Jump to content
NHL'94 Forums

ONGOING LADDER LEAGUE


LA Robbie

Recommended Posts

OK, here's an idea. Don't know if it's easy to set up on the website, or a practical idea, but let me know what you think.

2 problems with "rookies": 1) Often have to wait for a tournament to come round & lose interest 2) Have to stay on the classic waiting list kicking their heels & lose interest 3) Enter a tournament they're not really ready for, get slaughtered in their first few games and give up on it.

Also, for the rest of us there's a long time between classics, meaning months of exhibitions & just wanting the tournament to come round already!!!!

So what if there were a "ladder system" tournament, for GENS & SNES which was always ongoing, that everyone could play whenever they so wished?

Say there's, I don't know, 500 (just a random number) active GENS players in the community right now. They all register for the "Ladder League", and they get ranked 1-500, based on their classic records. So the Classic champ would start as No.1, runner-up as No. 2, losing semi-finalists as 3 & 4 etc etc etc.

When a new guy joins he automatically gets added to the bottom of the list (In this case he'll be ranked 501st).

Best of 5 series. Higher ranked player has home advantage.

You can only play a guy a maximum of 10 ranking points higher or lower than you. Only the 2nd-5th placed guys can challenge the No.1. (This stops people challenging a high ranked player who's 100 places above them, winning through luck of the day and gaining 100 ranking points in one go, leaving a guy ranked far lower than he should be.)

If you beat a higher ranked player you swap places with him. If you lose to a higher ranked player you lose one ranking place.

You must play a lower ranked player at least once every 14 days. If you don't you lose 10 ranking places.

You can use any team you want for a single series. You don't always have to be the same team every time you play, but you do have to use the same team for every game of the same series.

There are no limits to how often you can play, but if you don't play for 3 weeks you get kicked out of the league. If you join again, you come in at the bottom.

You can "book" a month off from the tournament each year without losing ranking places.

You are not required to play during Classics if you're signed up for Classics, but you can play if you want to.

Guys on the Classic waiting list can use the tournament to play competitive games, without having to get slaughtered by guys much more experienced than them in mismatch after mismatch, because they can only challenge those a maximum of 10 places higher than them in the rankings.

Using this system there's always a competitive league of sorts running to keep people interested, and you've always got something to aim for in terms of improving your ranking.

My question is: Is it easy enough to set up an automated system for this on the site, so that one guy doesn't have the responsibility of keeping track of it all, so, if you win a series both coaches input it on the 94online.com site and the rankings automatically update. The system would also have to prevent guys with differences of more than 10 places playing.

.................It's a decent idea right?

Edited by LA Robbie
Link to comment
Share on other sites

I think it's a great idea to have an ongoing ladder, but I'm not crazy about the "switch positions" way of ranking. One unlucky matchup and you can lose 10 spots. Sometimes a guy just has your number due to playing styles.

I think it would be better to move guys up/down a certain number of places based on the relative places of the two players. If they are 10 spots apart and the lower guy beats the upper guy, maybe move them up/down 3 spots each. Or maybe move them both to the middle but put the winner above the loser.

Another, more complicated, option is to use the chess Elo ranking system, which is a probability based system: http://en.wikipedia....matical_details

In Elo, the average player will have a rating of about 1500. if a player with a rating of 1500 plays against a player with a rating of 1300 (200 points difference), then 75% of the time, the player with 1500 should win. A 200 point difference is equivalent to a 75% chance of winning. When you beat a player, you get some of their ratings points, and they lose some, based on some formula...

EDIT: this could be automated on the site, given time (programming it up) and maintenance (adding new players..)

Link to comment
Share on other sites

It would be interesting to also factor in which teams you are using to play with, so if you use a weaker team than the other guy, the reward for winning is greater.

Another thing to keep in mind is that when it comes to playing league games, it would be necessary for the coaches to agree ahead-of-time if the game is going to count in the ladder rankings. It should count, but for whatever reason the coaches might not want it to.

Link to comment
Share on other sites

I think it's a great idea to have an ongoing ladder, but I'm not crazy about the "switch positions" way of ranking. One unlucky matchup and you can lose 10 spots. Sometimes a guy just has your number due to playing styles.

Well I have no problem with your suggestion of a ranking system based on points etc, but I also like my idea because A) it's simple - "If I beat him i'll be 32nd" for example, like in boxing: "If I beat this guy i'm No.1 contender", "If I beat the champ, i'm the champ", but B ) because you can only challenge the guys ranked close to you (10 up, or 10 down) you don't lose or gain that much from a loss or a defeat, so if you win 1 just 'cos your style doesn't suit an opponent you may not continue progressing afterwards. If you lose one for the same reason you're still ranked around other similar level players and have the chance to quickly advance again.

I also like that a new guy, who's an advanced player could quickly shoot up the rankings, by beating the guys above him, and not be having to play mismatches against other, less experienced noobs to build his ranking points up.

It was basically meant to be a simple idea to get noobs playing competitive games without major pressure, to gain experience and stop them losing interest, and also a simple way to keep everyone ticking over & interested between Classics, as i'm just dying for it to start right now, and there's still like 6 weeks to go or something?!!!

It wasn't meant to be a major important title that would be the be-all and end-all, but rather a way to play some competitive games between the ones that really do matter.

Me, I like the "beat the champ to be the champ" way of doing things. It's clean and simple, and would be less set up I imagine? but hey, that's why I posted this. To start the debate. Comments appreciated.

Edited by LA Robbie
Link to comment
Share on other sites

This, of course, would be great (along with updating the main page on the site). A paging system would be cool too. For example, instead of looking at my next 10 potential opponents and hoping I catch them online, I can click on their name and send a "request to play" or something, with some proposed times. I'll also be able to see received requests. This can also be used to monitor activity -- no responses to requests lower your ranking, etc.

Link to comment
Share on other sites

ya I like the idea of being rewarded based on the team you use as well.. maybe if we ranked the teams in order of best to worst, with some teams being given the same rating if they are viewed as equal.. then factor that into the formula?

Link to comment
Share on other sites

ya I like the idea of being rewarded based on the team you use as well.. maybe if we ranked the teams in order of best to worst, with some teams being given the same rating if they are viewed as equal.. then factor that into the formula?

Good idea -- as an aside, it would be great to get a bunch of people to rank the teams (or top 15) and compile that data to see what the consensus says. Has this been done in the past? If not, I may do this...

Link to comment
Share on other sites

ya I like the idea of being rewarded based on the team you use as well.. maybe if we ranked the teams in order of best to worst, with some teams being given the same rating if they are viewed as equal.. then factor that into the formula?

I think that by using past gdl/blitz results and creating some Elo ratings (with the assumption that GDL teams are equal and Blitz teams are fairly equal), then creating Elo ratings with classic results, we could infer the relative strengths of the different teams.

oh wait.. gdl results.. deleted..

We'd need separate Line Changes ON and Line Changes OFF ladders, due to depth differences in the teams.

Link to comment
Share on other sites

Wow this is gettin complicated! lol. People's brains have reallt started ticking over as to the possibilities.

In my idea it would just be playing classic rules, so as to prepare noobs, and everyone else ahead of classics.

If the guy you're gonna play picks Edmonton & you go ahead and select Chicago he has the right to say "Hang on, i'm gonna change my team then, cos I don't like that match up & wouldn't want that to count to my ranking".

I like how this has got people thinking, but for me I was just thinking purely of an easily accessible, easy to understand, straightforward ladder tournament to get new guys playing and up to speed, and experienced guys ticking over before classic seasons start.

Maybe my idea works well for noobs and pre-classic warm-up, and there's the scope there for an extra something more full-on for you hard-core guys in addition too?

Link to comment
Share on other sites

It could be possible to have both systems active at the same time. A challenge ladder, and Elo ratings. They could be programmed into one thing.

For the normal ladder (not Elo ratings), then it could just take into account the teams to say how many spots you gain, and if you beat a guy above you, but you're using a better team, then obviously you shouldn't get to move ahead of him. You need to beat him with an equal or lower team to move ahead (although there could be some wiggle room, where if the teams are within 3 rankings of each other, it counts as an equal team).

Link to comment
Share on other sites

Based on the "Building Lines with AJ" rankings, I took AJ's ratings for forwards, defensemen, and goalies [i filled in QUE (6.5) and WSH (3.0) goalies as the write-ups are not complete], weighed them 3,2,1, respectively, and spit out this result. Just for fun:

CHI 9.00

DET 8.75

BUF 8.25

MTL 7.67

CGY 7.67

VAN 7.42

TOR 7.42

BOS 7.33

DAL 7.25

WPG 6.92

QUE 6.75

EDM 6.42

LA 6.33

NYR 6.08

PHI 6.00

HFD 5.50

PIT 5.42

WSH 5.17

STL 5.17

NJ 5.00

NYI 4.75

SJ 3.58

TB 3.08

FLA 3.00

OTW 1.75

ANH 1.33

Link to comment
Share on other sites

I also like that a new guy, who's an advanced player could quickly shoot up the rankings, by beating the guys above him, and not be having to play mismatches against other, less experienced noobs to build his ranking points up.

If the site/algorithm could place new players fairly accurately, this wouldn't be a problem. Since Elo is probability-based, this isn't tough. If you win a series 3-2, then you start out with slightly more points than the guy you beat.. sweep him 3-0 then you start with around 200-250 points more than him.

(since 200 points is equivalent to an expected winning chance of 75% each game, and a single sweep corresponds to having about an 80% chance of winning each game..)

Link to comment
Share on other sites

Based on the "Building Lines with AJ" rankings, I took AJ's ratings for forwards, defensemen, and goalies [i filled in QUE (6.5) and WSH (3.0) goalies as the write-ups are not complete], weighed them 3,2,1, respectively, and spit out this result. Just for fun:

CHI 9.00

DET 8.75

BUF 8.25

MTL 7.67

CGY 7.67

VAN 7.42

TOR 7.42

BOS 7.33

DAL 7.25

WPG 6.92

QUE 6.75

EDM 6.42

LA 6.33

NYR 6.08

PHI 6.00

HFD 5.50

PIT 5.42

WSH 5.17

STL 5.17

NJ 5.00

NYI 4.75

SJ 3.58

TB 3.08

FLA 3.00

OTW 1.75

ANH 1.33

Good idea.

I also have a set of stats based off me and a buddy playing with all the teams ~19 times each (we were going to do 25, to get all the matchups, but never ended up finishing the schedule):


Team 		GP	W(OTW)	L	OTL	T	PTS	PTS%
CHI		38	27(2)	9	1	1	56	0.737
MTL		38	26(2)	10	1	1	54	0.711
WPG		37	24(3)	11	1	1	50	0.676
DET		38	23(1)	13	1	1	48	0.632
EDM		38	23(3)	13	1	1	48	0.632
TOR		38	23(3)	14	0	1	47	0.618
BOS		38	20(1)	11	6	1	47	0.618
QUE		38	21(2)	14	2	1	45	0.592
DAL		38	21(1)	15	1	1	44	0.579
PIT		38	20(0)	15	2	1	43	0.566
PHI		38	19(2)	14	4	1	43	0.566
NYI		37	20(4)	16	1	0	41	0.554
VAN		38	20(4)	17	0	1	41	0.539
LA		38	19(3)	16	3	0	41	0.539
CGY		39	19(0)	19	1	0	39	0.500
WSH		39	18(4)	18	2	1	39	0.500
BUF		38	18(2)	19	1	0	37	0.487
SJ		37	16(3)	19	2	0	34	0.459
TB		38	16(0)	20	1	1	34	0.447
HFD		38	15(1)	19	2	2	34	0.447
STL		38	15(2)	20	3	0	33	0.434
NYR		37	14(2)	20	2	1	31	0.419
ANH		38	13(2)	22	2	1	29	0.382
NJ		38	12(2)	22	3	1	28	0.368
OTW		38	11(0)	22	3	2	27	0.355
FLA		40	10(1)	25	4	1	25	0.313

Funny thing you might notice is how low Buffalo is. They are notorious for having chemistry issues. However, when I split up the games into 3 'seasons', BUF is actually 5th in the 3rd season, so it seems we got much better with BUF as we became better players..

Here are the results if I weight the results 1/2/3 by 'season,' so the last season is worth 3 times as much as the first:


Team	GP	Pts	PTS%
MTL	77	115	0.747
CHI	77	108	0.701
DET	77	104	0.675
WPG	74	98	0.662
TOR	77	97	0.630
BOS	77	96	0.623
EDM	77	96	0.623
QUE	77	91	0.591
DAL	77	89	0.578
PIT	77	89	0.578
PHI	77	87	0.565
VAN	77	83	0.539
NYI	74	79	0.534
BUF	77	81	0.526
WSH	80	84	0.525
LA	77	78	0.506
CGY	80	79	0.494
SJ	74	71	0.480
HFD	77	71	0.461
STL	77	70	0.455
TB	77	64	0.416
OTW	77	62	0.403
NYR	74	57	0.385
ANH	77	58	0.377
FLA	83	50	0.301
NJ	77	46	0.299

obvoiusly something funny is going on if NJ is last.. it could be that the matchups weren't totally fair (NJ played tougher teams in the 3rd season perhaps).

Link to comment
Share on other sites

It could be possible to have both systems active at the same time. A challenge ladder, and Elo ratings. They could be programmed into one thing.

For the normal ladder (not Elo ratings), then it could just take into account the teams to say how many spots you gain, and if you beat a guy above you, but you're using a better team, then obviously you shouldn't get to move ahead of him. You need to beat him with an equal or lower team to move ahead (although there could be some wiggle room, where if the teams are within 3 rankings of each other, it counts as an equal team).

I think it's fair just to say, if both players agree to play, it counts. Again, simple and accessible for the benefit of newbies. Don't scare people away with too many rules etc

Link to comment
Share on other sites

Good idea.

I also have a set of stats based off me and a buddy playing with all the teams ~19 times each (we were going to do 25, to get all the matchups, but never ended up finishing the schedule):

Great stuff -- I like the idea of using actual game data to generate results. SOOOO, I took the results of the classic regular seasons (Spring'08 - Spring'11) and aggregated the data:

Team	GP 	W 	L 	T 	Win %
NYI	 40 	 29 	 11 	 -   	 0.725 
CHI	 491 	 340 	 151 	 -   	 0.692 
MTL	 392 	 240 	 150 	 2 	 0.612 
DET	 503 	 307 	 196 	 -   	 0.610 
BUF	 486 	 281 	 204 	 1 	 0.578 
TB	 98 	 54 	 43 	 1 	 0.551 
VAN	 474 	 256 	 217 	 1 	 0.540 
BOS	 333 	 178 	 155 	 -   	 0.535 
LA	 394 	 209 	 183 	 2 	 0.530 
WSH	 36 	 19 	 17 	 -   	 0.528 
DAL	 412 	 217 	 193 	 2 	 0.527 
NYR	 316 	 150 	 165 	 1 	 0.475 
CGY	 472 	 223 	 247 	 2 	 0.472 
TOR	 304 	 141 	 159 	 4 	 0.464 
EDM	 107 	 47 	 60 	 -   	 0.439 
WPG	 379 	 166 	 212 	 1 	 0.438 
OTW	 32 	 14 	 18 	 -   	 0.438 
QUE	 439 	 176 	 261 	 2 	 0.401 
NJ	 68 	 25 	 43 	 -   	 0.368 
HFD	 131 	 48 	 83 	 -   	 0.366 
PIT	 312 	 112 	 200 	 -   	 0.359 
SJ	 145 	 48 	 97 	 -   	 0.331 
STL	 171 	 49 	 122 	 -   	 0.287 
PHI	 253 	 71 	 181 	 1 	 0.281 
ANH	 38 	 8 	 30 	 -   	 0.211 
FLA	 -   	 -   	 -   	 -   	 -   

It's pretty obvious that the NYI are the best team in the league :lol: . That was one season played by IceStorm, a top ranked A player (lost in semis to Carse's CAL). Not surprisingly, the expected teams are on top (CHI, MTL, DET, BUF). Smoz, looks like Buffalo is just not for you :D. TB has two seasons, played by AngryJay and Zalex. I like TB a lot too, and would have expected them to be higher than their rankings suggest.

Also interesting to note is CAL. I personally never do well with them, even though they are one of the better teams in the league, but I'm surprised they have a losing record. I guess it's just not me who doesn't do well with them.

Edited by kingraph
Link to comment
Share on other sites

Let's just use the BCS Ranking system. That way there's no controversy at all :blink:

Is that a handegg thing? How does it work?

This "BCS explained" page totally did nothing for me

even on wikipedia: "This section may be confusing or unclear to readers. Please help clarify the section"

EDIT: now realizing you were probably being sarcastic and BCS is known to be controversial?

Link to comment
Share on other sites

It's a college "handegg" system that they use to determine rankings. It's extremely controversial and just about everybody hates it. It's what they use instead of playoffs... There are different "polls" that determine the rankings of teams. There's a computerized ranking system, and then I think the media poll and the coaches poll factors in as well. In other words, personal opinion and bias largely determines who the "best" team is.

EDIT: AH, you got me on the edit...

Edited by PRoBob38
Link to comment
Share on other sites

OK, here's an additional idea:

Newbies automatically enter the ladder in the "C" division.

Other guys are ranked in "A" or "B" divisions, depending on their classic records.

Once in the top 10 of your division you can challenge players in the bottom 10 of the division above. Once you've won 2 series against 2 different higher division players you can join that division. There's no relegation though, so a "B" player who loses to a "C" player doesn't become a "C" player. Once you're considered a "B" player. or an "A" player you're always in that division for competitive games.

A "B" player can't challenge a "C" player to a competitive game just to try to beat up on someone. Only the top 10 "C" players can challenge the bottom 10 "B" players.

This way newbies start out only playing against newbies, and there's a clear "achievement" you have to gain to move up to playing tougher games regularly.

Link to comment
Share on other sites

I like it.

There may be group 'inflation', though, since you can never go back down from a higher level to a lower level. might have to occasionally move the 'A' bar up so 'A' doesn't get too crowded and 'B' too small.

Oh, and I'd say you have to win 2 out of your last x series (x=5?) to move up, not just 2 ever.

however, this requires some history-keeping, adding some complication.

Link to comment
Share on other sites

So this for everybody? I was thinking it was just for rookies. That makes things much more complicated.

For everybody, but with the aim of easily integrating newbies into online play. But it doesn't need to be complicated. Those vets that want something with complicated formulas and ratios and team ratings are welcome to push for it, but i'm just looking for a way that helps new guys practice, improve & stick it out until they reach a competitive level of play, like we did.

I just wanna see a "beat him, swap places" system that new guys can use to enjoy playing, against their own level, and progress. Nothing fancy, nothing complicated, nothing that needs mountains of commitment, just a pick up & play, "hit this guy up for a game" type of thing that keeps new guys excited about playing 'cos they're not getting blown out of the water every game, and keeps the rest of us ticking over between classics.

The only exception I like is the idea of divisions for the varied skill levels, and guys have to win a couple of series against a guy from the division above in a set amount of times to progress to the next division.

Lets make it happen. I really think the community would benefit greatly & grow through this.

Link to comment
Share on other sites

I feel like the divisions would be redundant. One ladder, where you can challenge the 10 guys ahead of you is simple enough and self-dividing. Wouldn't divisions just be drawing lines in the ladder? The added work for creating and maintaining divisions doesn't seem to equal the benefit.

Then again, I'm no programmer...so maybe its easier than I imagine.

Link to comment
Share on other sites

Yeah, I kinda agree with the idea of just doing one giant ladder rather than using the different divisions. If some new guy comes in and is a world beater, let him work up the ladder as quickly as he can. Like kingraph said, seems like the divisions are just dividing the ladder and not really serving much purpose otherwise. Just my thoughts.

Link to comment
Share on other sites

Join the conversation

You can post now and register later. If you have an account, sign in now to post with your account.

Guest
Reply to this topic...

×   Pasted as rich text.   Paste as plain text instead

  Only 75 emoji are allowed.

×   Your link has been automatically embedded.   Display as a link instead

×   Your previous content has been restored.   Clear editor

×   You cannot paste images directly. Upload or insert images from URL.

×
×
  • Create New...