Player ratings

[adion]

How are player rating established? When I get a card/player it has a rating on it. Where does it come from and does it ever change.

Why is the same player rated differently on 2 different teams?

[uschi_baerchen]

Originally Posted by adion

How are player rating established? When I get a card/player it has a rating on it. Where does it come from and does it ever change.

Why is the same player rated differently on 2 different teams?

The Player Rating determines the probability that it is included in a pack. It also has a high correlation to his Quality. So if you desperately search a very good starter and have an large amount of PP look at the top of the list of the Auction house. You will find Snydergards for as little as 10.000 to Pedro Martinezes for abysmal 180.000.


A Pitcher with Rating 94 is better than a pitcher with Rating 78 - normally. It is you duty to find out, if a pitcher with Rating 78 is not even better than a pitcher with Rating 94 (or at least almost as good but much cheaper since more often on the market). I shall give you a hind: compare Noah Snydergard, who is, if I remember correctly 97, to Michael Kopech, Len Barker or Forest Whitley (who is, if I remember correctly 78), as soon as they appear in the auction house. Is it klicking? (A further tip: when you find one of these and they have a buy at once Price, make shure yoju have enough Money to follow suit, that will be a bargain regardless)


Same Player on Team with different values are same named Players from different years. Josh Donaldson is such an example. He Comes in at least 2 different Settings. But These Josh Donaldsons somehow loose the indication who they are.

[adion]

Originally Posted by uschi_baerchen

The Player Rating determines the probability that it is included in a pack. It also has a high correlation to his Quality. So if you desperately search a very good starter and have an large amount of PP look at the top of the list of the Auction house. You will find Snydergards for as little as 10.000 to Pedro Martinezes for abysmal 180.000.


A Pitcher with Rating 94 is better than a pitcher with Rating 78 - normally. It is you duty to find out, if a pitcher with Rating 78 is not even better than a pitcher with Rating 94 (or at least almost as good but much cheaper since more often on the market). I shall give you a hind: compare Noah Snydergard, who is, if I remember correctly 97, to Michael Kopech, Len Barker or Forest Whitley (who is, if I remember correctly 78), as soon as they appear in the auction house. Is it klicking? (A further tip: when you find one of these and they have a buy at once Price, make shure yoju have enough Money to follow suit, that will be a bargain regardless)


Same Player on Team with different values are same named Players from different years. Josh Donaldson is such an example. He Comes in at least 2 different Settings. But These Josh Donaldsons somehow loose the indication who they are.

I understand what you are saying but since the game is not one season long but many seasons (years) why should a players rating, good or bad, be used for all season when the card is only based on one season? It seems to me that since the game plays (simulates) many seasons, not one year, that a players card rating should be based on his total stats for all years. That way it would level the playing field over the all the years.

[Orcin]

It is not a career simulation. It is a series of one-year simulations.

[zrog2000]

This crap again. There's another identical thread in the bug report forum.

I can't believe someone complains about the "realism" of a player's card being different for different seasons, while ignoring the fact that time travel doesn't exist.

[adion]

Originally Posted by Orcin

It is not a career simulation. It is a series of one-year simulations.

That's my point. It is a series of one year simulation. So, why should a players rating, good or bad, be based on one year performance?

[Fyrestorm3]

Originally Posted by adion

That's my point. It is a series of one year simulation. So, why should a players rating, good or bad, be based on one year performance?

The best way to think of it is, pretend this was actually happening in real life. You have a time machine and you are snatching baseball players out of the past to play on your team. You have to take them from some specific point in time.

And so does the dev team. Theoretically, yes, you could come up with ratings for cards based on their entire careers, but that becomes speculatory. Because what are you going to compare them against? Pedro Martinez, Sandy Koufax, and Cy Young were all amazing pitchers, but how can you say one had better Stuff than the others? Who do you put on top, and how do you decide that? They all pitched for different lengths of time, in vastly different eras, against vastly different players, under vastly different circumstances. But pick out a single year of a player's career, and you can then compare them against all the other pitchers that year to determine what their ratings should be. And that's a far superior and more reliable method.

[adion]

Originally Posted by Fyrestorm3

The best way to think of it is, pretend this was actually happening in real life. You have a time machine and you are snatching baseball players out of the past to play on your team. You have to take them from some specific point in time.

And so does the dev team. Theoretically, yes, you could come up with ratings for cards based on their entire careers, but that becomes speculatory. Because what are you going to compare them against? Pedro Martinez, Sandy Koufax, and Cy Young were all amazing pitchers, but how can you say one had better Stuff than the others? Who do you put on top, and how do you decide that? They all pitched for different lengths of time, in vastly different eras, against vastly different players, under vastly different circumstances. But pick out a single year of a player's career, and you can then compare them against all the other pitchers that year to determine what their ratings should be. And that's a far superior and more reliable method.

I have been told that the players have been adjusted for ERAs as in OOTP. Let's take Ted Williams for example. He a batted over .400 two different years.Let's say someone has a card for one of those 2 years. Why should he be able to use that card for multiple years when Ted did not? If you use his lifetime average of .344 with some algorithm that changes his average, up or down, between his high and low years at the start of each new year in PT, it seems to me that it would be more realistic. Then you start fresh with each new version of OOTP.

[Fyrestorm3]

Originally Posted by adion

I have been told that the players have been adjusted for ERAs as in OOTP. Let's take Ted Williams for example. He a batted over .400 two different years.Let's say someone has a card for one of those 2 years. Why should he be able to use that card for multiple years when Ted did not? If you use his lifetime average of .344 with some algorithm that changes his average, up or down, between his high and low years at the start of each new year in PT, it seems to me that it would be more realistic. Then you start fresh with each new version of OOTP.

Except outcomes in OOTP are based on ratings, not stats. You can't just plug in .344 for Williams' batting average and have him hit a standard deviation every year. You'd have to determine what a .344 average translates to in terms of Contact, Eye, and Avoid K - which goes back to my point of there being no real objective way to quantify that, especially when you're trying to compare him to other players across all eras.

All of this really comes back to the ratings vs stats thing, because I don't think you really follow how OOTP determines outcomes. Even if you somehow quantified each player's career into a range of ratings, that's not going to get you the results you want because it's not a hard ratio. In layman's terms, the game determines the results of each at-bat individually based on percentages drawn from the ratings of the players involved. So two players with identical ratings will end up having different stats at the end of the year. Even if a player's ratings fluctuated, year to year, it is not only possible, but outright probable that at some point, an 80-contact season will result in a higher batting average than a 90-contact season. So really, all that extra effort would be for nothing.

[adion]

Originally Posted by Fyrestorm3

Except outcomes in OOTP are based on ratings, not stats. You can't just plug in .344 for Williams' batting average and have him hit a standard deviation every year. You'd have to determine what a .344 average translates to in terms of Contact, Eye, and Avoid K - which goes back to my point of there being no real objective way to quantify that, especially when you're trying to compare him to other players across all eras.

All of this really comes back to the ratings vs stats thing, because I don't think you really follow how OOTP determines outcomes. Even if you somehow quantified each player's career into a range of ratings, that's not going to get you the results you want because it's not a hard ratio. In layman's terms, the game determines the results of each at-bat individually based on percentages drawn from the ratings of the players involved. So two players with identical ratings will end up having different stats at the end of the year. Even if a player's ratings fluctuated, year to year, it is not only possible, but outright probable that at some point, an 80-contact season will result in a higher batting average than a 90-contact season. So really, all that extra effort would be for nothing.

Okay, but I still don't think a players one season rating should be used for all season in PT. Aren't the ratings based on stats?

[zrog2000]

Originally Posted by adion

Okay, but I still don't think a players one season rating should be used for all season in PT. Aren't the ratings based on stats?

Yes, a player's specific year card is based on the stats from that year. What is so difficult about this topic?

[Orcin]

Originally Posted by adion

That's my point. It is a series of one year simulation. So, why should a players rating, good or bad, be based on one year performance?

Because you are simulating what that player did in that year and not his whole career. For the same reason, you don't use a player's career stats when doing a one-year replay of a season. You use the stats from that year.

PT is not a career simulation. Each year is completely independent of the previous year. You are starting a fresh league each season. That how the game is designed. If you don't like it, that's ok. But that's how it works.

[Fyrestorm3]

Originally Posted by adion

Okay, but I still don't think a players one season rating should be used for all season in PT. Aren't the ratings based on stats?

Are you saying that a player's ratings should be static, based on an average of all their career seasons? Or are you saying that PT should use random seasons from a player's career year to year? I just want to be sure.

[damientheomen3]

Zrog, there's no need to get testy about this. Perfect team and the Ultimate Team model can be a tough one to wrap your head around without some prior knowledge.

Adion, think of each player/card as a snapshot of a player, normalized to fit the modern era as well as possible, for the season they represent. The game mode is meant to be less of an OOTP simulation and meant more to use OOTP's variant engine to simulate how these players would perform against other snapshot players of all eras. Thst's why things like development and injuries are turned off, and why each league likely has multiple 100 overall Mike Trouts in the WAR leaderboards. It's not really simulating baseball, more like an arcade simulation of baseball combined with a card-collecting game

[zrog2000]

Originally Posted by damientheomen3

Zrog, there's no need to get testy about this. Perfect team and the Ultimate Team model can be a tough one to wrap your head around without some prior knowledge.

Adion, think of each player/card as a snapshot of a player, normalized to fit the modern era as well as possible, for the season they represent. The game mode is meant to be less of an OOTP simulation and meant more to use OOTP's variant engine to simulate how these players would perform against other snapshot players of all eras. Thst's why things like development and injuries are turned off, and why each league likely has multiple 100 overall Mike Trouts in the WAR leaderboards. It's not really simulating baseball, more like an arcade simulation of baseball combined with a card-collecting game

We had an identical conversation about the same topic yesterday on the bug report forum. He even got a reply from a dev. His response? Start a new thread here.

http://site.ootpdevelopments.com/boa...d.php?t=296751

[lboston1]

Dude, imagine your on a playground with all of your baseball card collection in your hand. And you and your friends, who also have their baseball cards want to play an imaginary game of baseball with those cards. Now you might have a Mike Trout rookie card AND a Mike Trout card from 2018, AND a Mike Trout card from 2016 and so forth. Each year the stats he had in those years were different. Your friends might even have the same card as you and there is no rule that states that only one of you can use it for your lineup.
Now if your imaginary game was based on the stats he produced on THAT card year, and those cards statistics never change, and thus the ratings assigned to those statistics don't change either, and you had to choose your lineups based on that and what cards you had in your own collection, that's OOTP Perfect Team.

With baseball being a bit random with outcomes and statistics, some players will perform more consistently than others over multiple seasons based on the ratings you derived from your cards player statistics for that year.


Its not simulating a career of players, the only simulation is the career of the manager (you) and how you fared each season with your own collection of cards. Add in the randomness of baseball and the difference in lineups (different players around Mike Trout in the lineup, different pitchers using a different pitch, the player being ordered to steal a base or hold, etc.) and no particular outcome being purely guaranteed based on the ratings. This means that your friend using the 2018 Mike Trout card will not have exactly the same statistics as you when simulated over a season, nor will they be the same statistics as the original card.

[Mudster]

Originally Posted by zrog2000

I can't believe someone complains about the "realism" of a player's card being different for different seasons, while ignoring the fact that time travel doesn't exist.

I was just talking about this next week.....