Hits are Terrible and L Should Have Won the Wylde... Here's Why

[Nykonax 1,550]

The VHL award ceremony is going on right now, and the Jake Wylde trophy for best defensive defenseman went to Randy Marsh with 5 votes. The other candidates being L with 4 votes, and Cowboy Prout with 2. For a while I've been saying hits are bad, but haven't actually done a full breakdown of why, so here it is

 

Brief intro, increasing checking means more hits means more penalties right. Discipline doesn't do anything just reduces your hits and thus probably penalties. (I haven't actually tested discipline increases, that's just my theory with some basis that it does reduce hits).

 

So looking at the 3 candidates, I compared their stats and added a couple of my own. The first being Hits/Pim, which is just a ratio of how many hits per penalty minute. It's found by just dividing hits and penalty minutes pretty self-explanatory. I think it's a decent stat especially to counter the "but hits lead to turnovers" argument. A higher hits/pim is good, since it means your hitting more and causing more turnovers without taking penalties.

 

Next stat is SH Goals Responsible For (SHGRF). This stat is just on average how many goals the other team would have scored while the player was in the penalty box. It isn't actually the real number since I can't be bothered to manually check every game for 3 players. But I still think it's decent. It's found by taking the players total penalties, then multiplying by 1 minus PK%. So for example, if we wanted to find Kyl Oferson's SHGRF you'd take his total penalties (8), then multiply by 1 minus Chicago's PK% (1-0.776), and you would get a total of 1.79 goals. Obviously goals can't be a decimal which is why it's just an average. I would assume the number would be higher since most dmen would be playing more PK minutes, and substituting them with a worse player or bot would only increase the number.

 

Without further ado, lets look at the stats.

Player	SB	PIM (No Fights)	Hits	Hits/Pim	Total Penalties	SH Goals Responsible For
Randy Marsh	162	230	312	1.36	115	22.21
L	164	147	244	1.66	72	16.13
Cowboy Prout	189	118	168	1.42	59	13.19

 

 

As you can see, Randy Marsh and L have similar shot blocks (giving this one to L since Malmo was a worse team with more shots against). Looking at Hits/Pim L is higher, suggesting he was more efficient in creating turnovers. Then Marsh was responsible for 6 more goals against than L. But Marsh still wins the award because his hit number was bigger, despite being less efficient at causing turnovers and costing his team more goals. Great defensive defenseman am I right?

 

I also decided to look at the other 3 names listed in the Wylde post.

Player	SB	PIM (No Fights)	Hits	Hits/Pim	Total Penalties	SH Goals Responsible For
Alex Letang	139	207	264	1.28	102	22.98
Robin Galante Nilsson	229	76	153	2.01	38	9.84
General Zod	140	154	282	1.83	77	16.08

Letang is the same thing as Marsh, just less efficient. Nilsson and Zod are interesting though, and are who I would've gave my votes to along with L. Nilsson has super good SB, and super high efficiency. Only thing is he suffers from a lack of volume in hits, so you could argue against him for that. Zod I think is super underrated and could be deserving of the award. Super efficient with hit volume similar to Marsh, and responsible for same amount of goals as L.

 

 

I think this award should have come down to L/Zod, and we really need to re-evaluate how we look at hits.

 

P.S. I kind of click-baited the title, looking at the data should probably be Zod winning it but gorlab gets more clicks.

 

 

[bigAL 2,176]

Damn I feel like we’ve been looking for a good way to quantify defence in STHS forever and you did it before the awards were over. I like rates, more is not always more when it comes to hits and PIMs. Tying the PIMs directly to team PK is so smart and effective. Damn. Nicely done. 

[Frank 4,764]

Which NHL analytics department do you work for? @Nykonax

[FrostBeard 1,975]

@Nykonax
I honestly agree with everything said. 
For a while we have been missing one crucial thing when it comes to defensive defenceman award. To me HITS are important but even more important than that are low PIM. 
With low PIM you are on the ice and playing your role as a defender, by taking a penalty and sitting in a box for taking ruthless and unnecessary hits you hurt your teams defence. For people voting on this - it SHOULD be taken into the account. 

Mainly the evaluation should go like this: 
1. Is the player on a team that made playoffs or at least was competitive in the season while fighting for a playoff role? Any player who is playing on a tanking/semi-tanking team could not really get that award as their stats are usually inflated. 
2. SB are the most important together with +/- as +/- rating is not only showing how well you score as a unit but also shows how well you are defending after getting those goals and quality shifts. 
3. Smart hits without a lot of PIM are taken into account. 

4. PIM should be the second main factor (After tanking/semi-tanking team) that eliminate a player from the award race. 

[oilmandan 686]

Hey @Nykonax

Interesting take on things. I found it odd I was even nominated, I dumped the remaining chunk of my tpe into checking and fighting to beat up everyone in the league. But no one wanted to dance. So blame the VHL for not wanting to fight Randy

[Nykonax 1,550]

5 hours ago, FrostBeard said:

Mainly the evaluation should go like this: 
1. Is the player on a team that made playoffs or at least was competitive in the season while fighting for a playoff role? Any player who is playing on a tanking/semi-tanking team could not really get that award as their stats are usually inflated. 
2. SB are the most important together with +/- as +/- rating is not only showing how well you score as a unit but also shows how well you are defending after getting those goals and quality shifts. 
3. Smart hits without a lot of PIM are taken into account. 

4. PIM should be the second main factor (After tanking/semi-tanking team) that eliminate a player from the award race. 

i dont think a player should be disqualified if they are on a bad team, can still be a good defensive dman if you look at efficiency stats as well. But I do agree there should be some sort of weighting on shots blocked. Not sure if I agree on +/-, you could make the same argument as shotblocks - that players on good teams have an inflated +/-. Not really a great metric because like real life, you still get a minus even if it's not even close to your fault.

 

I think my order of weighting stats would be SBs -> Hits/Pim -> Hits -> SH Goals Responsible For (if you want to use that metric despite it not being completely accurate). There's probably some stat you could make correlating shots against a team and shots blocked by a player which I think would be useful.

[FrostBeard 1,975]

1 hour ago, Nykonax said:

i dont think a player should be disqualified if they are on a bad team, can still be a good defensive dman if you look at efficiency stats as well. But I do agree there should be some sort of weighting on shots blocked. Not sure if I agree on +/-, you could make the same argument as shotblocks - that players on good teams have an inflated +/-. Not really a great metric because like real life, you still get a minus even if it's not even close to your fault.

 

I think my order of weighting stats would be SBs -> Hits/Pim -> Hits -> SH Goals Responsible For (if you want to use that metric despite it not being completely accurate). There's probably some stat you could make correlating shots against a team and shots blocked by a player which I think would be useful.

Yeah, I can agree with +/- after thinking about it, I still think it can be a tiebreaking value. 

When it comes to weighting stats, I do think that we have to be careful as sometimes, if you are really good player on a bad team with a nice build, you can just get extremely amazing stats. Robin Galante Nilsson is a great example this season - playing on a fairly weak team and just getting these easy numbers, we saw similarly with Viktor Alexei Kamenov on Malmo (Check when Malmo were pretty bad, compare his SB and you will clearly see an inflation) - I think that the only situation when a player on a worse team can get Wylde is if that player manages to do something extremely unique - like 280 SB compared to the next 210 or 200, when the margin is substantial enough. 

You did mention shots against a team/shots blocked - this is a very good metric and maybe there is a chance to develop a rating for that, so we have something that we can use in the future. Basically, we have to create an advanced stats system for VHL.

[Baozi 1,635]

I've never known Nyko to be this insightful. Very well done here.

[Prout 373]

Cowboy FISTED ANALLY BY A CIRCUS MONKEY confirmed. 

Edited July 17, 2021 by Prout

[Ricer13 2,592]

1 hour ago, Prout said:

Cowboy FISTED ANALLY BY A CIRCUS MONKEY confirmed. 

Second 

[dasboot 521]

Don’t give awards to inactives

[Victor 10,536]

Ngl I thought L did win the Wylde 

[CowboyinAmerica 2,708]

Review: 10/10

 

The work you put into this is evident from the jump, and I think you backed up your assertion extremely well. While people can agree or disagree with your ultimate conclusion (I'm personally not sold on hits/PIM being an apples-to-apples comparison that allows for an average to be taken since not all PIMs are a function of hit situations), I appreciate that you put in the calculations to back up your point. The presentation of this one was excellent as well, with well-formatted tables that made all of your stats easy to read. It was a relatively quick read even with all the stats as well, had a good flow to it. Great job overall!