Post by sj on Feb 2, 2011 17:18:10 GMT -5
If you're really playing your team, then you want to know this.
Okay, I might as well let the shit storm start now rather than wait until next week when the 2k11 file is posted.
The settings I'm listing below are the result of an exhaustive (almost compulsive) amount of testing. To test, I ran the file through to the end of the WS in 2k10 from several different dates in the season, imported it into 2k11, did the turnover, then ran it several seasons ahead multiple times.
First I'll give just the numbers or settings in each category, then I'll go into some explanation of what effect they have and why these settings were selected.
League --> Player Settings
Amateur Draft Talent:
Fictional - minus 100
Historical and College - zero
Talent correlations - no change to defaults of zero
Career Paths - no change to defaults of zero
Variation During Development:
Aging Randomness - minus 50
Draft Predictability - plus 100
Hall Of Fame Careers - minus 30
Talent Distribution (Per Team) [see note below]
Players with 90+ Overall - 8 (Default value is 0.5) [TMBL file versus BBM default = +7.5]
Players with 80-89 Overall - 19 (Default value is 6.5) [TMBL file versus BBM default = +12.5]
Players with 90+ Peak - 19 (Default value is 2.6) [TMBL file versus BBM default = +16.4]
Peak 80-89 Peak - 31 (Default value is 20.7) [TMBL file versus BBM default = +10.3]
NOTE: About the Talent Distributions (Per Team) settings.
1 - Multiply the number in each section by 26 teams to get the total number of players at each rating level. For example, "Players with 90+ Overall" set at 8 means there will be 208 total players in the file rated 90+. Using the BBM default settings for this category would result in a total of 13 players with this rating.
On average, we're set to allow talent levels roughly sixteen times higher (plus one thousand six hundred percent) than the default settings. The default settings were designed to keep talent at realistic levels, so this means we're allowing talent levels 1600% higher than real were players in 1987.
2 - Despite the wording used, these players are not evenly distributed per team. "Per Team" simply means that the base number (like 8 in the above example) will be multiplied by the total number of teams in the league. More teams means more players at these levels, less teams means fewer.
League --> League Settings
Difficulty Levels - all set to coach
Financial Engine
League Revenue - setting not yet finalized
Salary Demands - setting not yet finalized
Equalize Cities - On
Injuries
Injury Frequency - minus 30
Injury Severity - minus 30
Medical Staff Impact - zero
Team Expenditures (what we call expense spending)
Farm set to Very Good (85)
Scouting set to Excellent (100)
Medical set to Very Good (85)
These are absolutely equal for all teams.
Note that BBM charges teams for equalized expenses. For TMBL the initial cost is around 5.1M per category per team, total of around15.3M per team. You'll need to take the expense costs into account when signing players on this version. All teams will be paying this.
Give Days Off - simming 2+ days (this still doesn't work anyway)
Artificial Intelligence (GM) [remember, you're the owners, not the GM or manager]
Promotion to MLB minus 10
Promotion to AAA AA, A zero
Contract Renewals - zero
Contract Lengths - minus 30
Trade Frequency - minus 100
Mimic Historical Drafts - minus 100
Artificial Intelligence (Manager) [remember, you're the owners, not the GM or manager]
Platooning - zero
Stats (vs. scouting) - plus 50
Explanations of categories not previously defined[/u]
Aging Randomness - This determines if a player will have a predictable career or will change unexpectedly. Minus 50 favors a more predictable career without eliminating the small chance that he could crash.
Draft predictability - How accurate career predictions for players in the amateur draft are. It's at plus 100 because anyone can sim 6 - 8 seasons ahead to see how they turn out, and there's no reason to give more advantages to guys that have more time to spend doing this.
Hall Of Fame Careers - Determines if a superstar should always perform like a superstar, but that's at the expense of players with lower ratings who will then always be bad. The setting of minus 30 mixes things up a little bit. Not a lot because all our players have star and superstar ratings, but enough so that it's possible the predicted .275 BA hitter might actually hit above .250 a little more often.
Difficulty Levels - Pertain to finances and play by play. The financial part works, the play by play (AI manager) doesn't.
Financial Engine - How much teams make, how much players demand to be paid.
Injuries - How often a player is randomly hurt, and how long a typical injury will last
Artificial Intelligence (GM) Promotions - Remember when most of your minor league players ended up in R ball after the file turnover when they were all set to minus 100? Zero lowers the number of guys on your 40 man roster getting moved that far down in the minors.
Give Days Off - Excellent concept that the AI would rest a player who is tired so he doesn't get hurt. Unfortunately, it doesn't work. We set it anyway, and pretend it does. Maybe by 2k14 it will.
Artificial Intelligence (GM) Contracts, etc - Mostly not used in an all human league. Renewal lengths have a limited effect on how many years a player will ask for when it's time to re-sign him. The setting of minus 30 results in somewhat fewer 35 year old players demanding 7 year contracts.
Artificial Intelligence (Manager) Platooning - zero. At zero, the AI manager will neither favor a righty or lefty hitter versus a righty or lefty pitcher when playing a substitute or pinch hitting for you. Note that, although we like to pretend the AI uses players in the order we have them on the bench, it really doesn't.
Artificial Intelligence (Manager) Stats (vs. scouting) - plus 50. At plus 50 the AI manager is 50% more likely to use a player who is having a good season over a player who has great ratings but isn't doing well.
This is extremely important in TMBL because all of this league's players have what the AI would consider to be great ratings. That's how an AI lineup shift (like happens at file turnover time after releases) can result in your leadoff hitter being a player with great steal potential but only a predicted .280 BA, while a .340 predicted BA player sits on the bench. To the AI, .280 is quite good. To TMBL - not so much.
More information about talent rating limit settings[/u]
If you're one of the rare owners who has read this far, please read this last part.
I would appreciate it if whining or attacks or threats to quit over re-rated players is kept to a minimum.
I'm very aware that a percentage of the players in the file will be downgraded. Not a lot of them, but some.
I tested higher settings for the talent limits to try avoiding player downgrading, and they did work. However the side effect to that is talent inflation got even worse. That predicted .335 guy you just signed to a high dollar 7 year contract? Not so great four or five seasons from now when every team has 10 hitters predicted to hit .340+.
Am I exaggerating?
At the beginning of this season there were a total of 154 hitters predicted for a .320+ BA. As of right now, sim date October 20, there are 175. That's a 14% increase in less than one full season, and that's with draft talent set back to normal levels. Predicted .340+? April 1 had 64. October 20 has 79. Almost 25% higher after less than 7 sim months.
This file is the reason I'm not happy with TMBL. It's also the reason that a lot of smart owners are making less than smart decisions. This massive inflation took place, best as I can tell, between 1979 and now. Sometime around 1979 the amateur draft talent setting was raised to plus 100. Soon after, players predicted to hit .300 weren't very good anymore. Mid season 1979 there were 117 total hitters predicted to hit .300+. Now there are 282 of them. That's a 240 percent increase in less than 7 full seasons. Some of this was also due to the equalized A+ farm ranks, but most is from the influx of hyper inflated amateurs.
A lot of teams ended up with very long and very expensive contracts for what became below average players. This wasn't people making stupid mistakes when signing players, this was a pair of co-commissioners who didn't understand the long term effects who decided to make the draft more exciting by setting new talent to maximum levels in a game already infamous for a "talent creep".
As commissioner again, for the moment at least, it's my intention to ensure that this talent inflation does not continue to get worse. There would be too much crying if I tried to get the levels back down again so I'm not doing that now, but I can try to make sure they don't keep going up either.
Okay, I might as well let the shit storm start now rather than wait until next week when the 2k11 file is posted.
The settings I'm listing below are the result of an exhaustive (almost compulsive) amount of testing. To test, I ran the file through to the end of the WS in 2k10 from several different dates in the season, imported it into 2k11, did the turnover, then ran it several seasons ahead multiple times.
First I'll give just the numbers or settings in each category, then I'll go into some explanation of what effect they have and why these settings were selected.
League --> Player Settings
Amateur Draft Talent:
Fictional - minus 100
Historical and College - zero
Talent correlations - no change to defaults of zero
Career Paths - no change to defaults of zero
Variation During Development:
Aging Randomness - minus 50
Draft Predictability - plus 100
Hall Of Fame Careers - minus 30
Talent Distribution (Per Team) [see note below]
Players with 90+ Overall - 8 (Default value is 0.5) [TMBL file versus BBM default = +7.5]
Players with 80-89 Overall - 19 (Default value is 6.5) [TMBL file versus BBM default = +12.5]
Players with 90+ Peak - 19 (Default value is 2.6) [TMBL file versus BBM default = +16.4]
Peak 80-89 Peak - 31 (Default value is 20.7) [TMBL file versus BBM default = +10.3]
NOTE: About the Talent Distributions (Per Team) settings.
1 - Multiply the number in each section by 26 teams to get the total number of players at each rating level. For example, "Players with 90+ Overall" set at 8 means there will be 208 total players in the file rated 90+. Using the BBM default settings for this category would result in a total of 13 players with this rating.
On average, we're set to allow talent levels roughly sixteen times higher (plus one thousand six hundred percent) than the default settings. The default settings were designed to keep talent at realistic levels, so this means we're allowing talent levels 1600% higher than real were players in 1987.
2 - Despite the wording used, these players are not evenly distributed per team. "Per Team" simply means that the base number (like 8 in the above example) will be multiplied by the total number of teams in the league. More teams means more players at these levels, less teams means fewer.
League --> League Settings
Difficulty Levels - all set to coach
Financial Engine
League Revenue - setting not yet finalized
Salary Demands - setting not yet finalized
Equalize Cities - On
Injuries
Injury Frequency - minus 30
Injury Severity - minus 30
Medical Staff Impact - zero
Team Expenditures (what we call expense spending)
Farm set to Very Good (85)
Scouting set to Excellent (100)
Medical set to Very Good (85)
These are absolutely equal for all teams.
Note that BBM charges teams for equalized expenses. For TMBL the initial cost is around 5.1M per category per team, total of around15.3M per team. You'll need to take the expense costs into account when signing players on this version. All teams will be paying this.
Give Days Off - simming 2+ days (this still doesn't work anyway)
Artificial Intelligence (GM) [remember, you're the owners, not the GM or manager]
Promotion to MLB minus 10
Promotion to AAA AA, A zero
Contract Renewals - zero
Contract Lengths - minus 30
Trade Frequency - minus 100
Mimic Historical Drafts - minus 100
Artificial Intelligence (Manager) [remember, you're the owners, not the GM or manager]
Platooning - zero
Stats (vs. scouting) - plus 50
Explanations of categories not previously defined[/u]
Aging Randomness - This determines if a player will have a predictable career or will change unexpectedly. Minus 50 favors a more predictable career without eliminating the small chance that he could crash.
Draft predictability - How accurate career predictions for players in the amateur draft are. It's at plus 100 because anyone can sim 6 - 8 seasons ahead to see how they turn out, and there's no reason to give more advantages to guys that have more time to spend doing this.
Hall Of Fame Careers - Determines if a superstar should always perform like a superstar, but that's at the expense of players with lower ratings who will then always be bad. The setting of minus 30 mixes things up a little bit. Not a lot because all our players have star and superstar ratings, but enough so that it's possible the predicted .275 BA hitter might actually hit above .250 a little more often.
Difficulty Levels - Pertain to finances and play by play. The financial part works, the play by play (AI manager) doesn't.
Financial Engine - How much teams make, how much players demand to be paid.
Injuries - How often a player is randomly hurt, and how long a typical injury will last
Artificial Intelligence (GM) Promotions - Remember when most of your minor league players ended up in R ball after the file turnover when they were all set to minus 100? Zero lowers the number of guys on your 40 man roster getting moved that far down in the minors.
Give Days Off - Excellent concept that the AI would rest a player who is tired so he doesn't get hurt. Unfortunately, it doesn't work. We set it anyway, and pretend it does. Maybe by 2k14 it will.
Artificial Intelligence (GM) Contracts, etc - Mostly not used in an all human league. Renewal lengths have a limited effect on how many years a player will ask for when it's time to re-sign him. The setting of minus 30 results in somewhat fewer 35 year old players demanding 7 year contracts.
Artificial Intelligence (Manager) Platooning - zero. At zero, the AI manager will neither favor a righty or lefty hitter versus a righty or lefty pitcher when playing a substitute or pinch hitting for you. Note that, although we like to pretend the AI uses players in the order we have them on the bench, it really doesn't.
Artificial Intelligence (Manager) Stats (vs. scouting) - plus 50. At plus 50 the AI manager is 50% more likely to use a player who is having a good season over a player who has great ratings but isn't doing well.
This is extremely important in TMBL because all of this league's players have what the AI would consider to be great ratings. That's how an AI lineup shift (like happens at file turnover time after releases) can result in your leadoff hitter being a player with great steal potential but only a predicted .280 BA, while a .340 predicted BA player sits on the bench. To the AI, .280 is quite good. To TMBL - not so much.
More information about talent rating limit settings[/u]
If you're one of the rare owners who has read this far, please read this last part.
I would appreciate it if whining or attacks or threats to quit over re-rated players is kept to a minimum.
I'm very aware that a percentage of the players in the file will be downgraded. Not a lot of them, but some.
I tested higher settings for the talent limits to try avoiding player downgrading, and they did work. However the side effect to that is talent inflation got even worse. That predicted .335 guy you just signed to a high dollar 7 year contract? Not so great four or five seasons from now when every team has 10 hitters predicted to hit .340+.
Am I exaggerating?
At the beginning of this season there were a total of 154 hitters predicted for a .320+ BA. As of right now, sim date October 20, there are 175. That's a 14% increase in less than one full season, and that's with draft talent set back to normal levels. Predicted .340+? April 1 had 64. October 20 has 79. Almost 25% higher after less than 7 sim months.
This file is the reason I'm not happy with TMBL. It's also the reason that a lot of smart owners are making less than smart decisions. This massive inflation took place, best as I can tell, between 1979 and now. Sometime around 1979 the amateur draft talent setting was raised to plus 100. Soon after, players predicted to hit .300 weren't very good anymore. Mid season 1979 there were 117 total hitters predicted to hit .300+. Now there are 282 of them. That's a 240 percent increase in less than 7 full seasons. Some of this was also due to the equalized A+ farm ranks, but most is from the influx of hyper inflated amateurs.
A lot of teams ended up with very long and very expensive contracts for what became below average players. This wasn't people making stupid mistakes when signing players, this was a pair of co-commissioners who didn't understand the long term effects who decided to make the draft more exciting by setting new talent to maximum levels in a game already infamous for a "talent creep".
As commissioner again, for the moment at least, it's my intention to ensure that this talent inflation does not continue to get worse. There would be too much crying if I tried to get the levels back down again so I'm not doing that now, but I can try to make sure they don't keep going up either.