Need help understanding pitching scoring

[VolsRaysBucs]

I can only speak for myself, but again, those links provide me nothing I did not already know.  The point remains, our system of scoring does not accurately reflect VALUE.  I am not worried about compilation of stats, I am worried about how those stats are "rewarded."  Outs are not properly valued and hits are not properly punished.  The reason one won't be able to find published articles to fit "this side" of the argument is that the argument pertains to league scoring system.  Unless someone from ESPN, ect is familiar with our league, I doubt anything has been published.  Does that make sense to anyone but me.  Again, WE GET THE SCORING from a "real life" standpoint, everyone should understand FIP by now.  What I am saying is that we don't assign "point values" to the metrics used as accurately as we could.  In real life, outs earned are more valued than walk allowed are shunned.  Would a manager prefer Liriano's performance the other night over a pitcher who goes 5 1/3 with a couple K's, a handful of hits and 3 ER?  Of course not.  Yet, in our league, the second line is the premium scoring line.  You could spend your next 1,000 posts offering more examples of the metrics we use and their merits, but the fact remains that those metrics, and more importantly how we assign points to those metrics, is severely flawed. 

To summarize, no one is debating the metrics that go into figuring out FIP.  However, FIP gives you a single, clean number at the end.  We are not assigning points to that number, we are assigning points to the EVENTS that lead to that single, clean number.  Therein lies the problem, IMO.

[Canada8999]

The purpose of those links is to clearly define which metrics a pitcher has control over, and which they do not, since part of the premise of the league is that our scoring will only reflect categories that a player can control.  These categories are K, BB, and GB/FB or HR.  I've posted earlier in this thread that I agree IP could also be a good category.

The issue is that there is not clarity on which categories should be used since many here have scoffed at the thought of hits and runs being primarily made up of the above categories, defense and luck.  Once we've established which categories make sense we can focus in and discuss weighting, which I think at this point we have all agreed will be a worth while discussion.  But we can't then go back and defend the decision for the categories every few months, which seems to happen unless people agree with the principles of FIP, etc.

[Colby]

Klaw
(1:11 PM)
Because a pitcher doesn't control whether a ball he allows into the field of play becomes a hit or an out, WHIP includes a component (the H) that includes noise from defense and luck. There are several other stats that try to get around that issue, like FIP, which estimates a luck- and defense-neutral ERA based on a pitcher's strikeout, walk, HBP, and home run rates.
http://espn.go.com/sportsnation/chat/_/id/32103/mlb-insider-keith-law


Buster_ESPN Buster Olney
Looks like by the end of the night,Trevor Cahill will be 14-5, with the second-best ERA in the AL, No. 2 in WHIP -and No. 2 or No. 1 for CY?
25 Aug
in reply to @Buster_ESPN ↑
keithlaw
@keithlaw keithlaw
@
@Buster_ESPN Fangraphs has Cahill 31st among AL starters in WAR with a 4.07 FIP.
http://twitter.com/#!/keithlaw/status/22132994250

Not sure how we can use FIP in a points system.  I do think a certain percentage of hits should count against pitchers.

[Colby]

Good point Ben... let's aim at the stat categories that we should use...

HR
K
BB (perhaps slightly less weight)
HBP (is this part of BB?)
XBH (subtract out HR points, consider that a high percentage of 2B and 3B are due to bad pitches)
GB (perhaps slightly less weight)
FB
IP (perhaps we need more weight here)

We should post our total points scored for position players and pitchers to make sure our ratios are in line.  Some teams will be off due to an abundance of a certain position.  Ideally, the pitching staff should be 38% of the points.

Pirates' Pos. Players = 3183 -> 62.5%
Pirates' Pitching Staff = 1907 -> 37.5%

I just want to make clear that the ratios we use are solid.  I suppose we could consider the counting of hitting and defense statistics for pitchers, but all of the empty AB will hurt pitchers and their glove is not used that much.

[Dan Wood]

Instead of XBH, which could be considered bad luck or poor defense, we could give LD a go. I know that for the most part LD ratios will all even out. But in a bad start it should be reflected in the amount of line drives a pitcher gives up. LD has been requested several times in the past. But I really think, regardless of what we decide to do, we need to run a beta test on Fantrax to see what the scores would look like. Because even if we update the scoring, we could have the same problem that we are having now. The scoring needs to be put through the motions in order to know what the end numbers will look like. That is in my opinion. I don't think we should deviate too much from what the original idea for pitching scoring. Personally I don't think we should change anything, we all know what the values currently are in this system, and should adapt to it. We have all had a year to observe and react to what is good and what is bad. The types of pitchers that have value and the types of pitchers that don't.

[bravesfan4]

Xbh is a good category. If were saying xbh is luck then xbh should not be an offensive category. For example i have zack grienke.  In watching his game last nite i saw my yankees get 6 points because of back to back doubles.

All this talk about makes me sick. Walks should be eliminated. Each umpire is different and there are blown calls sometimes resulting in a walk. Thats not the pitchers fault. I think all hitters playing in texas should get less points hrs. And all red soxs shud get lesa points for doubles. A homer at SD should be worth 20. Do we like these ideas? Of course not.

Why dont we all stop over analyzing, stop being stubborn, stop looking at what is best for our team. And instead think about the game of baseball. What are important stats. Hits are important, but when a pitcher gives up a hit we get points. Something is wrong there.

[VolsRaysBucs]

The purpose of those links is to clearly define which metrics a pitcher has control over, and which they do not, since part of the premise of the league is that our scoring will only reflect categories that a player can control.  These categories are K, BB, and GB/FB or HR.  I've posted earlier in this thread that I agree IP could also be a good category.

The issue is that there is not clarity on which categories should be used since many here have scoffed at the thought of hits and runs being primarily made up of the above categories, defense and luck.  Once we've established which categories make sense we can focus in and discuss weighting, which I think at this point we have all agreed will be a worth while discussion.  But we can't then go back and defend the decision for the categories every few months, which seems to happen unless people agree with the principles of FIP, etc.

 

[bravesfan4]

So lets start a few seperate threads for categories and discuss them individually.

[rcankosy]

I think we all need to take a step back and start at the beginning.  WHY are we trying to eliminate luck in the first place?  Luck as it was defined by the links posted is as much a fabric of baseball as a home run.  Trying to eliminate luck is akin to changing history.  Can we declare the Yanks the champs in 2001, because the D-Backs had a lucky blooper in the 9th inning due to the Yanks playing the infield in?  Luck is part of the game folks.  The best players don't always succeed, and the best teams don't always win.  Billy Beane once said that the playoffs themselves are one big crap-shoot due to the limited sample size of the data.  Baseball didn't eliminate the post-season and declare a winner based on that argument even though 162 games is a more valid sample size than a max 17-game post-season.  I truly appreciate all the attempts here to make this as realistic as possible, but I don't believe that we are capturing the WHOLE essence of the simulation if we pick and choose which stats to use.  Hits allowed and heaven forbid earned runs may not be within the total control of the pitcher, but they are counted in the record books.  I'm sorry, but I could never support their removal no matter how much empirical data is provided to show that pitching is not an exercise in fairness.  I support the Game Score not because I think that the pitcher had 100% control over each of its underlying stat components, but rather because I believe that it's the most accurate way to capture the events of what truly occurred that day and weighs them in a meaningful way without removing any elements of the game due to luck or defense.  Luck and defense are as much a part of the game as the 7th inning stretch. 

[VolsRaysBucs]

The above was well stated.  The analogy I take from the post is that while a pitcher may finish the season with an ERA around 4, that does not mean each of his starts he yeilded 4 runs and 9 innings of work.  This whole thing about "luck" and "factors beyong the players control" is bastardizing a sound philisophy.  Unless one is playing an individual sport, there are an infinite number of factors a player can not control.  Results are the name of the game.  Teams don't care if a guy walks 10, did we win the game and did those baserunners come in to score.  To take it further, should a GB double play have more value if executed by the MI of, say the Twins (Hardy and Cuddyer) than one from the Yankees (Jeter, Cano).  Jeter and Cano are better fielders, combined, so their pitchers have a built in advantage.  So the walks of the Yankees pitchers may not be as "negative" as walks from a pitcher whose team are not glove wizards.  We can filter this all the way down to bogus proportions if we really want to.  I agree with the idea of beta testing a system.  It appears the majority, if not consesus, says this needs to be seriously looked at.