Myrgard

Are Tirri, Arzenic and Cruniac just an beatable trio?

Is Paris the new Rabican maneuvering his players like an expert chess player?

The word you meant to use was, unbeatable.

LOL TYPO

Take those ratios and divide them by two after you run into Team Asmo.

Hmm, that would leave Crun with a (5.4/2) = 2.7 Damage ratio

Interesting theory.

par73 wrote:Hmm, that would leave Crun with a (5.4/2) = 2.7 Damage ratio

Interesting theory.

That is his kill ratio, not damage.

Thanks for clarifying, I just wanted to make sure you could clarify cruniac will have a 2.7 kill ratio after he plays against your team. rgr


u hz No confidence in your team, lol

He shall kill 2.7 thrall for every 1 trow we kill

Damn, looks like he'll be leaving the ratios up for grabs for the other players on our team like how I usually play

I should just get rid of kill ratio... so useless.

Meh, I leave it for continued lolz

too bad you never included captain statistics or expanded on other possible areas of statistics for display

oh the fun we could be having

punkUser wrote:I should just get rid of kill ratio... so useless.

Meh, I leave it for continued lolz

mad cuz bad

par73 wrote:too bad you never included captain statistics or expanded on other possible areas of statistics for display

If there was anything particularly interesting it can be collected or even added pretty easily. But I've taken a good look at other statistics and not many achieve significance. Showing too many stats is not actually a good thing, as then people just cherry pick whatever they want to make their points, play non-optimally and generally misinterpret the numbers. Case in point, KDR... it's simply there because the game shows it, not because it means anything much.

Chohan wrote:mad cuz bad

Sounds like someone is volunteering for thrall brigade

punkUser wrote:If there was anything particularly interesting it can be collected or even added pretty easily. But I've taken a good look at other statistics and not many achieve significance. Showing too many stats is not actually a good thing, as then people just cherry pick whatever they want to make their points, play non-optimally and generally misinterpret the numbers. Case in point, KDR... it's simply there because the game shows it, not because it means anything much.

There isn't one stat in regular myth team games that has any true meaning and value when it is just the stat alone, besides Win-Loss. Damage is the same as KDR it's heavily influenced by what unit set you have and who you are playing against. WW2 was the only time where damage-kill ratio's were indicative of the top players.

Chohan wrote:There isn't one stat in regular myth team games that has any true meaning and value when it is just the stat alone, besides Win-Loss. Damage is the same as KDR it's heavily influenced by what unit set you have and who you are playing against. WW2 was the only time where damage-kill ratio's were indicative of the top players.

Agreed. Ratios are pretty tho

Chohan wrote: There isn't one stat in regular myth team games that has any true meaning and value when it is just the stat alone, besides Win-Loss. Damage is the same as KDR it's heavily influenced by what unit set you have and who you are playing against.

I should think that is obvious to everyone. (Btw it's also the reason why ranking systems should not take into account any result beyond win-loss, but I guarantee someone will start that bitch-fest when the time comes. )

It's pretty indisputable that damage ratio is a lot more telling than kill ratio though, for the simple reason that both team have an equal amount of "damage" regardless of trade, etc. Because trading is done in terms of "damage" values too (i.e. UT cost), this is already something that is optimized for at the team level. So sure you can get a crappy job that screws your ratios in any situation (ghol scouting, defence in certain cases, meat shield, etc.) with the regular unit distributions it at least roughly indicates whether in BC you're contributing positively or negatively to your team's overall relative %. KDR is pretty much completely meaningless as trades affect the total value available to each team. And... trow.

So yeah, obviously stats != skillz, but damage ratios is by far the closest to a useful stat that we have. I don't think there's really any reasonable argument against that, although I fully expect people to keep cherry-picking whatever stats make themselves look good

I think if we had a math major playing myth, they would be able to create a wonderful formula that represents true myth skill. however its like rating players in modern sports, there is no measurement with 100% true accuracy.

time spent/time wasted would be an interesting stat. except we'd all be tied with a 1:1 ratio.

i think we could all equally measure time spent, however time wasted introduces a lot of various values that is far from being measurable in a basic sense.

Chohan wrote:

punkUser wrote:If there was anything particularly interesting it can be collected or even added pretty easily. But I've taken a good look at other statistics and not many achieve significance. Showing too many stats is not actually a good thing, as then people just cherry pick whatever they want to make their points, play non-optimally and generally misinterpret the numbers. Case in point, KDR... it's simply there because the game shows it, not because it means anything much.

There isn't one stat in regular myth team games that has any true meaning and value when it is just the stat alone, besides Win-Loss. Damage is the same as KDR it's heavily influenced by what unit set you have and who you are playing against. WW2 was the only time where damage-kill ratio's were indicative of the top players.

Disagree, damage ratios mean something now that they are fixed. I'm not saying they are the most important thing, but if you are getting lower than a .8 dmg ratio as a non-capatain throughout a whole match [even playing within the strategy] then the chances are you did not play particularly well.

So basically dmg ratios aren't the main factor, but they generally do correlate with strong play.

No. Damage ratio is only the value of the units you have compared to the damage those units caused to the enemy. It's a very frequent occurance in 2t that one team gets its major striking force (power + mobility) wiped early, leaving anywhere upwards of 60% still running around getting mowed down with no recourse. It's also common in that instance that the players involved in starting that Domino of Decline ® will have excellent ratios (think 2 rushers with 20% in warriors and pus), and the inverse is true for their hapless teammates. It doesn't even have to be a complete wipe on the strike forces' part, just enough to take any kind of control over where, when, and vs who their teammates have to fight.

Early myth reps were very skewed because of this. There's a revered old schooler who was actually terrible, but because he got basically all the fast melee 99% of the time, his ratios would look great and his teammates maybe not. If his team won, he'd get the lion's share of the credit. If they lost, he'd point to his inevitably bad teammates' ratios and the blame would stick to them.

It's not so absolute anymore, but using damage ratios as even a hint of an indication of "strong play" is farcical in a 2t discussion. That you came up with some kind of ratio tipping point is insane.

(Only marginally relevant: what part of the country did you grow up in? I want to guess Rocky Mountains.)

wwo wrote:It's not so absolute anymore, but using damage ratios as even a hint of an indication of "strong play" is farcical in a 2t discussion. That you came up with some kind of ratio tipping point is insane.

Really? Go look at the mwc finals films or any 2 good tournament teams playing against each other since the new ratios have been put in place. Damage ratios are a very good indication of how well someone played.

I already stated that it is not a 100% thing. Perfect example is Zak getting good ratios vs WTC, but making major blunders that lead to his team getting flagged multiple times; but the point still stands, good damage ratios tend to correlate with good play.

Edit: Please do show me an example where a player got a .7 dmg ratio or lower for a whole match and played good. I realize it can happen on a game to game basis, but the chances of someone getting low damage ratios over a 5 game or longer series that played good is very unlikely.

Then qualify your statement by adding "for the entireity of a match in a competitive tournament environment in the last few years". It helps, but only a little.

I did mention match and "under the new dmg ratios" (which has only been recently), but you are right I should have clarified in a tournament setting.

par73 wrote:I think if we had a math major playing myth, they would be able to create a wonderful formula that represents true myth skill. however its like rating players in modern sports, there is no measurement with 100% true accuracy.

I have a masters in math... but indeed there's no perfect rating and there doesn't need to be because there's no perfect ideal of a player either. It's not even a well-defined concept.

wwo wrote:It's a very frequent occurance in 2t that one team gets its major striking force (power + mobility) wiped early, leaving anywhere upwards of 60% still running around getting mowed down with no recourse. ...

Sure, but that doesn't make the entire thing useless in the context of a reasonably competitive game. Obviously if two teams aren't a very well-balanced matchup then stats in general are pretty much meaningless. Anyone can hack down retreating forces as well as the next guy. That's incidentally why GKG's 1v1 tournament was split into different "divisions" for the second phase, because those second round of matches were much closer and thus produced better stats.

So yeah, not much point drawing too many conclusions from blowouts, but that's an orthogonal issue. For matches that are reasonably competitive, I think it's pretty clear that damage ratio correlates the most strongly with in-game play. You'll still tend to see the same players balloon to the top in dmg ratios, even when they get different sets of units. The bias of current captains to always give certain players certain units does of course make this a bit harder to separate, but if you glance over the stats from various rounds of the last bunch of tournaments trends do appear.

Anyways, not really disagreeing, but I think people are somewhat throwing the baby out with the bathwater here. Dmg ratio isn't always a perfect indicator of "skill" or anything (even if such a thing was well-defined), but it still holds some useful information in the context of a decently balanced match.

And yes, pre-1.8 even dmg ratio was pretty questionable since it also favored players with high-HP/value units like trow and was easily abused by detaching before death. I think people will agree the current setup is a lot better than it used to be.

crun ~mid being rushed you say? rgr flanking

East Wind tmnt wrote:crun ~mid being rushed you say? rgr flanking

Asmodian wrote:

wwo wrote:It's not so absolute anymore, but using damage ratios as even a hint of an indication of "strong play" is farcical in a 2t discussion. That you came up with some kind of ratio tipping point is insane.

Really? Go look at the mwc finals films or any 2 good tournament teams playing against each other since the new ratios have been put in place. Damage ratios are a very good indication of how well someone played.

I already stated that it is not a 100% thing. Perfect example is Zak getting good ratios vs WTC, but making major blunders that lead to his team getting flagged multiple times; but the point still stands, good damage ratios tend to correlate with good play.

Edit: Please do show me an example where a player got a .7 dmg ratio or lower for a whole match and played good. I realize it can happen on a game to game basis, but the chances of someone getting low damage ratios over a 5 game or longer series that played good is very unlikely.

I will Asmo, but first let me interrupt you.


Asmo is right that using damage ratio is a poor way to measure this. Kill ratio is also a poor way to measure this. There needs to be a combination of both ratios, plus other variables unaccounted for, for example: total percent of a team's units given should effect this and is just one factor that should be looked into. People have tried to make a system that measured overall MWC talent years ago, it was called the Wighty Score. If anyone still has/remembers the formula, you should post it here (it of course would have to be adjusted due to the damage x/2.56 change). Maybe PunkUser could implement wightscore on the statistics web page for traditional myth 2 purposes.

Yeah okay let me get into pari enflammedenflammed mode to give you some myth history lessons, is that alright with you?

Examples via myth history:

so you basically suck ass because you CAPTAIN and that means every captain ever is already predisposed to suck

Wwo is worse of a captain than a player, but his statistics will never reflect this because he will always have low statistics because he has no problem being the player relied on to trade and distribute units, heal wights, and watch over the team..

This does not count for gkg who would rather carry a team on his back and make his team eat his dick than experience losing, (military training pays off !) I can't really blame him or he probably would have "retired from myth" long ago

That idea of using damage ratio to measure overall talent at myth, means that every power player ever is already predisposed to be amongst the greatest players ever by this dmg ratio measurement.
heron guard heroes and trow shredding through a kicks-worth-of-health melee? yeah "great ratios"

it might be better to categorize players in their in game roles, i.e. Captains, Power Players, Participating Team Members; in order to get a better feel of "who does what the best" and all that jazz.

Another reason why it's unreliable and invalid to think that either the kill ratio or the damage ratio really reflect what goes on in a myth game is that the amount of unit trading points in dark maps is always much higher than light maps, thus the damage received and given will always be a lot higher.

it might be better to categorize these games by light/dark/(if necessary, neutral) statistically as well, depending on the map. All you have to do is be more efficient in dark maps to have first glance efficient statistics.

Light maps which combine numbers with dark maps in the ratios (without any sort of weighting) will have implications of "TRUE" good or bad play skewed.

Ok ok, enough tearing into possible reason for error in believing there is a correlation between good damage ratios and "good play".


My final point:

Example of where a player got a .7 dmg ratio or lower for a whole match and played "well"?
It happened last weekend, for example

Arsenal (The unlikely hero in the spot light)
Match Results vs Team Ratking
5 / 0 / 0 K-19 L-16 KLR-1.188 DG-41 DR-83 DGRR-0.494

Arsenal had a negative damage ratio for all 5 of the games that his team won.
He also had a positive kill ratio for 2 of the games, and negative for the 3 games he lost.
So technically Arsenal "Did nothing" and "Sucked ass" according to the idea of damage ratio being correlative to "good play".
However, I don't believe measuring his efforts with damage ratio (and a conflicting kill ratio) would be appropriate to determine how well he played last weekend.
I do believe, however, that Arsenal accomplished a lot for himself and his team last weekend.

In the first game (DrownedEmp, FR) , Arsenal had 5% of his teams entire units.
With 5% he was able to

• A) Be the only player who was alone within the first half of the game (he was the only player mid with 3 players on a N/S flank respectively)
  C) Tag an enemy defensive flag on his own
  B) Kill 6 units and only losing one

Game Two, Venice KotM
Arsenal gets one of four maximum warlocks and two warriors,
garnishing a relatively small 9% in light of 100%. (the highest was funky, with 18% of 100%)

with what he has, Arsenal accomplishes

• a) getting a negative damage ratio and kill ratio
  b) completely stalls ratking and captain, and the entire team ratking south so that the rest of the team can move into position and tear them apart, constantly applying pressure.
  c) he doesn't get a single kill or stroke of damage, and gets one loss.
  d) smart play and contributing factor to the win
  e) at the end of the game %wise, Arsenal has the 3rd most percent remaining.

Game Three, a poison holiday dark flag rally
Arsenal gets 13%, and is one of five players who receive 13% or 14% of the total units.

Arsenal accomplishes:

• a) Holding the enemy mid 24% with dantski and 6% Codex, with Funk who had a similar amount of % in units as Arsenal.
  b) Arsenal got one damage and one kill, but zero losses.
  c) The game is over at 4 minutes and 50 seconds when team ratking forfeits their final flags before ratios get even more skewed out of their favor.

Game Four, if i had a trow, stampede
Arsenal gets 15%, 3rd most % than any other player on Team Paris, and goes mid alone.

Arsenal accomplishes :

• a) Holding off pallidice and wwo, a combined 27%
  b) Keeping pallidice and wwo distracted long enough to never leave mid (to go after any stampede targets, or defend any of their own targets) and utlimately get the shit kicked out of them by Cruniac about 3 minutes into the game.
  c) tying tirri in 3rd place for top team kills with 7, tying funky in 6th with top team damage.
  d) getting 7 kills, but taking 7 losses.
  e) absorbing the most damage (and distraction) on his team, with 41 damage.

Game Five, acts of herongaurds, flag rally.
Arsenal, 12%.

• a) Defends the mid flag for his team in the closest game % wise.
  b) Absorbs two of dantski's puss
  c) informs his teammates of a potental flag to grab after selflessly sacraficing his units, which is the last flag his team captures
  d) absorbs 22 damage

If I wasn't able to bring Arsenal into this, I would not have bothered writing this at all.
Without Arsenal last weekend, who else would have been able to fill his void? No one, because no one else would have been there.
Arsenal proved himself to be a very valuable teammate and asset in a team effort for success.
Arsenal was a key contributor to the 5-0 romping of dominance which occurred last weekend.
If Arsenal asked me for more units next weekend, I would most likely grant him his wish for his good play.
Arsenal is the Clean Up Man.
Arsenal is considered a key player for Team Paris.
Arsenal played very well last weekend.
Be warned.

Statistics do not always tell the tale. You must look at how they are being measured, why they are being measured, why other things are not being measured? Is what you're measuring being measured reliably and accurately represents what you are measuring? Using math at face value is just numbers, you must expand from quantitative, to qualitative methods. This is what I call Money Balling.

More importantly, if you looked at "MWC Statistics All Time" you would notice that Asmo had a negative damage ratio in every MWC he entered except for MWC13, despite the fact he has been known to say that he was always a good player. And since now I start to be brash to people, after praising my teammate for his work as part of glory: I say this, hence

ego smash, up up back forward a b x y, blah

Wwo's point is right, Asmo's point is mostly wrong except for his saving grace of contradicting himself as well as bringing up the "possible" (obvious) correlation between a positive damage ratio and good play.
Wwo is a better captain and player than Asmodian, and Asmodian is a worse captain and player than Wwo.
Good game gents, and good night.

I just got paid 20 bucks to write this, 10$ an hour. Thank you for all mentioned and who participated in my invesitgation and critique of "Co-relational methods of Myth Ratios and Player Peformance"

b) Absorbs two of dantski's puss

One pus, which froze a bunch of his clumped units.

I guess I had a super series last week too by absorbing all those trow kicks.

Asmodian wrote:
Disagree, damage ratios means something now that they are fixed. I'm not saying they are the most important thing, but if you are getting lower than a .8 dmg ratio as a non-capatain throughout a whole match [even playing within the strategy] then the chances are you did not play particularly well.

So basically dmg ratios aren't the main factor, but they generally do correlate with strong play.

You just mentioned a lot of captains (Wwo,Gkg and I) who always have low ratios. :: for reading comprehension fail.

Really this whole reply is just full of reading comprehension fail. You are basically agreeing with me in most of the post, but do not realize it because you failed to understand what I had previously wrote.

The only valid point you brought up in your reply that had not already been mentioned was Dark vs Light. Also I highly doubt Arsenal played "good" according to those statistics. I will have to look at the films later. You like most of myth tend to hug your teammates nuts regardless of how bad someone may play. I feel like that is one thing that sets GKG apart as a captain. He is willing to criticize people for bad play (even in wins) to help them improve instead of hugging nuts to try and win someone over.

I feel all warm and fuzzy inside.

Thanks for the kind words, it's nice to be appreciated haha.

par73 wrote:Maybe PunkUser could implement wightscore on the statistics web page for traditional myth 2 purposes.

Nah, I looked into it ages ago and it was basically some totally made-up formula with several glaring errors. Good example of what's wrong with people trying to intuit statistics and why that doesn't work though

The existence of good team-play that don't result in good damage ratios still doesn't invalidate the metric as something reasonably correlated with BC skills over a large enough sample. No one is saying it's the be-all and end-all of Myth statistics in all situations, but it's often useful and it's unlikely that there exists something better (outside of straight wins/losses of course, but that's ultimately a team stat).

par73 wrote: Using math at face value is just numbers, you must expand from quantitative, to qualitative methods. This is what I call Money Balling.

Sort of ironic considering the whole point of the baseball scheme was the opposite... i.e. trusting the numbers directly instead of qualitative "feelings" about player value. Which brings us to an interesting question... what if we did a draft based on player stats from some set of previous tournaments? Would be interesting to see how "competitive" such a tournament would be vs. the captain's drafts

I would have to say the best stats for week 1 in reality go to...me
k/d 1.87
dmg ratio 1.266
This was done while being down 2 players on a team with an overall lower ball rating than Team Paris, and Team Paris had a player advantage in their match.
Go me

Asmodian wrote:

Asmodian wrote:
Disagree, damage ratios means something now that they are fixed. I'm not saying they are the most important thing, but if you are getting lower than a .8 dmg ratio as a non-capatain throughout a whole match [even playing within the strategy] then the chances are you did not play particularly well.

So basically dmg ratios aren't the main factor, but they generally do correlate with strong play.

You just mentioned a lot of captains (Wwo,Gkg and I) who always have low ratios. :: for reading comprehension fail.

Really this whole reply is just full of reading comprehension fail. You are basically agreeing with me in most of the post, but do not realize it because you failed to understand what I had previously wrote.

The only valid point you brought up in your reply that had not already been mentioned was Dark vs Light. Also I highly doubt Arsenal played "good" according to those statistics. I will have to look at the films later. You like most of myth tend to hug your teammates nuts regardless of how bad someone may play. I feel like that is one thing that sets GKG apart as a captain. He is willing to criticize people for bad play (even in wins) to help them improve instead of hugging nuts to try and win someone over.

I can't believe Asmodian just referenced himself in a quote, then argued with himself. Then again, I probably could.

This is hilarious.

Arsenal played a lot better than you last weekend. He won 5 games, you won 3. Are you mad?

P.s. Gkg has high ratios and very efficent ratios. he is a lot different of a myth player and phenomenon than "wwo" or "Asmo" so i don't know why you're even trying to bring this up.

he has 4 rings, you have one, wwo has zero.

your categorization of captains is a sad but amusing effort. please seek advice.

punkuser wrote:

par73 wrote: Using math at face value is just numbers, you must expand from quantitative, to qualitative methods. This is what I call Money Balling.

Sort of ironic considering the whole point of the baseball scheme was the opposite... i.e. trusting the numbers directly instead of qualitative "feelings" about player value. Which brings us to an interesting question... what if we did a draft based on player stats from some set of previous tournaments? Would be interesting to see how "competitive" such a tournament would be vs. the captain's drafts

Well I never watched the movie so I could "give a fuck less" what bias you are influenced by.
Anyway, I did the exact concept you are talking about...
So be prepared when the team I have drafted, filled with capable teammates with larger girth than Team Angry Face 2001, flush through this draft tournament because people are retardedly unaware of the resources available to them to be competitive myth players.

My version of Money "Balling" is different than what you perceive as moneyballing.

Kind of like how you (heterosexual), I (heterosexual) and Asmo (homosexual) perceive human sexuality differently.

Dantski wrote:

b) Absorbs two of dantski's puss

One pus, which froze a bunch of his clumped units.

I guess I had a super series last week too by absorbing all those trow kicks.

damn dantski, to think i thought you were egoless like the rest of the english boys*

what has NC done to you?


good lord,

Dantski circa NC 2007+ employment

*Elfoid and Empy Phil, not Enigman. M-T, Youngsy. Zer is Irish. Me? Welsh n` Scot!

Dantski wrote:

b) Absorbs two of dantski's puss

One pus, which froze a bunch of his clumped units.

I guess I had a super series last week too by absorbing all those trow kicks.

lol. Best post of this thread

Asmodian wrote:

Dantski wrote:

b) Absorbs two of dantski's puss

One pus, which froze a bunch of his clumped units.

I guess I had a super series last week too by absorbing all those trow kicks.

lol. Best post of this thread

INTERESTING NOW

Asmodian wrote: I already stated that it is not a 100% thing. Perfect example is Zak getting good ratios vs WTC, but making major blunders that lead to his team getting flagged multiple times; but the point still stands, good damage ratios tend to correlate with good play.

hilarious stuff, keep it up court jester.
there is none other as good at what you do, as you.


WE CAN'T ALL GET TROW AND HERON GUARD HEROES,
some of us are just chosen to do so
(and you never are)