  While the distributions overlap (it would be surprising if they didn’t), nevertheless the mean of distribution μ1 falls below μ2‘s 90% credible interval, and (interestingly) the mean of μ2 sits right at the upper value of μ1‘s 90% credible interval.

12. Feel free to tweet at me (@OilersNerdAlert) if you have any questions or counter-arguments.

nuge russell wowy posteriors 2.png

 

 

Advertisements

18 thoughts on “Comparing Nuge’s Oranges and Draisaitl’s Oranges

  1. I have said before, if the team needs to move someone who will get them a top 6 prospect, then trade Nurse. Nurse has high value and will never be a top 3 d man with the Oilers.
    The Oilers have depth at the d position, they also have spent their last two first round picks on a right winger. And with Bear and Jones just a year away, I think they can afford to trade Nurse. I’m not a nurse hater, I like the player, it’s just that the return would probably be equal to what you would get for RNH, RNH plays tough minutes including the PK, he is needed, also Nurse will need a new contract next season , if he has a good year , he will want a minimum four million, perhaps more. This will force the team to trade him or some other high paid player. When this happens the Oilers will lose the trade, so why wait until this happens. Part of cap management is being pro active, teams like Colorado and New Jersey would love to get a d man like Nurse, I’m guessing so would plenty of other teams.

    Like

    1. Your comment brings up a number of valid points and considerations. Not saying it’s right or wrong, but for sure the cap is going to force some tough decisions on the Oilers, and likely some fan favourite(s) will be heading out the door.

      Like

  2. G, do you think that there is a way to quantify the McDavid rocket sauce? By that I mean is there a way to mathematically show the average to which McDavid outlandishly boosts his teammates results?

    I think the effect would vary depending on teammate, but it would be useful to know that on average McDavid does “x” to his linemates Corsi%, DFF% GF% so that you can independently evaluate McDavid linemates.

    One thing that Pittsburgh now seems to be better at is not giving too much credit to Crosby’s linemates. They failed early on (see Kunitz contract), but seem to now have a better grip on what Crosby does for his teammates. This would seem to be incredibly important for the Oilers as they move along for salary cap purposes. Not overpaying McDavid’s wingers “because McDavid” will be a real thing (may already be with Draisaitl).

    Anyway just some thoughts for a future area to explore, loved the article.

    J

    Like

    1. Good questions, J!

      So a couple of ways to look at the rocket sauce.

      There’s a stat (hard to find now with all the sites that are down but I’m sure it’s out there) called CorsiRelTM. What this stat does is it basically adds up the total impact a player has on his teammates with/without him. So imagine McDavid plays with two players and those two players do +10% and +8% with him compared to without him. Then McD’s CorsiRelTm (assuming equal ice time) would be 9%. So that’s a very handy number to check out.

      The other thing that really shows the rocket sauce effect is Micah Blake McCurdy’s WOWY visualization over at hockeyviz.com. By drawing the with/without locations on the chart, you can dramatically see how elite players have this outlandish effect on their teammates really clearly. It’s probably my favourite analytical chart (which is why I stole it’s format for WoodWOWY and hopefully we’ll have those on the PuckIQ site at some point soon).

      Hope that helps, and thanks for the feedback!

      Like

      1. G, I’m wondering what your thoughts are on Maroon in the context of McDavid Rocket Sauce. He’s one of the few that doesn’t seem as affected by it.

        Maroon w/o McDavid: 53.8CF%, 50.3DFF%
        McDavid w/o Maroon: 52.3CF%, 52.8DFF%

        Pretty sizable CF% advantage, although DFF% looks more like what we’d expect.

        Maroon vs elites: 53.3CF%, 55.8DFF%
        McDavid vs elites: 53.0CF%, 56.9DFF%

        Maroon’s overall CF% vs elites is higher, so his CF% w/o McDavid must be higher than his percentage with him. Same thing applies to McDavid and the DFF%.

        It makes sense that McDavid makes Corsi events more dangerous, but it’s unusual for someone’s share of shot attempts to *decrease* when playing with 97. Maybe we’re dealing with a more substantial player than we thought?

        Like

      2. Yeah, Maroon is a super interesting study because he is pretty much the only player on the Oilers who does as well or better away from McDavid!

        So when I got “WoodWOWY” working (it will be on the PuckIQ site soon, but there are challenges with the data volume, which runs to ~15M records per season), he was the first player I looked at.

        Take a look at this WoodWOWY visualization (riffed directly off Micah Blake’s WOWY viz) and note where 19 sits when he’s without McDavid (the red box) vs Gritensity: http://i.imgur.com/dioR0Xk.png … you’ll see what’s interesting is that 19 makes a ton of hay vs Gritensity when he’s not with McD.

        That’s not really a “negative” aspect of Maroon in that I believe what we might have here is a *true* (i.e. rare) “player who can play up and down the lineup” – he can rip apart lower tier competition, yet still contribute at a top line level when he plays with gifted players.

        Now that separation might disappear in the future because when we start chopping the data down like this, the smaller samples make for volatility and you have to be cautious about the conclusions drawn.

        But if the effect holds next year too, I will be more confident that’s what we’re seeing.

        Does that make sense?

        Like

  3. I want to say how much I appreciate the post script. I loved the visualization of the distributions. And the explanation of how you determined significance.

    Like

  4. This is amazing analysis. Question – could there be a contextual element to this analysis? For example, you mention RNH and Russell face the toughest minutes. Perhaps the Russell factor drags everyone else down not because he is Russell but because he his assigned the most probable situations were shot blocking is required?

    Like

    1. Thanks Ryan!

      Yeah, that certainly could be the case. I’m definitely focused more on outcomes than on trying to explain why those outcomes happened (as I noted in the article, there are several things that only video analysis could really explain here).

      Now I’ll admit, the idea that the blocked shots are contextual is not what my eyes see – WARNING! STATS GUY WATCHES THE GAME! :-D!

      My visual impression of Russell is that there’s this kind of cycle that goes: opposition enters zone -> Russell retrieves -> Russell dumps out -> opposition retrieves -> opposition enters zone and as that repeated cycle progresses, Russell gets into more and more of a chase position until the enemy’s shooting position is dangerous and he’s forced to go down and block.

      In other words, it’s the nature of his conservative defense-first game and difficulty with exiting the zone under control combined with his willingness to sacrifice life and limb that leads to all those blocks.

      So (both by eye and by stat) I admire the never-say-die attitude and willingness to sacrifice that results in large part to Russell’s shot blocking, but have frustration with the process that gets him there.

      Note also that a lot of Russell defenders are absolutely correct in saying he’s actually decent at defending, because results wise he’s not bad as far as suppressing shots/danger/chances against.

      The real problem (and it’s really difficult to see this with the eye as we don’t really pick up rates of offense!) is that the ‘sacrifice everything’ process that lets him do a decent job of boxing out the middle prevents his own team from getting on the attack very much.

      So his poor shot metrics have less to do with him giving up a ton of shots (he doesn’t actually) but rather that his own team doesn’t get nearly enough when he’s on the ice.

      And at least part of why I wrote this article is that this effect, while it impacts pretty much all players including McDavid, seems to impact Nuge the most by far, and I felt that was noteworthy.

      Like

  5. RE: G

    Replying here because I can’t reply to your tree for some reason. Nice plots! Really looking forward to playing with that tool when it’s ready. Thank you for all your work on it.

    If I’m reading things correctly, Maroon does better vs elites w/o McDavid too, no? He’s right on the “good” line, whereas McDavid is below it. McDavid is better w/o Maroon vs muddle, and Maroon is better w/o McDavid vs gritensity. Really unusual player card. Quite intriguing. Small sample size seems the most likely explanation, because who the fuck gets worse with McDavid when facing the best?

    I’m not sure where we could fit Maroon’s ~$4.5M in our cap structure, but I’d really like to see it happen. By eye and number, he’s an intriguing player, and it would be a shame to lose him.

    Like

    1. Addendum: Could also be that when McDavid is on the ice, Maroon faces more elite players, e.g. w/o McDavid he faces one elite at a time, and with McDavid, he faces two. Prima facia, that should change the results.

      Like

      1. Tough to say. McDavid and Maroon face very similar levels of comp, and of course their comp level is identical when on the ice together, so it’s hard for me to think their comp levels are very different when apart.

        The thing to be most wary of is sample sizes – you kind of want to draw wide “error circles” around the data, so rather than look at nuances of where the boxes sit when close together, I tend to just look at where the mass of each of the box types (with, without, teammate without) sits as a way of assessing the overall impact of the player.

        Where I draw attention to differences in the individual boxes is with the outliers that sit noticeably far out from the mass, since that might convey some interesting information (though it could still also be noise).

        So when I look at the charts overall, the players worth paying interest to are 27 vs Elite, 62/25 vs Middle, 19 vs Grit, and most of all, 83 every damn where. Matt Freakin’ Benning everyone!

        Like

  6. I agree that Nuge has value, but he is nowhere near the player that Draisatl is currently or will be. It was evident in the playoffs when the Oilers were playing Nuge against top centres on the Sharks and Ducks (especially) with more size that he was not able to handle them. Getzlaf abused him all series when they went head to head, as well as the spurts they lined up against each other in the regular season. They had no answer in the playoffs until McClellan finally clued in that they need to put Draisatl at center going head to head with Getzlaf, but at a point where it was already too late (I called the need for change after game 1 – Getzlaf did not get the win but he dominated Nuge badly).

    Nuge’s problem is similar to what they said about Ryan Murray in his draft year – good to very good at most aspects, excellent in none. The one worrying fact at the draft is that Nuge had a huge reliance on PP points to boost his totals rather than EV. This has played out in the NHL. When you look at Nuge, he has good speed, good to very good hockey sense, good skill, and good to very good compete. However, he lacks elite hockey sense and skill, the size (height and weight) and man strength to compete with top centremen in today’s game (and I would argue the speed as well), and is still very weak on the dot (which is surprising this many years into his career). Nuge is a good complementary player in today’s NHL, but not a driver. His best years were with a winger (Hall) that could drive the play and create openings for him. Unfortunately, neither Lucic or Eberle fit the bill as drivers and it was reflected in his stats. I would like to see him as the 3C to start the year on a line with Jokinen (saavy vet) and Puljujarvi (potential driver of play) – they should do some damage against their opponents’ third lines.

    One thing I am curious to see this season is how much McDavid misses Draisatl on his wing. Everyone talks about how much Leon benefitted from Connor, but he is really the only other player on the team that has the speed, high end skill and elite hockey sense to match up with McDavid. Maroon is a good complementary left winger, but the team really has nothing close to a very good option, let alone elite, on the right side when Draisatl moves to the 2C slot. That could be an issue.

    Like

    1. Last season, Draisaitl without McDavid was just OK, McDavid without Draisaitl actually did just fine (very little change in scoring rates).

      Let’s hope Playoff Draisaitl Without McDavid is the one we get, rather than Regular Season Draisaitl Without McDavid. But Playoff Draisaitl’s numbers suggest strongly it was a hot streak and not something sustainable.

      As for Nuge vs Draisaitl, I think I made my reasoning pretty clear in the article as to why the gap between those two is much smaller than most think it is. If you don’t account for the McDavid effect (and ideally also the Russell effect, but as I noted in the article, you don’t need to), you cannot get a fair bead on any player. All of your impressions will be clouded over by McDavid vapour trails.

      Like

  7. Do your numbers only account for EV minutes? If most of RNH time with Russel is EV against high quality opponents or even on PK, and a significant amount of RNH time without Russel is on the PP, then wouldn’t this account, in part, for the large discrepancy in the numbers with and without Russell? That is, how can usage be factored into those numbers?

    Like

    1. All numbers I looked at are 5v5. PP and PK are each such unique situations and IMO need to be analyzed on their own (something that most analytical and statsy types, self included, perhaps spend far too little effort on doing).

      Like

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s