2004 Saberosters Release Announcement

[matskralc]

draven085, Aadik, SoxWin, and myself would like to announce the impending release of our 2004 Saberosters MLB roster set. draven085 and I have done much of the grunt work, but important contributions in the areas of finances, coaches, and prospects have been/will be made by Aadik and SoxWin.

What this set will include for next Monday's anticipated release:

-Current 40 man rosters

-4 year weighted player ratings. 2003 accounts for 40% of a player's rating, 2002 for 30%, 2001 for 20%, and 2000 for 10%. This is to ensure that a player's 2003 has the most impact on his future career, but also makes sure that a player who had a fluke 2003 doesn't become a god within a
league. It also helps boost players who had an off-year last year, by accounting for past success.

-Park adjusted ratings. This is something we're not sure we've seen any MLB roster set do. Using Diamond Mind Baseball's 4 year averages for ballparks, each season used to rate a player was adjusted for his home ball park. This means that players will not receive a double weighting for their ballpark. What I mean is this: a player's statistics have already been affected by his home ball park. By using those statistics to rate a player, one artificially inflates/deflates his ratings. Those ratings are inflated/deflated again within the game by a player's home ball park. The result is Colorado hitters and Detroit pitchers dominating the leaderboards, because they benefit from their ball parks being added into their statistics twice. Our roster set seeks to eliminate this by essentially making a player's ratings as neutral as possible, by eliminating his ball park from the rating process.

-Defense, baserunning, bunting ratings. These will be based off a combination of major league scouting reports, Diamond Mind's analysis, as well as UZR and other advanced defensive metrics.

-Ballpark ratings. Derived from Diamond Mind's analysis.

-Coaches and scouts.


What this set will include by final release:

-Full minor leagues, with prospects rated according to Sickels' book, Baseball America, minor league statistics, and other sources.

-Accurate salaries and contract lengths, with 70% modifier applied.

-Photo pack.

-Depth charts, lineups, and pitching rotations.


What this set will hopefully include by final release, but is no guarantee:

-Full team and league histories.

-Full player histories. This is something I'd always thought would be extra spiffy, but is enormously time consuming. Since this is my pet, whether or not it is included in the set is entirely dependent on me. I hope to get these in there.



This set will be made available before its final completion with the intention that you guys tell us what we goofed up. Point out any odd looking ratings. Tell us if a guy looks under or overrated. Most importantly, tell us why.

The four of us are excited about making a contribution to the OOTP community. We'd just like to get the word out so you guys know what we are trying to accomplish, and what sort of help we may require. We also think that our roster set could greatly benefit from any discussion that takes place prior to any releases. That's what this thread is for. Thanks!

[Gastric ReFlux]

One question: Will this roster set really know the players, or will it just be a bunch of ratings at heart?

Other than that, good luck and thanks.

[dougaiton]

Cracking stuff. Absolutely cracking!

[LivnLegend]

Good to hear!

Knowing the people behind it, I am sure it will instantly become the standard in roster sets and the one all others are measured against.

[draven085]

Originally posted by Gastric ReFlux
One question: Will this roster set really know the players, or will it just be a bunch of ratings at heart?

Other than that, good luck and thanks.

In all seriousness we believe our set will do a much better job at 'knowing the players' than any other we've seen because of the extensive work we've done on park adjusting statistics and ratings. Personally, I'm incredibly excited to release our project to the community and give everyone the most accurate MLB roster set yet.

[Steve Kuffrey]

I can't believe with this roster set the game still won't "know" the players....

Looking forward to it, thanks for the hard work.

[CBLCardinals]

Wow!

Sounds great fellas.

I know personally how much work it takes to create a roster set.

Can't wait to see the results!!

[jgross68]

Who is Jayback?


Er, I mean, nice work guys. Sounds like you've put a lot of thought into this and I'm looking forward to seeing your finished product.

[sporr]

It's about time you guys gave something back to the community...

[timlowery]

Wow! Can't wait. This will be an awesome addition to the roster sets out there!

Thanks Guys!!!

[ctorg]

I'm looking forward to giving it a whirl, even though I don't normally do modern rosters.

[jgross68]

I originally sent this as a PM to mats and that DRaven character, but then I got to thinking, maybe somebody else will see this, think it's a decent idea and then they'll do all the hard work leaving me to sit back and claim credit. So, here goes:

I know it's awfully late to bring something like this up, and you can file this under "shut your pie hole" or "where were you three months ago, dip-****?!" but have you considered age-weighting the four-year averages? What I mean by that is, a 26-year-old that had a big 2003 is more likely to have a great 2004 than a 35-year-old that had a monster 2003. Bill James' (maybe you've heard of him?) Win Shares research has gone a long way to plotting the average player's career arc (see Win Shares, pp. 199-202, "Aging Patterns Among Great Players"), James' Year of Birth Charts in Win Shares break out players' numbers into four age categories: Young (under 25), Prime (25-29), Post-Prime (30-34), and Final (35+). I haven't done any number-crunching to see how the numbers are distributed across the entire baseball universe, but maybe there's some way to leverage that?
Maybe not. I don't know. Tell you what: maybe, if I have some time, I'll see what I can do to tweak your numbers and see if this makes any sense and/or makes an appreciable difference.

[GForce22]

Originally posted by matskralc
[B]draven085, Aadik, SoxWin, and myself would like to announce the impending release of our 2004 Saberosters MLB roster set. draven085 and I have done much of the grunt work, but important contributions in the areas of finances, coaches, and prospects have been/will be made by Aadik and SoxWin.

What this set will include for next Monday's anticipated release:

-Current 40 man rosters

-4 year weighted player ratings. 2003 accounts for 40% of a player's rating, 2002 for 30%, 2001 for 20%, and 2000 for 10%. This is to ensure that a player's 2003 has the most impact on his future career, but also makes sure that a player who had a fluke 2003 doesn't become a god within a
league. It also helps boost players who had an off-year last year, by accounting for past success.

Basically extending by a year the TotalMinors method. Glad to see you went with that approach as well. It's been the core of TotalMinors' success, IMO.

-Park adjusted ratings. This is something we're not sure we've seen any MLB roster set do.

TotalMinors factored in park factors, but I hadn't seen others do so before it. Glad you're keeping this too, again extending what TM used by a year (I had used 3-year park factors).

Well done guys. With TM not coming out this year, I certainly feel your work will prove an aqequate successor. You're certainly using the right methodology

GH

[tim_uwrf]

draven085, Aadik, SoxWin, and matskralc -

really looking forward to this set. this sounds like the best set i've seen since i started using OOTP. i love the park effects, that is something that is very important and haven't really seen in any other set. If you got:

-Full player histories. This is something I'd always thought would be extra spiffy, but is enormously time consuming. Since this is my pet, whether or not it is included in the set is entirely dependent on me. I hope to get these in there.

that in your set, that would kick major ass. That would be so awesome. Thanks for the hard work guys.

[SoxWin]

You should definitely put Mats before Aadik and myself. This set is the brainchild of Draven and Mats, I'm just happy to provide a small bit of help.

[asamford]

I'm looking forward to these rosters!!

I fully expect that Ross Gload will be properly and realistically rated (you know, B/B/B, 7/9/10, that's about where I project him).

[Crapshoot]

Originally posted by SoxWin
You should definitely put Mats before Aadik and myself. This set is the brainchild of Draven and Mats, I'm just happy to provide a small bit of help.

exactly- me and Sox just meet the Affirmative Action hiring criteria for the project.

[Mel_Ott]

Originally posted by Aadik
exactly- me and Sox just meet the Affirmative Action hiring criteria for the project.

And we're all better for it. See, it works!

[sixto]

jeez look at this OTBL congratsfest.

[SoxWin]

Elitistsaloofs