SuperLeague Changes - Pre-Vote Discussion

[john4p]

Originally Posted by ED-209

However, Biscuit's score 20 system gets my secondary vote for sure.. and on that note - will we be able to cast a secondary vote when it is time?

I'll make a multiselect-poll. Personally I like Biscuit's idea the best, followed by fixed and 75/25.

Originally Posted by ED-209

I really dislike the idea of changing points awarded for first place.

Me, too.

[Biscuit]

Originally Posted by ED-209

Agreed! I think this is the best alternative method yet, well done coming up with that Biscuit
I especially like the way it incorporates the bonus points idea.

Alhough fixed 16 will still get my primary vote
Because it is will always provide the simplest, cleanest, fairest spread of points. Yes, the issue of players outside the top 16 all on 1 point is there, but I think this only matters if there is huge abnormal burst in player count. In most cases (if not all) would mean just a little extra dedication to break into the top 16.

The chat here massively changed my original opinion, which is a good thing. Fixed 16 is definitely a good option and fair. The main difference is the TIER system will make the league points closer and should provide some more excitement. I think that will be fun.

[lifeschool]

Biscuit - For the current league, the first 6 games had an average of 20 players playing. From Benefactor, the average was 14 per round - until the final two games, which the average was back to 20 players. So I can see why a 20 Tier system would help, and would encourage more than 16 to join. Newbies might not know about a bonus system, and might think their score would only get 1 point, so no point posting it.

Yes I still am not convinced, but there is time left to work things out. This thread is due to run for two weeks, so lots of time left to go. Its great that this thread is 4 pages in size after only 3 days.

[Biscuit]

Originally Posted by lifeschool

Biscuit - For the current league, the first 6 games had an average of 20 players playing. From Benefactor, the average was 14 per round - until the final two games, which the average was back to 20 players. So I can see why a 20 Tier system would help, and would encourage more than 16 to join. Newbies might not know about a bonus system, and might think their score would only get 1 point, so no point posting it.

Any system that keeps the league points close together and rewards players fairly has a good chance of retaining players throughout the year. There's always been a summer lull and it's hard to be sure what the cause is, could be the nice weather. Most of all I'd like to see fun competitive play. We gets lots already so whichever system we choose I'm sure it'll be great as usual.

[john4p]

@lifeschool: I think you misunderstand Biscuit's "Tier 20"-system. It doesn't mean that it will give different points to 20 participants. There will be a couple of same scores in the middle and bottom part of the table. No one will be below 5 points. Biscuit's system works just as well for 6 people as it does for 40 people taking part.

A catchy name for this system would be "Super 5" because the first 5 places are fixed and everyone gets a minimum of 5 points.

@Biscuit: The bonus points for places 6th+ will be calculated proportionally to the current 5th place's score, right?

[linesmachine]

Originally Posted by lifeschool

Biscuit - For the current league, the first 6 games had an average of 20 players playing. From Benefactor, the average was 14 per round - until the final two games, which the average was back to 20 players. So I can see why a 20 Tier system would help, and would encourage more than 16 to join.

Theory and reality are 2 different things. Real world outcomes don't always follow assumption. The above scenario from Dan could play out as he suspects. But it could also play the other way; people may well engage less or hold back in rounds with low turnouts and therefore less points on offer, that affect could compound round on round and ultimately kill the entire combo.

I personally believe the only thing we know for sure is that having a fixed number of points for (at least the top) positions provides transparency and simplicity to encourage engagement. Complexity will likely just put off casual gamers.

1st place should always win the same amount of points. The argument that your odds of a higher score diminish as more people join a round is mathematically accurate but it's not 'fair' in that if I'm good at Sensible Soccer and win the round, I get less points than another competitor who is good at Turrican II and wins that round. It suggests it's right to reward players who are best at the most popular games.

Take the Premiership football league again. If the size of the league shrank in number, would you promote it's growth by lowering the number of points for a win...or would that actually kill it off? And if it grew, would you increase the number of points for a win to make it somehow more beneficial to lower position clubs? I'm not sure it follows.

[john4p]

Correct, transparency and simplicity is key!

[john4p]

Missed this earlier:

Originally Posted by lifeschool

You are right, and this is why it would only make a difference round to round, as in some rounds the points would be lower. If a player is consistent and they get the similar position every round, then there would be not much relative change.

If the players take part in every round there would never be a difference. The relative change will always be the same regardless if he's consistently placing high or not because he'd always get the full bonus points even in last place. Thus the relative changes would be identical to the case that everyone gets 0 bonus points.

Originally Posted by lifeschool

But for the reasons you point out, I think giving one point per player is too risky and unpredictable, and I am more happy thinking about a => 12 or => 16 bonus system, which only rewards participation after a certain level.

If the fixed-16-system gets chosen then yes, Harry's idea would be worth thinking about.
That's another plus for Biscuit's system, its 1st place score never needs to change in order to reward every single player fairly.

Originally Posted by lifeschool

If my original 1 bonus point idea is totally bad, then I am happy to remove it from the options list for the vote.

It's not totally bad. Some people would surely be happy getting 21 points instead of just 1 even if it's just an illusion.
Biscuit's 5 points minimum are also just an illusion but there's the advantage that 1st place would still always get 20 points which keeps everything simple, transparent and consistent.

[Biscuit]

Originally Posted by john4p

@Biscuit: The bonus points for places 6th+ will be calculated proportionally to the current 5th place's score, right?

That's correct.

Originally Posted by john4p

Biscuit's 5 points minimum are also just an illusion but there's the advantage that 1st place would still always get 20 points which keeps everything simple, transparent and consistent.

Yes. I would prefer to use a system where the points on offer each round are the same. For me, the best part of the TIER/Super 5 system is that if the scores are very close then it's possible that everyone from 5th place and lower can get 12 points. But they only get those points by playing well. If they advance into the fixed points at the top they continue to get good rewards for their effort but the amount of points available for each player in every round is always the same.

[Biscuit]

I was curious to see how the scores come out for different types of results so I ran the numbers for a few more rounds. The most interesting is Benefactor which shows all of these point systems can benefit and hinder different groups of players. I didn't compare the bonus points idea because it would take too long for me to do.

I’ll leave them here and not go on about it anymore.

[john4p]

Looks good. I don't think anyone would complain about his points in the Super 5-system. The winners still get rewarded well but will be less uncatchable now.

[lifeschool]

Originally Posted by linesmachine

people may well engage less or hold back in rounds with low turnouts and therefore less points on offer, that affect could compound round on round and ultimately kill the entire combo.

Nothing like looking on the bright side. I cant see why anyone would hold back scores unless it affects the team score, or unless they save their score for later and forget to post it. I think if a game is appealing, players play it, and if not, they wont. But I like to think players are able to come and go as they choose, and each year we have people who only play for a couple of rounds. I dont think everyone will hold back their scores to kill the compo, given any system.

Originally Posted by linesmachine

but [1 point per player bonus] it's not 'fair' in that if I'm good at Sensible Soccer and win the round, I get less points than another competitor who is good at Turrican II and wins that round. It suggests it's right to reward players who are best at the most popular games.

And this is part of the reason why I think my 1 point bonus idea is not so good anymore. I mentioned on here and on Lemon that I think the 1 point idea has too many down sides that we have mentioned, so it should be taken off the table or at least into consideration. So that leaves the 12+ or 16+ bonus system if we go with the Fixed 16 idea. If these too are unworkable, then I think just having the top 16 places scores alone is perhaps too harsh in a 25 player round.

Originally Posted by linesmachine

I personally believe the only thing we know for sure is that having a fixed number of points for (at least the top) positions provides transparency and simplicity to encourage engagement. Complexity will likely just put off casual gamers.

I dont think complexity will put anyone off as long as the score matrix we have can take care of it. Working out how to score a game in the first place, using a complex formula, is hard enough with some rounds. If the system goes down, it will put me off simply because it would take a lot of maths to work out the entire table, especially when the 5th place score goes from 100k to 1M or 28M.

Biscuit - Interesting results, and its clear that the current system does have advantages under certain scenarios, but not compared to the overall spread of the other systems. Although its not hard to factor in the bonus systems:

Benefactor = 11 players = 11 points (1 point), 0 points (12+), 0 points (16+)
Blocky Skies = 20 players = 20 points (1 point), 9 points (12+), 5 points (16+)
Turrican 2 = 14 players = 14 points (1 point), 3 points (12+), 0 points (16+)
Apidya = 25 players = 25 points (1 point), 13 points (12+), 10 points (16+)

If we are saying the 1 point bonus idea is not going to work, then this leaves the Dynamic system based on 12 (to catch more players) or 16. If would be worth factoring all of this IF the whole season was worked out, and we can see the results of the systems.

[linesmachine]

Originally Posted by lifeschool

Nothing like looking on the bright side. I cant see why anyone would hold back scores unless it affects the team score, or unless they save their score for later and forget to post it. I think if a game is appealing, players play it, and if not, they wont. But I like to think players are able to come and go as they choose, and each year we have people who only play for a couple of rounds. I dont think everyone will hold back their scores to kill the compo, given any system.

Sorry Dan, I wasn't Dan-bashing in particular, but your point I think summarises the opposite side to my own so used it as the example! I was trying to get across a point without looking like an arrogant prick (something I am not well gifted in).

It's a kind of personal point because I started as a casual gamer and became a core gamer.

From my experience I can tell you that, as a casual gamer, I found any rule complexity stopped me entering the core of the competition.

And when I say about holding back, people aren't holding back scores so much as effort. As a core gamer I became slightly tactical about what I was doing and my point scoring. I started to invest in proportion to the points at stake (think double rounds for instance).

But it is only my experience. Others may differ.

[vagrant]

The only thing I don't like about the bonus points idea is that can push maximum points per round above 20, effectively changing what's been in place since the comp began (I think.. It's always been 20 for 1st right?). So like John said elsewhere it would destroy previous records/stats.

If we were willing to totally overhaul points awarded for first, why not just change to a 100 player fixed system and be done with it (which I am not endorsing btw )

Originally Posted by linesmachine

Complexity will likely just put off casual gamers.

Originally Posted by john4p

Correct, transparency and simplicity is key!

Yeh, I think overly complex system might have the potential to put some players off, or at least cause some initial confusion/frustration.
It might even be more of a deterrent then a 25 player round with 10 players sitting on 1 point.. because at least with a fixed system as a new player, you look at the leaderboard and know exactly what the points represent, and exactly what you need to do to improve, rather then wondering why your on the same league points as somebody with a lower score

Originally Posted by Biscuit

The main difference is the TIER system will make the league points closer and should provide some more excitement. I think that will be fun.

And then there is this point.. which is also a very good good one!!
I'm still preferring fixed 16, but Biscuit's system is in very close second

[vagrant]

Originally Posted by linesmachine

I was trying to get across a point without looking like an arrogant prick (something I am not well gifted in).

Well your a lot better at it than me, I am the master of arrogant pricks

[mihcael]

Originally Posted by ED-209

...(I think.. It's always been 20 for 1st right?). So like John said elsewhere it would destroy previous records/stats.
...

No it was 15 then we changed it to 20 (because it was better suited to the new scoring system)

Originally Posted by ED-209

...rather then wondering why your on the same league points as somebody with a lower score ...

The only system proposed that would address this issue is a dynamic score system.

[vagrant]

Originally Posted by mihcael

No it was 15 then we changed it to 20 (because it was better suited to the new scoring system)

Shows how much notice I took of that at the time
So what that means is that statistics have already been somewhat destroyed..
Well I guess previous records can be amended based on percentage of total points available for that season

The only system proposed that would address this issue is a dynamic score system.

It will.. but with dynamic, or even tier based on 20 maximum points, the more players that participate - the larger the 'blocks' of players on the same points become anyway.

EDIT: Right I've said some silly things in this thread so prepare for possibly another;
Fixed system based on 30 maximum points. (20 max doesn't matter so much anymore does it?)
Once-off bonus points for only new players to a season, depending on which round is entered.
round 2 - 1 bonus point, round 3 - 2 points and so on. Yes the bonus points are low, but it's better then nothing, and doesn't appear to have other issues mentioned.

Either that or same thing with 16 fixed.
Or Biscuit's suggestion.
Or maybe 16 + lifeschool's bonus system.
Either way it is definitely beer o'clock

[mihcael]

Originally Posted by ED-209

...
It will.. but with dynamic, or even tier based on 20 maximum points, the more players that participate - the larger the 'blocks' of players on the same points become anyway.
...

No, a dynamic system (with a fixed base) would grow with competitors.

11 players (F1GP)
20,17,15,14,13,12,11,10,9,8,7

25 players (Apidya) (every competitor above 16 adds 1 point)
28,25,23,22,21,20,19,18,17,16,15,14,13,12,11,10,9,8,7,6,5,4,3,2,1

[vagrant]

Ah ok, I thought 'dynamic' as in the current system.
Yeh it solves the spread problem, but I think keeping total points available each round consistent is more important.

[mihcael]

Current system is 50% Fixed + 50% proportional compared to 1st place score.