Discussions are currently being had on a majority of sports media in Baltimore regarding who is going to be in the Orioles’ starting rotation. One of those players being jockeyed back and forth is Miguel Gonzalez. As a player who was optioned last year, he appears to be a prime candidate for not making the rotation.
However, some say Miguel Gonzalez is constantly being overlooked and undervalued as a player. Over 435 IP, Gonzalez has posted a 3.45 ERA. In his final 11 starts in 2014, he posted a 2.19 ERA. Why would a player that has performed that well be excluded from the rotation?
With the advent of sabermetrics, there has become a dependency on the Fielding-Independent-Pitching metrics (FIP) as opposed to ERA to measure pitcher value. Analyzing his value with these metrics pens Miguel Gonzalez as a below average pitcher with a 4.59 FIP over his entire career. This discrepancy between ERA and FIP has led to countless articles and discussions (…including myself) on the topic. However, it is important to note that no other player in MLB history has this high of discrepancy with at least 400 IP.
However, Miguel Gonzalez’s greatest strength may be his ability to strand runners. This statistic called LOB% measures the ability for a pitcher to strand runners over the course of the season. The league average for this stat is normally around 70-72%. Miguel Gonzalez posts an incredible 80.3 LOB% as a starter, which makes him the greatest starter in this stat in MLB history with a minimum of 400 IP. There are a few other notable pitchers from this time period including:
The entire list of pitchers can be found on FanGraphs, but the interesting name that appeared to me was Jim Palmer at a 77.9%. Folks should know that I have a soft spot in my heart for Jim and I consider him to be one of the greatest pitchers of all time. However, his numbers have never been strikingly good for a Fielding-Independent-Pitcher as his K/9 rating sat a pedestrian level of 5.04. He also didn’t show a great deal of command allowing a 2.99 BB/9 which rates about average for a starting pitcher. In fact, he compares very similar to Miguel Gonzalez in terms of the LOB% and BABIP (Batting Average for Balls In Play). The largest discrepancy in performance is Palmer’s mastery in avoiding the longball (…especially grand slams):
One of the obvious reasons to explain the LOB% is due to the defense for both Palmer and Gonzalez. With a combination of Brooks, Belanger, and Blair, Palmer obviously had a great defense behind him. However, he still exceeds any other Orioles pitcher during that time in terms of his performance. So we have to search for other possible reasons why Palmer was so much better and use it possibly as a basis to explain Miguel Gonzalez.
One possibility might be weak contact that would allow for balls to be more easily played. Unfortunately, batted ball profiles do not exist for Palmer. However, Gonzalez does have this data and we have a subset of modern day pitchers that have a similar LOB%:
The numbers for Gonzalez compared to those other pitchers standout specifically in regards to strikeouts and walk rates of the other pitchers being much better. In addition, the other pitchers are posting average to above average ground ball rates, which theoretically should lead to a lower BABIP and LOB%. However, the other statistic that is intriguing is IFFB%, which measures the percentage of infield flyballs (which are normally always outs). With the league average being 9.7%, three of these pitchers display a propensity to to accrue an above average weak flyball.
For a long time, there has been assumption placed on groundballs being much better. However, recent work has to be changing that notion to a certain regard indicating that flyball pitchers can be effective. One of the most interesting quotes from this work was:
“ALSO, HIGH-IFFB% PITCHERS TEND TO OUT-PERFORM THEIR FIP AND XFIP, WHILE THOSE WITH LOW IFFB% RATES TEND TO UNDERPERFORM.”
If that’s the case, then perhaps Miguel Gonzalez being a flyball pitcher with a propensity to get IFFB% is an indication of the ability to get of weak pop-ups. Of course, this doesn’t always happen as he has been hit hard based on HR/FB (11.6%). Jim Palmer obviously did a good job at limiting the home run as well being a flyball pitcher so perhaps this indication of weak contact flyballs should be examined further as we watch the game.
Miguel Gonzalez stands as an enigma to many prediction models that rely on fielding-independent-pitching. However, we may need to re-evaluate outliers like Miguel Gonzalez who have consistently outperformed key characteristics such as LOB% and BABIP like other pitcher currently and in the past.