RotoWire Partners

Bogfella's Notebook: What the Formulas Won't Tell You

Brad Johnson

For more than 25 years, pitching guru Brad "Bogfella" Johnson has provided insightful evaluation and analysis of pitchers to a wide variety of fantasy baseball websites, webcasts and radio broadcasts. He joined RotoWire in 2011 with his popular Bogfella's Notebook.

Last week I promised to apply a famous quote by Albert Einstein - "Insanity: doing the same thing over and over again and expecting different results." - and apply it to some thoughts on scouting pitchers. The quote itself sounds pretty simple. When you think about it, there isn't much reason to expect different results when repeat something over and over again. So how does that apply to our favorite sport in general, and to pitchers in particular? This week's Notebook is going to look at the plethora (that sounds like an Einsteinian word) statistical formulas that are becoming increasingly popular in the fantasy baseball world, all intended to help you pick the best possible team. Go beyond ERA and WHIP, beyond strikeout to walk ratios, and focus in on xFIP, BABIP, and some other statistical results, and the world is your oyster, right? Without question, they will tell you a lot, but there are a few things about that pitcher which are extremely difficult, if not nearly impossible to quantify. I'd like to explore some of the variables you need to keep in mind when analyzing a pitcher - not to replace statistics, but to validate them!

What The Statistical Formulas Won't Tell You:


Let's start by taking a look at the past two seasons of a very recognizable major league pitcher, James Shields of the Tampa Bay Rays:


In 2010, he was, in a word, awful. He did somehow collect 13 wins and notched 187 strikeouts, but his WHIP of 1.46 and his ERA of 5.18, especially over so many innings (203), certainly cost most owners a place or two in the standings for those categories. Perhaps more importantly, that performance resulted in him being scratched off many 2011 draft lists. The advanced formulas indicated he had experienced some bad luck, and some analysts even mentioned the possibility of a modest bounce back season, but for the most part, Shields was someone to avoid. After all, the formulas are based on a pitcher doing the same thing over and over, so expecting significantly different results, at least beyond the impact of bad luck would be, well, insane.
So, in 2011, Shields turned in one of the best pitching lines of the season from opening day on. What happened? Even the most optimistic statistical gurus didn't see that coming. And, their approach wasn't flawed, and they didn't underestimate the luck factor. What happened could be called a variable. More simply, Shields didn't do the same thing he had done in 2010 over and over, so the results were different.

In Shields' case, the "something different" was a change in his mechanics. Working with coaches, he simplified his motion, making it both more efficient and more repeatable. His fastball had more life, and his breaking pitches had more bite. His change-up was more difficult to detect, and while it didn't look all that different, the results were undeniable. The important thing here is to keep in mind that no statistical formula can accurately predict an outcome if you artificially change one or more variables. In baseball terminology, Shields commanded the zone far better - instead of missing out over the plate with a flattened out offering, his pitches were nearly always on the black, and moving sharply to locations that made it difficult for the batter to make solid contact. A line drive hit like a bullet but held to a single because an outfielder made an excellent play to cut it off is not bad luck. Bad luck is a lame flare that just clears an infielder's glove after being golfed off the end of the bat when the hitter was completely fooled by the pitch. Statistically, they are the same, but over the course of a full season, if a pitcher throws the latter pitch over and over, his "bad luck" will be diminished.
Variables You Should Always Keep In Mind:

As I mentioned, for Shields it was a change in mechanics. However, while that is one of the most influential, there are several other variables you should always keep in mind when evaluating a pitcher on draft day, or when considering that new guy on the waiver wire who has strung together a handful of exceptional starts. Let's take a look:
- Change in mechanics:This happens constantly, even more so with younger pitchers as they get their initial mentoring in pro ball. One important thing to keep in mind, a change in mechanics can be intentional or unintentional and when it happens accidentally, the result is often negative. An example would be Michael Pineda in 2012. His mechanics are all out of whack right now, and his spring performance has be disastrous. It appears likely they will send him to the minors to try and clean them up, but right now he is not a good fantasy option. On the flipside, we have already covered James Shields and there are countless other examples. Some may remember how ineffective Roy Halladay was when he first appeared in the major leagues. He completely reworked his delivery and the results have been pretty positive so far.

- Change in venue:This is generally the easiest change to evaluate, although it may not be very easy to statistically predict. Mark Buehrle has been one of the most predicable pitchers in the game for years because he was able to do the same thing over and over. Now in Miami, he will probably continue that trend, but the results could be better because he will be doing it in a less home run friendly park. His WHIP and strikeout rate probably won't change a great deal, but his ERA could drop a bit if more balls stay in the yard. The most obvious venue factors would be a move from a hitter's park like Coors Field to San Diego, etc.
- Change in defensive support: This can be quite a challenge to evaluate because errors have very little to do with a pitcher's overall numbers. In fact, some of the most helpful fielders commit more errors because they have the range to get to balls other fielders would simply wave at as they bounded into the outfield for a hit. A few years ago, I predicted an improvement in the numbers produced by the Rays staff because the team had acquired Jason Bartlett, then in his fielding range prime, who could effectively limit opposing hitters.
- Change in repertoire:Pitchers learn new pitches and it can be a huge influence with regard to their overall effectiveness and/or role. A good reliever can get by with two above average pitches. As a starter, if he doesn't have at least three, and preferably four, that he can throw at any time, he will probably not last. It is not always easy to predict when a pitcher will be comfortable enough to use a new pitch in a game situation, and it can even make a difference when a pitcher changes how often he uses various pitches he already has in his arsenal.
- Change in physical ability: When pitchers pitch hurt, they often compensate for pain or weakness by unintentionally changing their delivery and mechanics. It can lead to substandard performance, or more concerning, risk of more serious injury. Another, usually less obvious, manifestation is the power pitcher who is simply losing his power arsenal because of injuries or age. That requires a different approach and can be very difficult for some former standouts to accomplish. And, these pitchers may underperform for years while being drafted based on their previous skills - a frustrating scenario for fantasy owners. This might be part of the reason A.J. Burnett has struggled so much recently.
There are other factors to be sure. Just keep in mind that your objective is not always to evaluate past performances as much as it is to predict the future. Statistical formulas can help you get a grasp on the past, and can suggest some statistical anomalies, but they usually only predict the future if the pitcher in question does the same things he has done in the past over and over again. Otherwise, the results can and do change.

Some Notable Rotation Happenings:

With Jason Marquis away for personal reasons and Scott Baker on the DL, the Twins have told Liam Hendriks he will be in the rotation. I think he's one of the best mound sleepers right now and wouldn't be surprised if he sticks. It sounds like early May might be optimistic for Andy Pettitte and that's good news for Phil Hughes. He too should find a permanent spot in the Yankees rotation. John Lannan will initially get the fifth spot in the Nationals rotation and Chien-Ming Wang should be back soon, but don't overlook Ross Detwiler - he's still their best option for that slot. Jeanmar Gomez won the battle with Kevin Slowey for the fifth spot in Cleveland. He was the better choice, but he probably won't help your fantasy team. The spinning role wheel of Daniel Bard has apparently stopped on fifth starter in Boston - at least for now. He has some appeal but he may still struggle with command so he is not a whole-hearted recommendation. While Johan Santana doesn't have the stuff he once had, if he's feeling healthy, and that's the word, you can't bet against him doing quite well in his return.

The Endgame Odyssey:

Boston may also find themselves in a closer quandary. Andrew Bailey (thumb) could start the season on the DL which should create a spike in Mark Melancon's value. But, there have also been rumors that swingman Alfredo Aceves could get into the saves mix too. The Reds are still keeping everyone guessing as to who will step in for the injured Ryan Madson. The early favorite is Sean Marshall, but Aroldis Chapman has to loom large in the picture, especially now that they have indicated he will pitch out of the bullpen. In New York, Frank Francisco has been dealing with reduced velocity and a sore left knee. He is the Mets closer but this isn't good news. The Royals seem to be leaning to Jonathan Broxton even though Greg Holland is probably the best choice. If Broxton gets off to a good start, don't be surprised if he is dealt. Drew Storen will start the season on the DL, albeit supposedly briefly, and the Nationals might be auditioning either Brad Lidge or Henry Rodriguez for a possible deal in their never-ending quest for a centerfielder. Notably, the presumed fill-in for Storen, Tyler Clippard - who they have said they don't want to trade - will remain in his normal set-up role.

For up to the minute updates on all things pitching, be sure to follow @bogfella on Twitter! Get your pitching questions answered, and my take on the latest role changes on the mound!