Tweaking Shandler

Theories, Concepts and Analytical Discussion (draft strategies, valuation, inflation, scarcity, etc.)
Post Reply
Message
Author
da_big_kid_94
Hall of Famer
Posts: 1574
Joined: January 3rd, 2009, 12:09 am

Tweaking Shandler

#1 Post by da_big_kid_94 »

Sorta like roto's version of Finding Forrester. I've take Shandler's pitching primer (Control ratio, K/9 Ratio and HR/9 ratio) and added a few things to it over the years (such as IP, Ratio and ERA). I'm going to add at least two more categories this year (probably three) so I'm soliciting opinions.
  • In 2012, I set the ERA range at 3.75 and below. Due to the 2012 season, I'm thinking to reset to 3.50 (don't know if this was a pitching year anomaly or what) - seems a LOT of pitchers had an ERA below that mark that year ...but they still weren't very good.
  • I'm also looking at changing the top end of the Ratio range from 1.30 to 1.275 - for same reasons as prior statement.
  • I am adding percentage of first pitch strikes - there were 119 pitchers who threw 124 or more innings last year - the median was 61% 1st pitch strikes in that group ... so I'm thinking I'm going to set the bar at 63.5% and above to start?
As an aside - there was only one pitcher who threw 60 or more innings last year who's 1st pitch strike % was UNDER 50%. Care to guess who without peeking? Give you a hint ...he's been one of my favorite whipping boys over the years.

All input/comments welcome.
These are my views based on my own opinions and observations - your mileage may vary.
"KNOW THY LEAGUE" - the Forum Funklord - 4/13/2009
Fantasy is managing stats ... roto is managing teams

User avatar
Todd Zola
Hall of Famer
Posts: 8260
Joined: December 25th, 2008, 12:45 pm

Re: Tweaking Shandler

#2 Post by Todd Zola »

What's the purpose of this exercise? Do you use it to generate a draft list, filtering out candidates that don't meet the cut-offs?

What is the league format (AL/NL/Mix) and # teams?

My initial reaction is if this is to categorically eliminate players from your draft list, then using ERA and WHIP is counter-productive to relying on skills. The crux of Shandler's methods are to identify players whose results do not belie their skills, thus if they repeat their skills the next season, better results should follow. Results such as ERA and WHIP that could be better or worse than the skills suggest don't make the best filters.

With first pitch strike percentage, I assume the premise is first pitch strikes lead to more outs.

If the idea is some pitchers results are not representative of how many first pitch strikes they throw, and if they throw the same percentage next season their results will improve, this is a very smart filter.
Catchers are like prostate exams -- comes a time where you can't put if off any longer, so you may as well get it over with and take it up the butt - The Forum Funklord

I'd rather be wrong for the right reasons than right for the wrong reasons - The Forum Funklord

Always remember, never forget, never say always or never. - The Forum Funklord

You know you have to seek therapy when you see one of your pitchers had a bad night and it takes you 15 minutes to find the team you have him on. - The Forum Funklord

da_big_kid_94
Hall of Famer
Posts: 1574
Joined: January 3rd, 2009, 12:09 am

Re: Tweaking Shandler

#3 Post by da_big_kid_94 »

Fair question...here is what i do. They are 11 team AL and NL onlys - $260 4x4 ultra style.

In prior years, a 6 point score was perfection and someone worth targeting. I have kept these lists since the 2005 season so I can , at a glance, notice trends of progression or regression.
  • First three are Shandler's categories (Control, HR/9, K/9)
  • Then I added 125 IP or more (helps split relievers from starters), Ratio under 1.30 and ERA under 3.75
  • I am going to add three more categories - BAA, % of 1st pitch strikes and team run differential (this one will require a lot of modification because of potential roster changes and some Kentucky windage).
The goal is to generate a list of, for want of a better term, pitchers of interest - pitchers who are really performing well but you may not notice they really are. It's an attempt to both include the pitcher's skills and account for the team he plays on. The first pitch category is not necessarily correlated to more outs. I would expect to see an indicative BAA for a pitcher wit a good 1st pitch strike %. I am a strong believer that the pitcher's most important pitch is strike 1. But I know these numbers aren't generated in a vacuum. One quick example - Bronson Arroyo. 2012? 68% first pitch strikes. Impressive - but now comes his BAA - .2673. THAT'S not good - especially in that park. So that tells me to look a little deeper. In 2006, he was a 5 point pitcher - he hasn't been better than three since. So here's a pitcher in a hitter's park who throws a lot of innings and is someone I'll leave for someone else.

On the other hand, Matt Harrison. 62% first pitch strikes - but a BAA of .2708! YET, his ERA is 3.29 and his Ratio is 1.26. To make things more interesting, he pitches in a VERY friendly hitter's park and has a lousy K ratio. That indicates to me that he strands a lot of base runners (especially with a mediocre K rate)- now I need to try and find out why - defense , hit luck, etc. He bears further investigation as well - he was a 6 last year and a 5 this year. he may be indication of one who's "talent" belies his numbers.

I don't need this to tell me about Sabathia or Lee or Verlander. I'm looking for guys like Vogelsong who i just might dismiss out of hand without really looking. it's not perfect - based on this , I was really high on Eric Bedard in 2012 - and you can see how that worked out. So this really isn't the definitive bid 'em or pass 'em list - it's just a tool that I'm looking to tweak. I use it to try and notice performances I may just as soon overlook without it - or, in the converse, prevent me from assuming things are present that really aren't.

Also, one other thing ...this process in not meant to be dismissive of relief pitchers - in fact, it is just as, if not more so, important when evaluating pitchers such as Jason Grilli who may have finally been used in the way best suited for him by the Pirates. The 125 IP helps separate the two disciplines.
These are my views based on my own opinions and observations - your mileage may vary.
"KNOW THY LEAGUE" - the Forum Funklord - 4/13/2009
Fantasy is managing stats ... roto is managing teams

User avatar
Todd Zola
Hall of Famer
Posts: 8260
Joined: December 25th, 2008, 12:45 pm

Re: Tweaking Shandler

#4 Post by Todd Zola »

Looks to me like your point system isn't a be-all-end-all filter, but more a means to flag guys to research more -- in either direction (avoid or target).

To that end I love the idea of adding first pitch strikes as I do think a greater percentage portends to more success, but I still would personally omit ERA and WHIP - but, as a wise man tells us, your mileage may vary.

There are some that might add GB% to the filter, which is fair but could flag some highly successful FB pitchers.

As we learn more about how a pitcher controls BABIP and HR/FB, those can be used as well.
Catchers are like prostate exams -- comes a time where you can't put if off any longer, so you may as well get it over with and take it up the butt - The Forum Funklord

I'd rather be wrong for the right reasons than right for the wrong reasons - The Forum Funklord

Always remember, never forget, never say always or never. - The Forum Funklord

You know you have to seek therapy when you see one of your pitchers had a bad night and it takes you 15 minutes to find the team you have him on. - The Forum Funklord

da_big_kid_94
Hall of Famer
Posts: 1574
Joined: January 3rd, 2009, 12:09 am

Re: Tweaking Shandler

#5 Post by da_big_kid_94 »

Exactly .... "Go get Johnny Cueto" ...now THERE'S a salient piece of advice. Don't need a system to figure out guys like him or Hamels or Felix. As I said, I'm looking for a tip off as to who I should be looking at ...and maybe I've overlooked. And where as I hear you (and have heard you) about ERA and WHIP, I don't want to remove it exactly because of guys like Harrison.

His K rate stinks, guys are hitting .270 off him and he pitches in Texas. So why are his ERA and WHIP so good? Now, to me, if you don't see those ERA and Ratio numbers right next to those others, you might think "BUM" and write him off ...and he just may turn out to be one. But you put those numbers back up against the others, you say "waitaminnit! Somethings not quite right here." I see the first set of facts - and they seem to be in direct contrast to the second set of facts - yet, by our standards, he's fairly successful. That's the purpose of my little exercise _ you look at 4x4 or 5x5 categories only and you say Harrison's worth quite a few bucks (16 to be exact) - but I want to look a little deeper (if possible) to figure out why the seeming disparity - and if it's something i should factor in to my evaluations.
These are my views based on my own opinions and observations - your mileage may vary.
"KNOW THY LEAGUE" - the Forum Funklord - 4/13/2009
Fantasy is managing stats ... roto is managing teams

User avatar
Todd Zola
Hall of Famer
Posts: 8260
Joined: December 25th, 2008, 12:45 pm

Re: Tweaking Shandler

#6 Post by Todd Zola »

Harrison is a nice example of the reason we need to discover the next level of analysis.

The standard sabermetric analysis is he has been lucky the past two seasons, outpitching his peripherals and ending up with an ERA "lower" than what it "should" be.

The what it should be part is the question as it is based on current means of determining expected ERA, assuming some aspects of the analysis are out of the pitcher's control and should be regressed to a global norm. While I believe this to be true, I also believe as we get better data, we will learn that each pitcher has his own norm, which may approach the global norm but not quite get there, meaning there is some skill involved where we now consider that luck.

Harrison reminds me a bit of Jon Garland as they profile similarly and Garland had a stretch where he outpitched his peripherals as ell.
Catchers are like prostate exams -- comes a time where you can't put if off any longer, so you may as well get it over with and take it up the butt - The Forum Funklord

I'd rather be wrong for the right reasons than right for the wrong reasons - The Forum Funklord

Always remember, never forget, never say always or never. - The Forum Funklord

You know you have to seek therapy when you see one of your pitchers had a bad night and it takes you 15 minutes to find the team you have him on. - The Forum Funklord

da_big_kid_94
Hall of Famer
Posts: 1574
Joined: January 3rd, 2009, 12:09 am

Re: Tweaking Shandler

#7 Post by da_big_kid_94 »

Garland is a good example of this sort of paradox. To me, offense is a binary state - every time up you were either successful or you failed. There may be one or two collateral factors that might influence a given at bat (game situation, runners on base, etc.) but for the most part, you either succeeded or you failed in each of your ABs. I mean, who ever heard of a tough luck at bat? Pitchers on the other hand - like you say - the next level. Defenses, quality of opponents, DH (influenced by roster makeup), bullpens, etc.

I'm not going to speculate I'm going to discover that next level ... I'm not unless someone can come up with a way to seemingly quantify the unquantifiable - hell, I'd take it if someone could just come up with some sort of luck percentile that could be factored in (hey, they came up with hit luck, didn't they?) - and Rickey never considered pitching of today when he said it ...so don't! :lol:

The problem I have with the sabermetric analysis in this case is that it just so happens to span most of his whole career, in essence. Trying to be logical and factual at the same time (yeah, I know ...but go with me for a moment), I can not support the argument of Harrison being lucky for two years in a row while excluding the thought that there are no 2nd half players - good or bad. There is just enough anecdotal evidence on either side of these arguments as not being able to disprove either - yet one is accepted as a possibility if not as a downright conclusion, while the other is dismissed as "Can't be. There's no such thing."

I believe what you say has more than just a kernel of truth - the next level of analysis. It may of may not speak to or clarify some of these thoughts. For now, I'll just go along with what I've got. 63% sounds like a good number.

And FYI - the pitcher I mentioned in the first post? Jonathan Sanchez.
These are my views based on my own opinions and observations - your mileage may vary.
"KNOW THY LEAGUE" - the Forum Funklord - 4/13/2009
Fantasy is managing stats ... roto is managing teams

da_big_kid_94
Hall of Famer
Posts: 1574
Joined: January 3rd, 2009, 12:09 am

Re: Tweaking Shandler

#8 Post by da_big_kid_94 »

I have been thinking about this topic for some time now ...the next level of analysis. The more I thought about it, the more the words of a certain Texas bred psychologist kept sounding in my head; "The best indicator of future behavior is past behavior".

Now I'm not saying I've discovered the cure for the common cold ... and guys, this idea only applies to auction leagues at the moment (possibly some one can extrapolate it to that genre if it passes some muster). What is one of the Holy Grail tools in this game of ours? The three year weighted average, no? We look at BA and HRs and RBIs and runs and hits and earned runs and all of that stuff over a three year period, do a little voodoo and come up with a number for the next season.

But are we leaving something out? The CRVC (and other like tools) seems to give us a piece of info we constantly ignore. It breaks down the player's value along the 4,5,6 or how many categories one may have in their league. However, the most important thing at an auction is what the guy is actually worth, right? So why have we not figured out that we are overlooking the track record of a player's value produced? We look at this and say 7 for BA 4 for runs and HR, 6 for RBI and 2 for SB equals a produced value of 23 dollars - but it seems we never look at the converse - what would a three year weighted average of the dollar value earned for each player show us? BJ Upton - big ticket item moving to the other league (especially in only leagues). You want to bump him up because you feel he's moved to a better situation? Fine - no problem - but how much?

You look at this stats on auction day and say he should produce this, this and this which is worth X. But what if a three year weighted average of his actual dollar value produced is Y ...and Y just so happens to be significantly lower than X. I hear that Texas psychologist again. Would that affect your bump number on BJ Upton? Could all of the intangibles in his new situation account for the number you had in mind? Another factor that was alluded to in another thread about projections - the oversight of not including a recently acquired player and the affect that acquisition may have on another player's projections. It may affect his future projection - but that's going forward. The three year weighted dollar value will indicate to you what he did with all sorts of X factors thrown at him in his recent past (team record, injuries, lineup changes, personnel trades, etc.). A record of what that player did in a constantly evolving environment.

It's not nearly fully developed .... just the germ of an idea being thrown out for feedback or one's perusal.
These are my views based on my own opinions and observations - your mileage may vary.
"KNOW THY LEAGUE" - the Forum Funklord - 4/13/2009
Fantasy is managing stats ... roto is managing teams

User avatar
Todd Zola
Hall of Famer
Posts: 8260
Joined: December 25th, 2008, 12:45 pm

Re: Tweaking Shandler

#9 Post by Todd Zola »

Some of this is incorporated into the projection.

I boil everything down to a neutral projection.
Park, team, etc.

Then I adjust it for the coming season. The new team, a change in the batting order and a better or worse surrounding cast (greater of diminished run scoring potential) can be accounted for numerically, and is.

If a manager has a later curfew and the player therefore likes him more and plays better, I can't account for that.

If there is a guy on the team that the player really likes and they talk shop and he improves, I can't account for that.

Shandler has a method that measures reliability, durability, predictability, etc. I like the game plan, but don't agree 100% with the execution, but that's more a philosophical thing.
Catchers are like prostate exams -- comes a time where you can't put if off any longer, so you may as well get it over with and take it up the butt - The Forum Funklord

I'd rather be wrong for the right reasons than right for the wrong reasons - The Forum Funklord

Always remember, never forget, never say always or never. - The Forum Funklord

You know you have to seek therapy when you see one of your pitchers had a bad night and it takes you 15 minutes to find the team you have him on. - The Forum Funklord

Post Reply