The Philosophy of OOTP Roster Creation

ZMan · 01-28-2005, 01:23 PM

I've been playing OOTP since v. 4., played High Heat during its 2001-02 glory days, even a little MVP Baseball last year, and in that time nothing has improved that experience more than the detailed roster work of some fine folks who have put hrs. into their roster projects, with we as a community reaping some great benefits for our own gaming. You know the major players...GForce, HaloFan, Pudge, Rolen17, et al.

I'm a sabermetrics admirer and am interested in tinkering around with an MLB roster set for OOTP6. So I thought I would start an item to discuss how the community's roster makers go about creating their rosters.

Some questions I have...

What sources do you use for making statistical projections? Stuff like the Major League Equivalencies? PECOTA or VORP? How do you take stat projections and translate them into OOTP's more cryptic rating system? I remember that GForce put a lot of emphasis on player age in evaluating minor leaguers. With major leaguers, do you just go with the ratings that the Lahman database spits out? If not, how do you adjust players?

HaloFan, I'm impressed by your array of techniques at evaluating minor leaguers. Anyone getting scouting reports from the Arizona Fall League for use in a roster set is one hard-working guy. Hats off. When rating players in your rosters, how much of your influence is from scouting reports and how much from stats? It sounds like the Single A players require the most subjective treatment.

I'm interested in input from everyone, since, if you're like me, it is really interesting how different people go about putting together their rosters.

halofan26 · 01-28-2005, 03:30 PM

Well when it comes to my scouting reports, about the only thing i dont get much of is carribean league reports. those are far and few between so when i get them sent to me i cherish them lol.......i get alot of scouting reports from the continental states and i get a hadful from foreign leagues across the Atlantic, like the leagues in Japan, Korea, Italy and a couple from Holland (really good pitching and a player named Greg Halman who can flat out hit). I have recently come across some reports from the Indie Leagues as well. Now when it comes to how much they influence how i do prospects and minor leaguers, if they are in the California League, Texas League, Eastern League, South Atlantic League, Pacfic Coast League, Pioneer League, Southern League and most of the International League, i use my scouting reports and i compare them with the ones i get and then i find a "majority" decision, but the real pain in the butt is when i have scouting reports on a certain player (perfect example is Jason Aspito of the Angels formerly of the White Sox and Scott Dunn of the same teams), the White Sox regret losing them yet they turn around and bash his ability to hit off-speed pitches and they say Dunn will never be able to be a starter and will have control problems. Well the Angels state that his control will get better as he goes along and that he CAN start. Well if you go by his stats, he walks over 5 batters every 9 innings, but when he gives up only 7 hits per nine innings, i say give him a shot and MAKE his control better. They dont influence my decisions totally but they do play a minor role unless i have scouted the players myself. I used to work for the Arkansas Travelers and i have made many friends and colleagues during that time so that also helps. I also have a pretty fair relationship with Tim Mead (Angels), Josie Arnold (Cardinals) and Eric Kubota (A's) and i have some "talking relationships with other teams but not as good as the execs i listed. Since OOTP uses stats to determine ratings and they use 550 ABs (i prefer 600 tho) to determine this, i have to go under each player and do the math as to what a player would do over a 550 AB season. Then here comes the hard part, i have to do more math (ask my wife, i hate math) to determine what (example) 30 HRs means in OOTP ratings which is what takes the most time when updating these sets every 3-4 weeks. What some people dont understand is that this roster set is not based solely on my opinion as i know some of these players personally and i know how they play. Like Jack Wilson of the Pirates is considered one of the top SS in the league and was awarded with his 1st all star nod (1 of many i think), but the game doesnt give him credit for his grittyness, his intelligence and scouts hammer his speed, but Jack has great ability in reading pitchers but he isnt paid to steal bases so McClendon doesnt run him often as i think he only has 19 SB's. Rick Ankiel is another player that we know of well. The game gives him anywhere from 1-1.5 stars but the game (and alot of people) dont register the fact that during his bouts of wildness and pitching troubles he was going through some horrendous family issues with his dad and the fact that during the regular season in 2000 in Sept., Tony LaRussa, Dave Duncan and Walt Jocketty wanted Rick to "adjust his delivery" so that the stress on his arm wouldnt be so bad. Well by doing this (in my opinion), they increased the stress on his arm causing his to tear his pitching elbow and cause him to end his 2002 season and part of 2003 season to surgery. They should have left his delivery alone because if its not broke, why fix it? Basically, i think i use to much info to create my set (that is what the wife says anyways lol). But i think that there is never "too much info" in a set that i want to be as perfect as possible. over 5000 unique downloads (just from my website as i dont know what other sites show) in a year have suggested to me that im doing something right lol

GForce · 01-28-2005, 05:50 PM

My main thing has always been, I don't care what scouts THINK he can do, I want to know what he's done. That's not to say I don't use scouting reports...I certainly use what Baseball America, Baseball Prospectus and John Sickels have to say and, being in the media, I have contacts all over that I can draw upon for scouting info. I've also traded e-mails with many minor leaguers over the years (a great resource...many sites have e-mail addresses for their players, many of whom get a kick out of the fact they are being done in a video game...many don't respond, but if you can get a few who do it's a nice little addition to have.)

My problem with scouts is how many of them look at "tools," as much or more than looking at what a guy has done. Tell me all you want this guy could hit 30 home runs...if he's hitting homers barely at or lower than his league average in low-A or A-ball, guess what...odds are he'll never hit 30 in the majors no matter how big, or how strong or what "Light tower power" he has. I believe in production, plain and simple, and rate guys accordingly.

I also don't believe in too much info, but there's info that just isn't too relevant, and that's the kicker. Mike talks about Jack Wilson, for example. IMO, people can talk about his grittyness or intelligence all they want, the numbers are the numbers...there's no reason to think he could steal 40 bases because he hasn't done it. If you try to extrapolate "well, this guy COULD do this and this if this and this happens" you'll make far more mistakes than you'll get right, IMO. Take a track record, compare it with track records around him, and evaluate.

My spreadsheets have been talked about a lot, and I only released them once, 2 or 3 years ago and they've evolved greatly since then. People have acted as though they must be rocket science or whatever, and perhaps my talking about them as the core of the TotalMinors franchise brought that about, but all they really are are statistical organization tools. League averages for every level, every league for the past 3 years in any significant offensive and pitching related category. Player blocks, which include places to enter how a player performed in those past 3 years and places to enter info to draw the proper league data in, and other spaces to run the calculations based on those figures and my formulas to spit out statistical ratings (both current and potential).

If you're into sabermetrics then you know the vital figures to look out for. BUt big things to consider are performance relative to the league and the age relative to the league. Is the better prospect the 22 year old performing 15% above Eastern League level or the 24 year old performing 22% over EL level...the 22 year old, even though the 24 year old had a better year, because the 24 year old is also old for the league. This is a HUGE factor and something that needs to be weighed in. This holds true for pitchers as well. Look at the vitals...K/BB ratio, K/IP...look at plate discipline for hitters (this goes WELL beyond strikeouts).

Also, know that NO formula is perfect. Expect outliers but know how to recognize them. This is the only time I really use scouting reports for a big outlook...if every scouting report says this kid will be a useful utility guy and maybe get a starting job at some point, and you have him at 25 hrs with a .380 OBP...you've probably done something wrong. I've had plenty of them and the more I did it the more I learned how to see them. You can find old feedback threads on TotalMinors and see what I mean.

* Loads of self-indulgent resume-type junk deleted *

I guess the best endorsement is that Markus enlisted me to develop the rosters for OOTP5, and I was to do the rosters for OOTP6 before other issues intervened (regrettably, of my own and others' doing). I've been fortunate to have a heck of a track record and have gotten a lot of praise along the way. I'm most thankful for that.

I'm always glad to discuss this stuff. Regrettably, time does not allow me to do TotalMinors anymore because life with a wife, 3 kids and a busy job doesn't let me work on the Opening Day deadline schedule anymore (at least not for free

). If you want to know more about methods, please let me know.

GH

halofan26 · 01-28-2005, 06:28 PM

perfect point to what i said, these players are not done by my opinion cause i know what some players can do and what some cant, but the stats are the stats and there is no denying it. NOT one person in the world would have ever thought that Beltre would hit 48 HRs this year cause he has never come close. great points tho!

GForce · 01-28-2005, 06:34 PM

Quote:

Originally Posted by halofan26

perfect point to what i said, these players are not done by my opinion cause i know what some players can do and what some cant, but the stats are the stats and there is no denying it. NOT one person in the world would have ever thought that Beltre would hit 48 HRs this year cause he has never come close. great points tho!

Sure. Beltre was one of those guys people said they were "waiting to see breakout," which was all well and good, but he hadn't yet and, in fact, he hadn't come close. I think a vital thing in roster creation is accepting these kind of things could happen and almost certainly will on occasion. If you had done a 2001 set, who would have had Melvin Mora devloping into a .330 hitter? Sometimes guys just click or fall apart. Major league GMs, scouts, analysts, whatever, don't get these right all the time. To act as though you will is just silly, and what you'll do is second-guess yourself so much that in order to change one guy to meet a new development you'll alter 10 guys you probably had right in the first place.

There's what you think and know, and there's what others think and know. Weighing all of that appropriately will bring the best results, but they won't be perfect.

GH

Raderick · 01-28-2005, 06:48 PM

GH,

I was wondering for a long time if or when you would do a follow-up to the pooular TotalMinors roster set from v5 for v6, or when v7 gets released, or are you done with it and going to committ to MVP Baseball?

GForce · 01-28-2005, 06:55 PM

Quote:

Originally Posted by Raderick

GH,

I was wondering for a long time if or when you would do a follow-up to the pooular TotalMinors roster set from v5 for v6, or when v7 gets released, or are you done with it and going to committ to MVP Baseball?

Actually, for time reasons, I'm done with modding "current" rosters. My All-Time Roster Project for OOTP is my only roster plan currently, as since it is not time sensitive I can do it at my leisure. I hope to continue to flesh it out into a very comprehensive franchise history set.

MVP was fun to do, but proved very frustrating when their player progression system was hopelessly f*'d up, despite efforts by myself and Bill Harris to fix it.

I never say never, but at least for the 2005 season, I won't be doing any current-set projects. I'll leave that to Mike

I'd love if Draven and MatsKralc ever got their Saberosters idea to fruition, but I fear the "not enough hours in a day" bug got them as well.

GH

halofan26 · 01-29-2005, 12:28 AM

Quote:

Originally Posted by GForce

Actually, for time reasons, I'm done with modding "current" rosters. My All-Time Roster Project for OOTP is my only roster plan currently, as since it is not time sensitive I can do it at my leisure. I hope to continue to flesh it out into a very comprehensive franchise history set.

MVP was fun to do, but proved very frustrating when their player progression system was hopelessly f*'d up, despite efforts by myself and Bill Harris to fix it.

I never say never, but at least for the 2005 season, I won't be doing any current-set projects. I'll leave that to Mike

I'd love if Draven and MatsKralc ever got their Saberosters idea to fruition, but I fear the "not enough hours in a day" bug got them as well.

GH

lol well as long as everyone wants it ill do it

ZMan · 01-29-2005, 02:23 AM

This is all great, great stuff.

I wonder, though, once you have a set of statistical projections for 2005, how does one begin to go about translating those statistics into the OOTP rating system? Were any studies ever done to sort of quantify this?

dougaiton · 01-29-2005, 08:57 AM

Quote:

Originally Posted by ZMan

This is all great, great stuff.

I wonder, though, once you have a set of statistical projections for 2005, how does one begin to go about translating those statistics into the OOTP rating system? Were any studies ever done to sort of quantify this?

Well, I'm coming at this from a slightly different angle...

My '69 roster set was a breeze because from '69 onwards, Retrosheet provides lefty/righty splits. I just banged these in, used the park factors to alter them to get a L/R adjusted singles/hr/doubles/bb/k stats for '69.

I then entered these in OOTP5, where the game asks for stats per 550 ABS. I then converted these into OOTP6.

You'll notice pitchers is absent from this. The conversion from OOTP5 to 6 screws up pitchers no end, so I had to tinker with some of these (but most were fine).

However, these days, its even easier. Jayzone has created a rating translator which translates ratings into stats. Working backwards from that, you should be fine. If you were creating a roster with ZiPS or PECOTA projections, that would be aces.

On defense, I use (what I think was) either Darkhorse's or Holyroller's strategy, which is using BP's Rate stats. I can't remember the formula, but it was pretty good.

Cowboys3356 · 01-29-2005, 10:46 AM

I believe one of the most important things about being a good MOD maker is that you work with a time period that you really know and love. I have stuck with the 1980s/1990s because I really know those players and how the game was played then. Scouting reports, formulas, etc, etc are all great but I think you really need to KNOW the time period you are making the file for and not just go off of other people's information.

BTW I think all of the MOD makers who have posted are good examples of what I am talking about. They are all real baseball people that have a love of the game and really know the players inside and out! They have all made great files and helped make OOTP more enjoyable for alot of people.

01-28-2005, 01:23 PM	#1
ZMan Minors (Double A) Join Date: Feb 2002 Location: Brooklyn, NY Posts: 107	The Philosophy of OOTP Roster Creation I've been playing OOTP since v. 4., played High Heat during its 2001-02 glory days, even a little MVP Baseball last year, and in that time nothing has improved that experience more than the detailed roster work of some fine folks who have put hrs. into their roster projects, with we as a community reaping some great benefits for our own gaming. You know the major players...GForce, HaloFan, Pudge, Rolen17, et al. I'm a sabermetrics admirer and am interested in tinkering around with an MLB roster set for OOTP6. So I thought I would start an item to discuss how the community's roster makers go about creating their rosters. Some questions I have... What sources do you use for making statistical projections? Stuff like the Major League Equivalencies? PECOTA or VORP? How do you take stat projections and translate them into OOTP's more cryptic rating system? I remember that GForce put a lot of emphasis on player age in evaluating minor leaguers. With major leaguers, do you just go with the ratings that the Lahman database spits out? If not, how do you adjust players? HaloFan, I'm impressed by your array of techniques at evaluating minor leaguers. Anyone getting scouting reports from the Arizona Fall League for use in a roster set is one hard-working guy. Hats off. When rating players in your rosters, how much of your influence is from scouting reports and how much from stats? It sounds like the Single A players require the most subjective treatment. I'm interested in input from everyone, since, if you're like me, it is really interesting how different people go about putting together their rosters.

01-28-2005, 03:30 PM	#2
halofan26 All Star Starter Join Date: Aug 2003 Location: Anaheim, Ca Posts: 1,491	Well when it comes to my scouting reports, about the only thing i dont get much of is carribean league reports. those are far and few between so when i get them sent to me i cherish them lol.......i get alot of scouting reports from the continental states and i get a hadful from foreign leagues across the Atlantic, like the leagues in Japan, Korea, Italy and a couple from Holland (really good pitching and a player named Greg Halman who can flat out hit). I have recently come across some reports from the Indie Leagues as well. Now when it comes to how much they influence how i do prospects and minor leaguers, if they are in the California League, Texas League, Eastern League, South Atlantic League, Pacfic Coast League, Pioneer League, Southern League and most of the International League, i use my scouting reports and i compare them with the ones i get and then i find a "majority" decision, but the real pain in the butt is when i have scouting reports on a certain player (perfect example is Jason Aspito of the Angels formerly of the White Sox and Scott Dunn of the same teams), the White Sox regret losing them yet they turn around and bash his ability to hit off-speed pitches and they say Dunn will never be able to be a starter and will have control problems. Well the Angels state that his control will get better as he goes along and that he CAN start. Well if you go by his stats, he walks over 5 batters every 9 innings, but when he gives up only 7 hits per nine innings, i say give him a shot and MAKE his control better. They dont influence my decisions totally but they do play a minor role unless i have scouted the players myself. I used to work for the Arkansas Travelers and i have made many friends and colleagues during that time so that also helps. I also have a pretty fair relationship with Tim Mead (Angels), Josie Arnold (Cardinals) and Eric Kubota (A's) and i have some "talking relationships with other teams but not as good as the execs i listed. Since OOTP uses stats to determine ratings and they use 550 ABs (i prefer 600 tho) to determine this, i have to go under each player and do the math as to what a player would do over a 550 AB season. Then here comes the hard part, i have to do more math (ask my wife, i hate math) to determine what (example) 30 HRs means in OOTP ratings which is what takes the most time when updating these sets every 3-4 weeks. What some people dont understand is that this roster set is not based solely on my opinion as i know some of these players personally and i know how they play. Like Jack Wilson of the Pirates is considered one of the top SS in the league and was awarded with his 1st all star nod (1 of many i think), but the game doesnt give him credit for his grittyness, his intelligence and scouts hammer his speed, but Jack has great ability in reading pitchers but he isnt paid to steal bases so McClendon doesnt run him often as i think he only has 19 SB's. Rick Ankiel is another player that we know of well. The game gives him anywhere from 1-1.5 stars but the game (and alot of people) dont register the fact that during his bouts of wildness and pitching troubles he was going through some horrendous family issues with his dad and the fact that during the regular season in 2000 in Sept., Tony LaRussa, Dave Duncan and Walt Jocketty wanted Rick to "adjust his delivery" so that the stress on his arm wouldnt be so bad. Well by doing this (in my opinion), they increased the stress on his arm causing his to tear his pitching elbow and cause him to end his 2002 season and part of 2003 season to surgery. They should have left his delivery alone because if its not broke, why fix it? Basically, i think i use to much info to create my set (that is what the wife says anyways lol). But i think that there is never "too much info" in a set that i want to be as perfect as possible. over 5000 unique downloads (just from my website as i dont know what other sites show) in a year have suggested to me that im doing something right lol __________________ Mike Lambert Sim League Central President Home of the Halofan roster set THE best baseball info and fan site ever The best Angels fan site "Because i got high" - Afroman

01-28-2005, 05:50 PM	#3
GForce Banned Join Date: May 2004 Posts: 3,458 Infractions: 0/2 (2)	My main thing has always been, I don't care what scouts THINK he can do, I want to know what he's done. That's not to say I don't use scouting reports...I certainly use what Baseball America, Baseball Prospectus and John Sickels have to say and, being in the media, I have contacts all over that I can draw upon for scouting info. I've also traded e-mails with many minor leaguers over the years (a great resource...many sites have e-mail addresses for their players, many of whom get a kick out of the fact they are being done in a video game...many don't respond, but if you can get a few who do it's a nice little addition to have.) My problem with scouts is how many of them look at "tools," as much or more than looking at what a guy has done. Tell me all you want this guy could hit 30 home runs...if he's hitting homers barely at or lower than his league average in low-A or A-ball, guess what...odds are he'll never hit 30 in the majors no matter how big, or how strong or what "Light tower power" he has. I believe in production, plain and simple, and rate guys accordingly. I also don't believe in too much info, but there's info that just isn't too relevant, and that's the kicker. Mike talks about Jack Wilson, for example. IMO, people can talk about his grittyness or intelligence all they want, the numbers are the numbers...there's no reason to think he could steal 40 bases because he hasn't done it. If you try to extrapolate "well, this guy COULD do this and this if this and this happens" you'll make far more mistakes than you'll get right, IMO. Take a track record, compare it with track records around him, and evaluate. My spreadsheets have been talked about a lot, and I only released them once, 2 or 3 years ago and they've evolved greatly since then. People have acted as though they must be rocket science or whatever, and perhaps my talking about them as the core of the TotalMinors franchise brought that about, but all they really are are statistical organization tools. League averages for every level, every league for the past 3 years in any significant offensive and pitching related category. Player blocks, which include places to enter how a player performed in those past 3 years and places to enter info to draw the proper league data in, and other spaces to run the calculations based on those figures and my formulas to spit out statistical ratings (both current and potential). If you're into sabermetrics then you know the vital figures to look out for. BUt big things to consider are performance relative to the league and the age relative to the league. Is the better prospect the 22 year old performing 15% above Eastern League level or the 24 year old performing 22% over EL level...the 22 year old, even though the 24 year old had a better year, because the 24 year old is also old for the league. This is a HUGE factor and something that needs to be weighed in. This holds true for pitchers as well. Look at the vitals...K/BB ratio, K/IP...look at plate discipline for hitters (this goes WELL beyond strikeouts). Also, know that NO formula is perfect. Expect outliers but know how to recognize them. This is the only time I really use scouting reports for a big outlook...if every scouting report says this kid will be a useful utility guy and maybe get a starting job at some point, and you have him at 25 hrs with a .380 OBP...you've probably done something wrong. I've had plenty of them and the more I did it the more I learned how to see them. You can find old feedback threads on TotalMinors and see what I mean. * Loads of self-indulgent resume-type junk deleted * I guess the best endorsement is that Markus enlisted me to develop the rosters for OOTP5, and I was to do the rosters for OOTP6 before other issues intervened (regrettably, of my own and others' doing). I've been fortunate to have a heck of a track record and have gotten a lot of praise along the way. I'm most thankful for that. I'm always glad to discuss this stuff. Regrettably, time does not allow me to do TotalMinors anymore because life with a wife, 3 kids and a busy job doesn't let me work on the Opening Day deadline schedule anymore (at least not for free ). If you want to know more about methods, please let me know. GH Last edited by GForce; 01-28-2005 at 06:16 PM.

01-28-2005, 06:28 PM	#4
halofan26 All Star Starter Join Date: Aug 2003 Location: Anaheim, Ca Posts: 1,491	perfect point to what i said, these players are not done by my opinion cause i know what some players can do and what some cant, but the stats are the stats and there is no denying it. NOT one person in the world would have ever thought that Beltre would hit 48 HRs this year cause he has never come close. great points tho! __________________ Mike Lambert Sim League Central President Home of the Halofan roster set THE best baseball info and fan site ever The best Angels fan site "Because i got high" - Afroman

01-28-2005, 06:48 PM	#6
Raderick Hall Of Famer Join Date: Oct 2003 Location: San Diego, California Posts: 2,737	GH, I was wondering for a long time if or when you would do a follow-up to the pooular TotalMinors roster set from v5 for v6, or when v7 gets released, or are you done with it and going to committ to MVP Baseball? __________________

01-29-2005, 02:23 AM	#9
ZMan Minors (Double A) Join Date: Feb 2002 Location: Brooklyn, NY Posts: 107	This is all great, great stuff. I wonder, though, once you have a set of statistical projections for 2005, how does one begin to go about translating those statistics into the OOTP rating system? Were any studies ever done to sort of quantify this?

01-29-2005, 10:46 AM	#11
Cowboys3356 All Star Starter Join Date: Sep 2003 Posts: 1,571	I believe one of the most important things about being a good MOD maker is that you work with a time period that you really know and love. I have stuck with the 1980s/1990s because I really know those players and how the game was played then. Scouting reports, formulas, etc, etc are all great but I think you really need to KNOW the time period you are making the file for and not just go off of other people's information. BTW I think all of the MOD makers who have posted are good examples of what I am talking about. They are all real baseball people that have a love of the game and really know the players inside and out! They have all made great files and helped make OOTP more enjoyable for alot of people.