Hybrid Attributes: Ending the Meta

[Daniel Janser 2,169]

16 minutes ago, BarzalGoat said:

A few points to touch on here. First that yes, our attributes being so high were a problem. We’ve discussed with Simon before and he has said that the engine isn’t really built to have players with max stats, let alone multiple max stats and let alone multiple of those players on a team. If you look at ratings for NHL players in STHS they are far lower.

 

Okay, I was not aware of the skill level being a problem, thanks for the clarification.

 

17 minutes ago, Beketov said:

The second point to keep in mind is that everyone is adjusted so it may look bad on paper to see only your stats but it’s important to remember that everyone is being hit the same way so It’s all relative. Comparing what you had to what you’ll have doesn’t make sense because no one will have what you had anymore. The system changed and everyone changed with it.

 

 

Well, except netminders...

So I guess with that in mind (i.e. Skaters massively nerfed, while tendies stay where they were) we will see a lot of soccer match results in the next couple of season.... which is not bad per se, just noteworthy...
 

Quote

 

The third important thing to note is that trying to build exactly the same way may very well be very TPE inefficient now because of how things are spread out. Playing with it to find what works instead of just trying to get exactly the same can do a lot to optimizing things.

 

 

Nothing is exactly TPE efficient with the new distribution, but it is difficult to convince myself to invest hard earned TPE in points which do not further the build I had in mind for my player (even less so when the cost for said attributes are at a 3:1 conversion rate AND in addition do nothing or close to nothing for the team or player (things like leadership, fighting, penalty shots and discipline)).

[Beketov 9,025]

13 minutes ago, Daniel Janser said:

Well, except netminders...

So I guess with that in mind (i.e. Skaters massively nerfed, while tendies stay where they were) we will see a lot of soccer match results in the next couple of season.... which is not bad per se, just noteworthy...

We tested lower scoring (the main effect on this) with current goalies and found the increase for them to be marginal, not even really up to usual levels compared to the insane lows goalies have been at for 20ish seasons. However with everyone adjusted evenly we can make “across the board” changes if we just need to up scoring for example by just increasing the scoring slider. The sliders would t work for fixing the meta because it’s just trying to fix one specific thing but for just “we need more goals in the league” they work fine. Additionally if we find that’s not enough or goalies need a re-work we can apply hybrid attributes to them as well.

 

15 minutes ago, Daniel Janser said:

in addition do nothing or close to nothing for the team or player

Do they though? The league has spent  so long focussing on the same 2-3 builds and ignoring everything that doesn’t fit the mold to the potential detriment of other attributes that have gotten ignored.

 

Ultimately aside from PURE meta there’s no build that’s been completely destroyed. They will take more TPE potentially but they aren’t ruined. Everything is still available to choose from.

[Steve 1,924]

A step in any direction is better than standing still. I am happy this is being implemented. Good job 

[Frank 5,210]

Does this come with a @Nykonax ban? If it looks like Van is going to three peat we should do em like the MLB did the expos.

[RomanesEuntDomus 442]

This looks nice, a very interesting and creative change approach to this problem. I could see this working out and am looking forward to seeing it in practice

[BOOM 8,746]

Fuck me. Whoever drafts me, please ensure that you have a mensa member/MIT graduate on board for build help. Thanks in advance. 

[Mr_Hatter 1,608]

Honestly, brilliant idea. I feel like it adds a LOT of flavor to player building, and gives us some easy knobs to tweak to spice up the meta should exploits be found/things get stale. Obviously things could not work out in theory, but it seems a great step that adds to the league without purely taking things away.

 

Great work all involved!

[Beketov 9,025]

2 minutes ago, BOOM said:

Fuck me. Whoever drafts me, please ensure that you have a mensa member/MIT graduate on board for build help. Thanks in advance. 

You’d be surprised. I expected it to be a lot messier to deal with but when I actually played with the spreadsheets I realized that as long as you have the details laid out in front of you as to what’s going what and you can see the numbers on real time (which you will) then it’s pretty simple.

 

Yes, it’s more complicated than the 1:1 we are used to but you don’t need to worry about the back end math much at all.

[BOOM 8,746]

I appreciate that and realise that change was needed. Kudos for the work behind the scenes. I just think that I'm going to struggle with this for some reason. 

[Daniel Janser 2,169]

1 hour ago, Beketov said:

Do they though? The league has spent  so long focussing on the same 2-3 builds and ignoring everything that doesn’t fit the mold to the potential detriment of other attributes that have gotten ignored.

 

 

Well I based my statement on the information available to me (mainly Gustav's build guide).

 

Okay even if penalty shots skill at 99 would guarantee that you score 100% of the times you take one... how often is this the case? first there must be a tie, then a goal-less extratime and then you need to be a dedicated shooter.... and I have yet to see a penalty shot awarded during regulation time (I am not saying they do not eixst, just saying in the 200+ games my player participated in so far, I have yet to see a penalty shot awarded). Whereas defence, Scoring, skating, puckhandling, passing etc. help you in every situation, every minute you play... I mean of course one can strive to be the next Pavel Datsyuk but I highly doubt it is worth the costs... 

 

AFAIK Fighting just elevates your chances to assemble a nice collection of 5minutes major (which I have plenty already, without putting any points into it as of yet), which I am sure every GM is excited to see when his player spends more time in the sinbin than on the ice.

 

Discipline is apparently a mitigating factor for taking penalties so not entirely useless (especially when you have high checking/fighting values), but again how much mitigation per point and is it really worth it?

 

Leadership is said to do jack all or at best has very minuscule effect on your and your team's performance (maybe a team with everyone at 99 LS could be interesting as an experiment, but again I doubt it is worth the time and effort to raise it to a meaningful level (there must be a reason why this attribute has the highest conversion rate on the new spreadsheet)...

 

Quote

Everything is still available to choose from.

 

That never was the problem though, was it?

 

Anyway, enough whining from my side, thanks to everyone involved in addressing the meta and lets see how things pan out...

 

[thadthrasher 1,692]

I've read through these comments and I'm excited to see how this changes things. I do have a question, though it's more to do with the history of the league and other things. 

1. How do we perceive this affecting things such as HoF considerations for players who must adjust halfway through their career?
2. Will already established records or future ones be asterisked? I'd imagine that these adjustments will change numbers produced?

[Enorama 2,038]

15 minutes ago, thadthrasher said:

I've read through these comments and I'm excited to see how this changes things. I do have a question, though it's more to do with the history of the league and other things. 

1. How do we perceive this affecting things such as HoF considerations for players who must adjust halfway through their career?
2. Will already established records or future ones be asterisked? I'd imagine that these adjustments will change numbers produced?

 

1. The same way other major changes like the update scale change or the weekly cap change. It changes for everyone all at once so when comparing people against their era it's the same as it's always been. 

 

2. Kinda the same as above. We've had major changes before that have effected scoring and nothing has been asterisked. Additionally, this change is just expected to return scoring to a "normal" level that we had for the first 60-70 seasons in the league. The last 10-20 have occasionally been wildly inflated compared to overall history. 

[thadthrasher 1,692]

8 minutes ago, Enorama said:

 

1. The same way other major changes like the update scale change or the weekly cap change. It changes for everyone all at once so when comparing people against their era it's the same as it's always been. 

 

2. Kinda the same as above. We've had major changes before that have effected scoring and nothing has been asterisked. Additionally, this change is just expected to return scoring to a "normal" level that we had for the first 60-70 seasons in the league. The last 10-20 have occasionally been wildly inflated compared to overall history. 

Gotcha, thanks!

[Ricer13 2,796]

New Meta:

 

Deking 

Stick Handling

Stick Checking

Defensive Coverage

Slapshot

Wrist shot

[Beketov 9,025]

3 minutes ago, Ricer13 said:

New Meta:

 

Deking 

Stick Handling

Stick Checking

Defensive Coverage

Slapshot

Wrist shot

Like max on all those? Because if so that’s over 1200 TPA so that’s gonna be a challenge to sustain; also you’ll have sub-70 scoring.

[Hybrid1486 577]

I got excited because I thought this was about me

[Phil 5,119]

This is awesome 

[Ricer13 2,796]

1 hour ago, Beketov said:

Like max on all those? Because if so that’s over 1200 TPA so that’s gonna be a challenge to sustain; also you’ll have sub-70 scoring.

Don’t get me wrong. I am very excited for a change and am happy you guys have implemented this. I am just very skeptical and it’ll take proof of concept for me to really believe in this change. 

[Ben 200]

10 hours ago, Beketov said:

Hybrid Attributes: Ending the Meta

 

Hello @Members!

 

As I’m sure you’re all aware by now, the VHL has a meta problem. If you aren’t aware…. where have you been? I’ll explain though: at its core, the “meta” in the VHL refers to building a player with max Scoring (as well as ideally max Defense) and absolutely no Passing, or as little Passing as you can possibly manage. This tricks the engine into taking an insane amount of shots rather than passing and more shots means more goals. An individual player doing this has minimal impact on the engine as a whole but an entire team doing it is not only salary efficient (since not much TPE is needed), but can also be very beneficial to that team as they outshoot and outscore their opponents rather consistently. As more and more users have started to discover this, more and more teams have come to adopt this “meta” strategy, intentionally or otherwise.

 

The SHL discovered this a few years ago and completely changed engines to eliminate it. Before I get into our fix I want to explain something to eliminate a misconception: we did know about this and we weren’t just ignoring it in hopes that it wouldn’t be a problem. At the time it was discovered, we ran tests and found the impact to be lesser with our engine version and roster composition. It was present but not extreme - combined with re-rolls being much harder to acquire here, we thought it best not to rock the boat believing changes would be more detrimental to league health than the problem which, at that point, had not really come up in the VHL.

 

Unfortunately times changed and as we know, the meta has been on our doorstep for a bit now. That is why the @BOG has been working very hard the last few seasons to come up with a solution that would not only eliminate the meta without changing engines (something that would be nuclear to the portal and unwanted if it can be avoided) but also improve build diversity, roster makeup, and hopefully league enjoyment.

 

Therefore, I present your new VHL Player Attributes:

 

  Reveal hidden contents

Deking

Stickhandling

Passing

Body Checking

Grit

Faceoffs

Stick Checking

Defensive Coverage

Offensive Vision

Speed

Slap Shot

Wrist Shot

Leadership

Poise

 

"But Beketov..." I hear you asking, "if we aren't changing engines and we can't change the attributes inside of STHS how do we have new attributes?"

 

Good question theoretical member! The short answer is: we don't. The above are the attributes you as members will be seeing but they are not the attributes STHS will be seeing. We are calling these hybrid attributes and their job is to give us more control over builds without directly limiting what can and can't be added to. They work by spreading out the TPE placed into them into different STHS attributes at different ratios. Specifically the ratios that are used are:

 

  Reveal hidden contents

Attribute	STHS Att. 1	Ratio	STHS Att. 2	Ratio
Deking	PH	0.24	DF	0.12
Stickhandling	PH	0.36	SK	0.12
Passing	PA	0.36		0
Body checking	DF	0.06	CK	0.36
Grit	CK	0.24	FG	0.3
Faceoffs	FO	0.54	ST	0.24
Stick Checking	DF	0.24	ST	0.18
Defensive Coverage	DF	0.18	SK	0.18
Offensive Vision	PA	0.3	SC	0.3
Speed	SK	0.3		0
Slap Shot	SC	0.12	ST	0.3
Wrist Shot	SC	0.12	PS	0.48
Leadership	LD	0.72		0
Poise	LD	0.3	DI	0.54

 

I know that looks like a lot of numbers and attributes but I promise it's not as complicated as it seems at first glance.

 

Essentially 1 point into a Hybrid attribute = a ratio of 1 or more STHS attributes. For example adding 1 point to Offensive Vision will add 0.3 points to PA and 0.3 points to SC (this sounds quite low but I promise it works out). To make it so that this doesn't completely break how much TPE is needed for a build all values will be rounded down (since STHS cannot accept decimals) but the points aren't lost, they simply aren't used until you hit a new whole number. I also want to assure you that these values were not randomly determined - they were the byproduct of many test sims to ensure they would mitigate all of the issues we were aiming to address.

 

These values can be more easily tweaked over time if a new meta develops and give us granular control over the STHS attributes in a way that simply having an update scale never has and never could. In the specific case of the meta, the eagle eyed among you have probably already noticed that there’s no way to achieve max scoring without adding to passing. In fact if you max all the scoring attributes you will end up around 64 passing by association. This was the crux of the meta fix, build attributes together in a way that making a purely “meta” player is impossible.

 

The additional bonus of course is build diversity, or at least that's the hope. I'm hitting this explanation over the head a bit and I promise I'll stop soon but I wanted to touch on this very quickly because I can see the comments now from that theoretical member again, "couldn't we have achieved the same thing by adding archetypes or simply a required gap between passing and scoring?" Yes, theoretical member, we could have. However we chose to take longer to make this than go one of those routes because the VHL's attributes have always been based around adding what you want, where you want. We do not have limits on specific attributes or requirements like gap; it would be easier but that doesn't inherently make it better. @Beaviss originally came up with an early version of this a few years back (so credit to him) and I shot it down because I didn't think it made sense at the time. However when looking at all of this again it seemed to fit perfectly. Fix the meta and add a whole new world of build theorycrafting. We don't have any new actual attributes but having different ways to get things up and having to balance certain trade offs inherently creates more opportunity to diversify. You can try this out for yourselves since @Spartan ohas graciously simplified our testing spreadsheet here (originally designed by @Enorama, credit where due because I certainly couldn't do all that math) which you can use to start designing your build right now before this all comes into play.

 

--- 

 

Which leads me out of the meat of this announcement and into the nitty gritty details: Implementation and what it all means.

 

First off, this is going into effect for the start of S83 which will likely mean extending the off-season slightly because we need to accommodate the fact that every skater in the league will be re-rolled down to TPA + banked after depreciation is run, which will happen early. This is done so that you can re-apply your TPE into the new attributes as you please. You can choose to change your build if you'd like or you can try and build exactly the same, the choice is ultimately yours. Inactive players will be re-rolled by their GM's and any players not re-rolled by the start of the season will be retired.

 

We recognize this is a large change though and builds may not be perfectly finalized even with the extra tools and preparation time. For that reason we will be offering a 3 season window (ending at the start of S86) in which you will be able to re-roll your player again for free. Everyone will be able to do this once regardless of if they have previously used the paid re-roll or not.

 

We do also recognize that this is a WC off-season as well as the other usual off season tournaments and those will not be affected. We will download a ratings file for them the last day before everyone is reset and run the tournaments off of that file. You may have noticed by now that none of those attributes above and nothing I have said thus far relates to goalies; that is because goalies shall remain untouched for now. They will not be re-rolled or affected in any way at this time. If changes to their attributes need to be made in the future we shall do so but for now we want to gather data without adjusting them.

 

The update scale and depreciation will also remain unchanged, again at least for now. From our testing depreciation should be roughly the same difficulty to fight as before for the higher TPE players and should be slightly easier for the lower TPE players. Additionally the update scale does not appear to require changing and will operate in exactly the same way as before with 1 caveat: attributes will start at 0 and go to 99 instead of the 40-99 that the VHL has ran with for 82 seasons. This only applies to the hybrid attributes with the ones inside STHS still having a baseline of 40. The update scale for 0-70 will remain at 1 TPE per point. The process of updating will also remain unchanged however it will look slightly different. We will be upgrading the update page to showcase not only which points go into your hybrid attributes but also the effect of your STHS attributes. It is worth noting, however, that there will be disheartening moments in this system on occasion. Because of how the ratios work and the rounding down there will be times where you add a point and see nothing change. Again I would like to clarify that every point added does get counted, it may just not be enough to push you over the rounding threshold. The TPE tool shall also be adjusted accordingly to help people figure out their builds and see exactly what adding to everything does. It will work the same as the above link but be a lot less manual.

 

I recognize this is a long explanation, a lot of words, and a lot to take in. It is not a decision we made lightly but it is a decision we are excited about. It should hopefully fix a lot of player attribute problems we have been facing in one move and allow the league to grow without destroying everything we have built with STHS. This will allow our attributes to be closer to the expectation of the engine without suppressing the creativity to do as you please, it will put an end to the meta build, it will make for an increase in build diversity, and it will allow the VHL to continue to thrive. Please know that we will be keeping an eye on things over the next few seasons as, regardless of the amount of testing we did, ultimately the biggest test is getting it in all of your hands. If we need to adjust sliders or change goalies or anything else we will do so. We ask that you all embrace this change and help to make the league better.

 

If you have any further questions, points of clarification, or concerns, please reach out to a Commissioner or a BoG member. We'd be happy to explain details to prevent incorrect information from circulating.

 

- @Commissioner

 

[diamond_ace 3,112]

7 hours ago, BOOM said:

Fuck me. Whoever drafts me, please ensure that you have a mensa member/MIT graduate on board for build help. Thanks in advance. 

I was in a group called SET (Study of Exceptional Talent) based out of Johns Hopkins that was largely considered a middle school/high school precursor to Mensa at the time. Obviously did not put in nearly the effort to maintain that sort of thing as I got older and so I'm pretty sure I was never officially a Mensa member (I did get a lot of their mailings at the time) but pretty close. Tbh now it seems somewhat trivial even though at the time it seemed like such a huge thing

[Victor 10,942]

I have to admit I tapped out of the BOG conversation months ago since I was taking a break anyway (and there were A LOT of words to read, even compared to Bek's post) but I'm glad this is where it landed and seems pretty digestible.

 

Anyway let me know in 10 seasons what the new meta is for my recreate.

[eaglesfan036 4,603]

You mentioned maxing shooting attributes results in 64 passing but I dont see that in the table you posted? Wrist shot + slap shot don’t add to passing in the table you posted

[Beketov 9,025]

2 minutes ago, Squidward said:

You mentioned maxing shooting attributes results in 64 passing but I dont see that in the table you posted? Wrist shot + slap shot don’t add to passing in the table you posted

If you only max wrist shot and slap shot you won’t even be at 70 scoring. To get the rest of the way you’d need to add to offensive vision which adds to passing.

[Thunfish 72]

That's certainly creative, I'll say. Hope it works out for y'all.

[Adrest245 129]

On 3/5/2022 at 1:44 PM, samx said:

I don't wanna read all that