nev,
Just for your entertainment here is a link to this team.
http://fantasygames.sportingnews.com/stratomatic/team/team_other.html?user_id=315818
Hope I posted the link correctly.
For anyone playing Live Draft leagues this is a must see for bad ideas.
Normalization MORE INFO - NEED CHARLIE
by sschu » Sun Jun 05, 2011 12:47 pm
It seems to me some player results are “normalized”, but teams are not. In order to get a team to win 110 games you need some unusually good results from players, stars and/or mid priced guys. Late season “collapses or revision to mean” are more about player normalization than team.
In my experience and anecdotally, I would theorize that there is “a lot” of normalization:
- For big $$ hitters
- For key stats (HRs, BA, 2Bs)
- For errors (player errors are usually close to their season stats, except catchers)
- Normalization for hitters seems to start around game 100
There is less normalization for:
- SPs, especially ERA and wins
- Team performance
- Any pitcher HRs allowed
It would not surprise me that there is normalization across seasons (Bonds for his last 3,000 ABs for example).
Win/lose performance over a season(s) is more about:
- Having some minimum amount of “ability” in all phases of the game, it is more difficult to win with pure bomber teams or extreme small ball teams (as compared to ATG II for example)
- Good pen/decent defense in big cap leagues assuming #1 is true
- Luck, 1 run games
I looked closely at actual results a number of times, hitters are remarkably consistent across multiple seasons, pitchers less so. Not sure what conclusion to draw from this info.
There is clearly normalization IMHO, how it works and on what parts of the game is still a mystery to me.
sschu
In my experience and anecdotally, I would theorize that there is “a lot” of normalization:
- For big $$ hitters
- For key stats (HRs, BA, 2Bs)
- For errors (player errors are usually close to their season stats, except catchers)
- Normalization for hitters seems to start around game 100
There is less normalization for:
- SPs, especially ERA and wins
- Team performance
- Any pitcher HRs allowed
It would not surprise me that there is normalization across seasons (Bonds for his last 3,000 ABs for example).
Win/lose performance over a season(s) is more about:
- Having some minimum amount of “ability” in all phases of the game, it is more difficult to win with pure bomber teams or extreme small ball teams (as compared to ATG II for example)
- Good pen/decent defense in big cap leagues assuming #1 is true
- Luck, 1 run games
I looked closely at actual results a number of times, hitters are remarkably consistent across multiple seasons, pitchers less so. Not sure what conclusion to draw from this info.
There is clearly normalization IMHO, how it works and on what parts of the game is still a mystery to me.
sschu
- sschu
- Posts: 55
- Joined: Tue Jul 03, 2012 2:34 pm
by FUDU » Sun Jun 05, 2011 1:16 pm
[quote:211033e0aa="sschu"]I looked closely at actual results a number of times, hitters are remarkably consistent across multiple seasons, pitchers less so. Not sure what conclusion to draw from this info.
There is clearly normalization IMHO, how it works and on what parts of the game is still a mystery to me.
sschu[/quote:211033e0aa]
IMO this point boils down to the fact that cards are based upon real life accomplishment/stats and in real life not one of these pitchers ever faced monster lineups as they can and do in strat.
Also the fact that BP effect is much more intensified in a dice roll game like strat as opposed to a BP effect in real life. So when you have a BP of HR rating of 1-15, and the roll falls on BP chance HR 1-8, you only need the dice roll to fall correctly to gain your 75% chance of a HR. Where as IRL lots of stuff needs to happens just right for the BP to cater to the hitter gaining a HR advantage.
There is clearly normalization IMHO, how it works and on what parts of the game is still a mystery to me.
sschu[/quote:211033e0aa]
IMO this point boils down to the fact that cards are based upon real life accomplishment/stats and in real life not one of these pitchers ever faced monster lineups as they can and do in strat.
Also the fact that BP effect is much more intensified in a dice roll game like strat as opposed to a BP effect in real life. So when you have a BP of HR rating of 1-15, and the roll falls on BP chance HR 1-8, you only need the dice roll to fall correctly to gain your 75% chance of a HR. Where as IRL lots of stuff needs to happens just right for the BP to cater to the hitter gaining a HR advantage.
- FUDU
- Posts: 55
- Joined: Tue Jul 03, 2012 2:34 pm
by sschu » Sun Jun 05, 2011 1:53 pm
[i:b47b1ec97b]IMO this point boils down to the fact that cards are based upon real life accomplishment/stats and in real life not one of these pitchers ever faced monster lineups as they can and do in strat. [/i:b47b1ec97b]
Of course the reverse is also true, Ruth never faced a rotation of big SPs everyday.
It is ATG. Perhaps the right idea is not normalization, but "dampening" the final numbers for the big $$ guys. Otherwise Kiner and Bonds would hit 125 HRs each season and if you got that pair, playoffs would be assured.
Hal should be subpoenaed to the ATG boards for some answers ... :-)
sschu
Of course the reverse is also true, Ruth never faced a rotation of big SPs everyday.
It is ATG. Perhaps the right idea is not normalization, but "dampening" the final numbers for the big $$ guys. Otherwise Kiner and Bonds would hit 125 HRs each season and if you got that pair, playoffs would be assured.
Hal should be subpoenaed to the ATG boards for some answers ... :-)
sschu
- sschu
- Posts: 55
- Joined: Tue Jul 03, 2012 2:34 pm
by 13Baseballs » Sun Jun 05, 2011 4:43 pm
[quote:3818fdae3c="13Baseballs"][quote:3818fdae3c="drfreeze49"]You know last night got me thinking about this [b:3818fdae3c]normalization issue[/b:3818fdae3c] so I checked all my [b:3818fdae3c]teams records (not really)[/b:3818fdae3c] and [b:3818fdae3c]discovered[/b:3818fdae3c] that [b:3818fdae3c]every league[/b:3818fdae3c] I ever played in all the [b:3818fdae3c]win & loss[/b:3818fdae3c] records in the [b:3818fdae3c]regular season[/b:3818fdae3c] always added up to [b:3818fdae3c]972 wins[/b:3818fdae3c] and [b:3818fdae3c]972 losses[/b:3818fdae3c]...wow now that's [b:3818fdae3c]amazing that Hal [/b:3818fdae3c]would be able to balance that out :)
And whenever I am in the [b:3818fdae3c]finals (not often)[/b:3818fdae3c]...and[b:3818fdae3c] I lose[/b:3818fdae3c]...I think to my self that [b:3818fdae3c]I finished 2nd[/b:3818fdae3c] while my [b:3818fdae3c]opponent finished next to last...
[/b:3818fdae3c]
The moral of this is...[b:3818fdae3c].Life is to short to sweat the small stuff[/b:3818fdae3c]...enjoy[/quote:3818fdae3c]
Looked at several of my old teams... and in every League... the win/loss record was 972/972... :roll: :lol:[/quote:3818fdae3c]
I looked at four teams, that just finished their season, this week... and in every League, the Won/Lost record is 972/972... I don't get it, what am I missing? Is this normal? I'm surprised that more has not been said about drfreeze49's findings. Sure would like an explanation. :shock:
And whenever I am in the [b:3818fdae3c]finals (not often)[/b:3818fdae3c]...and[b:3818fdae3c] I lose[/b:3818fdae3c]...I think to my self that [b:3818fdae3c]I finished 2nd[/b:3818fdae3c] while my [b:3818fdae3c]opponent finished next to last...
[/b:3818fdae3c]
The moral of this is...[b:3818fdae3c].Life is to short to sweat the small stuff[/b:3818fdae3c]...enjoy[/quote:3818fdae3c]
Looked at several of my old teams... and in every League... the win/loss record was 972/972... :roll: :lol:[/quote:3818fdae3c]
I looked at four teams, that just finished their season, this week... and in every League, the Won/Lost record is 972/972... I don't get it, what am I missing? Is this normal? I'm surprised that more has not been said about drfreeze49's findings. Sure would like an explanation. :shock:
- 13Baseballs
- Posts: 55
- Joined: Tue Jul 03, 2012 2:34 pm
by apolivka » Mon Jun 06, 2011 10:07 am
As usual, whenever this topic comes up, I trot out this team:
http://fantasygames.sportingnews.com/stratomatic/team/team_other.html?user_id=80041
Started 59-3
Then reeled off 100 wins in a row to finish it off.
Yes it was a theme league, and all teams played to lose, except this one.
Seems like HAL was unable to normalize....
http://fantasygames.sportingnews.com/stratomatic/team/team_other.html?user_id=80041
Started 59-3
Then reeled off 100 wins in a row to finish it off.
Yes it was a theme league, and all teams played to lose, except this one.
Seems like HAL was unable to normalize....
- apolivka
- Posts: 55
- Joined: Tue Jul 03, 2012 2:34 pm
by Palanion » Mon Jun 06, 2011 1:52 pm
[quote:62b871b394="nevdully's"][quote:62b871b394]Also the defense has the option of not throwing the ball. [/quote:62b871b394]
What????? Where???? How???? [b:62b871b394]Sounds like more black box stuff no?[/b:62b871b394][/quote:62b871b394]
It's not black box. It's just game engine. When managing a team on the CD-ROM, I am given the option on whether to throw for the basestealer on every SB attempt. If I have a high T-rating catcher and the success rate of the would-be stealer is 75% or better, I hold the throw and concede the stolen base. Why risk the error?
Since we cannot manage in-game here, I have no doubt that HAL has certain parameters set up to address this exact issue, much like we can in the CD-Rom using the Computer Manager settings.
Now, I'm not saying there is not normalization, but normalization and catcher errors are two distinctly different issues.
Is there normalization? I'm not sure. But...
There are things that we do not see here that are sometimes visible with the CD-Rom, such as board game excesses and double/triple ratings.
Again with the CD-Rom, I have had a player get a single, double, and home run in his first three ABs. And I think, "hey he is a triple short of the cycle!" However, once I actually saw this AND triple was rolled on the pitcher's card. Amazing! Except, my batter had ZERO triples that season, so HAL downgraded the triple to a double. It makes perfect sense - the game is designed with single season usage in mind - and so I was not upset by the result. I've also seen some double and triple results affected due to the ballpark.
Hey, the TSN SOM ATG record books have 90 and 100 homer seasons, 300 hits, 400 strikeouts, 140 steals, and so on. I think normalization would have prevented these excesses IMO.
As for team normalization (record-wise), I would guess that the only way the game could normalize records would be for the game engine to predict the team's W-L record before the season starts (based on players on roster, lineups, etc) and then react to this original prediction. Otherwise, what is it normalizing to? When a team starts hot (32-10, let's say), why can't the team maintain the this pace? Maybe it's anomalous. Maybe it's regression to the mean. Luck.
Lastly, regarding the whole infield defense thing.
[quote:62b871b394]1-10 on the X chart against a 4 2B [/quote:62b871b394]
If the 2b has no additional defensive respnsibility (infield in, holding a runner), then the chance of a hit from the range section is 6-20 or 30%. Beyond that, it's about e-rating. an e30 second baseman has a 38-in-216 chance of committing an error on those non-hit rolls.
Six x-chances to the 2b on a pitcher's card...
Roughly 6300 rolls over a season, half off the pitcher.
174 x-chances to the 2b over a season.
If the 2b never misses an inning, then the 4e30 would have 52 hits allowed (not factoring additional defensive responsibilities) and 30 errors. But of course there are many infield in and holding runner responsibilities that turn into hits and not error chances.
All in all, ouf of 167 chances, the 4e30 2b would likely allow roughly 82 baserunners. So maybe in the scenario above you might expect the result to have been more like 5-10, but 10 chances is a really small sample size.
What????? Where???? How???? [b:62b871b394]Sounds like more black box stuff no?[/b:62b871b394][/quote:62b871b394]
It's not black box. It's just game engine. When managing a team on the CD-ROM, I am given the option on whether to throw for the basestealer on every SB attempt. If I have a high T-rating catcher and the success rate of the would-be stealer is 75% or better, I hold the throw and concede the stolen base. Why risk the error?
Since we cannot manage in-game here, I have no doubt that HAL has certain parameters set up to address this exact issue, much like we can in the CD-Rom using the Computer Manager settings.
Now, I'm not saying there is not normalization, but normalization and catcher errors are two distinctly different issues.
Is there normalization? I'm not sure. But...
There are things that we do not see here that are sometimes visible with the CD-Rom, such as board game excesses and double/triple ratings.
Again with the CD-Rom, I have had a player get a single, double, and home run in his first three ABs. And I think, "hey he is a triple short of the cycle!" However, once I actually saw this AND triple was rolled on the pitcher's card. Amazing! Except, my batter had ZERO triples that season, so HAL downgraded the triple to a double. It makes perfect sense - the game is designed with single season usage in mind - and so I was not upset by the result. I've also seen some double and triple results affected due to the ballpark.
Hey, the TSN SOM ATG record books have 90 and 100 homer seasons, 300 hits, 400 strikeouts, 140 steals, and so on. I think normalization would have prevented these excesses IMO.
As for team normalization (record-wise), I would guess that the only way the game could normalize records would be for the game engine to predict the team's W-L record before the season starts (based on players on roster, lineups, etc) and then react to this original prediction. Otherwise, what is it normalizing to? When a team starts hot (32-10, let's say), why can't the team maintain the this pace? Maybe it's anomalous. Maybe it's regression to the mean. Luck.
Lastly, regarding the whole infield defense thing.
[quote:62b871b394]1-10 on the X chart against a 4 2B [/quote:62b871b394]
If the 2b has no additional defensive respnsibility (infield in, holding a runner), then the chance of a hit from the range section is 6-20 or 30%. Beyond that, it's about e-rating. an e30 second baseman has a 38-in-216 chance of committing an error on those non-hit rolls.
Six x-chances to the 2b on a pitcher's card...
Roughly 6300 rolls over a season, half off the pitcher.
174 x-chances to the 2b over a season.
If the 2b never misses an inning, then the 4e30 would have 52 hits allowed (not factoring additional defensive responsibilities) and 30 errors. But of course there are many infield in and holding runner responsibilities that turn into hits and not error chances.
All in all, ouf of 167 chances, the 4e30 2b would likely allow roughly 82 baserunners. So maybe in the scenario above you might expect the result to have been more like 5-10, but 10 chances is a really small sample size.
Last edited by Palanion on Mon Jun 06, 2011 3:13 pm, edited 1 time in total.
- Palanion
- Posts: 55
- Joined: Tue Jul 03, 2012 2:34 pm
by 13Baseballs » Mon Jun 06, 2011 2:50 pm
[quote:4da3a5b5d5="apolivka"]As usual, whenever this topic comes up, I trot out this team:
http://fantasygames.sportingnews.com/stratomatic/team/team_other.html?user_id=80041
Started 59-3
Then reeled off 100 wins in a row to finish it off.
Yes it was a theme league, and all teams played to lose, except this one.
Seems like HAL was unable to normalize....[/quote:4da3a5b5d5]
Looked at the numbers on this League... and again, the Win/Loss record is 972/972... how can each League always have 972 Wins and 972 Losses...
http://fantasygames.sportingnews.com/stratomatic/team/team_other.html?user_id=80041
Started 59-3
Then reeled off 100 wins in a row to finish it off.
Yes it was a theme league, and all teams played to lose, except this one.
Seems like HAL was unable to normalize....[/quote:4da3a5b5d5]
Looked at the numbers on this League... and again, the Win/Loss record is 972/972... how can each League always have 972 Wins and 972 Losses...
- 13Baseballs
- Posts: 55
- Joined: Tue Jul 03, 2012 2:34 pm
Return to Strat-O-Matic Baseball: All-Time Greats
Who is online
Users browsing this forum: No registered users and 26 guests