Page 1 of 3 1 2 3 LastLast
Results 1 to 15 of 41

Thread: Franchise All-Stars 2.0

  1. #1
    Join Date
    Feb 2002
    Location
    CT
    Posts
    224

    Franchise All-Stars 2.0

    Well, here’s my Franchise All-Stars 2.0 (aka Eternity League) database.
    It’s an update from a similar roster I did for Mogul ’04.
    What follows are some fairly long designer notes.

    (To be continued)
    Attached Files Attached Files
    Last edited by Casey59; 04-06-2007 at 05:35 PM. Reason: Moved text around to make it easier to find Mog. file

  2. #2
    Join Date
    Feb 2002
    Location
    CT
    Posts
    224

    Re: Franchise All-Stars 2.0

    It was a project. It was quite a project. It was quite a vast project. When I started, all my notes said Montreal Expos, not Washington Nationals. The more I went along the more I realized ways I could make it better, thus I spent a lot of time updating spreadsheet formulas and accounting for other factors, not to mention countless hours of data entry.

    Players are based on their career numbers, not just time spent with the club they are on. Last time around, I allowed the game to completely generate player projections with a .csv file, rather like what is happening on the board right now.

    This time, I crafted them myself. My algorithms included:
    -Raw career numbers
    -Historical ML averages per 700 PA since 1901 and OPS-plus
    -For pitchers it was ERA-plus, component ERA and historical norms.
    -Talent progression

    The historical ML averages for since 1901 are a significant part of the overall feel of the statistics. The average player in that time period batted .262 with a .710 OPS (once you factor out intentional walks) and a little under 13 homers per 700 PA. Offensive averages are much different in 2006, so you may see players you know with numbers that don’t quite look right, but it’s due to context. By contrast the average player today will hit over 20 homers per 700PA and have an OPS around .750. I considered basing the universe on those norms, but since I’m taking players out of the early 1900s, I figured their game conditions should be factored into the whole mess.

    Deadball era pitchers will no longer dominate based on game conditions during their careers; nor will modern pitchers suffer from home run era numbers.

    Some of the major changes since my first effort (in 2003):
    All players are 22 years old.
    All players are signed for 15 years.
    All players have a $2M salary. I played around with this using “calculate salaries” and some teams just got way too insolvent. Feel free to experiment.
    There are now 24 teams (up from 16) and previous rosters have been tinkered with to better represent careers and not one great season.
    Two pitchers were added for 17 total and 37-man rosters

    Fielding was a bear.
    Raw numbers are of some value, but they are subject to “false normalization”, that is every team has to make 24 or 27 outs (usually) and someone’s got to get the PO and A, so many times a bad fielding team will have similar stats to a good fielding team (except for the likelihood of more errors). In fact, losing teams tend to have more assists by OF and C and more DP turned (the fallout from more runners against them), making their stats look better than they really are. So while I used raw numbers (and comparisons to league averages), I also meshed them with Bill James’ ratings in the New Historical Abstract. I’m not real happy with this part of it, and I probably spent the least time fretting it, but I don’t know that there is any way to be comfortable quantifying the fielding skill of a ball player into three or four numbers. Also, I spent very little time on fielding for secondary positions. I let whatever the game generated for those positions stand, unless they were causing a player to get auto-sorted to a non-primary position.

    Speaking of auto-sort, it does some odd things. Sometimes players who are clearly weaker than an alternate choice wind up starting. It’s probably some sort of fallout from every player being so good.

    The standard qualifier was 10 major-league seasons to make this league. Some of the expansion teams haven’t had three players at a position (catchers a couple times) who have had 10-year careers, so a few 8-9 year players snuck in there. I always let a couple active players in with 8-9 years also. I didn’t really want injuries to be a problem (and these guys were usually durable, too) so most players have excellent health ratings. Ratings can go as low as 75, but most are from 80-96.

    Players were placed on the team for which they played the most games and generally were assigned to the position at which they played the most. But, there was some decision-making necessary. For instance, Rod Carew played the most games in his career at first base. However, he played more games at second base during his time with the Twins, the team he belongs on, so he is a second baseman. Need was a factor in many of these cases. Ernie Banks played more games at first base, but is assigned to shortstop, because it greater benefits the Cubs to get another first baseman in there rather than if I needed to get another shortstop in there.

    (To be continued)

  3. #3
    Join Date
    Feb 2002
    Location
    CT
    Posts
    224

    Re: Franchise All-Stars 2.0

    There were also regular issues about productivity vs. longevity. Say one player has a 17-year career at what I’ll call B+ level, while another player comes along and plays 12 years at an A- level. Well, the first guy will have his number retired and maybe get in the HOF, but the second player makes this league ahead of him. I wanted the teams to be as competitive possible and that seemed to be the way to do it. I did, however, keep the top five position players in games played for each franchise and the top five pitchers in innings and wins, no matter what (unless a guy didn’t qualify for a team, like Carew, the career leader in games played on the Angles, is still a Twin). I felt these players are identified as much as anyone with their team and most were going to get in anyway. Players with most of their careers in the 19th century were left out, because I just simply don’t know what to make of some of those numbers. I had started with 1,000 career innings as a cut off for pitchers, but that really shorted the modern closers. I allowed each team a few spots for guys with 750-1000 innings.

    Unknown stats: Caught stealing, intentional walks and GIDP are not known for much of the early part of baseball history. I estimated the best I could given what information is known, but kept an eye on historical norms. For instance, most players with unknown DP totals were probably very low, most single digits I imagine, as the players were out of the deadball era, when DP totals were low. Rather than just give all those players “8” or something like that I based it on stolen bases, feeling there is some correlation there with which to work. But even slow guys could steal 10 bases back then, so I kept most players in the teens, comparable to today’s totals. If you stole 30-39 bases/season, you got a 9, 40-49 an 8, 50-plus a 7 with others higher totals for lower SBs down to 20 GIDP for 5 or fewer steals.

    Intentional walks: Were really screwing with my numbers top to bottom. A lot of high walk guys were getting 200+ walks a season when I realized some had 100, 200 or (Barry Bonds) 600 extra walks in their raw totals. So I took them out, figuring the Mogul managers will issue intentional walks, too (hopefully at a realistic rate), so I had been, in effect, double-counting those. (So do some roster makers, so please take note). For players whose totals were unknown, I reduced them about 11%, the same total as the known players in this universe.

    I’d love to be able to somewhat normalize SB totals, because they fluctuate, based on the strategy de jour. In the deadball era, success rates were around 50%, today they are much higher. An old-time player might get 25 steals and get caught 22 times; today a player of the same relative speed might have 16 steals, but get caught six times. But I left them alone and based them on actual totals. Where data was unknown, I gave higher percentages to guys with more steals. Also the number of sacrifice bunts were left alone, so dead-ball era players tend to be better bunters. That’s another thing I’d like to take a look at, but I guess I need something to do for Version 3 (should be wrapping that up around 2011).

    Because of the league framework, you will see something of a re-contextualizing of most players. Here’s a comparison of two players; stats based on 700 PA.

    Ty Cobb: .367 BA, 16 3B, 6.5 HR
    Rickey Henderson: .279BA, 3.5 3B, 15.8 HR

    Now they look like this:
    Cobb: .322 11 3B, 11 HR
    Henderson: .292, 10 3B, 10 HR

    Is that right? Well, no one knows really. But their combination of triples and homers is roughly the same and where Cobb’s additional homers cost him some on-base points in my system, Henderson’s loss of homers gained him some.

    Some categories gave me fits, like triples. I ran many trials with different universi. After juicing up triples for most modern players I had one stretch where I was getting a couple dozen guys with 250 or 300 career triples. That wasn’t good. Back to the drawing board. Like most stats, it became a matter of finding a happy medium. Slow guys in 2006 with few triples would be slow guys in 1911 and probably not get too many triples. But, some players in 1911 were slow and hit six a year and stole 12 bases, that’s just how the game was. Joe Jackson and Sam Crawford averaged 21 triples per 700PA; they won’t be doing that in this league. Generally, deadball players lose a bunch of triples while modern guys gain them (unless they are painfully slow).

    Once you get past one season, the numbers are “Mogulized”, that is, players progress and regress. And when you start out with 864 22-year olds, sometime wacky things happen. I found you can’t start guys out too weak, because many will never get to where you expect, and likewise they can’t be too strong, otherwise that odd case of superior progression winds up being a .377 hitter with 58 homers for 15 years. But I think I’ve arrived at a pretty good starting place.

    All position players have 100 for Potential and Longevity and peaks of 28-34. Pitchers seemed to be retiring much earlier than hitters on the whole, so their peaks are 28-36. It doesn’t seem to help a whole lot. Maybe they need to be signed for 20 years or have a higher longevity. I guess in a perfect world, these would all be tailored to the individual. I hope to look at that one day, too.

    Other known issues include Vital stats not being cleared to start a game, too many outfield assists, some teams are insolvent, I’m sure there are more. Every team has A+ for Fan Loyalty. Scouting, Medical and Farm system are impossible to give an absolute rating to. There’s some sort of relativity factor there, so every time you set all the As to Fs, some of the Bs becomes As, Cs become Bs, etc., so it’s pretty scattershot. Editing the team you choose to play to all B- wouldn’t be unfair.

    I was getting too many doubles and discovered this fairly late in the whole process. So I set the Simulation setting at 80, rather than retool my offensive formulas after having done so much editing. I like to play around with those to simulate different eras. As it stands now, you’ll only get a few .300 lifetime hitters among the original players. If you want more, juice up the singles. The power is pretty good; there are usually a handful of 500 homer guys. If you want a deadball feel, change the homers to 10. It’s pretty much yours to fiddle with, do what you feel is right to enjoy it most.

    Here ends my dissertation on the Eternity Db. Have fun!!
    Last edited by Casey59; 04-07-2007 at 10:28 AM. Reason: cleanup

  4. #4
    Join Date
    Mar 2006
    Posts
    13

    Re: Franchise All-Stars 2.0

    Casey - I would like to give these a try, not real sure how to add them or install the rosters per se. Can you help? Thanks..

    Jeff

  5. #5
    Join Date
    Feb 2002
    Location
    CT
    Posts
    224

    Re: Franchise All-Stars 2.0

    I'm guessing there may be some variety based of version of the game, Operating system, etc. . . . but for me (Mogul '07; XP, it's like this):

    Click on the mogul file and choose SAVE
    You'll be asked where and choose the following:

    My Computer
    C
    Program Files
    Sports Mogul
    Baseball2007

    Save it there. when you start Mogul, Continue a Saved Game and UNIVERSE will be a choice. Choose it!

    After you choose what team you want (Commish mode) and enter your own name; SAVE AS something else. This way the original UNIVERSE will be untouched if you ever want to start with that file again.

  6. #6
    Join Date
    Mar 2006
    Posts
    13

    Re: Franchise All-Stars 2.0

    Thanks Casey!

  7. #7
    Join Date
    Feb 2002
    Location
    CT
    Posts
    224

    Re: Franchise All-Stars 2.0

    Sure. Let me know how out works out.

  8. #8
    Join Date
    Jan 2005
    Posts
    132

    Re: Franchise All-Stars 2.0

    I tried to run it but it froze my PC.

    What version was the universe saved under? Perhaps that is the problem.

  9. #9
    Join Date
    Mar 2006
    Posts
    13

    Re: Franchise All-Stars 2.0

    Worked great Casey, thanks again.

    Jeff

  10. #10
    Join Date
    Feb 2002
    Location
    CT
    Posts
    224

    Re: Franchise All-Stars 2.0

    Quote Originally Posted by GSpencer View Post
    I tried to run it but it froze my PC.

    What version was the universe saved under? Perhaps that is the problem.
    Mogul 07; 9.45

  11. #11
    Join Date
    Feb 2002
    Location
    CT
    Posts
    224

    Re: Franchise All-Stars 2.0

    Quote Originally Posted by Jliven61 View Post
    Worked great Casey, thanks again.

    Jeff
    No problem . . good to hear

  12. #12
    Join Date
    Feb 2002
    Location
    CT
    Posts
    224

    Re: Franchise All-Stars 2.0

    Over 100 views . . anyone having fun with this baby?

  13. #13
    Join Date
    Feb 2002
    Location
    CT
    Posts
    224

    Re: Franchise All-Stars 2.0

    200 downloads!
    Bueller? anyone?

  14. #14
    Join Date
    Dec 2006
    Location
    Pangea
    Posts
    6,985

    Re: Franchise All-Stars 2.0

    You Are Missing The Rockies!
    Quote Originally Posted by President View Post
    For some reason I thought rockies was a big black guy.

    I was wrong.
    Back at this dynasty thing again: Resurrecting The Rockies: 2001 Onward

  15. #15
    Join Date
    Feb 2002
    Location
    CT
    Posts
    224

    Re: Franchise All-Stars 2.0

    Ahhhh, true.
    Although I'm not sure 35-127 for a long, long time is a happy experience!!

Posting Permissions

  • You may not post new threads
  • You may not post replies
  • You may not post attachments
  • You may not edit your posts
  •