Login

juke · 03-13-2020, 11:30 AM

Ready for grading: 3031 words and some nerd shit
(x2 draft media bonus)

As the SHl draft looms near for our s53 class hopefuls, I thought it would be interesting to take a retrospective look back at our SMJHL entry draft, and see how well everyone stacked up for their first year, compared to what they were projected to accomplish based on their draft order. This piece includes the TPE gain trends of both our entire draft class and additionally on an individual level, TPE gains per team, and a re-ranking of the draft based on both TPE as well as rookie point production. And I will of course be using the opportunity to hype up my fellow Berserker rookies whenever the opportunity presents itself as well. Before I get started on the analysis, I have a few caveats/precursors to the data:

Redundancy: I’m a few days late actually writing this up at this point, and in that time, @"thecanadiancanuck" put together a fantastic, incredibly in depth article with basically a lot of the same information that I’ll have here (https://simulationhockey.com/showthread.php?tid=104172). So, if you haven’t already read that, you should check out that, because it’s better written, and a deeper analysis than I’ll be talking about here. And of course, my fellow Finnish defensive partner @MarlinManiac4 put out a huge article about a week or so ago, tracking all of the rookie’s TPE per team, and how well they’ve improved in their first year (https://simulationhockey.com/showthread.php?tid=103868). So yeah, this will be a little repetitive if you’ve read those, but if you give double media for draft articles, you’re bound to get similar breakdowns. So if you want more in depth, player by player breakdowns, please read those articles. If you’re simple minded like me, and like looking at scatter plots with colored dots, then I’m your guy.

Slightly out-dated TPE: All the TPE per player in this analysis came from the TPE tracker spreadsheet, which at the time of writing this is a little outdated. It’s still a good indication of well all of us are earning, but it’s missing a few RNG TPE events, like PBPs, and SHL predictions, which shot a few people up the rankings.

No free agents: Since this write-up is strictly based on a ‘re-draft’ setting, I didn’t include free agent signings in the analysis. This means FA standouts like Taylor Johansson and Santtu Rasanen won’t be included, despite having great TPE earnings for late additions. It’s strictly a breakdown of all the players that were available on draft day, and where they would go in a re-draft. Part of the reason I did this was so it would at least slightly differ from Canuck’s piece.

Drafting inconsistencies: When I say re-draft, I mean the term pretty loosely. When I re-rank a draft based on TPE or point production, the order is literally based on that arranged list only. It doesn’t take into account that teams might draft for a certain positional need, or the rule that I still don’t really know too well about GM/co-GMs having to go to certain teams in certain rounds, or if there’s any rules regarding drafting re-makes. Just pretend the draft is a pure free-for-all, and teams take the next person available on the list.

Interactive plots: If anyone has followed my previous media, I sometimes include links to interactive versions of the plots. I did this here as well for most of the graphs, since I felt like I should do a little extra since I’m repeating previously written concepts. Most of these graphs are scatter plots where each point represents a player, and it can be hard to keep track of specific values and players. The interactive plots let you hover over a point, and see the information that it contains, like which player the point represents, as well as the specific value that it’s graphing, whether it’s TPE, scoring, re-draft rank, etc. I recommend checking those plots out to get the most out of the graphs.

With that massive precursor out of the way, let’s get into the data. The first thing I wanted to look at was how much TPE our draft class has earned as a whole. The first graph shown here is a histogram of how much TPE each individual in our entire draft class has earned.

Each bar represents a range of 5 TPE, and the height of the bar represents how many players are within that range of TPE. As you can see, there’s an astonishing number of players above 250 TPE in our class. In fact, the most popular TPE range of our class is in between 275-280 TPE. It’s been mentioned frequently by some of our mentors that this is a strong and deep draft class, and this first graph backs that claim up. When I was looking back at the draft and the TPE list, I noticed that a fair amount of players don’t have an updated TPE amount, not even 155, so I presume they just went inactive and whoever updates the TPE form just didn’t bother to enter them. But even so, 29.6% of our draftees are either at 155, or the unlisted amount of TPE. Compare that to the 31.6% of our class that’s at 250 TPE or above, and I think it further speaks to the depth of this class. I don’t know how the previous draft classes compare, but having a higher percent of players at nearly 100+ career TPE earned compared to players inactive seems like a pretty impressive feat for the s53 class. But enough about our depth and success as a draft class, it’s time to judge us all as individuals.

The first section that looks at us as individuals, is the draft order vs. TPE. The first graph here, has a data point for every player based on where they were picked in the SMJHL draft, vs. their acquired TPE.

[Image: tpe_vs_pick.png?width=732&height=488]

Link to the interactive plot: https://chart-studio.plot.ly/~smalinowski7/5 (hover your mouse over data points to see more information)

The color represents every player’s re-draft value. The re-draft value is basically how much a player would have jumped or fallen in a re-draft, if the new draft order was TPE based. This means of course the earliest drafted players will mostly have negative re-draft values, since there’s only so much TPE that we can all earn, and in a re-draft they’re far more likely to drop a few spots than to take over the number 1 spot. Especially factoring in the RNG TPE like milestones, predictions, etc. That being said, I think there’s a very interesting pattern that we can see in the data. First, it looks like for the most part, the players drafted in the first two rounds tend to stick with updating more, and most of the highest TPE earners were picked in the first two rounds. An essential part of building a good team is making sure you hit on your first 2 round picks, as most of the other teams will be having success with them. Starting in round 3 is where things get interesting. You can see a very distinctive spread of TPE and career earnings start to form. About 40% of 3rd rounders (the green dots) and later continue updating at rates pretty similar to the higher picks, though not quite as high as still with some drop off with later picks, while the rest end up fizzling out. While hitting on your first and second round picks in essential to getting high end talent, getting lucky with a diamond in the rough after round 3+ is where teams really fill out their depth to build a championship roster. I want to take this time to hype up my first Berserker, Clean Andrei (@Clean Andrei Kostitsyn). I know a lot of people think he’s just playing a character only and not taking it seriously, but at a modest 240 TPE earned, he would be 15 picks higher in a re-draft. He’s proven that his time in the SMJHL is about more than just finding Sergei, and that he wants to be the best on the ice and win. Now that we have re-draft values for all the players, it’s time to give out some awards:

The Red Wings Euro Twins award for best draft value:

[Image: euro_twins.jpg]

There’s a little bit of confusion on my end about Mattias Birdstrom, but he technically has the highest re-draft value. But I’m not even sure if he was drafted or a free agent. He’s listed in Marlin’s article as a free agent, but listed as drafted on the SHL site. It also says he was drafted by Anaheim, but he played for Newfoundland this year. He does have the highest re-draft value if it’s legit, earning 184 TPE and going from pick 96 to 54 if you re-drafted by TPE. However, since there’s confusion, I’m instead going to give out the award to someone else. Ethan Duncan has the next highest draft jump at +35, followed by Separa Borra at +29. Huge respect to them for their grind, but I’m biased and this is my award to give out, so I’m giving it to Ryuuji Minamino (@Nhamlet) at +28, who was the 4th highest draft jumper. Ham was originally a 56th overall pick, but has absolutely grinded his TPE since then, getting every TPE possible to move up to pick 28 in re draft. The only TPE he’s missed are the pre-draft stuff when no one really knew what they were doing on the site anyways, and whatever SHL team drafts him will get a dedicated TPE earning machine, whose value is underrated if you just looked at career TPE. Steve and I even called it a few or so ago:

The Nail Yakupov “slow down the rebuild” award:

[Image: yak.jpg?width=732&height=549]

I wanted to be careful and not too harsh with my words, so I didn’t want to call this a bust award, but I mean that’s what it essentially is. It’s the award to the person who would fall the furthest in a redraft, and it sadly goes to my Finnish teammate Miro Slapskinnenen. Miro was pick 21, but peaked at just 199 TPE, and would fall 28 spots in a re-draft. Like I said, it’s essential that teams hit within the first 20 or so picks, so busting at pick 21 will surely slow down Vancouver’s rebuild, as they didn’t end up with too much high end talent from this entry draft, which you’ll see in the next graph.

The Dennis Green “They are who we thought they were” award: (I don't know how to insert GIFs, sue me)

[Image: greener.jpg]

Just for fun, there was one player who would’ve gotten drafted in the exact same spot based on a TPE re-rank, and that is Vince Chalut from my Berserkers. I actually won’t go into too much detail about him here, because I talk about him in another section, but he was the only 0 draft differential rank, so I thought he deserved an award. I also want to continue to flex my power on giving credit to anyone I think deserves it, so my bud Steve Harrington (@"NeonLime") gets a piece of this one as well, albeit for different reasons. I know it seems like an underwhelming title to be granted, but when you’re drafted at #4, getting what we thought out of that pick is essential to the team’s future success. We thought we were getting an elite talent, which is exactly what we did get. Steve is the premier young talent on this roster, and is projected to be a high pick in the upcoming draft. Whoever drafts him will surely continue to get what they expect; which is an elite playmaking forward, and an absolute riot in the locker room.

Similar to Marlin’s article going over TPE per team, I basically wanted to summarize it all into a graphic form. The following graph does just that, plotting each players TPE earned, sorted by team.

[Image: tpe_team.png?width=732&height=513]

Link to interactive plot: https://chart-studio.plot.ly/~smalinowski7/7

I don’t want to ramble on too much about this graph since you can always read Marlin’s piece, but the one thing that initially jumps out is “Damn Detroit.” I mean they clearly had the best draft and acquisition of players from the s53 class. An astonishing 7 players at or above 270 TPE, when the next best team has 3, they’re surely to be favorite to win the 4 star cup when a large chunk of their roster will be at the 350/425 cap. I already talked about Miro before, but Vancouver really could’ve used that 21st pick to increase some high end talent from this draft class, as they’re the only team with just 1 player above 250 TPE. Halifax, the last place team this year, has a few players at high TPE, but also a big chunk of inactives from this draft, so their future could potentially be rough once again when we all hit the 425 cap.

Moving on, I think it’s pretty obvious that TPE is the best way to re-rank a draft. It’s the only way to truly evaluate how active and successful players will be in the future, and for the most part, it’s the only thing that the players have control of. That being said, I thought it would be fun to look at the draft by how much point production the draftees gave their team, and essentially see who gave their team the most points at their respective draft position. It’s obviously going to be a crapshoot, one because of Simon, and also because the rookies got drastically different playing time depending on their team situation, but it’s an exercise in curiosity.

[Image: points_pick.png?width=732&height=504]

Link to interactive plot: https://plot.ly/~smalinowski7/17/#/

As you can see in the graph, there is indeed a slight trend of more points from the higher draft picks, but overall it is mostly all over the place. I was slightly surprised to see that the scoring trends were basically equal for forwards and defense, but I chalk that up to the fact that I think most rookie defenseman got more ice time than their forward counterparts. If you’re wondering why I included labels for the top 10 scorers, yes, it’s because it includes my name and I wanted to look good. But using these graphs and labels, it’s pretty clear to see that Paddy Kane-Reilly II and Vince Chalut would be the highest re-draft values based on points scored, with the former having the highest overall re-draft value. Funny enough, both of them lit it up with a relatively low amount of TPE, just earning 212 and 198 respectively. Chalut specifically hurts to see on this graph: he started the year absolutely on fire, and got most of these 22 points within the first 35 games or so. But then about halfway through the year he stopped updating, and seems to have vanished from our discord all together. With only 1 current active defenseman on our roster next year (me), Chalut would have been a shoe-in for a top line pairing defenseman. And with two of the top 10 rookie scorers, Newfoundland’s blue line would have been solidified in a season or two, and just in need of depth pieces to round out the lines. But instead, we’re back to square one and in desperate need of drafting some future defensive talent in this upcoming draft.

Like I mentioned before, scoring isn’t the best way to determine a player’s value, especially with Simon, due to drastic differences in playing time. So I tried to combat that, by normalizing everyone’s scoring based on time played, as well as by position. The following graph shows draft position vs. a relative scoring value, which ranges from about -2 to 2, with 0 being the average.

[Image: relative_scroing_pick.png?width=732&height=504]

Link to interactive plot: https://chart-studio.plot.ly/~smalinowski7/11

This makes the graph even more convoluted, and shows that there is almost no trend between draft order and scoring. This graph is once again colored by re-draft value, with the new value being the order of this relative point value. A lot of the higher draft picks are colored red indicating a lower re-draft, but remember this method is mostly a crap shoot and doesn’t mean anything, and a lot these picks are over 0 meaning above average anyways. Hovering the points in the interactive graph can give the exact values of each. Weirdly enough, and I don’t think this means anything, it seems like the scoring was clustered based on draft rounds. Most of the round 1 picks were clustered around 0 to 0.5, round 2 was clustered around 1 to 1.5, round 3 clustered around -1 to 0, and so on. I don’t think there’s any logic behind that, but it was a pretty repeatable pattern seen for the first 4 or so rounds.

The last thing I wanted to look at was the TPE that players gained, vs. the scoring they produced on the ice. This gives us a look at who may have over-performed or under-performed this year, which would have been a fun thing to follow along with for next year. But with the transition to FHM and new builds, player’s success next year will probably be highly independent of this year. Some of the higher ‘over-achievers’ include the top rookie scorers, which is intuitive since we’re all around the same TPE and they happened to score more. Lucas Hellquist and the previously mentioned Chalut and Reilly II also appear on the overachievers list. Some of the players that did not get blessed by Simon, and scored less than what they were expected based on TPE include Erik Johansson, previously mentioned Minamino, Hiroshi Ohira, and potential first round SHL pick William Hartmann. When repeating the normalization by position and playing time for a relative scoring value, we can see that Zoltan Topalo significantly got the biggest boost in points production from either Simon or maybe high TPE line mates, and Hartmann once again got the middle finger from our beloved and thankfully retired sim engine.

[Image: points_tpe.png?width=732&height=504]

[Image: points_tpe.png?width=732&height=504]

[Image: relpoints_tpe.png?width=732&height=504]

Link to interactive plots: https://chart-studio.plot.ly/~smalinowski7/13
https://chart-studio.plot.ly/~smalinowski7/15

That’s all the graphs and stats I have about what re-draft value each player provided from their draft position from the s53 class’s SMJHL entry draft. As always, if you read all of this, I applaud you for sticking with the somewhat dry material and my inability to write well. I encourage you guys to take a peek at the interactive graphs, so you can use it to harass teammates that would have been drafted lower in a re-draft, or to boost your own draft stock as we head into the big league SHL draft.

*puolivalmiste* · 03-13-2020, 11:39 AM

Damn you all crazy rookies writing deep high quality draft articles. Good stuff.

MarlinManiac4 · 03-13-2020, 11:51 AM

***Nhamlet*** · 03-13-2020, 01:10 PM

***Ace*** · 03-13-2020, 02:40 PM

Racist against goalies.

LordBirdman · 03-13-2020, 02:46 PM

This was awesome to read, Juuuuuuuuuuuuuuuukkkaa

**Lime** · 03-13-2020, 03:12 PM

03-13-2020, 11:30 AMSmalinowski7 Wrote: Steve is the premier young talent on this roster, and is projected to be a high pick in the upcoming draft. Whoever drafts him will surely continue to get what they expect; which is an elite playmaking forward, and an absolute riot in the locker room.

holding back tears

nice

Navigation

Extra Menu

About us