Jacob's Thoughts on Player, Line and Roster Construction

[jacobcarson877 2,536]

Hello VHLers! I have had a lot of thoughts over the last few weeks, some spurred on by reading other people’s wonderful thoughts and some I like to think are original. What I want to do here is compile some “theory” on player, line and roster building as a sort of intermediate-level document for those who have read through resources like Gustav’s Guide to Building and the TPE & Point Tasks Tutorial. Since those are already amazing, I will not be overlapping too much, so if you do not know how the STHS attributes work and do not know how to earn TPE please go read those they are great.

 

To start off then, I wanted to talk about player building. As a current VHLM GM this is an area that I think could use a lot of improvement right off the bat, and with the hybrid-attributes having been implemented for a couple seasons now, my view on building in the VHLM has changed a lot. Historically the idea was to get the Core-4 attributes (DF, SC, SK, PH) up to ~70 and then go from there.

 

Nowadays, with 1 TPE getting you much less than it used to, some concessions need to be made. Getting those stats up to ~50-60 is not much of an issue if you put your head to it. And for those earning well, this is probably still the way you’ll want to build. I do think one should try to prioritize pushing 1 of those Core-4 attributes at a time, starting with DF, so you always have the advantage at SOME aspect of the sim. The principle of min-maxing in sim leagues is still particularly important. Being great at 1 thing and okay at some others, is better than being good at everything.

 

This leads into my next, and most crucial point about building. This idea is pretty much a straight rip from Shindigs, but I want to compile it here as well. As much as this is important in the VHLM, it snowballs over time all the way up to the VHL. Building all of the Core-4 attributes to a point where you will be great at them is a luxury even the best earners are struggling to afford. What this means is that lower earning players (I will talk about my love for these players later), need to absolutely capitalize on being dominant in something. If they try to spread themselves as widely as the max earners, they will ALWAYS be at a disadvantage in the sim when it comes to every single comparison check. In order to not be blown out of the water in the VHL, all of those amazing depth players need to be competitive in some aspect as soon as 300 TPE with the new callup threshold.

 

But enough beating around the bush, what should those lower earning players be sacrificing in order to be dominant in something? The answer is Scoring. I’m not saying don’t invest in scoring unless you’re max earning, but what I do mean is that great DF will carry you much more offensively than mediocre DF and SC. Building a player that can literally have the puck and skate with it without losing it will lead to your player having more opportunities to do the things that get you offense, shooting or passing.

 

The VHL desperately needs players who can hold and carry the puck for their top stars. One of the leading theories is that only 2 forwards and 1 defender in a unit actually contribute to the offensive output, and the others largely don’t affect the results. Many teams have been riding high TPE duos with a rookie to remarkable success. Many of the top scoring units over the last few seasons have been exactly this. Not to dive into line and roster construction too much here, but as a lower earning player you have to give VHL teams every reason to want to play you in these gold mine roles. Learn to retrieve the puck and carry it, and let the stars drive you to offensive godhood.

 

My last point about player building is actually one I picked up in the SBA, and it has to do with final builds and order of operations. In your first few seasons, you will be tossed around to multiple teams in multiple leagues in multiple roles. For each team you will likely get that urge to completely change your final build to match the situation that you’re in. Your team needs more offense? Well, I will just plan for more offense in my build. That is not exactly a terrific way to handle it. Having a final build to work towards is essential to efficient and decisive building, and although it does not have to be set in stone, and can grow or shrink over time, it should remain stylistically consistent. That does not mean your player has to be stylistically consistent over their career, in fact I also discourage that. I am all for building around your team’s needs, but perspective matters a lot. You can, and absolutely should RUSH certain parts of your build in order to help your team out. In fact, you should always be rushing the next attribute that your team needs. Breaking your build into chunks helps you be able to adjust stylistically on the fly, while also working towards a significant enough milestone that the sim might actually give you appropriate reward. Working in 5-increase chunks should be the lowest you intentionally go for, but maybe early on when attributes are cheaper you work in 10-increase chunks.

 

That is it for player building! Remember to focus on DF and build one thing by a significant margin instead of slowly increasing everything to get that competitive advantage!

When it comes to line construction, I have had a lot of VHLM experience, and a little bit of VHLE-level experience with the WJC. But I have been around enough in the VHL to understand the schools of thought, and I have also found some interesting on other STHS forums about the engine.

 

First off, the non-VHL specific, objectively (although potentially not subjectively) correct way to build lines. GMs from leagues that use NHL players and such to simulate with use this method to enormous success. It is rather simple and makes a lot of sense. You want to have a skater (who is also generally the checker for the line), a playmaker and a scorer. It is ideal for the playmaker to be the C, with the skater and scorer playing the wings. With the overall lack of build variation in the VHL this is not necessarily possible, but there are still opportunities for this sort of line construction. This is something I want to encourage more at the VHLM level, as we work with new players to build for a specific playstyle. By offering build advice that separates our players into “classes” we can encourage build diversity in the VHL and work towards better strategic and build-focused roster construction.

I use forwards as an example, but this concept also works with D, having a higher SK D paired with a higher SC D.

 

The second way to construct lines, which is the more common way in the current VHL landscape is the dominant duo and competent filler strategy. This is what I mentioned earlier in the player construction section, and as much as I encourage the other line construction strategy moving forward, the “competent fillers” can easily be rebranded as the skater/checker on a line. But for now, having 4 high-TPE forwards and 2 high-TPE D is generally what this cap allows teams to afford. Some teams load their top unit and leave the other to the wind, but many teams have gotten a lot of success from splitting their talent evenly and filling the rest of their lineup with cheap prospect talent. I cannot really spill any theory for why this works, other than the difference in builds is not significant enough that “classes” can emerge, so players are simply branded as either “good” or “not good” and therefore a player who is good, is good at all the things and therefore fits all or any of the above roles.

 

Those are really the 2 schools of thought as far as I know for line construction, and both seem to have their success depending on the roster you have. As we know, most of what we do in the VHL is trying to break STHS in our favour instead of following its intentions, so theory and reality can be rather random, leading to the “Simon does what Simon does” phenomenon.

 

When it comes to roster construction, you have to commit to a set of values that your team is going to adhere to. Much like real NHL teams have “identities,” your team, regardless of the level, must have an identity. This often boils down to the players currently on your roster, but players come and go, and if you simply ride whatever players, you currently have, you will never build a pattern of long-term success. There are certain players that I pass over at the draft because they have not built in the style that I like, even at the VHLM level. Obviously ignoring players so new that they do not have the TPA to have gone in the wrong direction for me. Las Vegas under Shindigs has built an identity of dominant 2-Way forwards and rugged, tough defensemen. The Warsaw Predators under dlamb committed themselves to a high-flying offense, while sacrificing defense. This is how these teams were intentionally built, and although not everything was handed to them exactly how they wanted it, they committed to their identities and their line construction vision.

 

This is something I have really had to sit down and think about, watching my first attempt at a roster fail, then get traded off and now lost to the VHLM Draft and the VHLE. As much as I think all of my players have grown into talented players, and had a lot of success this season, I think I could have improved my advice to them over the course of my mentorship. I have perpetuated the meta, and although there is a need for meta-styled players in the league, not all of them will earn to the point where they will see the statistical success for which they might hope. My vision for the Marlins, and also for my player and the team I play on have changed. I hope to provide better tailored advice for my players and encourage players across the leagues to talk to their GM about their vision for the team. Every GM is going to have a different view than I will, so I will not spill too much about how I view building, to avoid misleading players according to their GMs plans.

 

There is room for a lot more build variation in this league, and STHS has the capacity to handle it effectively. We have abused this engine for so long, but I want this article to be a call for return to the ways that STHS has designed. Bring back line chemistry, interesting line/roster construction and strategic drafting selections. I bid adieu to the days of working in the BPA at the draft. To build a league free of the meta, we must commit to exploring a possibility that there are other things besides the meta. It starts in the VHLM, and with our top young talent in the VHL. Hockey is a bandwagon sport, so by demonstrating success, people emulate it. Let us show everyone that success can come from a variety of ways and philosophies. I am working to break my player from the mold of the meta that I fell into, and I hope to use my platform as a VHLM GM to build the next generation of builds from the new payers I am gifted with. It will take time to explore, and some commitment by everyone, but it is worth the risk for me.

 

Finally, I want to talk about player roles and really sit and be frank with everyone. Not everyone can be the star. And honestly, not everyone SHOULD be the star. We tried that, and then we broke STHS. But just because you are not the star (because if you are reading this it is more likely you are not the star than are), you have to accept that the best way to help your team is to embrace a role. What ties this article together is the fact that at a player building, line building and roster building level, we need to be ready to accept that role players are necessary and really the building blocks of every team. Get a few draft picks and you will get some star players. That is not really the question here anymore. Many teams have the stars they would need to compete, but they do not have the system around them to compete. Building a great prospect system full of a wide range of earners and build types is now necessary for long-term success. We need cheap talent to support holding stars. And honestly, the cheapest and most sustainable way to hold that window of winning open is through our lower earning players. I am not encouraging people to stop earning, but rather appreciate our welfare earners for the assets they are, enabling the top earners the security to stay with a team and a system that treats them well instead of being perpetual cap dumps in the offseason. Having a system of varied players, stylistically and earning-wise allows us to be able to replace talent as they retire and build a system of talent that follows our identity.

 

Look at real NHL teams like Tampa Bay. Whenever a player prices them out in contract negotiations, they always have someone in the minors eager to step into that role. They always have talent ready to shine, and always have a revolving door of talent ready to move out. Teams overpay for their cast-offs in free agency, and they laugh as they steal yet another project piece in the draft. This is how we have to look at our VHL franchises.

If build variety does not expand, then all we have to judge players on is TPA. Everyone who has ever talked to their GM builds pretty well. If TPA continues to be all that matters, then there is almost no reason to have attributes at all.

 

When it comes to prospects, there is also now an interesting dilemma about who to call up. Assuming people take my advice and lower earners build strong defensive players then it is almost a no-brainer to call up a lower earning forward of this mold and let the higher-earning talent develop and compete in the VHLE. This seems counter-intuitive, but the high-earning scorers only really provide that top-end value towards their peak, and at 300-500 TPE are probably worse off on a VHL roster than a 300TPE defensive forward. Much like the NHL, the young forwards who make a dent in the league are usually the high-energy puck retrievers, not the high-octane offensive guys. And generally, of those offensive guys, the ones that see the most success are on shallow rosters, where it would make sense in the VHL to call them up anyways.

 

As those role players grow into more expensive assets, they can be flipped to rebuilding teams looking to field a roster. There are more than enough roster spots in the league for those who are at the TPE for it. I order to continue to encourage players to reach that level, there has to continue to be opportunities to be useful pieces, and that boils down to how they build and how we approach roster construction. As someone who has been in that boat, although as a backup goaltender, my activity was directly tied to how much I was contributing to my team. If we treat our potential role players like the royalty they are, there will be more of them and better prepared for the VHL. How do we get more VHL-calibre talent? Allow for more than one route to being a successful player, and stop making our attribute system virtually worthless, by all building the same way.

 

I think that is probably enough to get some discussion going. Feel free to disagree, because well there is lots in here that could be wrong or be seen differently. But also, really try to understand what I mean, and do not take my frankness as an insult to your character. We are all out here trying to be better (or we should be at least). Please feel free to discuss below, I want to learn more, and I know some of you have some knowledge to dispense. This is well over 2000 words and I am sure I’ll probably only claim it once, but it was rewarding enough getting the thoughts out there.

 

Appreciate you all and best of luck!

Thanks to @Gustav, @Shindigs, @badcolethetitan, @Daniel Janser, @Spartan, @fromtheinside and @Ricer13 specifically for continuing to expand my perspective and teach me more and more about the different aspects of being a player, GM and member on the site! Most of these ideas stem from conversations we've had over the last couple years and have fundamentally shifted how I interact with the VHL.

2000+ Words, might claim at least once!

Edited August 24, 2022 by jacobcarson877

[badcolethetitan 921]

Great article and I can see what you’re saying in all of it. Honestly, it’ll be interesting to see the other people’s views on it. All in all, lots of great things to unpack here and it was a fantastic read. Good stuff Jacob. Now just don’t take over the M with how smart you are.

[leandrofg 1,159]

[Aimee 1,121]

I'll be bookmarking this for when I recreate in a year or so 

[Spartan 4,403]

I have never thought this much as a GM outside of figuring out how to make everything add up to 40 million (or 42 million now).

 

Jokes aside though, this is well written and insightful at looking at a lot of the recent trends around the league. I will say/admit that a lot of my moves are pretty instinctive and I don't ever go "I think this line composition will work." Maybe some people have a much more analytical approach, but I just like to bring in people who fit the culture and are down to have a good time. We figure out something to go on the ice later on lol. So just wanted to toss that out there in case any aspiring GM sees this and goes:

 

It's only as complicated as you want it to be. Anything works in the league! Except trading 1st rounders to Seattle for their aging veterans that they can't afford. Don't do that.

[jacobcarson877 2,536]

5 minutes ago, Spartan said:

I will say/admit that a lot of my moves are pretty instinctive and I don't ever go "I think this line composition will work." Maybe some people have a much more analytical approach, but I just like to bring in people who fit the culture and are down to have a good time

This is definitely how the league works today and I hope at the end of the day a lot of this sentiment still exists! Who cares about the numbers and all that if I'm not bringing in new homies to spend my downtime with. And to be honest, there has never really been a need for looking at line composition other than maybe Checking, but I think it would be kind of fun if the league went in the direction where player builds actually matter!

 

9 minutes ago, Spartan said:

Except trading 1st rounders to Seattle for their aging veterans that they can't afford. Don't do that.

Someday they'll learn. Someday.

[Spartan 4,403]

6 minutes ago, jacobcarson877 said:

And to be honest, there has never really been a need for looking at line composition other than maybe Checking, but I think it would be kind of fun if the league went in the direction where player builds actually matter!

I'd still say that I'm aware of builds and I have my own perceptions and biases of what works and what doesn't, I just don't sort of line it all up and plan for specific combinations and pairings. Rather what I think would work as a whole rather than at the micro level. It's not that the builds don't matter, I just think that I can make anything work as long as it's not blatant trolling lol.

[badcolethetitan 921]

One thing that I guess I didn't think about was the fact that when I draft, I do look at builds and I will sometimes skip on players that their builds don't fit my team. However, at the end of the day I for sure will just aim for the active players that are willing to interact with the team, then others who could be just as good, but they don't really care about interaction. I also would rather take someone who interacts more with the team and earns maybe slightly less, then someone who doesn't but does technically earn a bit more. However, I'm also in the VHLM and usually players are out in 1-2 seasons, VHL is probably way different.

[Ricer13 2,796]

37 minutes ago, Spartan said:

I have never thought this much as a GM outside of figuring out how to make everything add up to 40 million (or 42 million now).

 

Jokes aside though, this is well written and insightful at looking at a lot of the recent trends around the league. I will say/admit that a lot of my moves are pretty instinctive and I don't ever go "I think this line composition will work." Maybe some people have a much more analytical approach, but I just like to bring in people who fit the culture and are down to have a good time. We figure out something to go on the ice later on lol. So just wanted to toss that out there in case any aspiring GM sees this and goes:

 

It's only as complicated as you want it to be. Anything works in the league! Except trading 1st rounders to Seattle for their aging veterans that they can't afford. Don't do that.

All of this ^!!! Win or lose, it’s great to have a good group of people to go on this journey with. It’s why I enjoy this league so much. 

[ThePerfectNut 563]

Still not gonna apply any TPE to defence 

[Gustav 6,420]

Honestly, I'm not sure if ANYONE knows how lines work (beyond just "I'm going to look for players with builds that I personally like" as a general principle. Meta stuff doesn't count and I'm still salty about that). @Advantage was the all-time winningest GM in league history when I played for him and we'd run into things all the time that would either work or not work for no apparent reason, and no one really understood why. 

 

 

Another line-related story was my least fun season with Davos (which I think was S74? I don't care to check) where we had a great roster going into the season but got hit hard by Simon and ended up way out of a playoff spot around the deadline with me switching lines around quite a bit up to that point. I move a couple players around, nothing that should make a huge difference, and we go 2-0...so I don't touch it...and then I continue to not touch it as we put together some insane run where we go 15-2 or something like that and get right back into a playoff spot.

 

And then, with the same lines, and absolutely no changes, we get absolutely DEMOLISHED in just about every one of the handful of games we have left and end up missing the playoffs on a tiebreaker. Simon giveth and Simon taketh away.

 

In general, though, this is a cool read and there's a lot that a newer member should get to know.

[Advantage 2,891]

1 hour ago, Gustav said:

Honestly, I'm not sure if ANYONE knows how lines work (beyond just "I'm going to look for players with builds that I personally like" as a general principle. Meta stuff doesn't count and I'm still salty about that). @Advantage was the all-time winningest GM in league history when I played for him and we'd run into things all the time that would either work or not work for no apparent reason, and no one really understood why. 

 

 

Another line-related story was my least fun season with Davos (which I think was S74? I don't care to check) where we had a great roster going into the season but got hit hard by Simon and ended up way out of a playoff spot around the deadline with me switching lines around quite a bit up to that point. I move a couple players around, nothing that should make a huge difference, and we go 2-0...so I don't touch it...and then I continue to not touch it as we put together some insane run where we go 15-2 or something like that and get right back into a playoff spot.

 

And then, with the same lines, and absolutely no changes, we get absolutely DEMOLISHED in just about every one of the handful of games we have left and end up missing the playoffs on a tiebreaker. Simon giveth and Simon taketh away.

 

In general, though, this is a cool read and there's a lot that a newer member should get to know.

Truth.  Also hi.

[BOOM 8,748]

8 minutes ago, Advantage said:

Truth.  Also hi.

 

[Shindigs 1,771]

The interesting stuff about team identities and just what work is that you'll end up with funny stuff. A lot of which has changed a lot with the hybrid attributes to totally turn what works/doesn't work on its head.

 

Like you more or less nailed down properly, Vegas identity is based on Physical DFD and two-way forwards. I'd like to say power forwards, because honestly the thing we call a two-way forward in the VHL isn't a two-way forward. It's a power forward. The funny thing about this being our identity is that I personally dislike both physical DFD and Power forwards in all sim games I've every played and I go so far as to literally remove all of them from my draft board in EHM no matter what I know their Potential Ability to be. Because simply put Penalties lose games, and hits aren't good enough to make up for it.

 

So why am I doing literally the opposite of that with Vegas? Because in STHS the way you make hits "not worth it" is by having enough SK/ST/PH that your player absorbs >70% of all hits without actually losing the puck. Which VHL level centers with their fancy 99 ST from faceoffs all have around those numbers or more of absorbed hits. The hits you see in the statline of STHS aren't "successful hits" it's "hits that connected". So an absorbed hit, that basically does close to nothing, still counts. But it doesn't actually help your team gain an edge. While still giving a metric tonne of penalties. So I still don't like excessive checking at the VHL level.

 

But since at the VHLM level you have nearly all hits actually be "successful" due to the inability to push SK/ST/PH high enough, it means hits are simply put overpowered. A DFD without hitting gets about half as many net positive possessions per game as a hitting DFD at the VHLM level, and the buy in is 28 TPE for 28 BC. That's absurd TPE efficiency, and the reason why Juan Ceson, who has spent most of the last two seasons at around 120 TPE has consistently been one of the best DFD in the entire VHLM. Because when you have something that efficient, you don't even need 200 TPE.

 

We haven't really had *that* many Two-way forwards though. Not really, we randomly had 2 in my first season because two of our players basically put 15 TPE in every attribute, and that kinda gives you a Power Forward of sorts. This season we only had Larry Abass Jr @RileyL as an actual Power Center, because I talked to him and that was the type of player he wanted to build towards. So I helped him put together a build that would reach that goal. Which it did, as he's probably more or less guaranteed the best two-way forward award in the M this season. Despite playing the regular season with like 90-170ish TPE. Because it's just a stupid powerful archetype at the M level, even on a budget. And it builds into anything from a SDF, to a gritty scorer to an actual VHL Power Center. Depending on earn rate. So it's a very versatile base to work off.

 

That last part leads into what my philosophy is when helping VHLM players with builds. Which is to make something that fits into just about anything. That's the blessing in disguise of how "underdeveloped" VHLM players are in the hybrid era. You can't really get past 70 DF, and 70 DF fits into any VHL/E build. You can't really get past 70 SC, and 70 SC fits into any VHL/E build. You can get to like 80 PA, which would be silly and wouldn't fit into most VHL/E builds. But then I haven't seen anyone actually do that yet (thank god). So as VHLM GMs when we're helping with builds I think the following is an important thing to keep in mind:

 

Set your player up for success in the future by giving them a solid base on which they can build almost anything, based on what their first season(s) of experience will tell them that they actually want long term.

 

I've been playing around with what order players should actually build in when helping them. To try and figure out what order to actually do things in at the VHLM level so I an give more correct advice in the future. This was basically how that went:
S83: James Kagelberg comes in with 89 OV and 0s in all defensive attributes. Ends up the top goal scorer. (Hypothesis: Players can now skip DF until after they get 60+ OV and still produce offense)
S84: Reid Johnson D and Fred Hampton C both build OV first, then Defense. Fred Hampton ends season with 40+ Goals 99 points. Reid Johnson struggles both offensively and defensively until he finally gets his DF higher. (New Hypothesis: Forwards can skip DF until after their OV gets into the 60s, Offensive/Two-Way Dmen cannot)

S84: Grigoris Polyhronidis Jr. D builds a gritty DFD starting with nothing but PC/DC/GR and still actually mostly keeps up even scoring with Reid until the playoffs where Reid has the DF to support his Offense and takes over. (Current Hypothesis: Forwards can skip DF until after their OV gets into the 60s. Dmen who want offense still need to build up their DF to at least 50 before even touching OV)

 

The above is basically how I approach anything I'm trying to get good at. Create a hypothesis, test it, reject it, create an alternate hypothesis, repeat. Eventually you'll get close enough to functionally perfect understanding that you can almost guarantee the expected outcome.

 

And just as an aside, I don't really take build into account much then drafting. Sure if they have trolled their build to the point where it would be weak even at 200 TPE. That could be a red flag. But most builds can be salvaged. James Kagelberg fell to the 4th round of the S83 draft (presumably) because his build at the draft was literally 89 OV and nothing else. I realized that as long as he split his remaining TPE between DC and PC and played as winger instead of center. The raw amount of offensive stats and bare levels of DF should be enough to make him a very good player. So when I scouted him, I asked him if he was willing to play winger instead and build that way to be more successful. Which he was happy to do. That lead to him winning the Goal Scoring title in his first full season, as well as being an instrumental part of our back to back cups.

 

For those who have trolled their builds, I just talk them through why, explain how they could do it better. And make them aware that the free re-roll is a thing, should they want it. Which they usually do.

Edited August 25, 2022 by Shindigs

[fromtheinside 1,290]

The level of effort and detail in this article is fantastic. Really, really great job my dude! This will be useful for a ton of new users as they start up. 

[jacobcarson877 2,536]

claim 2

[jacobcarson877 2,536]

claim 3

[jacobcarson877 2,536]

claim 4