Jump to content

Welcome to Card Game DB
Register now to gain access to all of our features. Once registered and logged in, you will be able to create topics, post replies to existing threads, give reputation to your fellow members, get your own private messenger, post status updates, manage your profile and so much more. If you already have an account, login here - otherwise create an account for free today!
Photo

LS / DS Actual Statistics From Store Champs/Regions

* * * * * 1 votes

  • Please log in to reply
127 replies to this topic

#1
tierdal

tierdal

    Advanced Member

  • Members
  • PipPipPip
  • 2344 posts

So people keep throwing out roughly equal numbers for LS and DS wins - but it occurred to me that this is a very poor way to look at balance.

 

If the majority of your DS wins come from sweeps VS splits - you are going to get pretty similar results between both sides with player skill coming into effect.

 

Just looking at our own NYC region we had 3 DS wins that were not sweeps.  (IE DS splits)

 

Thats roughly 10% win rate vs LS for non-sweep DS wins. That does not bode well for balance imho.

 

Does anyone have their raw results from regionals/store champs they would be willing to share? Very curious how things netted out across the board



#2
BobaFett

BobaFett

    Advanced Member

  • Members
  • PipPipPip
  • 3037 posts
Tierdal, you can't twist numbers around to get the outcome you're looking for.

The meta is not unbalanced to the tune of 90/10 like you're now purporting. It's much closer to 60/40 just as the numbers are telling us.
  • KennedyHawk, MonteKev and CharlieFilms like this

#3
tierdal

tierdal

    Advanced Member

  • Members
  • PipPipPip
  • 2344 posts

Tierdal, you can't twist numbers around to you get the outcome you're looking for.

The meta is not unbalanced to the tune of 90/10 like you're now purporting. It's much closer to 60/40 just as the numbers are telling us.

 

Well - not if you use proper math. A sweep and a draw are not equivalent and thus can't be weighted the same. That's just basic statistics. But feel free to believe what you wish.



#4
KennedyHawk

KennedyHawk

    Advanced Member

  • Members
  • PipPipPip
  • 1890 posts

Just keep this discussion in one of the other 70 threads crying about imbalance. It wasn't nearly as bad as the vocal minority are telling us at least that was my experience.



#5
tierdal

tierdal

    Advanced Member

  • Members
  • PipPipPip
  • 2344 posts

Just keep this discussion in one of the other 70 threads crying about imbalance. It wasn't nearly as bad as the vocal minority are telling us at least that was my experience.

 

this wasn't really about crying about imbalance...just wanted other TOs to share their data as I was curious



#6
KennedyHawk

KennedyHawk

    Advanced Member

  • Members
  • PipPipPip
  • 1890 posts

this wasn't really about crying about imbalance...just wanted other TOs to share their data as I was curious

 

I was going off of your original post. Which talks about balance more than once and insinuates that it is not boding well.

 

So people keep throwing out roughly equal numbers for LS and DS wins - but it occurred to me that this is a very poor way to look at balance.

 

...

...

...

 

Thats roughly 10% win rate vs LS for non-sweep DS wins. That does not bode well for balance imho.

 

Does anyone have their raw results from regionals/store champs they would be willing to share? Very curious how things netted out across the board



#7
dbmeboy

dbmeboy

    Advanced Member

  • Contributor
  • PipPipPip
  • 3910 posts

Just looking at our own NYC region we had 3 DS wins that were not sweeps.  (IE DS splits)

 

Thats roughly 10% win rate vs LS for non-sweep DS wins. That does not bode well for balance imho.

This implies that you had 27 LS wins that were not sweeps (ie LS splits).  I don't have the data to know if that's true or not, but seems unlikely from what I've seen from other regionals.  Especially given 12 players x4 rounds means 6 matches per round x 4 rounds = 24 total matches.  Except there were 2 byes used, so only 23 total matches.  To get to 3 DS wins you have to have 1 DS split and 1 DS win and then time, leaving 21 matches.  Of those, we'd need 13 LS splits and a LS win and then time, leaving only 7 matches to be sweeps.  And that's assuming that you didn't mean there were 3 DS splits, which would literally not leave enough matches for the 27 LS splits required to get to a 90/10 split you're claiming.



#8
Scottie

Scottie

    Advanced Member

  • Members
  • PipPipPip
  • 1968 posts
Why don't you just post the raw stats if you want to analyze them. I think you are going to find a few posters not willing to accept your analysis because in the past you've shown a habit of misrepresenting event results in order to further your agenda.
  • fliptheforce and Jackofhearts like this

#9
tierdal

tierdal

    Advanced Member

  • Members
  • PipPipPip
  • 2344 posts

Roughly half our games were sweeps. So 12 sweeps/ 12 splits. There were 3 DS Splits/ 9 LS Splits.

 

That is a 25% Win Rate for DS IN splits and a 75% Win Rate for LS IN splits

 

If you assume that the majority of SWEEPS are based on skill (IE the person won both games) ... it does not paint a pretty story.

 

All things being equal you would assume a 50/50 split between splits.



#10
tierdal

tierdal

    Advanced Member

  • Members
  • PipPipPip
  • 2344 posts

Why don't you just post the raw stats if you want to analyze them. I think you are going to find a few posters not willing to accept your analysis because in the past you've shown a habit of misrepresenting event results in order to further your agenda.

 

I can't even think of one time that is true. Miss-remembering sure, misrepresenting implies done purposefully and is a bit insulting. 

 

Id love to see the All Things Fun data as you had commented that alot of the splits were DS but I only remember 4 from that day - but I dont have the results.



#11
yankeefan1355

yankeefan1355

    Advanced Member

  • Members
  • PipPipPip
  • 1225 posts

IMO if the if the balance is off by %10 like tierdal said than you have a perfectly imbalanced game which is what you're aiming for anyway.   If BobaFett's %20 is correct than you have a perfectlyy imbalanced game that's off by about %5 from where you'd like to be.

 

Also, you're don't want 50/50 splits in these type of games.  You really whant a little bit of umbalance around %10-15 in favor of the most dominanat side at that moment which is LS.



#12
tierdal

tierdal

    Advanced Member

  • Members
  • PipPipPip
  • 2344 posts

I dont think you want splits. I think you want each side to have a relatively good chance of winning. Right now from the results I have they do not. And people know this - almost nobody chooses Dark Side if they have a choice in a top cut ...unless it behooves them at a later point (IE they get to play LS in a more important game).

 

I saw this happen in NYC Reg, and I saw this happen at the NYC store champ and both jersey store champs...although Scottie disagrees with me on this ..which is probably what he is referring to up above.

 

Game will never be perfectly balanced but it was never this imbalanced before imho.



#13
yodaman

yodaman

    Advanced Member

  • Members
  • PipPipPip
  • 2741 posts

Personally, it would be easier to report that data in its rawest form:  #LS wins, #DS wins, # timed games.  

 

Tierdal said there were 12 sweeps, 3 DS splits and 9 LS splits. 

 

Each sweep gives 12 LS and 12 DS wins. 

 

The 9 LS splits give 18 LS wins 

 

The 3 DS splits gives 6 DS wins. 

 

So for those results there were 30 LS wins, 18 DS wins.  The ratio is: 62.5% LS vs. 37.5% DS  Not sure if the top cut LS/DS results were listed anywhere to add to the results.

 

For reference, there were 37 LS wins, 19 DS wins and 5 timed games at Hickory.  Ignoring the timed games that was 66% LS - 34% DS.

 

Like I posted in another thread.  What amount of discrepancy indicates TOO much dominance by one side?  I think different people have different opinions about that.


  • TheBaronFel likes this

#14
dbmeboy

dbmeboy

    Advanced Member

  • Contributor
  • PipPipPip
  • 3910 posts

Now then, 3 vs 9 is a lot different from the 3 vs 27 your original post stated, is it not?

 

 

It's been a long time since I've had much reason to use statistics... I'm thinking a Chi Squared analysis may be appropriate here?  I know there are people much better at math than me lurking around here.



#15
tierdal

tierdal

    Advanced Member

  • Members
  • PipPipPip
  • 2344 posts

Right but that ratio factors in sweeps which really paints an entirely different picture. A great player can beat a bad player with a worse deck.

 

But put two good players together and it quickly becomes apparent that currently DS is struggling.

 

 

Now then, 3 vs 9 is a lot different from the 3 vs 27 your original post stated, is it not?

 

 

It's been a long time since I've had much reason to use statistics... I'm thinking a Chi Squared analysis may be appropriate here?  I know there are people much better at math than me lurking around here.

 
Indeed! Apologize for that. 


#16
yodaman

yodaman

    Advanced Member

  • Members
  • PipPipPip
  • 2741 posts

Tierdal, I agree with your theory about matching up good players and seeing what happens to try to get an indication about the balance of the game. 

 

 I'm a math professor so I always like looking at the raw data and numbers.  I'll leave it up to others to interpret them however they wish.

 

It is worth noting that at three regionals last weekend (the ones in Arizona, Hickory NC and FFG), the winner came from the loser's bracket of the top cut and ended up sweeping in the finals to win the tourney.  That means they each had to win a game with DS in the finals against what was clearly another good player in order to win the tourney. 

 

I think it's worth looking at 3 things to get data since each tells interesting things.

 

1. LS wins vs. DS wins in swiss.

 

2.  LS wins vs. DS wins in the top cuts.

 

3.  Total LS and DS wins in tourneys.


  • Jarratt likes this

#17
tierdal

tierdal

    Advanced Member

  • Members
  • PipPipPip
  • 2344 posts

Tierdal, I agree with your theory about matching up good players and seeing what happens to try to get an indication about the balance of the game. 

 

 I'm a math professor so I always like looking at the raw data and numbers.  I'll leave it up to others to interpret them however they wish.

 

It is worth noting that at three regionals last weekend (the ones in Arizona, Hickory NC and FFG), the winner came from the loser's bracket of the top cut and ended up sweeping in the finals to win the tourney.  That means they each had to win a game with DS in the finals against what was clearly another good player in order to win the tourney. 

 

I think it's worth looking at 3 things to get data since each tells interesting things.

 

1. LS wins vs. DS wins in swiss.

 

2.  LS wins vs. DS wins in the top cuts.

 

3.  Total LS and DS wins in tourneys.

 

Id love to get all the raw data we could and really look at it. Even start looking at strengths of schedules to understand (atleast at a proxy level) who were the "better" players

 

So please feel free to post results !!!


  • yodaman likes this

#18
yankeefan1355

yankeefan1355

    Advanced Member

  • Members
  • PipPipPip
  • 1225 posts

"this game has never been this imbalanced before"   Really tierdal, I dont think we're anywhere close to Gamor/Jedi or Dash/Freeholder territory.

 

In all honesesty after readomg all you're posts it sounds like you're just upset at not winning the tournament and doing as welll as you thought you deserved to so you're trying to manipulate statistics with imcomplete information and blame  the way the game for how you did in the tournamnent.  IMO.



#19
Jarratt

Jarratt

    Advanced Member

  • Members
  • PipPipPip
  • 3660 posts
This is a slightly different issue but gets back to the point.

I know why in Swiss 2 games are played against the same opponent. It is to make pairings easier in an already long day of tournament play. And in events with no top cut it seems to work a bit better.

However while they appear to be matches they really aren't and shouldn't be considered as such. I know some people don't like double elimination and feel the old system is better. I'm not advocating either here but having one part of the tournament appear to matches and the other being single games is definitely confusing and muddying the issue. Swiss would almost certainly be more balanced with a more complicated pairing system. It's just not practical.

With that in mind I just don't think analysing data as matches in Swiss, sweeps vs splits, is actually useful. The only affect the matching system has is timed games and possibly they should be thrown out the window. A draw tells us nothing.

Additionally without deck concept then raw data isn't that useful. If you want to see how well Mono Jedi is doing then you can't compare all LS wins and losses. Deck choice is often just as important as player skill.

I'm far more interested in the success of diversity rather than the pure numbers of LS > DS. If LS is having success with a bunch of different deck types then it's not Jedi that are the problem, it's why has DS got no tools for this. Or maybe most winning LS decks are running MTFBWY which they may or may not be but it's useful to know how the power objectives are as well.

Raw data might be what we have, but we need to treat it carefully because simple wins and losses isn't really going to get us very far in any analysis.

I for one was pretty excited by what I saw on the various regional streams over the weekend and the variation that is out there.
  • Wrath87, yodaman, KennedyHawk and 1 other like this

#20
KennedyHawk

KennedyHawk

    Advanced Member

  • Members
  • PipPipPip
  • 1890 posts

This is a slightly different issue but gets back to the point.

I know why in Swiss 2 games are played against the same opponent. It is to make pairings easier in an already long day of tournament play. And in events with no top cut it seems to work a bit better.

However while they appear to be matches they really aren't and shouldn't be considered as such. I know some people don't like double elimination and feel the old system is better. I'm not advocating either here but having one part of the tournament appear to matches and the other being single games is definitely confusing and muddying the issue. Swiss would almost certainly be more balanced with a more complicated pairing system. It's just not practical.

With that in mind I just don't think analysing data as matches in Swiss, sweeps vs splits, is actually useful. The only affect the matching system has is timed games and possibly they should be thrown out the window. A draw tells us nothing.

 

Agreed with this.

 

A sweep doesn't mean it was unevenly matched skill or any of that. Those games shouldn't be discounted. I suffered a few sweeps in really close games, and swept in some games where the final edge battle determined the win. Throwing those games out as uneven skill leveled mis matches seems off.


  • Wrath87, BobaFett, yodaman and 1 other like this