A year of mistakes, in review

Around these parts, I do a whole lot of evaluation of others. I evaluate players, I evaluate general managers, I evaluate trades, I evaluate commenters... rarely, though, do I take a look in the mirror to see what I'm doing wrong. And that's probably not a good thing. You see, self-evaluation is incredibly important. And it's time, I think, for me to do some of my own.

The first week of classes in college, I was told by practically every professor to "fail early and fail often", because failure analysis is the first step on the path to failure prevention. (Engineering joke!) So while this list of failures is in some ways a "mea culpa" and in other ways a "haha I can't believe I actually said that," the actual main point of the exercise is to learn from my mistakes. With that in mind, let's run through all of the articles I posted this year, find the stupidest things I said, and make fun of them!

Ready?

#1: Mariners Lose 10-5 Slugfest To Indians

What I Said:

Danny Farquhar pitched again today. He is still, somehow, inexplicably, in major-league camp. It's not like he's been having a very good spring, and last year he bounced around several different organizations, so I'm not sure what to make of his continued presence. Most likely he's just hanging around in case another reliever gets injured, but if the inevitable Wells trade also ends up sending a reliever the other way he could actually start the year in the bullpen. Danny Farquhar: the new Jeff Gray, maybe!

What Actually Happened:

Danny Farquhar turned out to be the team's best relief pitcher. He deployed a new arsenal of crazy good pitches, crushed it in Tacoma, came up to the big leagues and stole the closer job from Tom Wilhelmsen with a ridiculous K/9. His curve was the most unhittable pitch in the major leagues, and his xFIP was the ninth-best in baseball. He put up 1.9 WAR.

Lesson Learned:

Don't discount scrap heap relievers; they can overhaul their repertoires in an instant. MLB scouts generally know best when it comes to this kind of acquisition. There's not a whole lot you can do with the stats to say that Tom Wilhelmsen, or Steve Delabar, or Danny Farquhar won't be good. (More on Wilhelmsen later, though).

#2: The Mariners' Bullpen and Specialization

What I Said:

While specialist-heavy bullpens may not be inherently better, they don't have to be: as long as they're inherently cheaper and just about equally good, they will be a superior option for teams with limited financial resources. "No correlation" is a result after all. This study endorses the Mariners' approach to bullpen-building.

What Actually Happened:

The Mariners' bullpen was pretty bad. They got a lot of strikeouts, but walked far too many dudes and allowed far too much hard contact. Part of the problem was its "specialist" nature; because the team had a bad rotation, the bullpen was frequently called upon to go multiple innings, which is a thing that specialists are bad at. Tom Wilhelmsen's implosion didn't help.

Lesson Learned:

Consider players in the context of the rest of the team. Sometimes bench players get called upon to fill a bigger role than they're capable of handling, and things end badly - witness the disintegration of Kelly Shoppach and (more relevantly) this particular bullpen's implosion. No part of the team, be it the bullpen or the bench or the rotation, should really be analyzed independent of the rest of the players.

#3: Kyle Seager Tightens Up The Strike Zone

What I Said:

There is also the question of statistical significance. User gogurt rightly suggested that I should add marks for the standard deviations of each group on the chart. I slapped my forehead and went to go calculate them... and when I did, discovered that these ISO and wRC+ values are all within one standard deviation of each other. Welp. Turns out I'm still not good at this job yet! So while we can say that there seems to be a trend, we cannot say that it is a statistically significant one, which casts further doubt on the whole scenario. Bleh. Sorry, guys, I promise I'll get better at this as I go.

What Actually Happened:

This one wasn't about my conclusion being wrong as it was about my methodology being dumb. Because of the size of the "buckets" I divided players into for the purpose of comping, my results ended up being not remotely statistically significant - and yet I used them to support a point. When a couple of commenters correctly pointed out my mistake, there were multiple pies on my face. Whoops.

Lesson Learned:

Check your math before you hit publish.

#4: Whom Should Michael Saunders Replace?

What I said:

[The answer is] Raul Ibanez. Stick a fork in him; he's done. [...] Numerically, the power is gone. Visually, the power is gone. His O-Swing% is .2 percentage points off of his career high, his Z-Swing% is at a career low, and his contact rate is 1.1 percentage points off of his career low. His BABIP is awful and it's not bad luck. His defense has been predictably putrid. Raul Ibanez is quite possibly the worst player in the major leagues right now. Dumping Mike Carp has never looked like a worse decision than it does at this precise moment.

What Actually Happened:

Raul Ibanez hit twenty-nine home runs. He wasn't good - his defense ensured that - but he was better than Carlos Peguero, Endy Chavez, and Jason Bay, all of whom I suggested the Mariners keep over him. His power, at least, was absolutely not as "gone" as I predicted.

Lesson Learned:

Don't overreact to small sample sizes of batted ball data, and check your biases at the door. I was dead set against  Ibanez before the season even began, and when he hit badly in a small sample size I pounced. I tried to back up my bad practice by using more advanced data (distance figures from Baseball Heat Maps), but it turns out that that data is pretty unreliable over small samples. In the future, I'll be more careful with the data and less eager to bandwagon against unpopular acquisitions.

#5: Should The Mariners Call Up Nick Franklin?

What I Said:

Despite sample size concerns, Nick Franklin's approach changes and Brendan Ryan's possible decline are enough to make me believe that Franklin is the best shortstop in the organization right now.

What Actually Happened:

Franklin got off to a hot start playing second base, but his numbers collapsed after one month in the majors. Part of this was due to his inability to make contact with breaking pitches. The contact rate spike he'd experienced in AAA didn't carry over to the major leagues, and he started striking out like a madman. Also, his defense wasn't good enough for shortstop, and Brad Miller shortly thereafter came up and performed much better than Franklin.

Lesson Learned:

Take minor league "plateau leaps" with several grains of salt, don't assume defensive proficiency, and consider possibilities beyond a binary choice. (In this instance, the answer to "Ryan or Franklin?" was "Miller".)

#6: Do The Mariners Need To Extend A First Baseman?

What I Said:

Gun to my head, I'd say that the Mariners are going to extend Michael Morse pretty soon. [...] He'll get something that probably makes you slightly uncomfortable. 3/40, say. But you shouldn't freak out when that happens, because there really isn't any good alternative.

What Actually Happened:

Michael Morse imploded. The Mariners didn't extend him, thank God, because that contract would've been horrible. He's probably going to get 1/4 from someone this offseason. Corey Hart began to look a lot more available and appealing, and Kendrys Morales declined the QO, giving the Mariners some nifty 1B/DH options.

Lesson Learned:

It's not smart to overpay players in advance, especially when they're injury prone, just because the market doesn't fit your needs.

#7: The League Catches On To Tom Wilhelmsen's Curveball, Maybe

What I Said:

If the question is "Is Tom Wilhelmsen worse now?", the answer is "maybe very slightly, if you believe in a small sample size". Or, in other words, "no, not really."

What Actually Happened:

Tom Wilhelmsen was worse. The K's went away, the BB's went up, the runs started coming in, and he lost the closer job - to the very same guy I disparaged in #1. Double whoops.

Lesson Learned:

Don't have too much faith in established relief pitchers. Even awesome ones can very rapidly become not awesome.

So. It was an up-and-down year, for me. I got a writing gig at my favorite baseball blog, which was pretty freakin' sweet, but I didn't always use my pulpit well. Sometimes my methodology was bad, sometimes I was too quick to trust small sample sizes, and sometimes I was just plain wrong. Hopefully, this exercise - this self-evaluation - will help me to be a better sabermetric analyst and writer in the future. Maybe, one day, I'll go a whole year without mistakes.

Maybe.

Onwards and upwards!