|
Hey, if throwing away 8 points a week is what you like to do, that's fine by me.
|
Not a good week to have jaedong on anti-team.
|
Braavos36370 Posts
On December 04 2009 05:56 integral wrote:Show nested quote +On December 04 2009 01:44 Hot_Bid wrote:On December 04 2009 00:43 integral wrote:
So instead of having jangbi (-2 points last season under new scoring) or baby (2 points) or horang2 (1 point) on my team, I have sair and sungsun and nbs. I don't know these players, I've never even seem them play, I just know they're less of a risk than someone who plays and loses. That's a good way to not get negative points but it also basically guarantees you of a middle of the pack finish among thousands of owners. If you want to finish high this season, you will have to pick players you think will win, not just pick players who will "not lose." On that note, why the hell does jangbi cost 2 points when he ended up negative last season? Because we adjusted a lot of the costs. Since you think Jangbi costs too much, you're putting him on your anti team right?  Anyway, I don't really like this way of scoring. Starcraft is zero-sum, so if you're not winning you're losing -- but the way I see it, the real accomplishment is just getting to play at this level. Penalizing players for playing games is silly. How are we penalizing players for playing games? We're simply removing the automatic point for playing. There's no constitutional law stating that every player in fantasy proleague is guaranteed an appearance point. I disagree -- the real accomplishment is not "simply getting to play." The real accomplishment is winning. And with 4 points to a win and -2 for a loss, players break even if go 1-2 (which probably means they still score, because of ace and team wins). From an owner standpoint, predicting who gets to play is very, very easy. I can predict with about 90% accuracy (as can everyone else) who will be sent by most teams. Especially since I can trade after seeing lineups. But its much harder and more skillful to predict who will win games, especially without the safety blanket of knowing that theres no penalty for just picking players who play a lot (which is easy by just looking at lineups). Originally I really liked the new rules for scoring, particularly since anti-team players could gain you points. The only reason I spoke up in the first place was because the "expected point gain of a 1-pointer" constant should change to appropriately fit the new scoring rules, and the only reason I continue to post is because Abydos1 apparently did not understand that the new scoring changes would lead to an average decrease in scores, not an increase. Obviously this impacts the scoring and trade values of players not just this round, but the next round as well -- set too high of a constant for the number of points distributed and player costs will deflate from round to round. It's only after thinking about this a long time that I'm starting to dislike the new scoring rules. I'm not some conservative who thinks rules should be a particular way just because that's the way they were done before, nor do I feel like there should be some sort of charity giving points to players who lose. My point is that mathematically, 1-pointers who play will on average do significantly worse than their non-playing counterparts. That's what I mean by a penalty for playing. 1-pointers, if they play, are going to lose. Yes, there is a skill in picking players that are going to do well, but the emphasis right now is not on winning but on win percentage.
So pick 1-pointers that don't play, or don't pick them at all. I don't see how this changes anything, because if you pick players "on average" you're going to end up in the middle of the pack. What the new rules do is rewards owners for choosing / trading for 1 and 2 pointers who will break out, and punishes owners who are wrong about it.
Also, I really don't understand why you would change a players' cost relative to the previous round. Player initial costs are really important in determining who goes on what team, and if a player did so poorly as to score negative points the previous round they really shouldn't be seeded at 2 points the next time. It's ridiculous to arbitrarily pick and choose which players' costs to change -- you might as well be using a completely different criterion for assigning player costs, like "skill", and do away with math altogether.
Reasons:
1. Unpredictability of 1-Round Results
Before, we used a formula that was split (2/3 cost based on most recent round; 1/3 cost based on two rounds ago). Unfortunately this is the start of the season so "last round" was like half a year ago and thus one single round is not predictive of player value going forward, especially with the variance that comes with matchups in SPL.
2. Diversification of Teams
Add to this that while each players cost theoretically should be the same value to your team (ie Jaedong at 8 should be worth the same to your team as Type-B at 2, because Jaedong scores 4x the points), owners almost never diversify picks. Its always the same players being chosen round after round. One goal obviously is to diversify teams -- if everyone has the same team or 80% the same team (JD Flash Bisu etc), then FPL becomes simply about who guesses the right 2-point player. This is not an ideal situation, and thus why we "adjust" players according to the demand of owner choices.
There would be an uproar if you arbitrarily assigned a player points or inflated their trade value in the middle of the season, it isn't much different to do that in the offseason.
We don't set every players value, we only adjust a few of them, for the reasons posted above. For instance, Las Vegas sets betting lines for many sporting events, but does not actually set the line at what they think it will be. They set the line at what they believe people's perceptions are. Jaedong's actual price based solely off PL 09-10 R1 is 6. Last round, almost half the owners in FPL picked Jaedong @ 10. If his value was 6 this round, just about every single serious team would have Jaedong. Add to that Bisu @ 6 (we also adjusted him up), and that essentially makes FPL a 18-point, 4 player team. Do you want 99% of FPL to have Bisu, Jaedong, or Bisu+Jaedong?
The point of this league is to also compete with each other and make PL more interesting. If every single team was the same this would drastically kill the enjoyment value of FPL. Thats why we "adjust" most players. Everyone sees the stats from last time, and diligent owners can figure out pretty easily from last rounds ending Trade Values where the differences are.
Also, since its easy to see which adjustments were made, you can easily just put the adjusted players on your Anti-Team if you believe their values are too high. You are welcome to put Jaedong or Jangbi on your Anti-Team if you feel their prices are not fair.
Obviously FPL is still about picking players and making trades and doing all that better relative to everyone else that is playing FPL. I get that. No matter how the scoring and player costs change, everyone's teams are still scored the same. But for all your talk of skill, it's getting pretty annoying trying to play FPL seriously.
You may think everyone's teams are scored the same so its "fair" but that's not the only consideration we have. It'd be equally annoying to play FPL if 90% of the owners had the same teams. You can only make so many teams with 6+3 slots, especially if there are "easy" picks. This round, there will be greater diversification of choices because of the adjustments we made (less "easy" picks), and thus that increases the skill factor and enjoyment factor of FPL.
Bottom line is, would you rather have Jaedong at 6 or 8? For your team personally maybe he's better at 6, but for FPL overall? Would it really be good if 95% of the teams had Jaedong? Its not like we made him cost 12 -- plenty of people will stick pick him.
|
Movie such popular pick once again, don't trust him guys, even if he only he has to prepare for PvZ~!
|
I enjoy this team a lot, but it took a bit of fiddling with to get it right. Mostly, I just really like the anti-team. I don't think Sea is bad, just that 9 points is too high of a price - he's probably more like a 5 or 6 point player. Really is almost certainly a better player, for example- too bad I didn't have room for him on my main. I laughed when I saw that hydra was a 3 point player, but I'll probably swap him out for anytime the second he has a zvz in the lineup.
I have some doubts about the main team, though. Ameba I picked because I feel like he is the only 1 pointer likely to win points if they get played rather than lose them. Light is consistent, been playing really well. Same thing for Stork and Violet. Type-B/Shine are both zergs who have lots of potential, and deserve to be the go to players on both of their teams rather than the backups they currently are.
Gubbinal
Main Team
1  Ameba
4  Light
4 Shine
5  Stork
2 Type-B
7 Violet (captain)
7 KT Rolster
Anti Team
1 Herb
3 Hydra
9 Sea
|
Wow. So many people picking BeSt for their AT... He's been quite raping lately.
|
|
|
On December 04 2009 07:37 Manit0u wrote:
Wow. So many people picking BeSt for their AT... He's been quite raping lately.
yeah, but he wont play in the first week.
|
On December 04 2009 07:37 Manit0u wrote:
Wow. So many people picking BeSt for their AT... He's been quite raping lately.
he's also not playing this week..
|
|
|
On December 04 2009 06:48 Hot_Bid wrote:Show nested quote +On December 04 2009 05:56 integral wrote:On December 04 2009 01:44 Hot_Bid wrote:On December 04 2009 00:43 integral wrote:
So instead of having jangbi (-2 points last season under new scoring) or baby (2 points) or horang2 (1 point) on my team, I have sair and sungsun and nbs. I don't know these players, I've never even seem them play, I just know they're less of a risk than someone who plays and loses. That's a good way to not get negative points but it also basically guarantees you of a middle of the pack finish among thousands of owners. If you want to finish high this season, you will have to pick players you think will win, not just pick players who will "not lose." On that note, why the hell does jangbi cost 2 points when he ended up negative last season? Because we adjusted a lot of the costs. Since you think Jangbi costs too much, you're putting him on your anti team right? Anyway, I don't really like this way of scoring. Starcraft is zero-sum, so if you're not winning you're losing -- but the way I see it, the real accomplishment is just getting to play at this level. Penalizing players for playing games is silly. How are we penalizing players for playing games? We're simply removing the automatic point for playing. There's no constitutional law stating that every player in fantasy proleague is guaranteed an appearance point. I disagree -- the real accomplishment is not "simply getting to play." The real accomplishment is winning. And with 4 points to a win and -2 for a loss, players break even if go 1-2 (which probably means they still score, because of ace and team wins). From an owner standpoint, predicting who gets to play is very, very easy. I can predict with about 90% accuracy (as can everyone else) who will be sent by most teams. Especially since I can trade after seeing lineups. But its much harder and more skillful to predict who will win games, especially without the safety blanket of knowing that theres no penalty for just picking players who play a lot (which is easy by just looking at lineups). Originally I really liked the new rules for scoring, particularly since anti-team players could gain you points. The only reason I spoke up in the first place was because the "expected point gain of a 1-pointer" constant should change to appropriately fit the new scoring rules, and the only reason I continue to post is because Abydos1 apparently did not understand that the new scoring changes would lead to an average decrease in scores, not an increase. Obviously this impacts the scoring and trade values of players not just this round, but the next round as well -- set too high of a constant for the number of points distributed and player costs will deflate from round to round. It's only after thinking about this a long time that I'm starting to dislike the new scoring rules. I'm not some conservative who thinks rules should be a particular way just because that's the way they were done before, nor do I feel like there should be some sort of charity giving points to players who lose. My point is that mathematically, 1-pointers who play will on average do significantly worse than their non-playing counterparts. That's what I mean by a penalty for playing. 1-pointers, if they play, are going to lose. Yes, there is a skill in picking players that are going to do well, but the emphasis right now is not on winning but on win percentage. So pick 1-pointers that don't play, or don't pick them at all. I don't see how this changes anything, because if you pick players "on average" you're going to end up in the middle of the pack. What the new rules do is rewards owners for choosing / trading for 1 and 2 pointers who will break out, and punishes owners who are wrong about it.
I'm just pointing out the pattern. Excluding team wins, 1-pointers that play will do worse than 1-pointers that do play. Obviously I'm adjusting my team accordingly, I've already mentioned that. The reason I even have 1-pointers on my team is because this season rewards win % and ace appearances -- not necessarily winning -- so I've gone with what I think are the "most guaranteed" win % players who will show up in the most ace matches. Picking 3 top players leaves me with 8 points left over to distribute between a team and 3 other players. If you see no problem with no-name no-skill players earning more points than players that actually get the opportunity to play, I guess we're just not going to see eye-to-eye on this.
On December 04 2009 06:48 Hot_Bid wrote:Show nested quote +Also, I really don't understand why you would change a players' cost relative to the previous round. Player initial costs are really important in determining who goes on what team, and if a player did so poorly as to score negative points the previous round they really shouldn't be seeded at 2 points the next time. It's ridiculous to arbitrarily pick and choose which players' costs to change -- you might as well be using a completely different criterion for assigning player costs, like "skill", and do away with math altogether. Reasons:1. Unpredictability of 1-Round ResultsBefore, we used a formula that was split (2/3 cost based on most recent round; 1/3 cost based on two rounds ago). Unfortunately this is the start of the season so "last round" was like half a year ago and thus one single round is not predictive of player value going forward, especially with the variance that comes with matchups in SPL. 2. Diversification of TeamsAdd to this that while each players cost theoretically should be the same value to your team (ie Jaedong at 8 should be worth the same to your team as Type-B at 2, because Jaedong scores 4x the points), owners almost never diversify picks. Its always the same players being chosen round after round. One goal obviously is to diversify teams -- if everyone has the same team or 80% the same team (JD Flash Bisu etc), then FPL becomes simply about who guesses the right 2-point player. This is not an ideal situation, and thus why we "adjust" players according to the demand of owner choices.
Team diversification is mostly limited by the lack of slots for players and the limited number of high-cost players relative to the number of low-cost players. Right now there are 23 players 4 points or higher, and 179 players 3 points or lower, the vast majority of those being 1-pointers. People pick the same players because there aren't that many players. Gotta get to 30 somehow, and you're not gonna get 30 with all 3-pointers. You'd see a lot more diversification if the number of players you could pick were increased (e.g. to 8), but just adjusting the costs of cheaper players won't necessarily impact diversification much.
On December 04 2009 06:48 Hot_Bid wrote:Show nested quote +There would be an uproar if you arbitrarily assigned a player points or inflated their trade value in the middle of the season, it isn't much different to do that in the offseason. We don't set every players value, we only adjust a few of them, for the reasons posted above. For instance, Las Vegas sets betting lines for many sporting events, but does not actually set the line at what they think it will be. They set the line at what they believe people's perceptions are. Jaedong's actual price based solely off PL 09-10 R1 is 6. Last round, almost half the owners in FPL picked Jaedong @ 10. If his value was 6 this round, just about every single serious team would have Jaedong. Add to that Bisu @ 6 (we also adjusted him up), and that essentially makes FPL a 18-point, 4 player team. Do you want 99% of FPL to have Bisu, Jaedong, or Bisu+Jaedong? The point of this league is to also compete with each other and make PL more interesting. If every single team was the same this would drastically kill the enjoyment value of FPL. Thats why we "adjust" most players. Everyone sees the stats from last time, and diligent owners can figure out pretty easily from last rounds ending Trade Values where the differences are. You are welcome to put Jaedong or Jangbi on your Anti-Team if you feel their values is too high.
Obviously you're exaggerating to make your point about diversification, but I'm not impressed. The highest %-owned players last season were: Jaedong 47%, Iris 43%, Jangbi 34%, Movie 33%, and skyhigh at 28%. If you want to include both/or percentages with only 23 players above 4-points I think that's a little disingenuous... how much can be attributed to player costs and how much to fanboyism or bandwagoning based on recent results? Isolate the variable of player cost and your diversification argument would be a lot stronger.
On December 04 2009 06:48 Hot_Bid wrote:Show nested quote +Obviously FPL is still about picking players and making trades and doing all that better relative to everyone else that is playing FPL. I get that. No matter how the scoring and player costs change, everyone's teams are still scored the same. But for all your talk of skill, it's getting pretty annoying trying to play FPL seriously. You may think everyone's teams are scored the same so its "fair" but that's not the only consideration we have. It'd be equally annoying to play FPL if 90% of the owners had the same teams. You can only make so many teams with 6+3 slots, especially if there are "easy" picks. This round, there will be greater diversification of choices because of the adjustments we made (less "easy" picks), and thus that increases the skill factor and enjoyment factor of FPL. Bottom line is, would you rather have Jaedong at 6 or 8? For your team personally maybe he's better at 6, but for FPL overall? Would it really be good if 95% of the teams had Jaedong? Its not like we made him cost 12 -- plenty of people will stick pick him.
This part just annoys me, apparently the only issue you have with what I wrote is that I talked about adjusting player costs, because the rest of it you don't address. Would you care to address the issue of the constant used to determine trade values being inaccurate and causing deflation?
As far as I'm concerned my main point hasn't even been addressed: at the end of the round unless something is changed the point distribution will be totally fucked up. Top players with no losses and with ace appearances will be miles ahead of 4 and 5 point 50/50 players that with 10 lineup appearances would come in at 10 points + team wins. All players have to exceed 33% winrate to break even with their teammates who don't play -- which 12 of 25 1-pointers that played did not do, 12 of 26 2-pointers did not do, 9 of 14 3-pointers did not do, 4 of 12 5-pointers did not do, and which 2 of 4 6-pointers did not do. This is a significant change to the scoring that obviously impacts player choices as well -- there are a lot of low win % players that are not going to be earning any points next round. If you're concerned about team diversification you should be concerned about this, because if you're not an idiot you'll take your chances with a high cost high win% player over two lower cost lower win% players. On second thought, this is actually good for diversification; owners who really understand the new scoring will be picking up a lot of players they've never heard of before to fill out the other half of their Jaedong/Effort/Bisu/Flash (pick any two) teams.
A simple "hey you make some good points about the new scoring system" would make me feel a lot more heard. Right now even the most basic thing I pointed out, that overall points will decrease with the new scoring, hasn't even been acknowledged -- despite Abydos1 having already apparently increased the value of the constant instead of decreasing it.
This thread has been really annoying, having to bump my own posts through the spam so someone acknowledges them and whatnot. Just thought I would say that in case it impacted how you perceive my tone right now.
|
|
|
So this doesn't get reiterated, I understand why you adjust player prices based on demand, I just don't agree. It just seems pretty arbitrary to me given the limited player pool. Plus, many of the adjustments made are due to the increase of the constant used to calculate value and relative decrease of the point costs. Almost everyone is cheaper than they should be with the new constant being too high relative to players' expected point gain. With an accurate constant and purely based on the last round results, flash would be 12 points, zero would be 11, and sea 10.
Also, just for clarification, I didn't mean to make it sound like only 3-pointers will be doing poorly next round because so many did so poorly last round, that would be retarded. What I meant to say was, there are a lot of low win % players that are not going to be earning any points next round. gonna edit this into the other post since it can be misinterpreted.
|
Why can't I pick Frozean? He's not in the drop-down menu.
|
Braavos36370 Posts
To answer you on the 50-50 10 lineup appearance guy versus the teammate that doesnt play: you realize that you are playing FPL to gain points not avoid losing points right? If a player does not play he may earn the same points as someone who goes 3-6, but its NOT the same thing because someone who doesn't play gives you no opportunity to gain points.
I understand your argument that this might lead to people going for one or two of the 8+ guys and the rest 1-pointers that don't play, but that is not without risk. High Win% people are by no means locks. Jaedong, Bisu, and Leta all did very poorly, dropping ~3 or more points in price. Last round, two very low cost players (Sea and Zero) more than tripled their cost, and several midrange guys (like Luxury, Violet, Free, etc) also had large gains. The skill is NOT in predicting who will get 9 or 10 appearances, that is easy. It's predicting the guys that will go 7-2 over 4-5.
As for diversification, I think we can all agree that Jaedong, Iris, Movie, Jangbi were farrrr to prevalent, even if you factor in that Jaedong was only one of two 10 pointers. The other three cost 2 or less, so if we had no diversification issue, they "should" be picked in somewhat equal amounts with other 1 and 2 pointers.
I'll let Abydos answer the factor question and why the number is higher than 4.6 this time. As for you getting annoyed or frustrated, remember that we are a) trying our best and b) all trying to have fun here. There is no need to get angry about something like this, especially with how we discuss it. I'm not trying to offend you and I don't see where you're getting that, obviously we're reading and considering your posts because I'm taking the time to respond to them.
This is more of an experimental round with new scoring; we're aiming to make Fantasy a skill game but not completely a skill game. There must be a factor of luck for casual owners to enjoy playing. In fact, I'd argue there must be a relatively significant factor of luck.
|
On December 04 2009 09:14 Blyf wrote:
Why can't I pick Frozean? He's not in the drop-down menu.
CuteAngel
|
3 Lions
United States3705 Posts
|
On December 04 2009 09:41 Hot_Bid wrote:
[...] you realize that you are playing FPL to gain points not avoid losing points right?
As far as I can tell I'm not an idiot and I realize that I'm playing to gain points and not lose them. But when points are zero sum, not-losing-points and gaining points amounts to the same thing. I don't know why you would say something like that except if you are operating on the assumption that I'm an idiot; maybe you're used to being more intelligent than most of the people you talk to, but I don't really appreciate the implication. FPL is ultimately a crapshoot and a silly game that doesn't indicate or mean anything, so no I'm not raging over here, it's just I'm not sure why I'm doing all this math and analysis if it's not addressed with the same level of thought I put into it.
You're correct that cost is not a predictor of results -- players will exceed their cost, but I cannot predict who those will be and those players might be the ones that I have picked just as well as they might be two random 3-pointers. That's why I'm going with the strategy that maximizes win% and likelihood of ace appearance per cost. If you want to play the "you never know" card you have to acknowledge it goes both ways. It's foolish to take a risk that statistically does not reward you unless you are confident that that player will defy the odds. Find me a 1-pointer that's playing this week with above a 33% winrate in their matchup and that isn't playing an A or S-class player and I'll put them on my team.
So consider, if you will, my attempt to build a statistically optimal team with the premises and approach I've laid out in my previous posts. I want to maximize win percentage, win opportunity, and ace match potential. So, I look at not just players' costs but their lifetime win percentage and current ELO across all 3 matchups, and ace match likelihood. This gives me a short list of candidates: Flash, Jaedong, Effort, Bisu. Calm, Sea, Leta, Fantasy, Stork, Zero, and (surprisingly) Violet are 2nd tier candidates. I think Jaedong and Effort will have more appearances and play more ace matches than Bisu will, while not having a significantly worse win rate. But oops -- I can't include Flash because there's no worthwhile team I could select worth 1 point because Oz and Khan's prices were arbitrarily inflated. So now I'm going with Leta and CJ instead of Flash + and another team and it feels like settling for 2nd best for what seems to me like no good reason. Oz would be worth 1 point if costs were accurate. Khan would be -1. Of course I'm going to post about the cost discrepancy.
Lastly, 50% being too high is just, like, your opinion, man. So no we all obviously cannot agree that Jaedong or any of those players were "far too common" the previous round. He won the golden mouse and was generally destroying everything. 50% ownership for a player like jaedong is not a problem to be solved when there are so few comparable players. There's no reason to think everyone would pick a player as long as price accurately reflects the scoring from the previous round. I checked round 3 as well, turns out jaedong had 57% back then. 39% the round after that, and flash was on 51% of teams that round. If you're trying to push percentages lower than what you said was "farrrr to prevalent" in the previous round at 34% for Jangbi and 33% for Movie, you're not going to do it by modifying the costs up one or two points, you're just annoying anal retentive players like me. I'm pretty confident that when the signups are done the number of Jaedong owners will still exceed 33%. If you want to ensure diversification there are other, more effective ways of going about it.
|
Braavos36370 Posts
On December 04 2009 11:53 integral wrote:Show nested quote +On December 04 2009 09:41 Hot_Bid wrote:
[...] you realize that you are playing FPL to gain points not avoid losing points right? As far as I can tell I'm not an idiot and I realize that I'm playing to gain points and not lose them. But when points are zero sum, not-losing-points and gaining points amounts to the same thing. I don't know why you would say something like that except if you are operating on the assumption that I'm an idiot; maybe you're used to being more intelligent than most of the people you talk to, but I don't really appreciate the implication. FPL is ultimately a crapshoot and a silly game that doesn't indicate or mean anything, so no I'm not raging over here, it's just I'm not sure why I'm doing all this math and analysis if it's not addressed with the same level of thought I put into it.
Its a crapshoot and yet its not. That's what makes fantasy fun. Every fantasy sport -- be it basketball, baseball, football, etc has huge elements of luck but everyone always does ridiculous amounts of analysis about it. I don't know where you think we're not considering things thoroughly, because we've put a tremendous amount of thought and work into the fantasy system. Just because we're not agreeing with your opinion does not mean we're not "addressing fantasy with the same level of thought" as you.
You're correct that cost is not a predictor of results -- players will exceed their cost, but I cannot predict who those will be and those players might be the ones that I have picked just as well as they might be two random 3-pointers.
This is why fantasy is fun. Some people believe that they can predict those who will exceed their cost each round. If you look at results historically, I'm sure some owners have picked several players each round that exceed their cost.
That's why I'm going with the strategy that maximizes win% and likelihood of ace appearance per cost. If you want to play the "you never know" card you have to acknowledge it goes both ways. It's foolish to take a risk that statistically does not reward you unless you are confident that that player will defy the odds. Find me a 1-pointer that's playing this week with above a 33% winrate in their matchup and that isn't playing an A or S-class player and I'll put them on my team.
Exactly this -- sometimes you are confident that a player will defy the odds. And that isn't done by looking at math, its done by watching the games.
So consider, if you will, my attempt to build a statistically optimal team with the premises and approach I've laid out in my previous posts. I want to maximize win percentage, win opportunity, and ace match potential. So, I look at not just players' costs but their lifetime win percentage and current ELO across all 3 matchups, and ace match likelihood. This gives me a short list of candidates: Flash, Jaedong, Effort, Bisu. Calm, Sea, Leta, Fantasy, Stork, Zero, and (surprisingly) Violet are 2nd tier candidates. I think Jaedong and Effort will have more appearances and play more ace matches than Bisu will, while not having a significantly worse win rate. But oops -- I can't include Flash because there's no worthwhile team I could select worth 1 point because Oz and Khan's prices were arbitrarily inflated. So now I'm going with Leta and CJ instead of Flash + and another team and it feels like settling for 2nd best for what seems to me like no good reason. Oz would be worth 1 point if costs were accurate. Khan would be -1. Of course I'm going to post about the cost discrepancy.
They aren't "arbitrarily inflated" because like I stated before, we NEVER solely based price on one previous round. It was always a combination of the last two rounds. Except this time, the "previous round" was R5 which was too long ago. Oz and Khan's value dropped after R1 -- they both performed poorly. Their prices just didn't drop as much as you wanted, because as I said, we don't base everything off one round. Having teams at drastically different prices due to a single round of different performance has not been how we historically handle it.
Lastly, 50% being too high is just, like, your opinion, man. So no we all obviously cannot agree that Jaedong or any of those players were "far too common" the previous round. He won the golden mouse and was generally destroying everything. 50% ownership for a player like jaedong is not a problem to be solved when there are so few comparable players. There's no reason to think everyone would pick a player as long as price accurately reflects the scoring from the previous round. I checked round 3 as well, turns out jaedong had 57% back then. 39% the round after that, and flash was on 51% of teams that round. If you're trying to push percentages lower than what you said was "farrrr to prevalent" in the previous round at 34% for Jangbi and 33% for Movie, you're not going to do it by modifying the costs up one or two points, you're just annoying anal retentive players like me. I'm pretty confident that when the signups are done the number of Jaedong owners will still exceed 33%. If you want to ensure diversification there are other, more effective ways of going about it.
I'm by no means saying that we'll push Jaedong ownership at below 33%. Jaedong's reputation is precisely why we needed to bump his cost up. Jaedong can be doing horribly, cost 10, and still be picked by 50% of teams. If owners are willing to pick him at 10 no matter what, there is NO benefit to having him at 6. None at all. Because that just means an even larger percentage of owners will have him, so why even play fantasy if everyone picks the same team? It becomes even more of a crapshoot. If we adjust prices to a point where people may consider NOT having Jaedong, then we've done our job with the price increase.
You say pushing prices up "only annoys anal retentive owners like you" but this is simply untrue. The fact is, changing player costs definitely impacts % ownership on teams. Modifying a lower cost player's price up will push ownership %s down. Jaedong and Bisu at 6 will put them on the majority of teams, to the point where it will be impossible to be competitive without one or both on your team. Is that what we want? The only way adjusting their prices up would be negative is if the cost is too high so that almost nobody picks them. But as you said, their reputations are so high, they will still get picked even at the higher costs.
Ask yourself, do you think Jaedong is overvalued at 8? Or that Jangbi is overvalued at 2? Are you going to put these players on your Anti-Team at these prices?
|
Braavos36370 Posts
Also, I'd like to hear your ideas about how to diversify player ownership without changing costs, because Jaedong @ 6 = autopick for just about anyone trying for Top 10.
|
|
|
|
|
|
EPT
