Cactus Game Design Message Boards

Redemption® Collectible Trading Card Game HQ => Playgroup and Tournament Central => Redemption® Official Tournaments => Topic started by: Warrior_Monk on May 17, 2014, 06:18:11 PM

Title: Proposition: Strength of Schedule as Secondary Tie Breaker
Post by: Warrior_Monk on May 17, 2014, 06:18:11 PM
Normally the tie breaker is first head to head, and then lost soul differential. My proposition is that it goes from head to head to strength of schedule, and then lost soul differential.

How is this calculated?
You add the points your opponent has each round to the tournament tracker.  This will be dynamic, filled in when the total points cell is filled in.  Additionally, between the master total points and lost soul differential column, there will be a column for Strength of Schedule.  You then add up all of the points of the opponents you faced throughout the tournament and divide it by 3(Y) where Y is the total number of rounds (opponents).

Thus Strength of Schedule=[(X1+X2+X3...)/(3Y)]/Y where X is the total number of points your opponent has and Y is the current round.

Let's look at Nationals 2012 as an example, after, say 7 rounds.
Martin Miller had played against Jonathan Greeson (18), Blake Maust (15), Connor Magras (12), Matt Townsend (15), John Earley (15), Mark Underwood (14), and Rob/Roy (12). If you add up all of these points, you get 101 total points. You then divide that by 3(7), and you get 4.8095248095, and then divide that again by 7 to get 0.68707482993197

The reason you divide by (3Y) and then again by Y is to give the opponent's total win percentage (in this case, 68.7%). It technically isn't necessary though, it's just more information.

Why is Strength of Schedule superior to Lost Soul Differential?
Let's keep looking at the same year, same round, as that's the required number of rounds with 65 people to have at least one undefeated person.  In this case, however, 4 players were at 6-1. Here are the numbers.
PlayerTotal PointsOpponent's Win PercentageLost Soul Differential
Martin Miller1868.7%19
Josh Brinkman1865.3%13
Alex Olijar1857.5%12
Jonathan Greeson1863.9%11

Had the tournament ended here, Greeson would have had every right to have a fit.  Head to head wasn't sufficient, as Alex hadn't played any of the top cut, and Martin and Josh didn't play.  Greeson actually beat Josh and lost to Martin.  Under current rules, Martin would take first, Josh second, and Alex third.  However, Alex had a comparable easy time in the tournament, avoiding the other three in the top four and playing three with 15 points. Greeson played against two people in the top four, and one with 15. It's no wonder Alex had a better lost soul differential: he had easier opponents!

The Objection
The primary argument against strength of schedule is that it's not in your control. It's really not Alex's fault that he didn't play more difficult opponents, that was just the luck of the draw.  You do have control over how many souls you rescue in a game, and how many your opponent gets. 

However, that argument breaks down because your opponent has just as much control over how many souls the get and how many you get.  The caliber of the opponent is really want matters.  Against easier opponents, you should be able to have a better lost soul differential. If you have easier opponents, you should win more often. Thus Strength of Schedule is the better tie breaker

A Compromise
Now, if you really want, you could multiply the soul differential by the strength of schedule.

PlayerTotal PointsAdjusted Soul DifferentialOpponent's Win PercentageLost Soul Differential
Martin Miller1813.05368.7%19
Josh Brinkman188.48965.3%13
Alex Olijar186.957.5%12
Jonathan Greeson187.02963.9%11

You'll notice the adjusted soul differential comes up with the exact same standings as the Opponent's Win Percentage. There are a few cases where this would change the standings (most extremely when someone has a negative lost soul differential), but in general I think it's easier if we just stick to strength of schedule.

Too Long; Didn't Read
Olijar is bad.
Title: Re: Proposition: Strength of Schedule as Secondary Tie Breaker
Post by: Chris on May 17, 2014, 06:45:01 PM
One other reason differential is a poor tie breaker is that one or two souls here or there can be based entirely on luck. Following the statistics that Westy gave, suppose Greeson plays a perfectly balanced deck, draws one soul, but no defense on the first turn, allowing his opponent to get a rescue. He then keeps his opponent from getting another soul for the remainder of the game. Should he really be penalized by not even placing because of such a random factor?

Additionally, giving differential precedence over strength of schedule can affect how people play in tournaments. Suppose my opponent played Son of God and New Jerusalem early in our game. The score is now 3-2 with me ahead, and I have Son of God and New Jerusalem in my hand. I also know that Falling Away is somewhere in my draw pile, and I have the ability to block my opponent for the next few turns. With differential having such a major impact, from a competitive perspective, I should wait to try maximize my differential. However, this is arguably an unsportsmanlike play, as it's drawing out the game longer than necessary. Regardless of the ethics of such a decision, the fact that the mere tournament structure could have a large impact on how I play the game doesn't seem right to me. Going a bit further, differential can impact whether or not I play Falling Away as opposed to some other dominant (and traditionally, I have played Falling Away because of the impact differential has). Again, the fact that tournament structure can dictate what cards are included in my deck doesn't seem to be right.

tl;dr DAE think Westy is bad?
Title: Re: Proposition: Strength of Schedule as Secondary Tie Breaker
Post by: LukeChips on May 17, 2014, 07:47:32 PM
Wow, you put a lot of work into this! Thanks!
Title: Re: Proposition: Strength of Schedule as Secondary Tie Breaker
Post by: Alex_Olijar on May 18, 2014, 10:42:16 PM
I would have been third. Just like westy.
Title: Re: Proposition: Strength of Schedule as Secondary Tie Breaker
Post by: Mr.Hiatus on May 19, 2014, 06:25:08 AM
Great work, I completely understand where you are coming from, but never gonna happen. I don't think Bany would care for that much work when we just implemented top cut. Why would you need this when top cut is used, as that would have solved the lack of opponents  for your 2012 sample.
Title: Re: Proposition: Strength of Schedule as Secondary Tie Breaker
Post by: Chris on May 19, 2014, 09:59:31 AM
The main reason we're arguing for the change is that we believe strength of schedule better measures how one has done in the tournament than differential does, so in the case of point ties after that seventh round, it definitely matters, as somebody is likely going to be cut.
Title: Re: Proposition: Strength of Schedule as Secondary Tie Breaker
Post by: LukeChips on May 19, 2014, 10:52:57 AM
I agree with Chris.  :)
Title: Re: Proposition: Strength of Schedule as Secondary Tie Breaker
Post by: TheJaylor on May 19, 2014, 11:52:29 AM
Quote from: LukeChips on May 19, 2014, 10:52:57 AM
I agree with Chris.  :)
For future reference, if this is all you're going to say then you should just plus 1 their post. Unless you have any additional reasons you agree, it's usually not worth the post.

I like the idea of strength of schedule and I think it has the potential to happen given that someone can put the confusion equation into Excel to figure rankings. I think it won't be as important considering that we already have Top Cut but I guess it would be a better determinant of who gets cut and continues.
Title: Re: Proposition: Strength of Schedule as Secondary Tie Breaker
Post by: Chris on May 19, 2014, 01:58:15 PM
Quote from: KoalaKingoFA on May 19, 2014, 11:52:29 AM
Quote from: LukeChips on May 19, 2014, 10:52:57 AM
I agree with Chris.  :)
For future reference, if this is all you're going to say then you should just plus 1 their post. Unless you have any additional reasons you agree, it's usually not worth the post.

I agree.
Title: Re: Proposition: Strength of Schedule as Secondary Tie Breaker
Post by: Mr.Hiatus on May 19, 2014, 05:39:16 PM
What I'm saying is, that if top cut is implemented again this year then you wouldn't need to use this, as you can't do the equation in an elimination style. Either top cut, or this. I would prefer this equation with no top cut.
Title: Re: Proposition: Strength of Schedule as Secondary Tie Breaker
Post by: Drrek on May 19, 2014, 07:29:28 PM
Quote from: Mr.Hiatus on May 19, 2014, 05:39:16 PM
What I'm saying is, that if top cut is implemented again this year then you wouldn't need to use this, as you can't do the equation in an elimination style. Either top cut, or this. I would prefer this equation with no top cut.

No you cannot run an equation style with top cut (except to determine who makes the top cut, which is also very important), but only one tournament in the year runs top cut, whereas the other important tournaments run straight swiss. Even with top cut (rightfully) determining the highest tournament, it would be good to have a better equation for regional and below level tournaments.
SimplePortal 2.3.3 © 2008-2010, SimplePortal