Best criteria for Matchmaking

Share your ideas and Suggestions about Blood Bowl 2.
User avatar
VoodooMike
Posts: 1614
Joined: 14 July 2009, 07:44
Location: Calgary, Alberta
Contact:

Re: Best criteria for Matchmaking

Postby VoodooMike » 18 September 2013, 21:52

So if a coach is good their team might stabilise at say +5. At n = 90 that is -450tv. They will get to -45O eventually whatever n is it is just a matter of time.
Sort of. It more or less means that the combination of the coach's skill (both in terms of on-the-field play, and off-the-field team development decisions) and the roster's inherent strength benefit (ie, Chaos Dwarf is inherently better than, say, Goblins) equates to around 450 TV of value added to the team. Remember that anyone they face will have the same additional value (or.. subtracted if they happen to be a bad coach playing ogres, say) from their team's general TV value.
Ok, that is not 'why'. But zSum is much more intuitive and easy to do method than TV++. I think that the benefits of this TV++ MM & inducements system will not match the suspicion it will receive.
Though zSum will be strong and fair enough compared to the TV 'alone' MM system we have right now.
You probably don't understand the physics behind air travel or even the engineering behind modern automotive technology, but I'm guessing you don't walk everywhere. Sometimes people just need to say "I'm not willing to learn how this thing works, but I trust that it does because people I know understand it say so" or they need to take the time to understand it. There is never a justification for saying "I don't understand it, and am not willing to learn, so it simply shouldn't be".

In this case, zSum is weaker than TVPlus and TVPlus is pretty damned intuitive - its a very simple calculation! You also need to use a self-limiting rating for matching, and in your suggestion that isn't happening. zSum is wins - losses... but someone's zSum doesn't get used for inducements.. The best and the worst coaches will simply "fall off the edges" of your system, as they'll just keep winning or losing until their zSum is totally meaningless and so far from anyone that they'll never be matched with anyone. In a TVPlus system that uses TVPlus rating for inducements, the rating is self-limiting... each time your rating adjusts up or down, the difficulty of continuing to progress in that direction is increased. If you lose match after match, it gets harder and harder to lose (unless, y'know, you're trying to lose I guess).... if you win match after match it gets harder to win.
Friendly Reminder: Correlation does not equal Causation - tattoo it on the inside of your eyelids if it'll help.

erestor
Posts: 116
Joined: 27 January 2009, 20:42
Location: Debout sur le terrain à gueuler sur l'arbitre, et le sort qui s'acharne...

Re: Best criteria for Matchmaking

Postby erestor » 19 September 2013, 08:00

There is never a justification for saying "I don't understand it, and am not willing to learn, so it simply shouldn't be".

Which is not at all what I have written.
In this case, zSum is weaker than TVPlus and TVPlus is pretty damned intuitive - its a very simple calculation! You also need to use a self-limiting rating for matching, and in your suggestion that isn't happening. zSum is wins - losses... but someone's zSum doesn't get used for inducements.. The best and the worst coaches will simply "fall off the edges" of your system, as they'll just keep winning or losing until their zSum is totally meaningless and so far from anyone that they'll never be matched with anyone. In a TVPlus system that uses TVPlus rating for inducements, the rating is self-limiting... each time your rating adjusts up or down, the difficulty of continuing to progress in that direction is increased. If you lose match after match, it gets harder and harder to lose (unless, y'know, you're trying to lose I guess).... if you win match after match it gets harder to win.
One more time, I agree with that.

BUT... but TV++ will definitively separate the game from the board game cause TV++ involves too much calculations for most of the players. And many board game players want the CRB nothing but the CRB so a method wich is good for MM without changing the inducements rule will be more acceptable than the (better) TV++ method.

(Are you the guy who wrote this : "So, zSum is still the champion method for deciding matches in a matchmaking environment. Too bad nobody uses it (yet)!" :mrgreen: ) (only a joke)

EDIT : So, it's not "I don't understand it, and am not willing to learn, so it simply shouldn't be". No. The correct recap would have been : "i understand, VM is aweseome, TV++ is awesomer, BUT in practise, i don't want to use it because it's too much calculation OR too far from the board game rules".
As the long departed Commissioner Three-horn famously said on many occasions “This is Blood Bowl, buddy, anything could happen!”.

User avatar
dode74
Posts: 7041
Joined: 11 December 2008, 11:18
Location: Nr. Reading, UK
Contact:

Re: Best criteria for Matchmaking

Postby dode74 » 19 September 2013, 08:26

I get your point, erestor, but ranking calculation in MM is also very different to anything like I have seen in the board game. Matching by TV is also different, as is the sheer number of possible opponents (teams and coaches) available relatively easily. There are far bigger differences between the board game and the PC game than this.
The aims of the game are different: with TT it's a far more social and far less strictly competitive game compared with online MM where you might never meet your opponent again so people tend to concentrate more on the game itself. Such differences need to be acknowledged and, if possible, worked with.

And it's really not that complex to implement for TT. Add the following paragraph to your post-match sequence:
"If you won the match add 30 to your TV; if you lost it then subtract 30; if you drew make no further adjustments".
I think it would be very simple to implement it should people want to, but am not suggesting that it should necessarily be implemented in the TT rules. No reason why it shouldn't be there in the "house rules" section, though, were any change to the LRB possible (HA!).
Image

erestor
Posts: 116
Joined: 27 January 2009, 20:42
Location: Debout sur le terrain à gueuler sur l'arbitre, et le sort qui s'acharne...

Re: Best criteria for Matchmaking

Postby erestor » 19 September 2013, 08:40

I get your point, erestor, but ranking calculation in MM is also very different to anything like I have seen in the board game. Matching by TV is also different, as is the sheer number of possible opponents (teams and coaches) available relatively easily. There are far bigger differences between the board game and the PC game than this.
The aims of the game are different: with TT it's a far more social and far less strictly competitive game compared with online MM where you might never meet your opponent again so people tend to concentrate more on the game itself. Such differences need to be acknowledged and, if possible, worked with.

And it's really not that complex to implement for TT. Add the following paragraph to your post-match sequence:
"If you won the match add 30 to your TV; if you lost it then subtract 30; if you drew make no further adjustments".
I think it would be very simple to implement it should people want to, but am not suggesting that it should necessarily be implemented in the TT rules. No reason why it shouldn't be there in the "house rules" section, though, were any change to the LRB possible (HA!).
You're right. So is VM.

And I would be very satisfied with the TV++ system (much more than with the present one). Especially as I dont play anymore the TT since BB has been released (yes I'm ashamed... but I play now a lot more ^^). So I was more arguing about a general feeling that some players could have than my own one.

What I wanted to underline was the fact that something hidden like MM would not disturb TT players though something noticeable like TV++ could.
As the long departed Commissioner Three-horn famously said on many occasions “This is Blood Bowl, buddy, anything could happen!”.

User avatar
VoodooMike
Posts: 1614
Joined: 14 July 2009, 07:44
Location: Calgary, Alberta
Contact:

Re: Best criteria for Matchmaking

Postby VoodooMike » 19 September 2013, 10:00

BUT... but TV++ will definitively separate the game from the board game cause TV++ involves too much calculations for most of the players.
The MM environment already separates the PC game from the board game in the same way things like wikipedia are quite separate from old physical encyclopedias - its 1000x bigger than the physical environment, and that level of zoom shows off the wrinkles and blemishes that nobody was seeing when it wasn't so damned big. The result, for BB, is the exact opposite of what many BB enthusiasts have claimed: the environment doesn't fail the game, the game fails the environment because it breaks under the lode.

The MM environment itself can be considered a "house rule" league system, but people have been terrified of applying changes to accommodate the much different environment while preserving the "spirit" of the game. The spirit of the game will be fine in the long run - it isn't somehow violated by houseruling... a good chunk of the tournaments involve much heavier houseruling than any you've yet seen applied to any online version, from different team creation/development rules, to rez setups where injuries and death simply don't apply, etc.

The calculation for TVPlus is so trivial that it is shocking... you could even use it in TT by using a crayon on a napkin if you were so inclined. That said, there's no reason to do so - TT leagues are just a social affair, for the most part, and as stated, its not really tournament-friendly, so..
And many board game players want the CRB nothing but the CRB so a method wich is good for MM without changing the inducements rule will be more acceptable than the (better) TV++ method.
If you don't use it for inducements then don't use it at all - it will lack the self-limiting aspect, and that will lead to the rating getting increasingly inaccurate and worthless over time, rather than more, which TVPlus used for both does.
(Are you the guy who wrote this : "So, zSum is still the champion method for deciding matches in a matchmaking environment. Too bad nobody uses it (yet)!" :mrgreen: ) (only a joke)
Probably.... that's the thing with champions, they need to beat all challengers or they lose their place. zSum lost out to TVPlus.
What I wanted to underline was the fact that something hidden like MM would not disturb TT players though something noticeable like TV++ could.
Know what really seems to disturb TT players? The results of a MM environment that is based on TV... facing the same type of teams over and over, being unable to advance and maintain agility teams at high TVs, etc. No matter which way you go, people will be bothered by something... I'd rather they be bothered by having to deal with some change, than they have legitimate, verifiable grievances with the results of the old system.

MM is a huge thing for BB... possibly the biggest thing for it since the game began. It may not feel conceptually huge, but it's what lets people find a game any time they want, and face a new opponent every single time they play. MM environments are to BB what internet porn has been to masturbation.
Friendly Reminder: Correlation does not equal Causation - tattoo it on the inside of your eyelids if it'll help.

Antonlunau
Posts: 28
Joined: 24 August 2011, 15:32

Re: Best criteria for Matchmaking

Postby Antonlunau » 19 September 2013, 10:25

I know this has been a long debate but i want to make sure i understand it:

*The current inducement system, in most cases, favour the team with the higher TV

*Teams with a good record (because of coach skill or optimized teambuild aka min/maxing) will be matched against teams with a lower win% but a higher TV or against teams with a similar coach skill/build

This should result in more even matches, am I getting it right?

If that is the case then i have one question:

The premise is that the inducements will not be enough to balance the teams, but that excludes coaching skill since the stats does not take into account if the inducements are given to the team with the better coach/elo rating. With this system, the inducements will most likely be given to the coach with superior skill since there is a connection between skill and win%. Will the system work If a good coach is able to use the inducement better than an average coach?

User avatar
Koadah
Posts: 1211
Joined: 08 April 2009, 16:17
Contact:

Re: Best criteria for Matchmaking

Postby Koadah » 19 September 2013, 10:59

The way I see it the inducements are awarded according to TV+ not TV. The better coach will probably not get the inducements unless he has just had a bad run.

Which could be nasty though.
CaRBB

Antonlunau
Posts: 28
Joined: 24 August 2011, 15:32

Re: Best criteria for Matchmaking

Postby Antonlunau » 19 September 2013, 11:41

The way I see it the inducements are awarded according to TV+ not TV. The better coach will probably not get the inducements unless he has just had a bad run.

Which could be nasty though.
Really, is that how it works?! So a great coach with an 1100 TV team will face a 1800 team with a lousy coach? That cant be right

User avatar
dode74
Posts: 7041
Joined: 11 December 2008, 11:18
Location: Nr. Reading, UK
Contact:

Re: Best criteria for Matchmaking

Postby dode74 » 19 September 2013, 12:30

Why not? If you're that good (by the rating) then your team will have a high zSum and will likely have played a good number of games.
With x=30 (i.e. each point of zSum is worth 30TV) for this, your example would mean the better coach has a 700/30 = approx 24-win advantage over the worse, high TV coach in order for there to be no inducements, be that 12-0-0 vs 12-12-24 or any other combination. That goal is to make all the games even, and for everyone (even the poor coach) to have a good chance of winning. It's nothing more nor less than matching by performance and applying a handicap.

Yes, it could be nasty, and you could lose the game to a worse coach. That's the idea. Your zSum score will effectively be your "performance" figure, and as you reach the point where the handicap the opponent gets means that you start having a win% of about 50 then you will be winning as many as you lose, and that zSum figure will stabilise. Each game will start out as predicted to be "close" when you achieve your actual zSum (although it will, itself, be a moving number).
Image

User avatar
Koadah
Posts: 1211
Joined: 08 April 2009, 16:17
Contact:

Re: Best criteria for Matchmaking

Postby Koadah » 19 September 2013, 12:51

It wouldn't be as bad as Antonlunau thinks but I would still prefer that a match between teams with similar TV & similar zSum get a higher weighting than a match where largish TV + zSum differences cancel out to give similar TV+.

Or is that trog thinkg? ;)

Are you int the 30 is better than 90 camp Dode?
CaRBB


Return to “Ideas and Suggestions”

Who is online

Users browsing this forum: Trmbdunalt and 1 guest

cron