Warrior General Games

Originally posted by InRomoWeTrust
You're talking two completely separate things here with your suggestion thread item. Besides New Jersey Generals, I'm pretty sure there are no 'zombie' teams that the Virgins have passed this season.

NY Nightmare and Legacy are terrible. So yes, there are a few teams left they should have passed by now, 20 games in. Look at Nightmare's and Legacy's wins as stated previously. They should have been near the bottom or past the top 25 by now, 20 games in. Consecutive blow out losses need to have an amplified loss formula. Period.

Also, these "zombie" teams should have fell out after the 10-15th game. It shouldn't take 2/3 of the season to get them to move down 10 spots for being realistically a 50-70 rank team.

Blow out losses is not how ELO works. Score differences is not how ELO works. At all. You're creating a completely new system that you think is good without actually creating something that can be represented.

If Queen City or the Stunners blows out a quality team who happens to not be able to tackle, does that team deserve to be massively punished like you desire? Being blunt, I scored more points against Ministry of Dotball than I did Boogie Men (an inactive/crap team).

And not a chance Legacy should be penalized in a way that you would a 'zombie' team. They may not be winning, but they put up competitive games on a consistent basis. New York Nightmare likewise lost to AA by just 7 points. Air Raid by 2. They're another team that isn't winning but shouldn't be penalized like a 'zombie' team. That's ludicrous. With a higher k-value, they'd be moved lower and passed up with other teams. But no way those are teams that should be down at 35 or w/e is spinning in your head.

wish there was an "overall" record for teams. I know that wouldn't be any kind of real indicator of where teams should be in the ladder -- but it would be interesting to see how it lines up.

I'm in the camp that says the virgins are moving up too slowly considering their (vet) 20-0 record.

looks like last season the virgins were (pro) 28-4 (lost championship game)
year before: (journey) 26-6
year before: (seasoned) 26-6
year before: (soph) 22-9-1
rookie: 22-9-1

(could be slightly off on a bit).

So going back looking at their last 116 games (3 full seasons plus 20 from this season) they have lost 16 games total.

Looking at the same period (3 seasons + this one) for Queen City: QC has lost 48 games (exactly 3 times as many as the virgins)

sure - makes perfect sense that queen city is ranked higher...(I understand how ELO works and I understand that pure win/loss record doesn't necessarily mean a lot...but COME ON)

Originally posted by TxSteve
wish there was an "overall" record for teams. I know that wouldn't be any kind of real indicator of where teams should be in the ladder -- but it would be interesting to see how it lines up.

I'm in the camp that says the virgins are moving up too slowly considering their (vet) 20-0 record.

looks like last season the virgins were (pro) 28-4 (lost championship game)
year before: (journey) 26-6
year before: (seasoned) 26-6
year before: (soph) 22-9-1
rookie: 22-9-1

(could be slightly off on a bit).

So going back looking at their last 116 games (3 full seasons plus 20 from this season) they have lost 16 games total.

Looking at the same period (3 seasons + this one) for Queen City: QC has lost 48 games (exactly 3 times as many as the virgins)

sure - makes perfect sense that queen city is ranked higher...(I understand how ELO works and I understand that pure win/loss record doesn't necessarily mean a lot...but COME ON)


Right, k-value adjustment would address the above NOT some crazy overhaul with score difference dynamics.

Does adjusting the k value provide larger movements in the ELO for wins and losses? What is the argument for not adjusting that? I guess I don't understand why it currently is so low.

You really hate competition don't you? So you are saying that adding a 'if' formula to the current formula is impossible? Then rewrite your own to keep the same system but with an addition to move shit teams down quicker.


If Queen City of the Stunners blows out a quality team who happens to not be able to tackle, does that team deserve to be massively punished like you desire? Being blunt, I scored more points against Ministry of Dotball than I did Boogie Men (an inactive/crap team).
Yes, they deserve a higher elo loss more then usual for being blown out by 70+. As stated in the suggestion forum, the loss ins't ZOMG I am dropped 10 spots for being blown out. It should be about 125-150% of a normal elo ooint loss.

And not a chance Legacy should be penalized in a way that you would a 'zombie' team. They may not be winning, but they put up competitive games on a consistent basis.
So you are saying they are a deserved top 15 team for wins against:
Coondogs (.000%)
GVP (.526%) And are bad. Their wins against opponents record is (.319%)
Goldminers (.053%)
MoD (.473%)
Dookie (.350%)
Generals (.000%)
Lue (.400%)
Tuskenriaders (.053%)

Thats 1 win over a team that has a winning record (.526%). You really think they should still be top 15 20 games in? Thats honestly what you think?

New York Nightmare likewise lost to AA by just 7 points. Air Raid by 2. They're another team that isn't winning but should be penalized like a 'zombie' team. That's ludicrous.
Your argument here is that they had close games so they deserve to be ranked a top 8 team? Really dude? You made sense when you said "They're another team that isn't winning" but that's about it. Because they are not winning and with wins only coming from terrible teams, yes they deserve to be near the bottom of the top 25 by now, 20 games in.

Originally posted by bhall43
Does adjusting the k value provide larger movements in the ELO for wins and losses? What is the argument for not adjusting that? I guess I don't understand why it currently is so low.

Yes it does. I don't think it's low (teams are moving, hence Virgins +10 and Queen City +8 in the earlier posts), it's just needing a tweak upwards.

There are dangers with increasing it too much, as the system becomes too volatile, so it's important to work in small adjustments to make sure we hit a sweet spot without a huge guess and check mentality.

Virgins last 116 games (3 full seasons plus 20 from this season) they have lost 16 games total.

Queen City (3 seasons + this one) for Queen City: QC has lost 48 games (exactly 3 times as many as the virgins)

Stark: 41 losses in that time

NY Nightmare: 44
Legacy: 51

ninja, u mad?

You're not reading things in context. Never am I stating those teams should be exactly where they are in rank today.

Originally posted by InRomoWeTrust
Yes it does. I don't think it's low (teams are moving, hence Virgins +10 and Queen City +8 in the earlier posts), it's just needing a tweak upwards.

There are dangers with increasing it too much, as the system becomes too volatile, so it's important to work in small adjustments to make sure we hit a sweet spot without a huge guess and check mentality.

what are "the dangers" if things get volatile? who says that is bad?

Originally posted by InRomoWeTrust
ninja, u mad?

You're not reading things in context. Never am I stating those teams should be exactly where they are in rank today.

To a point, yes I am a little mad. As should others. We pay for this game. We get a ladder game ever 3 days. For 80% of the ladder games to be shit matchups is horrible. This can be fixed, and hope it does.

Originally posted by TxSteve
Virgins last 116 games (3 full seasons plus 20 from this season) they have lost 16 games total.

Queen City (3 seasons + this one) for Queen City: QC has lost 48 games (exactly 3 times as many as the virgins)

Stark: 41 losses in that time

NY Nightmare: 44
Legacy: 51



In an ELO model, a newer team like the Virgins are at a distinct disadvantage (purely age and competition-based). Really compounded by a k-value that is too low. I don't foresee a plausible situation where a new vet team will ever be basically instantly top 5'd. We'd likely either have to have less teams or a crazy k-value. Definitely getting big upward movement is needed, but I imagine the top 5 will always be a tough nut to crack.

Bort was probably on K when he set the value...

Originally posted by TxSteve
what are "the dangers" if things get volatile? who says that is bad?

Common sense says it's bad. Biggest danger is the system overcorrects itself too fast. A team on a streak, regardless of elite competition, would be getting more ELO to their name for it. With the way ladder scheduling works, the rankings could get REALLY funky, not resembling something 'accurate'.

In theory, if you really jacked the k-value, you could create a system that reflected a ladder based on like the last 3 games, lol. That's the polar extreme. Now that wouldn't happen, but that gives you an idea of how that could flow.

The realistic side of having a k-value that is too high is that it rewards the streaky team over the team that has consistent success (consistent in the form of 'that season').

Sounds like we are all kind of agreeing: the k value is too low.