Saturday, August 3, 2019

An update on the project, and where you can find all of my work!


Hello again, readers. I've finally finished pondering on, and subsequently building, the vehicle that I want to use to deliver the ever-expanding body of work I began to undertake on this blog in July 2018. This post is going to detail what my thought process was on the way here, and what the New Deal is when it comes to how you're going to be able to access and interact with my work. If you just want the short version, and to skip to viewing the content that's been released publicly now, go to patreon.com/petropraxis. Available for free, to anyone, are the Player Cards of Cooper Marody, William Lagesson, and Mitch Callahan. Player Cards are an aggregation of on-ice/individual shot/goal/points stats from my tracking project  - shots, shot assists, on-off shot shares, etc; and the player's ranking among the entire team in each of these metrics. Below, in the text box, are some notes about their deployment, who their linemates were, and my thoughts on their performance as the statistics describe it.

Here is an example:




As for the schedule of free content going forward, it's flexible; I'll be unlocking other Cards slowly over August and September - of course 16 more of them are available right now to anyone who becomes a Patron. 

Moving on, the rest of this post is going to go into detail the process I'm going to publish through moving forward, and how we got here.

...

The original aim of this blog was to see if we could take the mainstream discussions we were having as a community, deconstruct them, and either validate them on their merits or go beyond them. Take conversations that were happening over and over in a feedback loop; ping-ponging popular talking points, and see if the ground they're standing on actually holds under duress. In this sense, that experiment is over. The experiment of finding out if people thought that was worth reading, and worth having, is also over.

As I poured time into this project, I faced two main problems. One, the issue of wanting to do something better than what was possible without more labour time. Trying to continue writing essays and doing data analysis on the Oilers, while doing the same with the Condors, in addition to tracking, packaging and analysing micro-stats wasn't gonna happen. Two, I was going to have a deployment problem. Without designing some kind of central hub for the work, people were going to have as many navigation problems as I had organisational problems, and figuring out how to properly present the work was going to eat into these precious hours.

I've been doing this planning for awhile, thinking about how to open up ways to support me now that I know there's a lane for me and my work; as well as massively expand its scope and deliver it in more effective ways, and landed on Patreon. It's a content-hosting platform that lets you release different content to different tiers of subscribers; mine will just be the $4USD tier and the free tier. A lot of hockey content creators are on Patreon, but I want to do something a little different than any of them. Here's why I'm going in this direction-

1. I really didn't want to do ads or click-based revenue. There's more superficial reasons for this - I don't like the presentation of it, the last thing I want is distractions on the page. There's also the really important structural reason, which is that garnering unique page views would become such an inevitably heavy input on editorial decisions -which is lethally toxic for the type of work I'm trying to give to people. Beyond that, putting out the hundreds of hours of work for free is damaging to every other creator (many of which I respect deeply) that should be fairly compensated for their own work, should it be in demand.

2. I want to open up the data from the Condors project in a more direct sense that just injecting it into my pieces, or sharing bits of it in my conversations. This way, it makes more logistical sense on a human level for me to spend hours upon hours meticulously tracking it and opens up more time to do so, speeding up the timeline in which I can inform you radically more than anyone else on this critically-timed farm team season. There's nowhere else to find the level of depth that will be provided on what is likely to be the most important facility to the transition from the current Oilers team to a championship one.

The needs that need to be met - cheap skill to surround the thermonuclear core that's in place can only be provided by a strong farm system; any other way costs much too much in assets. The Oilers are up against strong development teams in the Pacific Division. Becoming Syracuse-West all but guarantees some version of a championship window. Dissecting this season will provide, among other things, multiple measuring sticks:

Marody's zonal-transition artistry will challenge every centre that comes through the pipeline. How will, say, McLeod's exit/entry carrying and passing compare?

Benson's primary shot contribution shares (shot and shot assists) will inform us on how each and every playmaking forward is progressing in their rookie season.

A mature utility prospect must approach Gambardella's forechecking recovery rates, resultant shot impacts (on-ice and individual) and goalscoring in order to earn an NHL shot from a long way out.

Ethan Bear's exit passing will help us contextualise what will almost certainly be exit-passing dominance from Bouchard. How far ahead will the 10th overall be of gifted passers past?

What's also critical here is the amount of players (and their distinct ability) vying for NHL jobs in the fall from this Condors roster, the lack of potent challengers for those spots currently in the system, plus the difficulty in creating strong enough barriers for the more likely players to earn those roster spots by outside acquisitions - you can't sign a Tyler Benson in free agency. Because of the likelihood of some of these players to make the NHL next season, we can really underline the value of the benchmarks set by those players in this AHL season, and figure out how translatable each and every level of performance and facet of the game is from AHL to NHL.

3. There's a ton of areas I want to improve in that are crowded out by the amount of labour time that's needed to put into what's on my plate already. Expanding the video content here beyond the basic level of what I know how to do right now has been an ambition of mine for awhile, as well as getting better with Excel and Tableau. The Condors project data will be so, so good with some visualisations. Whether I learn that myself, or can use support funds to pay another person to make some vizzes for us, this work can get a lot richer and cleanly accessible in short order.

4. I'm wanted to open up some editorial input to you. Particularly the readership that gets enough value, entertainment, and thought provocation from the work to be compelled to support me. That's the pre-publishing input you guys are going to have: since I'll always have more theories, ideas and impulses for content to create, than time to create it with the level of effort I want to, we're going to conduct polling to collectively decide which article or project ideas for the bonus content are the most popularly anticipated and I'll execute on that. Beyond that, I'll have the final say on the constitution of the content schedule in order to balance the Serious Analysis pieces (player profiles, hard target searches, projections) with, say, the irreverant and occasionally antagonistic media critique and other lighter, less serious work. 

...

With all of that process in mind, I had to figure out what I was going to open with, what I was comfortable with asking for support in return for. There had to be a good enough backlog of content to be worth subscribing for at first, and an easily-communicable program that will be built on.

The rewards for being a patron will always be visible on the front page, but here's a rundown for launch:
  • Access to the constantly-updated Condors Data Project works, which at launch includes all 21 regulars' player cards and in-depth overviews of full roster charts and rankings. 
  • Access to the exclusive constantly-expanding shift-by-shift and clean highlight video archive, beginning at launch two videos - a 16:50 minute long-form compilation of Ryan McLeod's defensive and transitional game and a video of all points scored by Evan Bouchard.
  •  A vote in the polls to decide what the bonus written content is, what the next video archive addition is, and a question of your choice to be answered in-depth in the mailbag posts.

Details from what content is available now, also excerpted from the Patreon front page:

At launch, we currently have a near full set of player cards - those are compilations and visualisations of individual (shots per game/per hour; shot assists per game/per hour; shot contributions per game/per hour; % individual contribution of total on-ice shots-for) and on-ice (shot share %s; on-off shot share %; shots for/against per hour; shot differential per hour, goals for/against per hour, goal differential per hour, GF%Rel) and the player's ranking among the 14 forwards and 7 defensemen who have more than 20 games tracked in each of these metrics.

All of these stats are found on one page, as well as an explanation below of what their deployment history was, and my general thoughts on their season along with any other little details that are pertinent to getting an accurate overview on where the player's at.

You can click on the Player Card button on the left side of the front page under Featured Tags and it'll filter everything else out so you can scroll freely among just the Player Card posts. From there, each player will have a tag you can use to filter out everything but posts that involve that player. Ergo, if you want to find everything that's been written about that player, click their name among anywhere tags are, on the left panel or at the end of any post that mentions them. I'll be cleaning up my blog posts about Jones and Benson and posting them over here to centralise that here as well.

We also have two video archive posts - a long-form (16:50) gathering of Ryan McLeod's work in the defensive and neutral zones in checking and in transition and a compilation/replays of all points by Evan Bouchard with the Condors during the playoffs last spring. There's also a tag for Video Archive posts, meaning you can click on that and all of the Video Archive posts that have been uploaded will appear.

And finally, I wrote a rundown on the full rankings of each Condor by their individual rate metrics, through charts and typical musings, as well as some notes on deployments.
One last thing though: It's inevitable that a lot of the content behind the paywall is going to get smuggled out. If you see it copy+pasted in some Reddit thread or Twitter or anywhere else, don't bother telling me about it - just reply to them drop a line or link this Patreon page for anyone else who sees it, letting them know whose work it is and where they can support the author and find more.

A final note on the material reality of this work: were 10% of the unique visitors I typically get on my articles to chip in that small subscription fee, I could do this work full-time. If you're at all interested in what I've outlined above, or just enjoy my work as it is, please consider becoming a part of the support network that allows me the ability to provide massively more of my current canon of content, as well as go far beyond it.

Either way, as always, thank you for reading.

Wednesday, March 20, 2019

The Post-Chiarelli Puzzle Has Been Solved Once Before


The symmetry between the condition in which Chiarelli left the Bruins and the condition in which he left the Oilers is both striking and repellent.

However, studying the journey the Bruins have taken from that wreckage all the way to two strong chances at the Stanley Cup Final in back to back years is always informing and at times downright educational. Certainly it is for us as observers - and hopefully it will be for both those who are searching now for his replacement as well as whoever that replacement ends up being.


...


One of the reasons this comparison is often easily made is that, counter to a popular opinion, Chiarelli is said - by all sources that have spoken on the issue - to have had autonomy and trust from upper management in both cities. Trust to expend futures (with seemingly nothing off limits) and trust to build a team in his vision.

He expended futures, with seemingly nothing off limits, and he built those teams in his vision. The league landscape took more kindly to what Chiarelli thought the ideal hockey team would be in 2008-2014, and he was supplied with more pieces of his liking through the pipeline in Boston than in Edmonton.




Via Sportrac, This is Boston going into the summer of 2013 versus Boston going into the summer of 2015. The difference in roster size is a function of one being at the end of a playoff tournament, and the other being a team that didn't enter the tournament, where obviously the some of the roster limit rules being lifted is used to full effect.

When you look at this think of what the decisions about who was kept, who was disposed of  (and the price attached in each decision) tells you about the disposition of the decision-maker. Seem familiar?

How about the end result, do you recognise that?

Let's start with the result and work our way back. The 2014-15 Bruins were over two million USD over the cap pre-adjustment, and the Oilers today are too. Boston shelled that (3rd highest in the NHL) value out for the 17th highest point total that year - and Edmonton pays the highest total in the league for their spot at 24th in points this year.

Boston spent:

  • 19.24% on non-ELC replacement-level (or close), bottom-half players;
  • 10.59% on a pair of 33-year-old bottom-half players with trade protection;
  • 31.43% on three core players (Bergeron, Chara, Rask) with full no-movement clauses;
  • 2.63% on buyouts.

Edmonton is projected to spend:

  • 19.82% on non-ELC replacement-level (or close), bottom-half players;
  • 12.29% on a pair of 30+ year-old bottom-half players with trade protection;
  • 33.18% on three core players (McDavid, Draisaitl, Nugent-Hopkins);
  • 2.01% on buyouts.




Yet again via Sportrac, this is end of 2015-16 and 2017-18 for the Oilers. The degradation in skill is profound, especially since there is just one real addition positionally in roster spots that matter, and it was obtained at a nauseating cost.

One entry of each list is not like the others. Having the core locked in wasn't a weakness of either club (and on Edmonton it will be a massive blessing - ridicule anyone who points to those three contracts as a problem), but it does describe the level of maneuverability available and the general makeup of the rosters being similar.

Chiarelli's last months in Boston saw him spend - among other (insignificant) things - a 6th and two 2nds for Max Talbot and Brett Connolly, in a hurry to improve waning playoff chances. The latter would be more useful than the former - but neither would remain with the team past summer 2016. In what would be his last months in Edmonton, several strange pick and young player expenditures netted him four players; two are already gone and the other two are likely to leave as well - one by free agency and the other by buyout.

You can draw parallels between the 2015 Bruins and 2018 Oilers hand over fist when it comes to target acquisition and structural makeup, especially in terms of salary cap expenditure and preference of position. Off-ice, you can also notice the distrust of data on one hand, and the misuse of it on the other. Signs of outright illiteracy in that field were there, best exemplified with a statement marking Kris Russell as a top zone exit defenceman league-wide.


Via an Edmonton Journal transcription of an Oilers Now interview with Peter Chiarelli in October 2016; author David Staples.

It's pretty easy to reverse engineer from a conclusion like that back to an error in interpretation of data, collection of data, or both, with considerable precision. To follow up that conclusion with statements like you can derive analytics any way you want at the end of the day, and there's been a lot of moves that are based on analytics that don't work out, is pretty obscene for obvious reasons. Possessing a contradiction of both distrusting data's utility and being willing to have it reaffirm your most dangerous self-sabotaging biases is to sit very far into the most dangerous quadrant of possible dispositions in regards to managing a hockey team.

You can also consider his longstanding bullishness on the team - and the players on it - that suggests the pride of autonomous ownership. General managers in the NHL have to dance that dance to a degree, but there's a certain tact that went unused by Chiarelli. This was not due to errors in communication, but rather of substance. The only targets of Peter Chiarelli's criticism (among his acquisitions) seemed to be Cam Talbot and Todd McLellan - the latter rumoured to not be Chiarelli's explicit interview-and-hire (which I buy) and the former similarly rumoured about (which I don't). Lucic was celebrated for the most meagre of perceived improvements over the past year; Larsson was spoken of glowingly far beyond even typical pro-sports protocol for fluffing incoming assets. On that last one, it's informing to look it up and read his words on Larsson again. It's really something.

Pampering of Chiarelli hires extended into contract negotiations. Consider the treatment of Jeff Petry, after signing a one-year $3 075 000 contract, by Craig MacTavish to the treatment of Kris Russell, after signing a one-year $3 100 000 contract by Peter Chiarelli. 


...


The future that laid in front of the Bruins in the summer of 2015 has many differences from the one that awaits the Oilers this summer, but something I'll touch on now (and repeatedly harp on from here on in) is that the ability to sweepingly reconstitute an NHL roster and organisation in a short period of time is much more available than what is commonly said or seen.

A lot of bearish takes on the likelihood of future success for this organisation depend on the long history of NHL teams changing incrementally - and typically being awarded strokes of genius by fate as opposed to aggressively engineering the team they want and that team winning the way they want it to win. I have time for this as a general perspective, it's how I view the entire story of the modern NHL up to like, fifteen years ago.

But it's actually ahistorical and pretty much an outright fabrication to draw from that that teams are inherently these slow-turning freighters that can't radically alter the direction of their team-building. The reason you don't see massive salary-dumping summers or the type of big trading years that resemble your armchair-GM sessions is because teams (management and ownership) straight up don't want to do that most of the time. Not because it's ill-advised or impossible. It's ahistorical to put forth any other conclusion because of the historical examples that contradict it. 


...


This is a good time to get back to Boston, Massachusetts. Let's go back to June 25th, 2015. The general manager had been replaced, his last deeds had been a blow to the draft capital the team should have possessed, which was inconvenient because the draft was the next day.

The first post-Chiarelli trade is a classic cap-dump: Carl Söderberg for a 6th round pick in the draft after tomorrow's and that's it. This'll go well, because they'll later spend that pick on Oskar Steen, the kind of smart late-round gamble you take if you pay attention to the Swedish U20 league, the SuperElit. It's since paid off.

Söderberg would sign with Colorado and take of 6.51% of their cap space the next year in the first fifth of his new deal.

The second trade is the Dougie Hamilton deal the next day at the 2015 NHL entry draft. He's another RFA that had to sign, and he did for the Flames at 8.05% of the 2015-16 salary cap. The return was the 15th, 45th, and 52nd overall, all 2015ers.

The third ships out Milan Lucic with Boston retaining 45.80% to make the money nearly balanced when Martin Jones, Colin Miller and the #13 overall comes back. Lucic was at $6 000 000 (making 45.80% $3 250 000), Jones would sign for $3 000 000, and Miller $602 500. Here's that in visual format  (mercifully) via Capfriendly:





Then on June 30th Boston flips Jones to the Sharks for a 2017 1st round pick and the rights to Sean Kuraly - in a way dumping the entirety of the amount that Lucic would take up by not having to pay the $3 000 000 that Jones would sign for. 

The fifth, final significant transition trade is the coupling of the tail end of Marc Savard's back-diving contract (owed $1 150 000 cash that's counting as $4 021 428 cap for each of two seasons; though spending them on LTIR of course - fuck Matt Cooke and concussions) with pending RFA Reilly Smith for the rights to Jimmy Hayes. Where season-opening cap is concerned, this frees up $7 446 429.


...


And there you have it - I'm going to ignore a couple things (Jimmy Hayes' signing and the trade of a 3rd round pick for Zac Rinaldo that went unmentioned but occurred on June 29th of that timeline) for the purpose of stripping down the events to simply illustrate how much cap had been dumped. Of course those spots need to be filled, and they were by Hayes and Rinaldo, but the point here is to see where things were after being stripped bare, by a real team in the real NHL a mere four years ago.

Remember as well, that this summer was highly, highly controversial in Boston and around the league when it came to Bruins fans and hockey fans in general. Don Sweeney was instantly ordained as one of the wilder general managers, and it wasn't just because of the draft selections. Lucic was a fan favourite. Hamilton was given the hatchet job by media while also being meaningful to the fanbase as the other major piece of the Kessel return.

The reason I make this point is to distance the understanding of Boston's 2015 summer from it  being a stroke of genius that turned the team around, or aligning of the stars otherwise. This is super important; it makes my whole point that major roster surgery isn't just possible, but plausible. It's  not to be understood as a terrifically difficult undertaking or an unspeakable risk but a modest and open-ended template.

That's not a commentary on the personalities or managerial styles of those that sit up high in Edmonton, by the way. Another thing I'll repeatedly stress is that this is not what I think the Oilers are going to do, if anything I'm getting out in front of a possible narrative that'll hit the airwaves and online spaces this fall if the 2019-20 season starts off poorly.

It'll go like this: Big Bad Man Chiarelli boxed them in (true) and so there were no good answers to return to competitiveness in sight this summer and New Guy can't be held accountable yet (false).

...

Back to the (true) story. Here's what Boston removed, by percent of the incoming 2015-16 salary cap:

Carl Söderberg - 6.51%
Dougie Hamilton - 8.05%
Milan Lucic - 4.55%
Martin Jones - 4.20%
Marc Savard - 5.63%

Total - 28.94%

That's a lot. In 2018-19 cap space, it's a shade over $23M. In 2019-20's, it's a hair over $24M.

Most  low-effort attempts at a straight-across analogy from Edmonton's prospective 2019 summer to Boston's 2015 past will be incoherent. You need to shave things down in some areas and do some addition-by-tinkering in others. Edmonton doesn't have a Savard-like LTIRable back-diver. Pending RFAs are few and not as valuable as Hamilton.

Here's an expendable (consider that two players the Boston sent out wouldn't be 'expendable' by the following arbitrary standard) asset list by cap (2019-20; $83M) percentage.


Sam Gagner - 3.80%
Zack Kassian - 2.29%
Kyle Brodziak - 1.39%
Andrej Sekera - 6.63%
Kris Russell - 4.82%
Brandon Manning - 2.71%
Matt Benning - 2.29%

Total - 23.93%

And their 2019-20 buyout savings / / bury-ables where applicable - where the player possibly wouldn't be claimed on waivers and there's savings in the buyout- no totals because that would be silly.


Sam Gagner - 2.57% / / 1.30%
Zack Kassian - 1.61%
Kyle Brodziak - 1.39%
Andrej Sekera - 3.61% / /
Kris Russell - 3.71%
Brandon Manning - 1.61% / / 1.42%
Matt Benning - 2.01%


(Notice how Milan Lucic is missing? That's because there's no use in bringing him up as an asset-out in a broad view of what can be removed. It's empty fantasy unless you narrow down the conversation, which I'll do in a bit. 
-
For inquisitive minds, his contract is 7.23% of the cap, and his buyout would save 2.36% of the cap. The percentage of total cap that would be moved if his contract was moved along with the rest of the list above is 31.16% - just past the amount that the Bruins were able to shed.)

The point here is that there's a practical liquidity to this roster, in the sense that a one-summer teardown been done before by a team that was both in a similar position (strong core locked in; overpaid ineffective supporting cast) with similar goals.

Setting exact values and conjuring up precise exchanges is a whole other pile of labour (which I will probably do) but what we're going to do here is take the historical examples of cap-shedding transactions from the Bruins' 2015 summer, as well as recent moves by other NHL teams, and process the Oilers' possibly out-list through that encyclopedia.

The point is to combine all the different types of deals and strategies in recent NHL history with the amount cap space that has been opened up in our historical example in order to model the possibilities a retooling summer could have in a way that's realistic and valid.


1A. Cap-dump for Draft Capital

A player with (some) positive value, for picks, with much lesser value than the players' actual impact on the ice. It's difficult to exactly gauge who fits where here on the pay scale of Edmonton's asset list: Andrej Sekera and Kris Russell are closest to taking up Söderberg's cap-space percentage (Sekera a near-perfect match) but there's trade protection available to each individual and an age/health concern with the much better player. Zack Kassian and Matt Benning are both permissible logically, Kassian almost certainly had a price-tag set at the deadline. Those two together represent a possible 3.62% of cap space instantly available - though unlike at defence, there's not enough forwards (realistically) coming for Kassian to be automatically replaced.

The key signifier here is that the asset earns a return that is obviously negatively influenced by their cap-hit to contribution ratio or constrained bargaining position of the selling team.

I'll note that the list I perused on Capfriendly of simple, two-asset cap-altering deals depopulates significantly the closer you get from seven years ago to today. The richness of contracts being dumped in this way has also been reduced in terms of dollars of cap itself and not just cap percentage, making the case for an argument that teams have become hyper cap-sensitive profound. Consider some of those older deals against a 2017 example of retention being needed for a much less weighty cap burden.





1B. Player for Prospect

There's off-ice strata to this, with the slide rule, waiver eligibility and signing status itself as significant factors. For example if you have a signed player that's outperformed their draft position, but has already signed their ELC at a lesser price-point than they would have if they signed after their uptick in production or otherwise perceived production, that's a value modifier. Waiver eligibility is a massive value modifier.

These deals are more complicated than straight 1A types are. There's also a kind of stylistic slant to them; I've noticed they're favoured by two types of management teams. Those led by former pro/ama scouts/scouting directors - Jim Benning loves trading for prospects instead of picks - and those led by timeline-conscious actors, like Peter Chiarelli circa February 2018 when he pivoted one return into another to turn Pat Maroon into Cooper Marody and Mark Letestu into Pontus Aberg through player-to-pick-to-player transactions.


2. Asset Deconstruction

This protocol was used in the Lucic-for-Jones, Miller, and a 1st trade.

In practice, most 'package' returns are bad. A good way to avoid suffering the typical fate of selling player for package is to move on from the out-player at the right time; scout another organisation's assets well; achieve a cap-related objective; or all of the above.



This is a an optically uneven example outside of illustrating the basic design for that reason. Miller wasn't an NHLer at this time and was coming from a long way back, and Jones wasn't a starter. You could theoretically get this package for a player of Lucic's former prestige in the form of a young, capable defenceman and a matured and talented backup with starter potential, but that doesn't mean they're going to turn into Colin Miller and Martin Jones. And that's without the 1st. It's important to understand that this trade isn't being used as a template because it was a winning one.

Anyways, one part that is good about using this couple of deals as a practical example is that the cap objective wasn't meant to be achieved in the first deal. Most of the time it won't, if only by virtue of the shallowing pool of cap-imbalanced trades year-by-year as of late. In the modern NHL, every team is in on cap-neutral hockey trades, but salary-shifting ones have a much smaller market. The cap-neutral trade with Los Angeles involves most of the moving pieces in the sequence, and then the simpler sale of Jones is the one that results in relief of 3M.

You're probably wondering what heavy-cap asset the Oilers can afford to deconstruct this way. I'd argue there isn't one - Ryan Nugent-Hopkins qualifies especially as a player on a career year, but he's not only inner-circle in effectiveness on the team, but he's also a rare (on this roster) proven skill forward. You would essentially have to trade him for a wing that can score and a secondary asset, and then pacir that secondary asset with picks and prospects for another wing who can score in order to end up with a net gain in forward scoring. Possible, but requires some dexterity and for those inclined to care the optics would be awful.

Here's the real answer for modelling this trade to fit the needs of the current Oilers cap situation-

Invert the values of the assets.

We can even use the same initial player. The current Lucic contract is probably the same distance in value from replacement level as the version the Kings traded for, except in the opposite direction.

So in this model, instead of trading a major positive value asset for a group of them, we're trading a major negative value asset for a group of them

For example, trading a majorly overpaid player for a minorly overpaid player and a junk contract.

You then play the minorly overpaid player in the former's roster spot, and buy out the junk contract or trade it again. Nobody's taking Lucic's contract straight up, not even if Puljujärvi is attached. You could be trading with a budget team, so you could also be trading for a LTIR version of the same dead cap, or a player that the trading partner would like to bury in the minors but the ownership won't let them, there are tons of possibilities here.

The general idea is to replace the frontloaded, massive values required in order to get teams to take a larger contract with several smaller taxes.

3. Asset and Anchor

This template is one used in almost every speculative summer I've read (and ridiculed) on Capfriendly's publicly published Armchair-GM pages for the Oilers, typically involving one Milan Lucic. This means it doesn't need much explaining, but the history of these deals is the most important point for understanding what makes our guesses realistic or not.

Brooks Orpik's contract in the Colorado trade is showing up on Capfriendly as his current, post-buyout contract. This is incorrect - at the time of the trade his contract was 5.5M through 2018-19.

Let's add some context quickly to our Post-Chiarelli Boston example: Jimmy Hayes was coming off of a 19-goal, 35-point performance in a 72-game 2014-15 season. He was (and is) also a six-foot-five BC-alum born in Dorchester, MA. The Bruins weren't trading for a warm body here, or they didn't think they were, which is all that matters when we're learning trade values from history. It's about what GM's think players' 'true' value is.

I included a lot of trades with LTIR players here. Can you guess why?

That's right, because those deals can be processed as a hypothetical leg to the Asset Deconstruction in the hypothetical Oilers post-Chiarelli summer.

...

Plenty has been written about who the Oilers need to hire - general manager and otherwise. Brainstorming about who should be hired, what kind of departments inside the organisation need to be built or bolstered - fans and media combing through interviews and press availabilities of different candidates.

These things change on a whim (with possible candidates being confirmed or denied) and it's hard to really say who unequivocally should take these positions with just an everyman, spectator's amount of information on these executives.

In addition, regardless of who Edmonton hires, their goals and objectives should be the same, in a broad sense. That's why talking about and detailing them is a beat with much more meat on the bone.

On one hand, yes, the Oilers need to do a Toronto-style front-office housecleaning.

On the other hand, they would also be served well by a Boston-style book-cleaning.

Think of this piece as a throwback to the original purpose of this blog: grab everything that's taken for granted in the conversations being had by Oilers fans - and played out in the media - deconstruct it all and see if they (or, how they) still ring true.

Is your twelve-trade armchair-gm summer realistic? What are the ways teams have solved problems like this in the past? What's the historical context for your attempted Lucic trade?

I hope what's been written here answers those questions; I hope that more questions are prompted as well - and feel free to pose those to me in the comments, or in Twitter DM, or through email.

As for what I'd do, who I'd trade for, sign, and how I'd make it all work?

Coming soon.

Wednesday, February 20, 2019

Caleb Jones is the Bakersfield Condors MVP

I was going to wait to call something like this until I had every player's data worked out; on ice and individual; with-you and without-yous, video review too and so on - but by the time I got to sorting everything out for Jones I'd seen enough and started to alter the outline and frame of this piece to reflect his status as the most valuable player for the Bakersfield Condors this year.

It was a bit of a worry, Jones's player-identity lines up with my preferences in a defenseman (the kind I would ice six clones of) and we'll get into what that means, but the trouble was essentially that I was getting ahead of myself based on the eye-test before I had the data to challenge and adjust the perception that was forming, which is always something you want to avoid in either direction. This was also after the early part of the NHL audition he earned, where I was very impressed with his third pairing work and had no expectation of success to disappoint when it came to his entirely too early 'promotion' to Larsson's pairing.

In fact, I remember the exact play when I realised the player had - crucially - taken his eyes and his sense with him to the NHL, while keeping the heart rate in a healthy range. It's that point when you realise that someone gets it.




It's kind of funny that at the time, and still now, I can't really tell if this is something of little significance that bias is blowing up because of the trauma we know as Oilers fans from watching defensemen puck-watch or puck-chase or be puck-paralysed in this exact situation so many times, or if instead it was a cue that an unbiased, normal, non-Oilers-watching-brain-having person would take note of.

After tracking another 16 games where the shot-share dominance persisted (probably more-so than I have recorded because of the score-effects of the recent win streak) I started to want to pick out the strong-links, find who the drivers were. Marody and Benson were obvious suspects, but they've always been outscored understandably by the glued-together veteran line that was better at busting cycles in the defensive zone, and generally more consistent at puck-recovery as a trio all of elsewhere in the rink.

But I also had noticed Jones many times, doing things that you seem to pick up on when you're a tracking project like this. A lot of the stuff that we went through in my Sekera post, the back three-quarters. Picking the puck out of traffic in the defensive zone and wheeling away, testing forechecker's cardiovascular facilities, solving the puzzle of the neutral zone with a neat pass. The type of play that you encounter if you're watching a strong possession defenceman for the first time in awhile.


...


A fortunate part of my confidence in the identification of Jones as the main engine on the blueline at 5-on-5 is courtesy of the fact that his tracked games were split up in such a way that's sometimes rare for regular defenceman. In my sample - which is an accurate reflection of his season; all but five of Caleb Jone's games in the time period are tracked, and the time period tracked is up to February 9th - he's played on a default pairing with Stanton ~43% of the time, Lagesson 48% of the time, and the other games were with Gravel.

The reason that this is fortunate is because of a problem that's always run into: Whether you're model-building, pursuing a wider view by informing yourself of a player's stats (advanced or traditional), or even just using the eye-test, is that the difficulty of accurately dividing up blame and praise for any player's contribution to any given situation is a large part of the fundamental and daunting barrier of analysing the game - that's much more easily scaled when you have both a large sample size to draw from, in which there's a large array of teammates (of varying qualities and identities) to help isolate the player's effect on the game.

So, what'll happen a lot with defensemen is that they'll end up playing a ton with just one partner, meaning their time apart with which you're trying to solve a question based on an assumption (they're both good together, but who is better apart is the better player) often has a couple major maladies: the sample size is small, and the player that is better in the coach's eyes plays the harder minutes away from their regular partner. Say there's a defense pairing with one savvy veteran and one talented rookie - when the rookie is already in the process of sheltering by virtue of playing with the smart established player, it logically follows that that rookie's situations away will be soft ones where the coach is capitalising on the tactical vulnerabilities of the opposing team, and the veteran's ones will be mostly spot duties where the vulnerability is his team's own.

This is a solved problem in my sample when it comes to Jones's defense partner, and with the forward line teammates. Jones' pairing is typically the top/tough comp. one (my partially-educated guess, though, is that Woodcroft runs a top-four style Dcorps as opposed to there being a top-pairing and two others) and so they do play more with the top-six forward group the Condors ice by virtue of both who they're deployed against and the fact that they play the largest portion of game-time and so do the top-six forwards.

In terms of shot-events, I manually went through and counted each that occurred where Jones was on the ice with one centre and no others. Around 16.9% of such events with Marody, 20.1 with Currie, 30% with Malone, 17.7% with Vesel, and 15.3 with Esposito.

Given that our sample then has a good division between both defense partners and forward lines, I'll assert that if, for the majority of them, their better minutes are with Jones as opposed to without him, we can safely assert that (within our sample) Caleb Jones is the man most pushing the possession margins for the team on an individual basis.

Let's find out if that's the case, starting with the two defense partners that Jones started and spent the vast majority of his 5-on-5 minutes with for 90% of the games in our sample:





This is pretty stark. I imagine that it's somewhat likely the inequality of possession in the AHL between teams and their players is greater than in the NHL, as teams have raced (for longer) to close those gaps in the highest league where they were first noted, discussion, and subsequently addressed to the point where CF% is less likely to predict GF% than it was in the past - and this discussion (in public) hasn't really occurred in regards to the AHL.

What I mean is that a player with Jones's ridiculous shot share (64.47 CF%) is perhaps a lesser standout (percentile-wise) in the AHL than the NHL.

I'm attempting not to fawn.

Below is the same deal, but with the centres. One detail of note is that since these are events that occurred when each centre was the only centre on the ice, excluded is the minutes spent the recent line of Benson-Marody-Currie (which were probably positive shot-share) and the season-long phenomenon of DZ faceoffs with whatever centre plus Luke Esposito taking the draw (which were probably negative shot-share). Another is a reminder that we saw Benson, who played on one line for a majority of my sample in my previous posts (the profile on him and the update on the team) crank up the shot share for the team in a major way. Does he or his line with Marody pull up Jones's numbers?




There's a sore thumb here (and in the next) that I'll address real quick: The Josh Currie numbers are a result of a few very successful lines deployed in the beginning of the season when the Condors were especially dominant on the shot clock, combined with four later games where the Benson-Currie-Puljujärvi line essentially torched every 2nd shift for 3-7 shot attempts, resulting in dumb stat-lines like 26-5 CF-CA goosing some already strong numbers to the point of absurdity. It should be mentioned, though, that most of those dominant shifts were (as we see here) spent with Jones and given what we know so far, it's fit to wonder how much the defenseman figured into that equation.




Again, the uniformity here does a lot of the work for you in terms of figuring out who's doing what. The important bit here is that the two taller blue pars are representing a pairing that, as we've seen above, the centres do worse away from. This is a boat-floating player, in terms of possession.

Let's wade into goals. More specifically, shot-shares against goal-shares, team-wide, Jones-on versus Jones-off.

Something that has to be noted before we do, though, is that the numbers I'm now going to be using are going to be the ones that are recorded officially by the AHL itself. They're not limited to my sample, and so I won't be speculating on the difference between the shot and goal-share as if the delta is from the exact same set of games, just games from the specific time period (start of season up to and including the game on February 8th, 2019). Essentially the question to be answered is, how much were the shot-shares from my sample representative of goal-share of the team over that period of time in totality?




Interestingly enough, goals outperform shots when Jones is off the ice, and shots outperform goals when he's on. The shot-share dominance is still too large - the goals can only wander so far from it (typically goaltending only goes between .890-.930, shooting 5-11, you're not going to end up south of 50% goals if you're pushing 60% of the shots very often once the sample size grows large enough) for it to go into an awkward place where the possession is essentially for naught. Still,  Jones not standing out in 5v5 primary points per hour either (that stat ranks the Condors regular six in the order of Day, Stanton, Jones, Lagesson, Bear, Lowe) prompts an exercise in establishing whether or not there's a bunch of inflationary corgis baked into this thing offensively.

Turns out, in terms of conversion, Jones has the typical talented young defenseman effect of increasing the GF/60 to a great extent and the GA/60 to a lesser one en route to his net beneficial effect to the team's goal differential over time. Credit should be given to Woodcroft for cashing this, because it's a gambit that commonly has been rejected outright by traditional hockey coaching and management staffs - to their profound disadvantage - for at least the past decade-plus that we have the data to reference, and likely many before.



(For those interested in how I got these numbers - you're not going to find them by googling - these estimations were the product of grabbing the estimated TOI and player-level and team-level on-ice goals for numbers from prospect-stats.com, and then using the fact that NHL teams played 48 minutes 5v5 per game on average this season and using that as a rough reference to give the Condors 2352 5v5 minutes played given their 49 games played at time of writing. Then I just divide and conquer the on/offs from there.)

Remember the breakdown of partners from earlier? The defensemen Jones was partnered with for 90% of the games I tracked (and, since I still watch the games I don't track, I can tell you that ratio applies to all Condors games, or close) are 4th and 6th among the Condors six regular defencemen. That primary implications (if these estimations are close to accurate) of that, in my view are population two:

  1.  Jones' GF/60 would likely be even greater if he were partnered with better offensive facilitators.
  2.  The theory of Jones' numbers being inflated by the possibility that he could have been deployed in the most favourable situations (most minutes with the top-six, offensive zone draws, OTF abuses, weak competition) takes a bit of a hit when you see a lead this large over his partners in those minutes - for the record, I'd also wager the third pairing with Logan Day gets the most red meat in terms of deployment. 
Even I didn't expect that he alone would be the primary driver of actual goals per hour on the blueline.

...

An exercise of interest for me, under the umbrella of post-mortem team analysis, will be to dig into how much of the Condors shots-for at 5-on-5 come from defensemen, or from the point in general - and then slice that up into each different player and how they effect that. From a visual standpoint, Jones doesn't have the best shot of the Bakersfield Dcorps but he certainly finds himself or others in tighter on a more regular basis and is exceptionally (among this group) aggressive in jumping up in the play. Only Logan Day is similar in that regard, but he certainly doesn't pick his spots as well.

It'll be interesting to observe whether or not Jones' on-ice GF% will come closer to his CF%, and if it does whether or not that's via the share of the shots decreasing or the conversion-rate/save-percentage increasing; and all of the same in regards to the inverse situation of the Jones-off hours.

Last thing on the hard possession numbers: There's a split in performance from before he departed to the NHL, and after. There was some talk from Ken Hitchcock right after it was announced Jones was to be sent back down, the coach putting forth that the player had lost confidence and was in a bad place on the ice and off. That could be one explanation for the splits, but Jones has also put up 9 points in 10 games since returning to the AHL. I'll suggest the team is playing different games now, they're leading more, their overall possession advantage has eroded slightly (but goal differential has improved) as you'll learn about in my next team update if that difference is still present at the time of publishing. This likely combines with a readjustment to the game he plays in the AHL as compared to the one he does in the NHL; the shift that's likely more comfort than confidence based.

Let me give you a look at the early game logs; and mind this is also a good demonstration of what the game-to-game CF-CA splits of a player who puts up near 3:2 ratios really looks like. We'll move on to the video after this - and you'll see what I mean when I talk mentioned the different 'game he plays in the AHL' - but check out this run:





...



I want to start by reintroducing the division of a defenseman's game that I used to contextualise my look at Sekera's return to pro hockey, that you can read about in the beginning of that piece.

Jones lays waste to the AHL in this area, but I want to cut the back three-quarters in half again - though not by area. I want to instead mark the difference between a set, proactive play and a broken, reactive play. It would be significantly more difficult to communicate this distinction so cleanly with one clip of one ~30 second run of hockey time, but luckily there was a great example within those parameters to be found.

To quickly brief you on it, the first part is a clean breakout by the Condors where they lure each of three Roadrunners, pulling them far enough to make a backcheck mightily inconvenient and sendin Jones on his way for a clean-to-clean exit-to-entry. The brain is there on this play, but so is the head-up, four-way mobility by Caleb himself, especially in the last pass out of the zone where he seperates from his check far and fast before receiving the puck. Later, you'll see an example of Jones taking advantage of the timing and spacing of the opposition, on a more broken play, that he must react to.




When I'm tracking games, the primary goal is to get all of the data down accurately and quickly. Aside from that, I've also made a habit of marking down specific time-stamps of plays to return to and clip, with just short words or phrases making an umbrella under the massive array of unique sets of play that happen in the game of hockey. One of them, is 'stop'. It's when a player that's in front of the puck stops the flow of play in their direction; 'stop and go' is when in that same stopping motion there's an immediate and successful pivot that sends the play safely (for their team) and dangerously (for the other) back the way it came.

One can break down every little part of a player's game on video in equal-ish partitions, but it's much more efficient to take a combination and blow it up for the majority of the analysis if your goal is to give the best impression of both a player's identity and how effective they are with it and I'll show you now how it works: Here's a compilation of Caleb Jones stopping, going, and his 'stop and go'.




There's a lot of common themes here. The opposition generally doesn't know if he'll pass or skate; when or where he'll do either. There's a useful hesitation (Oilers fans will know what useless hesitation looks like) that fakes out opposition without Jones' forward linemates running out of track. This area of the game (the reactive half of the back three-quarters) is just so important in the environment that is the neutral zone in today's professional hockey game.

An argument that plays like in the clip above are the majority reason that Jones drives possession is a convincing one; the same if you're arguing that he sets himself up well for an NHL future by doing this at twenty-one years of age. Add to that the commonality between what Jones brings and what the Oilers lack and it's hard to imagine a 2019-20 season that doesn't see Jones in Oilers silks for much of the campaign, barring injury.

...


The offensive zone game is where Jones's ability is less unique, but still effective enough to become an NHLer - though if he'll be powerplay-playing defenseman is better left to a scout or one of those sort than I. He's not being groomed to be one, I haven't done a ton of work on the Condors powerplay but often they'll play two defencemen and in points it's a split between Logan Day (who's essentially deployed as an offensive specialist), Caleb Jones, and Ethan Bear. Of the 26 PPP earned by defensemen on the team, those three make up 22 of them: Day with 9; Bear with 8; Jones with 5.

I'll top off my doubts with a two-parter that I have of Jones' powerplay game that I feel pretty accurately represent what his impact is at the 5-on-4 game-state. Phrased shortly, he's as good as you'd assume (based on his 5-on-5 back three-quarters game) at facilitating the zone entries, which is critically important, but just regular good as a distributor once everything's set up. Here's an example of the Condors entries, two unique-ish ones regular speed and the default, go-to ones sped up:




...


Below is an opening clip of what illustrates Jones' ability to take and make a pass that's beneficial on at 5-on-4, followed by a series of clips that gives a good reflection of what's the game-by-game, more unassuming participation of Jones on the Condors' powerplay. He's not Erik Brannstrom (who has a billion powerplay points in this league as a teenager) so it's more a matter of getting your teammates to the latter stages of what becomes a scoring play than directly participating in them.



...

I want to make a quick pivot to the player's shot, and goal scoring. Jones' goals come from hard slapshots, and that's primarily the setup for his shots - where he leads the Condors with 2.29 shots per game all-strengths - as opposed to trying to sift wristers through, and he'd rather use a slapshot for an inside-out deflection (from the centre-slot, say) than a slap pass for a outside-in deflection - backdoor tap-ins and the like. Here's the player's clean goals.



...

As you could probably guess seeing as the player facilitates offense as opposed to being a point-producing passenger, he's pretty close to full value on his assists. Some of them are more remarkable than others; being plays that essentially span the whole ice surface, turning a blocked shot or pass against into a fast break goal. Here's some of my personal favourites.



...

At this point, I hope to have hypnotised the reader to exhaustion, and that their brains are now fertile to any conclusion I'm willing to put forth due to the sheer enormity of labour in writing this piece overpowering any opposition . That the mountains of evidence are so vast that even if I've interpreted them incorrectly, there's no easy defense against a turning-handle mounted machine gun spitting out innumerable falsehoods.

Seriously though, I do hope that I've made a persuasive argument for Jones as the most successful Condor, but before that or anything else I hope the most that I've given a good snapshot of what Jones has been this season, and the type of AHLer he is. I hope that someone who's never watched Jones or the Condors at all this season gets a good impression of what the prospect does well and how well he does it - and that regular observers of the team or closer examiners of Oilers prospects have their knowledge base expanded as well.

I'll end now, as in my profile of Tyler Benson, with my prescription: It'd be wise to not outright gift a third-pairing role on the Oilers to Jones outright, but believe your eyes if you see that he torches all competition for that slot; between NHL hopefuls and NHL veterans alike.

Tuesday, February 12, 2019

Slick Reggie Wheels; Stunts on Stint

Imagine you're Andrej Sekera.


You stubbed your toe (actually, got your toe stubbed by some strange sadistic man) and then fell down a staircase immediately after recovering and then have to patiently recover while your team rolls around in a puddle of gasoline. Pressure mounts as your return comes closer on the calendar, the fanbase's collective anxiety builds a house of cards of emotional stability.


Then, you escape the frigid landscape to California, play your first hockey-game hockey in relative ages, and some dude on the internet records every one of your shifts and writes down everything that happened in them.


What a world.


...


The early seen-him-good takes are out, and I'm not about to contend with them (I seen it too) but rather contextualise where he was good, what I know and don't know, and for those inclined provide the picture the data invokes in its admittedly miniscule sample.


It's a lot easier to do this deal with a guy we know because I'm not exactly starting from scratch in terms of my understanding of the player, and my understanding of the fanbase's understanding of the player.


Every good son & daughter loves Andrej Sekera's work as an Oiler for the same reasons, more or less. He's (arguably?) Chiarelli's best addition on the blueline, he was the first free agent to say "Connor McDavid", and he was the first top-four defenceman the Oilers just got in a while. When he or Klefbom took on tough competition for the majority of the 5-on-5 time that type of competition was on the ice for in 2016-17, fans could watch the team in the back three-quarters of the ice wondering (with some degree of calmness) how they were going to get out of this one, as opposed to wondering how they were going to get out of this one. Sekera and Klefbom gave the impression that if they were ever in possession of a hand-grenade, they were at least going to handle it deftly.


The back three-quarters is a distinction I'm going to use to split the focus of the video review of Sekera's game. Before we check out the February 8th game, let's get one thing out of the way:

  • I'm not trained, educated or particularly well-read on  anything skating conventions and form, nor am I particularly adept at sussing out where the exact line is for each individual player's skating that, should they fall below it in ability, significantly hurts their capacity to help the team when they're on the ice. I can tell a fast skater from a slow one, I can tell who's quick, who has a "plus" first couple steps, but in the most populous part of that bell curve of NHL skater's skating level, I can't tell the 55th percentile from the 45th. I'm not going to pretend to be the ultimate arbiter on this player's skating and whether it will or won't be an issue because that's not my current capacity.

What I can tell - because of its undeniable demonstration - is whether or not a player gets somewhere on time.


...


When I watched the Condors' games this weekend, I was thinking about what's missing on the NHL team's blueline, and what would be the antidote for the game-to-game issues that ail the team.


Part of the opening barrage of verbal from Ken Hitchcock in his first month of press availabilities (or perhaps it was in the videos that I looked up) was the introduction of a tempo-based phenomenon in hockey he called the three-quarter ice game. It was described as something that arises when one team is hemming the other in, abusing change timings and reducing their attempts to move up the ice to uninspired, soft chips out past centre ice before high-tailing it to the bench. This probably sounds familiar to anyone mercilessly punishing themselves with regular viewings of the Oilers for the past weeks, as the Oilers have the league-worst score & venue adjusted shots-for rate at 5-on-5 in the NHL over the past 25 games.


It's a team concept, and one that's effectiveness you can't exactly bring with you in your suitcase from farm to big-leagues as any particular individual player. So we're not looking for, say, how often the Condors ground-and-pounded the other team when Sekera was on the ice. Rather, we're looking for what he did individually to react to each time the puck (intended to) come out of the offensive zone, or started there. How did he help to defuse rush attempts, muffle carry-in attempts or outwit forecheckers on the recovery. What's the reaction - and is it a good one - to the puck coming down the tracks at him, or when he has the puck and a forechecker is on his way? Perhaps most importantly, is he up to the pace of a professional game?


(Of course, this also has great utility for my untrained eyes to figure out all things mobility, given that how quick you close - or separate - is half of everything that happens)


Here's a compilation of plays that are lumped together for a reason you'll implicitly understand (all of my readers are sterling intellectuals) as you watch. If you read or watched nothing else in this article, this video would give you a good idea of what people mean when they're writing or talking about Sekera looking good so far.


Bonus theatre, from here to the end, is getting a good look at #37 William Lagesson as well.






They're all plays that, whether or not the situation demanded it or not, the middle of the ice is used to make a play that is on its surface perhaps more risky, but is in fact the safest, cleanest play to get the puck moving the other way.


As is familiar, Andrej Sekera has this kind of nifty way of tricking checkers, like in our profile of Benson (and with all intelligent players), he wants the first layer of checking to come as close to him as possible. It makes their road back to the play longer, and the gaps in their frame (five-hole, triangle) larger by virtue of them being closer and therefore easier to pass through.


Obviously Sekera being able to make these plays shouldn't be surprising and is largely irrelevant, but I feel the frequency with which he's making them speaks to confidence and trust in a newly recovered body.


Let's pivot to three-quarter ice shifts where possession was established by the opposing team; shots or chances against occurred; etc. Essentially the collective failure to regain possession, and the role Sekera played in it. Shallow pool, by the way, he was on for 4 shot attempts against at 5-on-5 total.


(These will all be colour-coded like past projects because there's no common thread and I need to note specifics)




Red has Sekera toiling away, playing inside on his checks pretty well with maybe a second and a half of letting a bad screen sit, but gets the first step on his man when the puck actually comes his way. A bit too casual on any distribution from there and he turns it over, but he still tailed the puck carrier enough for it to lead to nothing.


On orange an opposing forward definitely has positioning and space on Sekera, but a combination of Sekera closing quickly and the forward having seemingly no interest in even trying to use his speed has the play passed out to Lagesson's check and Andrej just follows his man effectively.


With blue, Yamamoto and Sekera end up with an awkward situation and giveaway a pretty easy exit opportunity, but you'll recognise the later part of the clip where Sekera gets pass past the opposing point man. Shortly after, Lagesson makes a nice outlet pass for an Actual Possession Exit.


Black marks a smart, already aggressive gap being closed and the support Andrej knows he has lets him step in and just knock the puck off of the carrier's stick. Sekera's D partner Lagesson kind of just gets baited by the Barracuda floating and threatening to split the D at the blueline and the outside guy is wide open for the pass with speed. Lagesson then makes up for his mistake quite well and takes him to the wall, but ends up with a holding penalty in exchange for the extra sauce.







Red's a play you might have seen on Twitter, where Sekera just seemingly accidentally crushes someone. Before all that, though, he's got a good stick on everything and never wanders, which is a theme.


The puck hops over Starrett's stick in the play marked orange, which is fun. The key part here is the recognition to stay inside on the Barracuda closing in case the puck bounces their direction, and then the wherewithal to make a better play than just blasting the puck up the boards with his head down (which is a deficiency of the major league team) and fakes like he's going to, to boot.


Starrett tosses Sekera an absolute grenade in the next play, late and far out from the wheelhouse. A bit of a freeze follows but he emerges with the puck and gets pretty clearly held, so an unfortunate sequence. The next time the puck arrives, Sekera ends up taking two Barracuda out of the play physically (?) and the sortie is solved from then on.


...



Moving onto the question that should always tag along with watching for this or that: In total, what happened? We can use the eye test to examine each event, but what we last saw and what was the most memorable play can cloud our recollection of what happened in aggregate. The tracking project this data is (exclusively) from has doubled its sample size to 32 games now; so the numbers from this explainer-post aren't up to date but this is a good briefing on what I'm trying to do.


Also, I'll note that nearly every shift was spent with William Lagesson, whom Sekera has talked about playing alongside in Jim Matheson's piece here, and that these numbers are from both games combined at 5-on-5, not just the February 8 game featured in the video.

Metric - Sekera on / / Sekera off

CF - 38 / / 76
CA - 20 / / 55
CD - +18 / / +21

FF - 20 / / 51
FA - 10 / / 33
FD - +10 / / +8


...



And there you have it, folks, a psychotically detailed report on two games played by a hockey player who's just trying to get his legs under him; warm up the hamstrings.

We'll be going back to player profiles now, but I can also post the shift-by-shift compilation of both of Sekera games on YouTube if it's requested since I gathered it already for this piece. I'll link it here and/or post it on Twitter if so sometime after work today; 8-10 MST.



Monday, January 28, 2019

Woodcroft's Corsi Condors & the Guts of the Project


About a month ago I realised that if I only published the numbers from my tracking project for the Bakersfield Condors after every single player was up to date with each of the data points I was tracking, I'd be doing this stuff way further past real-time updating than I wanted to be, maybe even into the summer.

I found this unacceptable; so instead we're going to build this thing on the fly.

For those who have no idea what I'm talking about, over this AHL season I've been manually tracking certain events that I find invaluable for evaluating both what a team is attempting to do and how much they're succeeding at it, and sussing out the same in terms of individual player's identities and how much they're succeeding with them.

This post is both an update on the project and an attempt to get out in front of some of the questions readers may have about the data I'm citing (sometimes as the focal point) in upcoming player profiles like the one published yesterday featuring Tyler Benson.

I've been compiling the following data-points:


Team Totals of-

  1. Shot Attempts For; Against
  2. Unblocked Shots For; Against
  3. Offensive Zone Carry-in Attempts; Successful Carry-ins; Failed Carry-outs/Turnovers
  4. Offensive Zone Pass-in Attempts; Successful Pass-ins; Failed Pass-in Attempts
  5. Offensive Zone Dump-in Attempts; Set Play Dump-ins; Recovered Dump-ins; Set Play Recoveries; Failed Dump-ins/Turnovers
  6. Defensive Zone Carry-out Attempts; Successful Carry-outs; Failed Carry-outs/Turnovers
  7. Defensive Zone Dump-out Attempts; Set-Play Dump-outs; Recovered Dump-outs; Set Play Recoveries; Failed Dump-out/Turnovers

Individual Totals of-


  1. Shot Attempts 
  2. Unblocked Shots 
  3. Shot Attempt Assists 
  4. Unblocked Shot Assists 
  5. OZ Carry-in Attempts; Successful Carry-ins; Failed Carry-ins/Turnovers 
  6. OZ Dump-in Attempts; Recovered Dump-ins; Failed Dump-ins/Turnovers 
  7. DZ Carry-out Attempts; Successful Carry-outs; Failed Carry-outs/Turnovers 
  8. DZ Dump-out Attempts; Recovered Dump-outs; Failed Carry-outs/Turnovers


How I define each of these may vary from others' understanding of them, so I should flesh that out. A lot of these are based in my individual philosophy of what the optimal play in hockey is, and what is the true importance of events as well as other tweaks that are purely my arbitrary distinctions that are not necessarily and airtightly based on empirical data and undeniable evidence. I'll willingly internet-argue about them until either of us dies or my mind is changed; the latter I promise is possible.

I'll also note that the following is entirely optional reading for this piece. This is pure nuts-and-bolts stuff, and isn't at all necessary to understanding what I'm doing and what the data is or means. If you're familiar with micro-stats concepts and don't really care about the minutiae of what's under the hood, skip it. To easily get past it, ctrl-f the phrase 5-on-5.
Shot Attempts - On-ice and Individual

One place where I may stray from others in shot attempts is that I think an individual fanning on a shot is still a shot attempt. I think this is important because if a player makes a good pass or the team is controlling play, I don't want poor execution by the individual shooting to deflate an event that was earned by the other four players on the ice or the passer. It also adds to the distinction between players who get more of their shots on net, players who can work with hot passes and operate within checking distance. It also adds to a distinction between guys whose passes end up legit shots, getting passes in a guys wheelhouse where it's easier to get a good shot off. Another reason I think this is important is because you're going to have to arbitrate this either way at some point along the spectrum between fanning on the puck so much it barely moves, and fanning on the puck but it still hits the net. A player can flub a shot completely, or they can not get all of it. When does it turn into a shot attempt? If it hits the net or not? What happens to fanned shots that miss the net? Aren't they still valuable in terms of being able to redirect? You run into this issue at some point whether you want to or not.

Additionally of note; I count shots from outside the blueline as successful dump-ins (not necessarily un-recovered or recovered though) and not individual or on-ice shots, even if they hit the net - unless they result in goals. I understand this is heretical because those are absolutely still shots technically and that they've resulted in goals before. I compare it to how the NHL doesn't count shots from behind the net as shots on goal unless they bank off of the goalie/a player and go in, resulting in a heatmap of scoring likelihood (blue is shots that have  below 6-point something percentage of resulting in a goal and orange are shots above that, forget how it was constructed) looking like this:




(Via Emmanuel Perry, by the way)

A third difference is an on-ice distinction - This may be more heresy, but if a player leaves for the bench and a shot for or against happens immediately after another player replaces them, I'm giving that event to the player(s) who left the situation unless the replacement makes it to the blueline. This works in the DZ to punish the bad changer and not the replacement, and in the OZ to reward the defensemen who were on for the successful break-out and not the guy who changes on into a favourable (in shot event terms) situation. Obviously change effects aren't eliminated by this choice of mine, but the more extreme change-based punishments/rewards are.

The reason I don't record actual SOG is because one, I don't typically use SF% in evaluation for a variety of reasons; and two, it's already recorded by the AHL and so I can use it without recording it manually and to inform where big gaps from FF and SF exist. There's also a second witness in this way.



Zonal Transition


Let's define the distinction between a success and a failure. To me, a successful carry-out or carry-in means the player fully crossed the line both technically and in effect. What I mean by in effect is this: by NHL-style, video-review, photo-finish maybe a player crosses the line completely with the puck before getting rubbed out along the boards without any puck support and turning the puck over, but that's still a failure to carry the puck in because the effect is the same. If you don't make it five feet into the OZ before turning the puck over, I don't think any hockey coach is going to consider your turnover not a blueline turnover, the fatal kind, just because you technically did or didn't get across the line. I decide at what point the carry-in is successful on a case-by-case basis, with a focus on whether or not the puck supports existence or inexistence is of the player's volition. Meaning, is he attempting an carry-in entry to the offensive zone while his entire team is changing and turning the puck over, resulting in a headman pass to an opposition forward who hard-punishes the wholesale change for a breakaway chance? Or does he get separated from the puck at the exact moment his puck support arrives, who picks up the loose puck with ease and the going gets going?
In this way, a recovered puck is never a failure. I split credit for individuals who recover the puck on dump-ins, because that's the way forechecks work most of the time. The first forward comes in, disrupts the defending team's recovery of the puck and attempt to break out the puck by separating the recover-er (typically but not always a defenseman) from the puck and the second forward collects. In that situation, both guys get 1 credit for a recovery. In the rarer situation where it's a one man show busting an attempted breakout and recovering the puck all by himself, I give him a 2


What follows is the team totals data that I feel comfortable publishing right now. What I mean in terms of comfortable, is that I've vetted these numbers fairly heavily and feel they're at a decent enough sample size. In terms of distance from being published, I'd say more shot-events data is closer to being here, and zonal-transition team totals are a bit further away, and zonal-transition individual totals are the furthest away in our timeframe. Right now, a lot of stuff (all the zonal transition data) isn't to my standards of publishing, but that also means it's not under the hood when I'm posting player profiles and pieces like that. The only information I'm citing in those posts is the on-ice numbers that follow, as well as the individual players' shots, shot assists, and on/off data that'll be released as each player is profiled and all fit into what we see here. None of the data that's unpublishable is being used or cited in a black-box manner.


5-on-5 Shot Attempts


The games I have tracked are purposely diverse in terms of opponent and date. I have games from each month fairly evenly, none of the months have groups of games all in a row tracked, except for the four games Jesse Puljujärvi played because I was very curious about them. To be transparent, each of the game reports are listed below.

(I'll also add the AHL-recorded all-strengths shots-on-goal totals along with the amount of powerplay opportunities for each team to 'show my work' in verifying that my records aren't out of the realm of reasonable from what the league themselves have on record. The way this works is we're going to look at where any of my 5v5 shot attempt totals are much different from what the AHL-recorded all-strengths shots-on-goal ones are, and then check if the amount of powerplay opportunities given to each team could have resulted in the total numbers getting goosed either way by one team getting, like, 10 more powerplay shots from being awarded 3 more powerplay opportunities. Remember that this is a fairly crude instrument because not all powerplays or powerplay opportunities are equal in shot generation rate or duration. To be clear: The 5v5 data is mine, the All-Strengths totals are from theAHL.com.)


October 20th vs. San Diego Gulls

5-on-5

CF - 72
CA - 33
CD - (+39)

All-Strengths (PPOP)

SF - 49 (5)
SA - 23 (3)
SD - (+26)


October 27th vs. Stockton Heat

5-on-5

CF - 48
CA - 34
CD - (+14)

All-Strengths (PPOP)

SF - 30 (7)
SA - 24 (1)
SD - (+6)

November 16th vs. San Diego Gulls

CF - 76
CA - 34
CD - (+42)

All-Strengths (PPOP)

SF - 47 (2)
SA - 40 (7)
SD - (+7)

November 17th vs Colorado Eagles

5-on-5

CF - 41
CA - 20
CD - (+21)

All-Strengths (PPOP)

SF - 42 (5)
SA - 25 (1)
SD - (+17)

November 20th vs Colorado Eagles

5-on-5

CF - 43
CA - 54
CD - (-11)

All-Strengths (PPOP)

SF - 30 (3)
SA - 30 (3)
SD - (0)


November 23rd vs. Ontario Reign

5-on-5

CF - 58
CA - 49
CD - (+9)

All-Strengths (PPOP)

SF - 33 (4)
SA - 30 (6)
SD - (+3)


November 24th vs. Ontario Reign 

5-on-5

CF - 41
CA - 29
CD - (+12)

All-Strengths (PPOP)

SF - 29 (5)
SA - 33 (5)
SD - (+4)


November 29th vs Ontario Reign

5-on-5

CF - 46
CA - 36
CD - (+10)

All-Strengths (PPOP)

SF - 37 (4)
SA - 25 (7)
SD - (+12)


December 5th vs. Manitoba Moose

5-on-5

CF - 50
CA - 56
CD - (-6)

All-Strengths (PPOP)

SF - 31 (1)
SA - 32 (1)
SD - (-1)


December 8th vs. San Jose Barracuda

5-on-5

CF - 55
CA - 22
CD - (+33)

All-Strengths (PPOP)

SF - 35 (7)
SA - 18 (1)
SD - (+17)


December 19th vs. Stockton Heat

5-on-5

CF - 56
CA - 41
CD - (+15)

All-Strengths (PPOP)

SF - 42 (2)
SA - 31 (4)
SD - (+11)


December 22th vs. Colorado Eagles

5-on-5

CF - 45
CA - 48
CD - (+3)

All-Strengths (PPOP)

SF - 35 (4)
SA - 35 (4)
SD - (0)


December 31st vs. Ontario Reign

CF - 58
CA - 34
CD - (+24)

All-Strengths (PPOP)

SF - 31 (4)
SA - 31 (4)
SD - (0)


January 5th vs. Tucson Roadrunners

5-on-5

CF - 59
CA - 37
CD - (+22)

All-Strengths (PPOP)

SF - 35 (5)
SA - 24 (3)
SD - (+11)


January 9th vs Texas Stars

5-on-5

CF - 49
CA - 54
CD - (-5)

All-Strengths (PPOP)

SF - 30 (3)
SA - 26 (2)
SD - (+4)


January 18th vs Stockton Heat

5-on-5

CF - 47
CA - 31
CD - (+16)

All-Strengths (PPOP)

SF - 36 (5)
SA - 20 (6)
SD - (+16)


Totals:

5-on-5

CF - 844 (52.75 per game)
CA - 612 (38.25 per game)
CD - (+232) (+14.50 per game)
CF% - 57.97%

All-Strengths

SF - 572 (35.75 per game)
SA - 447 (27.94 per game)
SD - (+125) (+7.13 per game)
SF% - 56.13%

PPOP - 66 (4.125 per game)
TSH - 61 (3.81 per game)


I figure another important point here is to find out whether or not I've selected games where the Condors generally have a better powerplay opportunity differential than in the rest of their games that aren't tracked. This could also pollute the sample; what if the games I've selected have, in aggregate, the Condors getting more PPOP for and less against (and from there more all-strength shots-for than against) than they usually do, which would cover up the possibility that I've been too generous a shot-tracker to Bakersfield? The AHL website's team stats section has records of this and has the Condors are 161 powerplay opportunities, and 162 times shorthanded. Over 42 games, that's a rate of 3.83 times on the powerplay per game, 3.86 times shorthanded per game. Given that our sample's rate of PPOP is above and TSH below, we can infer that the group of non-tracked games have an even lower rate of PPOP and higher rate of TSH than in our sample in order for the total rates of the tracked and non-tracked games to land in the middle like that.

So, if we assume there's a normal distribution in regards to AHL team's shot generation on the power-play (meaning the Condors, and their opponents in aggregate, all get effectively close to the same average amount of shots-on-goal per powerplay opportunity) the all-strengths and 5-on-5 numbers seem fairly close and allow for a small amount of a great many possible factors, like perhaps the Condors getting more of their shot attempts from the point or from sources unlikely to register shots-on-goal (more on that in the future) result in the difference there. Perhaps the penalty kill is poor and allows more shots-on-goal than the powerplay registers and that pulls the all-strengths numbers down. If none of these (or a multitude of other factors) are true to any degree, then the only remaining one is that my tracking is generous to the Condors. This could always play a factor, but given the fact that the AHL-tracked numbers are in the same neighbourhood it's unlikely that many points I'm going to be making by drawing on the data I've tracked is fully invalidated by this effect.

Either way, it's useful to the readership to spill the guts on the results of an audit like this (however flawed) so that they (you) can better inform your own decisions and takeaways from my writing.

And now, what we've all been waiting for, Actual Hockey Goals, even-strength, from the AHL themselves.


Even-Strength Goals for and against


GF - 102 (2.45 per game)
GA - 84 (2.00 per game)
GD - (+18) (+0.45 per game)
GF% - 54.84%


There you have it, folks. Even with the considerations listed, I'd posit that the Condors are a shot volume team and that this has translated into a strong even-strength goal differential, though some gets lost along the way. By video, this is a McLellanite team through-and-through. I don't mean this as an insult, but if you watched the early 2017-18 Oilers team, the Condors operate a lot like them in the way they pursue their goals. The breakout is along the boards, and the offensive-zone play often consists of going low-to-high after a dump-in retrieval and hammering the opposition with point shots.


We'll get to a video and zonal transition data based analysis and critique of these things in time - in terms of the team and individual players - but since I'm launching my profiles of individual players using some of the stuff I've tracked so far, I figure now was a good time to update everyone on my processes, and on the early results in totality to give you a good idea of where I'm at in a macro sense of dissecting this team.


I hope this pre-emptively answers most of the questions people have about where I'm at with everything and how I'm managing the endeavour; and gives a better idea of what I'm setting out to do and some of the obstacles that stand in the way. I look forward to publishing more player profiles as things go along and updating them as the project starts to more resemble its finished form.