2011 Projection-Based Rosters: Methods, Formulas, & Discussion ...

[Knight165]

Re: 2011 Projection-Based Rosters: Methods, Formulas, & Discussion ...

That's the thing, for as many people as there are that play years and years of a franchise, I bet there are more than that who will only play one season, or even part of a season, and then start over... Or like me, that will move on to '11.

I think the franchise guys are much more vocal because they are constantly telling people about there franchise, and giving progress reports, or have a full thread dedicated to it.

But it's the majority I believe, that only play one year max, and for them, progression means jack. What we need is how the player plays today.

So if you're only playing one year....what do you need prospects at all for then? They aren't playing in the MLB....and while their fielding/speed skills may be at MLB levels...the rest of their play is certainly not. So are you saying you want to see a AA player with an 80 fielding rating, but his proper 40 batting ratings?....because that is what his MLB level is likely to be....so that you can bring him up and field with him and have him hit .220?
I'm not trying to dissuade anyone from anything, but to me it makes ZERO sense to make a minor league roster for anyone EXCEPT those who are playing multiple seasons. That's what the minors are about. Development.
Now...if you are making only a minor league roster to play just minor league games...sure....then I could see rating the players at their current levels as if they were going to play as you see in the game at the MLB level....but if everyone is say...halved at the minors anyway....isn't it really the same thing?


M.K.
Knight165

[TripleThreat1973]

Re: 2011 Projection-Based Rosters: Methods, Formulas, & Discussion ...

So if you're only playing one year....what do you need prospects at all for then? They aren't playing in the MLB....and while their fielding/speed skills may be at MLB levels...the rest of their play is certainly not.

Great question, let me answer ...

[1] There are some "minor leaguers" that are "MLB ready" and will be in MLB lineups at opening day. They are rookies.

[2] There are some minor leaguers that could play in MLB if they weren't kept in the minors to delay arbitration and service time and/or if they were not blocked by a veteran player with a big contract.

[3] You can now trade for prospects you can use, or trade valuable prospects for a veteran/star you can use.

With "minor league" rosters using projection/MLE's there are approximately 10 MiLBers that could "start" in the MLB. Some people will start them, some won't. But, the option is there. I'm talking about guys proected to be 60 CON 70 POW type guys. Gutierrez (CIN), Belt (SFG), Trout (LAA, 70 CON 30 POW) ... there are some corner mashers in AAA just waiting to play in MLB.

Some people like to play the minor league games.

If a starter gets hurt and you have to call up a minor leaguer, he won't be horrific, and you won't have to use "Craig Counsell".

In the end, it's not a make or break because, by far, most of the MiLBers will still be in the 40s, and it won't matter for 99% of them.

I don't want to make it out to a bigger deal than it is. I'm editing MiLB'ers ratings because I have projections for everyone from MLB through R- ball. If I didn't, I'd just do MLB.

Edit: The issue is not wanting to field with an 80 fielder while his batting is 40. The issue are the MiLBers that are 60-70 hitters, that have 30-40 fielding. Adam Dunn is the worst fielder in MLB history (in the OF), with -42 fielding runs in a single year. That would be a 30-40 in the game. 65 to 68 is basically "league average". So bringing the MiLBers (whose fielding rating are way low) up to 55's across the board, just makes them "replacement level fielders".

It may all be beside the point, because I notice that the guys that are MLB ready (Trout, Johnson, Belt, Gutierrez, etc) do have fielding ratings that are already above the "replacement level".

Again, I don't think it is something that is going to be make or break, unless for some reason, someone wants to start Hicks over Revere or Span.

[TripleThreat1973]

Re: 2011 Projection-Based Rosters: Methods, Formulas, & Discussion ...

I think I've found something very useful for individual pitch values for individual pitchers. I'll be editing those (if needed) in these rosters as well.

MLB fielding will also be edited. C&P into spreadsheets is a beautiful thing.

Will be done with AL bEAST tonight, Start AL Central as well.

[Knight165]

Re: 2011 Projection-Based Rosters: Methods, Formulas, & Discussion ...

Great question, let me answer ...

[1] There are some "minor leaguers" that are "MLB ready" and will be in MLB lineups at opening day. They are rookies.

[2] There are some minor leaguers that could play in MLB if they weren't kept in the minors to delay arbitration and service time and/or if they were not blocked by a veteran player with a big contract.

[3] You can now trade for prospects you can use, or trade valuable prospects for a veteran/star you can use.

With "minor league" rosters using projection/MLE's there are approximately 10 MiLBers that could "start" in the MLB. Some people will start them, some won't. But, the option is there. I'm talking about guys proected to be 60 CON 70 POW type guys. Gutierrez (CIN), Belt (SFG), Trout (LAA, 70 CON 30 POW) ... there are some corner mashers in AAA just waiting to play in MLB.

Some people like to play the minor league games.

If a starter gets hurt and you have to call up a minor leaguer, he won't be horrific, and you won't have to use "Craig Counsell".

In the end, it's not a make or break because, by far, most of the MiLBers will still be in the 40s, and it won't matter for 99% of them.

I don't want to make it out to a bigger deal than it is. I'm editing MiLB'ers ratings because I have projections for everyone from MLB through R- ball. If I didn't, I'd just do MLB.

Edit: The issue is not wanting to field with an 80 fielder while his batting is 40. The issue are the MiLBers that are 60-70 hitters, that have 30-40 fielding. Adam Dunn is the worst fielder in MLB history (in the OF), with -42 fielding runs in a single year. That would be a 30-40 in the game. 65 to 68 is basically "league average". So bringing the MiLBers (whose fielding rating are way low) up to 55's across the board, just makes them "replacement level fielders".

It may all be beside the point, because I notice that the guys that are MLB ready (Trout, Johnson, Belt, Gutierrez, etc) do have fielding ratings that are already above the "replacement level".

Again, I don't think it is something that is going to be make or break, unless for some reason, someone wants to start Hicks over Revere or Span.

Sure...I understand a few of them...heck...I know guys want to see different guys in their season play...I'm one of 'em!
If he meant just a sprinkling of the upper tier players...sure I get that. I just wouldn't understand doing it for guys who are 4 years away....(still have the good glove though )....but would be embarrassed in the MLB at the plate.
Like I said....and something I'll be e-mailing Kolbe about.....there absolutely should be some set attributes which should be pretty constant until age regression.....fielding being one of them....

...and I'm sorry....I absolutely do not want to hi-j your thread....like I said, I just wanted to let you know where I was at when creating those guys. Mine is definitely a "in the future" set...always has been.

I PM'ed you about the pitching.

M.K.
Knight165

[TripleThreat1973]

Re: 2011 Projection-Based Rosters: Methods, Formulas, & Discussion ...

Sure...I understand a few of them...heck...I know guys want to see different guys in their season play...I'm one of 'em!
If he meant just a sprinkling of the upper tier players...sure I get that. I just wouldn't understand doing it for guys who are 4 years away....(still have the good glove though )....but would be embarrassed in the MLB at the plate.
Like I said....and something I'll be e-mailing Kolbe about.....there absolutely should be some set attributes which should be pretty constant until age regression.....fielding being one of them....

...and I'm sorry....I absolutely do not want to hi-j your thread....like I said, I just wanted to let you know where I was at when creating those guys. Mine is definitely a "in the future" set...always has been.

I PM'ed you about the pitching.

M.K.
Knight165

Not a hijack at all. If anything it was me jumping the gun and seeing low minors guys with incredibly low fielding ratings, and assuming that all minor leaguers were like that.

They aren't.

The guys that are "mlb ready" already have fielding stats higher than the "replacement level". So, as you said, your rosters are "in the future" currently.

The confusion started because of something I said. As I went further through the process, there were maybe 4 or 5 guys (total) that "might" have a lower-end MLB bat but far below the replacement level fielding. These are often "power" guys that lack any sort of discipline or contact.

[emath]

Re: 2011 Projection-Based Rosters: Methods, Formulas, & Discussion ...

Yeah Knight, I should have explained better. I meant guys who have a shot of making the team at some point this year.

Heck for me, it's the top guys who might even garner a call up in September.

Plus, I am not always about perfect realism. If I have a MLer who is stinking it up, I might call up my best prospect at that position to see what he can do.

[TripleThreat1973]

Re: 2011 Projection-Based Rosters: Methods, Formulas, & Discussion ...

Here's an example ...


I just edited the Seattle mariners with the 2011 projection-based ratings.

Justin Smoak and Dustin Ackley had fielding ratings in the 30s. These twoi are major league players. They may spend some time in AAA this year, but even in 2010 they were major league players.

The fielding rating are too far from the average MLB fielder, and it also affects their overall rating, which could make them far too easy to obtain through trades.

I have been just "adding 15" to each of the MiLB'ers fielding ratings (so people can reduce it if they want), but for Smoak and Ackley I "added 30".

There are about 12-15 MiLB'ers who are going to be opening day starters, that need their ratings "adjusted" to the MLB level. There are about 45-50 that have a "shot" at making the 25-man team.

[TripleThreat1973]

Re: 2011 Projection-Based Rosters: Methods, Formulas, & Discussion ...

Update ...

I went back and combined 5 projection systems (instead of just using Zips ... and Oliver/THT), to get as accurate of a baseline as I could. By combining ZiPS, Oliver, Marcel, Guru, and CAIRO, I have a very good balance, and fair treatment of minor leaguers. If I had to do it all over again, I would just combine Oliver and Cairo, since they both project for 2000+ players, release them all at the same time, release them in spreadsheet form, and balance each other. But, i kept finding projections at different times, and blah, blah.

I ran into several problems when entering the ratings.

Problem #1: There's not an updated OSFMv4
-----------------------------------------------

In order to use the original OSFM rosters, one would have to go back through and make a season's worth of transactions. Ugh.

Problem #2: Unnanounced overhauls
-----------------------------------------

This is where someone puts up an "updated roster", but fails to mention that they changed a bunch of the ratings, and you don;t find the changed ratings until you get through 1/3 of the entering process.

Problem #3: Bogus edits
-----------------------------------

This has been the big problem. I downloaded a roster set updated to Feb 5, so i would not have to go through and make all the trades, etc. I downloaded 6 rosters, and then looked at certain guys to see if they were OSFM or not. These were. HOWEVER, what was not advertised is that someone went through and edited the Twins and made them the "Super Twins", and then someone else, or their little brother, went through and made a bunch of ridiculous edits (seriously, Clay Buchholz all pitches 90+ based on 172 IP of BABIP inflated performance in 2011?). It's **** like this that pisses me off, because the bogus edits are not advertised in the description.

Not only that but the edits are absolutely stupid and completely inaccurate. The Twins, for example, are plagued with DH'ds disguised as outfielders. Cuddyer, Kubel, etc are terrible fielders. The roster I downloaded has them all jacked up to high 80s or 90s on all 4 fielding rating. Their pitchers' pitch ratings were even worse. Evidently Kevin Slowey and Baker have surpassed Lee and Halladay. Ridiculous, simply ridiculous.

Anyway, so rather than go back and use OSFMv2 or v4 and make all the trades, and then go bakc and re-enter all of the ratings I had already entered. I cam up witjh a way to calculate control and break based on a pitcher's stats and their pitch value ratings at Fangraphs (which are based on outcomes and movement from pitch fx data).

All AL batters are entered, and I REALLY like what I see so far. There aqre some young guys that are "ready" to play and there are some "old guys" that need to stop surviving on reputation from seasons past.

The rosters won;t be done when i wanted them to, namely because I decided to add more porjection systems, which takes a while (but will be more accurate), and because the rosters I started editing were "FUBAR'd" at the ML level. Of course they didn't edit MiLBers (which I checked), but screwed up the MLB'ers using whatever half-assed method they did to drastically inflate the ratings of their favorite teams and/or favorite players.

I still cannot figure out why anyone would edit rosters in the way. The Twins outfielders all 90+ fielding. That kills me.

[TartanElk]

Re: 2011 Projection-Based Rosters: Methods, Formulas, & Discussion ...

I'm finding that a lot of these rosters aren't exactly spot on anyway. So many trades and things are missing.

It's also hard to find ones that use OSFM any version. How are you getting lucky to find those rosters?

[Knight165]

Re: 2011 Projection-Based Rosters: Methods, Formulas, & Discussion ...

It's the weak point of the Vault IMO.
You get a bunch of guys....most absurd...but even some..who start out with all good intentions, but just don't have the dedication to follow through.
You already know...it's a TON of work. Guys start....figure..."that's good enough".....then post 'em up. You don't find out until much later.

I usually do an end of season set...which would have helped a bit...but I was so busy this year...it was impossible.
Sorry about that.

I said it before...and I'll say it again...I'm thoroughly impressed with your research.
You wouldn't happen to be a fireman also, would you?...You seem to have more time than me!

M.K.
Knight165

[TripleThreat1973]

Re: 2011 Projection-Based Rosters: Methods, Formulas, & Discussion ...

I'm a junior high principal and pitching coach who hatched this idea over Xmas break.

Now school is full force and the HS and 10U pitchers are practicing and the time is no longer there. I'm able to do this bc I don't sleep

[TripleThreat1973]

Re: 2011 Projection-Based Rosters: Methods, Formulas, & Discussion ...

Fangraphs has stats regarding [1] pitch repitoire (velocity, % thrown, etc) and pitch value (can convert to movement, etc), as well as, fielding information (can use as many years as you want) that can be converted to ratings using "Bahnzoesque" formulas.

At this point, the strong temptation is just to take all of the work put in, and put it into MLB11 rosters, since I know I'm gonna buy that game no matter what I've said previously (grin) ... instead of putting out what may be accurate, but incomplete rosters for MLB10 and the 2011 season.

The time invested would be the same, and I'm not a real big fan of doing the same thing twice. I just don't have that kind of time.

Combined with rosters coming out for OOTP, Baseball Mogul, and Diamondmind, the "research" for minor leaguers is already done for you, you just have to take the information and apply it to "league average" ratings for The Show.

I feel like I have really good formulas, and background stats, and all of that jazz ... just trying to fight the urge to aviod doing the same process twice. But this activity (2011 rosters for MLB10) has really allowed me to work out the kinks ... which you really see as you are entering the ratings into the game and see how different players compare.

Dan Johnson for the Rays should be pretty darn good, btw. They won't miss Pena.

[TripleThreat1973]

Re: 2011 Projection-Based Rosters: Methods, Formulas, & Discussion ...

The other temptation is to learn how to use Microsoft Access, and make my own pitch fx database, like Bahnzo did, and then use the formulas for control and break that are in his pitcher tools.

Just have to decide whether to learn some quick and dirty access skills or not. Probably much easier than doing it in 10K+ rows/columns of excel.

anyone know of any database or spreadsheets that have pitch fx data, along with mlbam_id's? Just by name would be fine, with ids would perfect.

I know texas leaguers has lots of useful information, but you either have to click on the pitcher while editing or type it into excel and apply formulas.

[MRoach27]

Re: 2011 Projection-Based Rosters: Methods, Formulas, & Discussion ...

I must say...

I'm new to the Show... moving from 360 to PS3, but... I am a math nerd and have to admit that this thread is the most interesting thread I've ever had the joy to read.

I'll open excel and just play with formulas and see how they translate and I'm blown away at how amazing and flawlessly they fall into place.

TT and Knight, thank you for numbers...

Now I just need to explain my obsession to my wife... Any advice?

[K_GUN]

Re: 2011 Projection-Based Rosters: Methods, Formulas, & Discussion ...

Now I just need to explain my obsession to my wife... Any advice?

yes......it keeps you out of the bars....strip joints....clubs....trouble.....and keeps you in the *man cave* downstairs

:o)...it's what I tell my wife.....she agrees!