Things I've learned about myself while playing OOTP

[The Game]

Originally Posted by The Wolf

Go read and learn. Come back when you understand about what statistics correlate year to year and how much and which ones are better predictors. Also pay particular attention to outliers.

I have read and have not learned anything new. Why don't you attempt to teach me?Ii am not against using sabermetrics but I have yet to see evidence that shows that looking at sabermetrics proves how a player will do year to year or that relying solely on sabermetrics translates into winning ballclubs. Oakland has drafted well and relied heavily on several known steroid users. Same with Boston. Probably not this year for either. Very few had either team making the playoffs. i am all for looking at OBP when evaluating hitters, i just dont see how FIP is the end all be all for pitchers vs opp avg BB, ERA, K/9
My opinion can be changed. Quest and others changed the way I manually vote for Gold Gloves in my EBL. I relay more heavily on ZR vs fewer errors. Others have changed my opinion based on sabermetrics vs dinosaur stats for potential HOFErs.

[The Wolf]

I'm not your private school. Read the book Baseball Between The Numbers and the FanGraphs website and learn. You don't have to accept sabermetrics: you can join Hawk Harrelson and Joe Morgan and any of several dozen sportswriters who can't understand math and reject it. That doesn't affect me an iota, nor does it stop sabermetrics' steady march into MLB front offices. You can keep up or not. Many physicists went to their graves unwilling to accept quantum mechanics; many geologists went to their graves unwilling to accept continental drift. Galileo's heliocentricism was rejected. In the long run none of that made any difference, and neither will you.

Do and say what you want, but don't expect me to tutor with you or argue with you. If you want to learn, learn. If not, then don't.

[D-BacksJosh]

Wow this thread has been hijacked and derailed.

[The Wolf]

Originally Posted by D-BacksJosh

Wow this thread has been hijacked and derailed.

Yes, and I am likely the most guilty. Sorry.

[David Watts]

Not sure if a thread like this can be hijacked. More or less just an open discussion that's free to go where it may.



Said by someone that's secretly jazzed to actually start a thread that's now 5+ pages long Damn, how shallow has my life become?

[Le Grande Orange]

Originally Posted by The Wolf

Galileo's heliocentricism was rejected. In the long run none of that made any difference, and neither will you.

[nitpick]

Heliocentrism actually dates as far back as Aristarchus in the 3rd century BC, but it was popularized by Copernicus in the 16th century.

[/nitpick]

[The Wolf]

Yes, LGO, I know.

[RchW]

Originally Posted by pjh5165

Looking at Dickey's stats, he has consistently outpitched his FIP through his career. I'm not even going to say that sabr could have predicted his great 2012 season or him falling off a cliff in 2013, although his 3.27 FIP in 2012 showed that his 2.73 ERA was reasonably lucky.

The 2013 Dickey regressed back a little past his career FIP of 4.29. Even given that, the explanation is simpler. He spent 3 seasons in one of the most pitcher friendly NL parks. Then he moved to a notorious AL hitters park that was made even more like a band box because the Jays removed Windows Restaurant after the 2012 season. The resulting open space now allows medium power hitters to hit 2nd deck HR.

Park factors for Rogers Centre in 2013 showed 4th in runs (15th in 2012), 1st in doubles (9th in 2012) and 3rd in HR (15th in 2012). In 2012 Citi field was 23rd in runs, 29th in doubles and 12th in HR. Any organization with an ounce of sense should have questioned whether Dickey was a good pitcher for the Rogers Centre even before the effect of the renovation was known. Add in a shoulder injury and loss of his "fast" knuckleball and he became a sitting duck.

[The Game]

Originally Posted by D-BacksJosh

Wow this thread has been hijacked and derailed.

Sorry I wasn't expecting this to get drawn out, I was going to start a new thread but Wolf is too stubborn to explain things. Wolf I reserved the book at my library I will check it out, but every example 3everyone has posted sabermetrics especially for pitchers are 50/50. What was Verlander's FIP in 2012? He had a bad season this year I am sure people predicted him to be great as normal. For every example anyone can provide for FIP being a way to judge how well a pitcher will perform next year I can also provide examples of how that won't matter. As i said it is a crap shoot trying to predict what will happen season to season.

[The Game]

Originally Posted by RchW

The 2013 Dickey regressed back a little past his career FIP of 4.29. Even given that, the explanation is simpler. He spent 3 seasons in one of the most pitcher friendly NL parks. Then he moved to a notorious AL hitters park that was made even more like a band box because the Jays removed Windows Restaurant after the 2012 season. The resulting open space now allows medium power hitters to hit 2nd deck HR.

Park factors for Rogers Centre in 2013 showed 4th in runs (15th in 2012), 1st in doubles (9th in 2012) and 3rd in HR (15th in 2012). In 2012 Citi field was 23rd in runs, 29th in doubles and 12th in HR. Any organization with an ounce of sense should have questioned whether Dickey was a good pitcher for the Rogers Centre even before the effect of the renovation was known. Add in a shoulder injury and loss of his "fast" knuckleball and he became a sitting duck.

So basically FIP can't predict how a player will do especially on a new team.

[RchW]

Originally Posted by The Game

So basically FIP can't predict how a player will do especially on a new team.

It did. He regressed to his career FIP. His time with the Mets was the anomaly. This is why I'm mad at the Blue Jays Management. Didn't any of them look at Dickey's career and see the obviously high probability that he would regress.

[MrDov]

Originally Posted by The Game

So basically FIP can't predict how a player will do especially on a new team.

No stat will be able to accurately predict the future, but FIP does a much better job than ERA can ever do. ERA relies on things outside the pitchers control, like defense and BABIP. FIP is in place to isolate the pitcher's performance as much as possible. In a simple comparison, ERA is a stat that's dependent on a pitcher, his defense, and luck. FIP is a stat that's dependent only on the pitcher. For this reason FIP is a more accurate gauge of a pitcher's performance over a season, and is a better indicator of future performance than ERA.

[boilermaker]

I find sabermetrics fascinating and really enjoy how it enriches the discussion of baseball, though some of it still is black magic to me.

I don't fully understand WAR and pitching WAR in OOTP really confounds me. For a "catch all" statistic, I'm see inconsistencies between a good WAR year but otherwise statistically lousy. This is an example I've spotted just this very minute in one of my save games:



Looking at ERA+ and the BB/K, I can't make sense of any correlation to WAR as a catch all. I don't know, maybe this is a lousy example as those numbers are all over the place in general. The year with the lowest WAR looks like the best season this pitcher has had....?

As someone uninitiated into sabermetrics, I do say I quite like wOBA. I pay attention to that quite a bit these days.

[The Game]

Originally Posted by MrDov

No stat will be able to accurately predict the future, but FIP does a much better job than ERA can ever do. ERA relies on things outside the pitchers control, like defense and BABIP. FIP is in place to isolate the pitcher's performance as much as possible. In a simple comparison, ERA is a stat that's dependent on a pitcher, his defense, and luck. FIP is a stat that's dependent only on the pitcher. For this reason FIP is a more accurate gauge of a pitcher's performance over a season, and is a better indicator of future performance than ERA.

I don't disagree with your stance on ERA, I look more at K/9 BB/9 opp avg GB %. ERA is great for the current season not for predicting success. I had a player with a 1.87 ERA win the POTY, but he has 4 GGers behind him and the best defensive C to ever play in my league. Put him on the team I just took over and he has an ERA in the low 3 range. i just dont see how FIP predicts success.
As far as Dickey is concerned , he is a knuckleballer, they are a different breed. He could go 20-0 to start the year then 0-15 to end the year. Have an ERA in the low 2's then 20+. I agree that his 2012 season was a fluke, but based on that he should have been good in 2013. Playing in the Skydome is not a factor when looking at FIP.
I am going to run a test league of the career leader in FIP 1 yr deals then manually draft the top 5 SP and top 7 MR/CL in FIP and see how they do over a 5 yr span. Now obviously not RL so results will be different then RL.

What stats do you guys recommend for hitters? WAR? VORP? OPS?

[boilermaker]

Originally Posted by The Game

What stats do you guys recommend for hitters? WAR? VORP? OPS?

Re-stating a small line from my most recent post, I'm finding myself as a big fan of wOBA.

This is the argument for wOBA's importance:

wOBA | FanGraphs Sabermetrics Library

I think it captures quite a lot of information for a single statistic.

[The Wolf]

ERA does not measure a pitcher's performance; it measured the performance of the pitcher plus the bullpen (whether inherited runners score or not) plus the defense plus luck. It does not correlate well to future pitcher performance. These are known facts.

FIP directly measures a pitcher's performance in The Three True Outcomes : K's. BB's, HR's. (Yes, just like OOTP's engine, it is DIPS-based.) FIP (and it's adjusted version xFIP) correlates much better than ERA to future pitcher performance. These are also known facts.

Now you can be a dinosaur and look at a team-inclusive stat that doesn't predict future performance well or you can keep up and look at one that measures actual pitcher performance of the things that they do control that correlates with future performance much better. Your call. You have the right to fall behind the times if you want. Those of us that haven't have the right to tell you that.

[The Wolf]

I find that hitters are best evaluated by wOBA, wRC+ and WAR. Pitcher WAR is a little iffy (there are big threads on this subject on FanGraphs and on Tom Tango's forum), so for me with pitchers it's FIP/xFIP.

Before DIPS I was very attached to using WHIP to evaluate pitchers because even then I knew the ERA was tainted by the unearned run concept. After DIPS I realized that even WHIP was just another garbage stat. It was a hard moment. My education in sabermetrics killed a lot of personal sacred cows. I understand the game statistically pretty much to the current boundaries of sabermetrics now and I see the game much more clearly for the dynamic contest and series of tradeoffs, balances, difficult decisions and hard choices it is now. Wes Westrum had it more right than he knew when he said "Baseball is like church. Many attend but few understand."

[The Game]

Originally Posted by The Wolf

ERA does not measure a pitcher's performance; it measured the performance of the pitcher plus the bullpen (whether inherited runners score or not) plus the defense plus luck. It does not correlate well to pitcher performance. These are known facts.

FIP directly measures a pitcher's performance in The Three True Outcomes : K's. BB's, HR's. (Yes, just like OOTP's engine, it is DIPS-based.) FIP (and it's adjusted version xFIP) correlates much better than ERA to future pitcher performance. These are also known facts.

Now you can be a dinosaur and look at a team-inclusive stat that doesn't predict future performance well or you can keep up and look at one that measures actual pitcher performance of the things that they do control that correlates with future performance much better. Your call. You have the right to fall behind the times if you want. Those of us that haven't have the right to tell you that.

I don;t see how FIP predicts anything. Verlander, CC, Cliff Lee, Roy Holiiday, Smardjiza, along with a few others you could not predict their success or failures no matter what stats you look at. Got any links for some fangraphs guy that has a better then 75% accuracy rating?
BTW I am all for looking at the total picture and deciding on how well a player is or might do. Looking at FIP/wOBA alone will not decide that. Too many factors come into play overall.

[RchW]

Originally Posted by The Game

I don;t see how FIP predicts anything. Verlander, CC, Cliff Lee, Roy Holiiday, Smardjiza, along with a few others you could not predict their success or failures no matter what stats you look at. Got any links for some fangraphs guy that has a better then 75% accuracy rating?
BTW I am all for looking at the total picture and deciding on how well a player is or might do. Looking at FIP/wOBA alone will not decide that. Too many factors come into play overall.

Parsing words may be the problem. FIP is more predictive of a pitchers future performance. There were no shortage of analysts who thought that Dickey was a high risk and way overpriced acquisition for the Jays. Unfortunately none of them worked for the Jays.

It still doesn't work in a vacuum. If I'm evaluating Roy Halladay then one has to look at age and injury way more than FIP. How did Lincecum get his most recent contract? because the Giants think that his FIP is more predictive than his ERA. They could be wrong but that doesn't invalidate FIP.


http://www.fangraphs.com/blogs/tim-l...-death-of-era/

[The Wolf]

What Starting Pitcher Metrics Correlate Year-to-Year? - Beyond the Box Score

ERA has a year to year correlation of .38

FIP has a year to year correlation of .46

You guys keep on using ERA to make year to year predictions. Those of us who use FIP will keep on owning you.

FYI: the SF Giants just showed how much they believe in FIP over ERA - $35 million dollars worth.