Dynamic Pricing Feedback Topic

contrarian's point about the issue with having the total salary be static is a great callout. I don't know why they did that to begin with — players should have the ability to have their salaries lowered by a lack of usage. Any implementation does need to be careful with this, though, because MOST players never get used and could really get down to extreme cookie levels. With enough league/team volume this wouldn't be an issue, but there are simply too many player seasons available for the market dynamics to control for this at the current amount of leagues/teams.
12/25/2020 3:30 PM
Is there a way to limit across the board how much a salary could increase or decrease? Acquire a baseline through use of a test period? The new guys need some fresh data to work with.
12/25/2020 5:20 PM
Posted by ozomatli on 12/25/2020 3:30:00 PM (view original):
contrarian's point about the issue with having the total salary be static is a great callout. I don't know why they did that to begin with — players should have the ability to have their salaries lowered by a lack of usage. Any implementation does need to be careful with this, though, because MOST players never get used and could really get down to extreme cookie levels. With enough league/team volume this wouldn't be an issue, but there are simply too many player seasons available for the market dynamics to control for this at the current amount of leagues/teams.
i wonder if it's even possible to find every player year
12/26/2020 9:40 AM
This post has a rating of , which is below the default threshold.
easily
12/26/2020 10:21 AM
This post has a rating of , which is below the default threshold.



All of this is wrong, IMO.

  1. I'm not proposing setting the base salary higher than 80M — I'm rejecting the concept of a base salary.
  2. Who is "we"? Existing players? This is the wrong framing to think about attracting new players from. To your point, this is why I recommend high cap leagues to new players.
  3. 80M is not centered, cap-wise. The midway point between the highest and lowest caps available is 147.5M.
  4. Your point about shifting caps rather than players is a short term view. Yes, that would happen (as it should), but then the market begins to self-regulate in the new (correct) state.

Without a base salary level set, there’s no way to set salaries to begin with. It’s just a shot in the dark. We have historical usage and performance data that can serve as a relative baseline, but relative to what? Each other, sure... but that doesn’t tell us anything about a price point. There has to be a set baseline value for at team/league level or there is no way to even begin setting a salary structure in place. It could work in theory only if aside from being untethered from a baseline, OLs weren’t $80m, but also floated to the point relative to where realism and all-stars were still useable, but this would be a constantly changing target as salaries would continuously drive upwards towards high cap usage of players like we already have seen. It would just create an unmanageable mess.

“We” in that context is the collective “we” of site management, existing players, and new players. Management, as they want the most engagement and re-engagement, existing players because we want new players to come back, and new players because the sim needs to make sense or there’s no point. It’s why I also disagree with shifting a new user’s first experience into higher caps. Yes, the limited player pool will help in some regards, but not enough to overcome the biggest hurdle for new players which is the lack of perceived realism. The farther from $70m you get, the more of a problem that becomes. It’s already enough of a challenge at $80m explaining why Bonds and Arod aren’t hitting 70 and 50 HRs, at $180m trying to explain why they aren’t hitting 30 and 25 is going to be a harder pill to swallow. It adds to the complexity of the matchups. It’s easier on a learning scale to see normalization work at realistic levels and then grow into quality of matchups than to take both at once. the goal here is to drive engagement across the board, but especially in retaining new owners. The closer to a realistic experience players get on their first try, the more likely they’ll try again. Seeing Bonds hit 40-50 or Maddux with 2.40 ERA is easier to grasp historically than Bonds hitting 25 and Maddux putting up a 4.35 because in their mind Bonds was already facing Johnson and Maddux etc... Maddux was facing Bonds and Gwynn etc... the historic aspect of of normalization and relative greatness hasn’t been fully exposed. And on retaining new users, I think $60m would be a better option to start out as they’d get those realistic results still, and their stars will play like stars. They may not be able to afford as many all-stars, but they can still grab 1-2, and those they get will play like all-stars. They’ll feel better about their players performance and feel like they grasped more of the game since it’s not so widely divergent from real life or expectations, while still getting to see the effects of normalization when they see guys like 1900 Donlin hit 35 HRs compared to his 10 in RL. This will allow for an easier exposure to these concepts before also subjecting them to relative greatness.

$80m is centered cap-wise around performance. A true center would be closer to $70m. $80m is at the low end of the mid-caps (<$80m=low cap, $80-120m =mid-cap, >$120m=high cap), so it’s below the true average cap point, but that’s mostly meaningless since the high cap range is extends almost infinitely in terms of functional utility). It’s also not about the average cap price, but about player performance. Which ties into the last point, as well as into the points previously on perceived value.

The dollar value itself is meaningless and it absolutely will gravitate to where we are now, it’s not just a short term view. It’s literally the equivalent of inflation. Prices increase but value hasn’t changed. There would be nothing different other than the price tag attached to the player. But leagues and players would be setup the same as now but with higher totals so that they play the same as the current caps do. The market will adjust to equilibrium between a players value and perceived value regardless of the baseline price-level. You can’t just make up numbers out of thin air or out of perceived relativity.

My point about Joss was that he wasn’t even a bargain once you factored in defense, but everyone used him anyway because they perceived relative value out of him. He was priced appropriately relative to others. If he continues to be used the way he was then his price goes up and the perception of his value changes. He was never more valuable, just perceived to be. Once that perception changes he’ll find a balance between real value and perceived value. Which solves the cookie problem, not the value problem. The current base salaries account for the value problem. But nobody likes facing the same players over and over and over again.

Dynamic pricing creates equilibrium between the real value and the perceived value. It doesn’t show a players true value. But most of your points aren’t about dynamic pricing, but just pricing in general. Wholesale increasing salaries, or selectively increasing better players salaries to limit use, doesn’t create equilibrium between real and perceived value.
12/27/2020 5:34 PM (edited)
Hey sorry I don't have time to respond to all of that but from a quick glance I don't think you understand how pricing strategy works — you understand how current pricing works

Different takes :)
12/27/2020 5:51 PM
Posted by ozomatli on 12/27/2020 5:51:00 PM (view original):
Hey sorry I don't have time to respond to all of that but from a quick glance I don't think you understand how pricing strategy works — you understand how current pricing works

Different takes :)
I get it, and it was written from mobile, so the formatting is rough to boot! Probably lots of typos, as well....

Anyway, I’d like to think that as an economist by education and a finance/pricing program/product manager by trade that I understand both practice and theory on pricing models. Maybe I’m missing a nuance you’re going for, but from I’ve seen presented, I feel like you’re missing the nuance of difference between the model you’re describing and what that means in practice. Part of that seems to be from not separating value from price and the pricing model from value aspirations.

Economic and pricing models are something I spend a ton of time on and enjoy working with, so if I am missing something I could happily spend all day working through it.
12/27/2020 7:05 PM
the chicago school

oh boy
12/27/2020 8:06 PM
my job involves working with complex pricing models and even I'm struggling to follow this discussion, I can't imagine what everyone else is thinking.

i think WIS could follow the previous dynamic pricing logic, while fixing the following issues with the first implementation:
  • Reset the player usage counter after each pricing change
  • Stop counting identical combined seasons of the same guy as separate (aka the 1902 Bill Bernhard problem)
With just those two changes, you'd have a chance of eventually reaching some sort of equilibrium (although it might take a very long while).

Even if some of the popular cookies aren't actually as valuable as the masses think, it's very bad for retaining new users when they see the same guys on everyone's team in an Open league (or any league). I remember I played one Sim Hockey open league and quit after that because there were several cookies on almost everyone's roster in that one league, and I figured what's the point of this, I guess I just have to copy everyone else to win, how boring. Maybe I was wrong about that but it's the impression I got.

12/28/2020 1:59 PM
Posted by 06gsp on 12/28/2020 1:59:00 PM (view original):
my job involves working with complex pricing models and even I'm struggling to follow this discussion, I can't imagine what everyone else is thinking.

i think WIS could follow the previous dynamic pricing logic, while fixing the following issues with the first implementation:
  • Reset the player usage counter after each pricing change
  • Stop counting identical combined seasons of the same guy as separate (aka the 1902 Bill Bernhard problem)
With just those two changes, you'd have a chance of eventually reaching some sort of equilibrium (although it might take a very long while).

Even if some of the popular cookies aren't actually as valuable as the masses think, it's very bad for retaining new users when they see the same guys on everyone's team in an Open league (or any league). I remember I played one Sim Hockey open league and quit after that because there were several cookies on almost everyone's roster in that one league, and I figured what's the point of this, I guess I just have to copy everyone else to win, how boring. Maybe I was wrong about that but it's the impression I got.

I quit hockey for the same reason. Played one time. That was enough. Hockey is my favorite sport.

On the Dynamic pricing, I'd like to see it live. When a player is selected, his salary goes up a certain percentage. Like 3% for players $5-10m. 2% 10-15, 4% $2-5M. And the rest of all players go down (maybe grouped by primary pos) by the same total amount.

Ex: Draft a $7 mil 3B. He becomes 7.21 for next owner. If there are 2000 3B in the pool, they each go down $105.

Have the free market determine the prices.

If a guy drops so low, that he becomes a bargain, fantastic. Rewards the guy that finds him.

The net result be a system where there will be a zillion more clicks on the site and a zillion more players researched than now. And that will keep new players interested. And even the old ones too.

Players would play more OL in this situation. I've been playing since March. I've gone from six months of Open league (10) to Theme Leagues (10 and growing fast in last 3months). WHY???? Because the Theme's promote searching for bargains with their various rules.

12/28/2020 4:37 PM
Posted by oldtimer59 on 12/28/2020 4:37:00 PM (view original):
Posted by 06gsp on 12/28/2020 1:59:00 PM (view original):
my job involves working with complex pricing models and even I'm struggling to follow this discussion, I can't imagine what everyone else is thinking.

i think WIS could follow the previous dynamic pricing logic, while fixing the following issues with the first implementation:
  • Reset the player usage counter after each pricing change
  • Stop counting identical combined seasons of the same guy as separate (aka the 1902 Bill Bernhard problem)
With just those two changes, you'd have a chance of eventually reaching some sort of equilibrium (although it might take a very long while).

Even if some of the popular cookies aren't actually as valuable as the masses think, it's very bad for retaining new users when they see the same guys on everyone's team in an Open league (or any league). I remember I played one Sim Hockey open league and quit after that because there were several cookies on almost everyone's roster in that one league, and I figured what's the point of this, I guess I just have to copy everyone else to win, how boring. Maybe I was wrong about that but it's the impression I got.

I quit hockey for the same reason. Played one time. That was enough. Hockey is my favorite sport.

On the Dynamic pricing, I'd like to see it live. When a player is selected, his salary goes up a certain percentage. Like 3% for players $5-10m. 2% 10-15, 4% $2-5M. And the rest of all players go down (maybe grouped by primary pos) by the same total amount.

Ex: Draft a $7 mil 3B. He becomes 7.21 for next owner. If there are 2000 3B in the pool, they each go down $105.

Have the free market determine the prices.

If a guy drops so low, that he becomes a bargain, fantastic. Rewards the guy that finds him.

The net result be a system where there will be a zillion more clicks on the site and a zillion more players researched than now. And that will keep new players interested. And even the old ones too.

Players would play more OL in this situation. I've been playing since March. I've gone from six months of Open league (10) to Theme Leagues (10 and growing fast in last 3months). WHY???? Because the Theme's promote searching for bargains with their various rules.

I’d think you love roster twist leagues. Where else do you see people playing guys like Placido Polanco or Jeff Blauser?
12/28/2020 7:09 PM
I like seeing cookies on all of my opponents' teams, personally. If all of my opponents are using the same players or very similar players, then the easier it is for me to figure out how to beat them.
12/29/2020 12:39 PM
why

how

you don know who's on that team till the season opens
12/29/2020 5:45 PM
◂ Prev 1...12|13|14|15|16|17 Next ▸
Dynamic Pricing Feedback Topic

Search Criteria

Terms of Use Customer Support Privacy Statement

© 1999-2024 WhatIfSports.com, Inc. All rights reserved. WhatIfSports is a trademark of WhatIfSports.com, Inc. SimLeague, SimMatchup and iSimNow are trademarks or registered trademarks of Electronic Arts, Inc. Used under license. The names of actual companies and products mentioned herein may be the trademarks of their respective owners.