Looking at Devon Travis Through His Peers

(Title photo credit to Keith Allison: https://www.flickr.com/photos/keithallison/)

When the Toronto Blue Jays traded Anthony Gose to Detroit for Devon Travis over the winter, opinions on Travis’s future were as mixed as I’ve ever seen. Some evaluators believe he has advanced bat-to-ball skills and can be a capable second baseman right now, while others (notably ESPN’s Keith Law, who considers him a non-prospect) do not think he has major league tools. Blue Jays blogger Andrew Stoeten did a good job encapsulating the spread of opinions in a trade reaction piece at the time of the trade. Based on all the information and opinions available, I came to the conclusion that Travis would likely be a tolerable but unexciting starter for the Blue Jays this season, landing in the ~1 WAR range that makes him better than Ryan Goins & Co. but leaving second base as a hole that the team would look to upgrade in July.

Flash-forward to the end of April and his hot start makes me even less sure of what Devon Travis actually is. His performance has been truly sensational — he currently sits twelfth among all position players with 1.2 Wins Above Replacement — and is backed up by an encouraging set of peripherals. Travis’s ability to avoid strikeouts (16.9% K%) has carried over from his minor league record, he’s walking at a decent 7.9% and he’s not popping out an extreme amount of the time (8.0% IFFB% is just below average). Obviously, Travis’s statistics are heavily inflated and he’s not going to hit for a 179 wRC+ all season, but his entire production profile isn’t jeopardized by a fluky BABIP. This isn’t Anthony Gose, the player Travis was traded for, who is hitting .314 based on a ridiculous .500 BABIP.

But we’re still not even a month into the season and while Travis has already banked as much production as I expected him to accumulate all year and we have an idea of what his production profile looks like (contact hitter who can play the position), it’s too early to say that his true-talent production level deviates from what what expected before the season.

What makes it hard to evaluate Devon Travis is that his situation isn’t a typical one. Prospects who are not of the top 100 variety and have no major league experience are typically not given full-time roles out of Spring Training. Devon Travis was named Baseball America’s #84 prospect prior to 2014, but was not listed prior to 2015 and that opinion was backed up by other prospect rankers too.

To find comparable players, I used Lahman and Retrosheet data to find players who debuted in March/April at age 25 or younger, made at least 25 starts in April and May, and were not Baseball America top 100 prospects prior to the season. Since Baseball America’s rankings go back to 1990, the seasons 1990-2014 were considered. After finding the relevant player-seasons, I gathered some standard Fangraphs stats on them.

31 players fit the description. They generally fit the profile I had in mind when thinking about Travis. Guys who were seen as marginal prospects that teams, for one reason or another, were compelled to thrust into everyday roles and hope for the best. Here are the players and the rookie seasons they produced.

Apr-May GS
Robbie Grossman2013282880.080.2430.10197-9.9-0.1
Kirk Nieuwenhuis2012393140.080.3120.12494-3.30.7
Freddy Galvis2012432000.0350.1450.137648.10.6
Tyler Pastornicky2012401880.0530.170.08366-10.3-1.2
Jerry Sands2011292270.110.2250.136108-1.90.7
Brennan Boesch2010295120.0780.1930.15996-10.10.7
Scott Sizemore2010301630.0920.2450.11271-3.6-0.3
Blake DeWitt2008424210.1070.1620.12915.91.6
Brian Bocock200825930.1290.3120.013180.6-0.6
John Bowker2008303500.0540.2110.15383-8.8-0.8
Gregor Blanco2008325190.1430.1910.05889-1.41.2
Josh Barfield2006475780.0520.140.143933.72.3
J.J. Hardy2005334270.1030.1120.137848.91.6
Luis Gonzalez2004283510.0430.1910.17793-12.9-0.9
Jose Castillo2004294140.0560.2220.11268-6.6-1
Chad Tracy2004335320.0850.1130.123898.61.6
Matt Holliday2004394390.0710.1960.198104-6.41.3
Eric Hinske2002446500.1180.2120.2011214.64.8
Donaldo Mendez2001311270.0390.2910.0598-1.5-1.3
Chad Allen1999425230.0710.170.11982-15-0.9
Bobby Smith1998274160.0820.2640.1469910.42.4
A.J. Hinch1998373910.0770.2280.11677.40.5
Deivi Cruz1997464670.030.1180.073464.3-1.3
Sal Fasano1996271600.0880.1560.14583.70
George Arias1996262740.0580.1820.111584.60
Jason Bates1995273680.1140.190.15283-4.5-0.2
Quilvio Veras1995275380.1490.1260.1111104.33.1
Bobby Higginson1995264860.1280.220.16887-3.40.4
Mike Lansing1993415550.0830.1010.0819862.7
Phil Hiatt1993382630.0610.3120.14773-8.4-0.9
Pat Listach1992346490.0850.1910.0591015.93.4

These players experienced a very wide spread of results in their rookie campaigns. Five, led by 2002 AL Rookie of the Year Eric Hinske, were above-average hitters right away. Many, such as Tyler Pastornicky in 2012, didn’t do much of anything and washed out after the first half. Matt Holliday and JJ Hardy went on to churn out very productive careers.

Here are some summary notes on the 31 players:

– The average wRC+, unweighted for plate appearances, was 81. The unweighted average is relevant because the data from each player represents an individual impression; players who exceed the talent level that we’re attempting to capture will play more and may be over-represented by playing time weights. However, players who do not hit at all out of the gate may get demoted before they can realize their true-talent production level. Donaldo Mendez’s 8 wRC+ didn’t reflect his abilities. The average wRC+, weighted for plate appearances, was 87.

–  A slightly elevated strikeout rate was typical. The average K% of players with 300 PA (i.e., guys who kept playing for at least half the season) was 18.4%, which exceeds the Major League average of every season prior to 2009.

– There isn’t a tendency for these players to be particularly oriented to or averse to defense. The total Def score (Fangraphs’ value metric that accounts for position adjustment and defensive performance relative to positional average) was a very neutral -0.7.

– The players produced at a 1.0 WAR/600 clip. The average player who recorded 300 PA produced at a 1.2 WAR/600 rate. Here’s a histogram describing the production distribution of all 31 players:

Distribution of Travis-like hitters

Distribution of Travis-like hitters

The results don’t appear normally distributed but that could change with a larger sample and they appear slightly more normal when the bins are widened.

So, if going into 2015, we knew that the Blue Jays were comfortable with Travis manning second base full-time and that Travis was not considered to have noteworthy tools, it would have been reasonable to expect him to produce at an above-replacement level this season. Despite not coming with much fanfare, players like him are generally okay and most do not look as mismatched as some expected them to. However, few have truly excelled and only Eric Hinske and Quilvio Veras were impact players in their rookie seasons. And neither Hinske nor Veras held on to have particularly bright careers so they may have produced above their talent level in their rookie seasons. Hinske had his moments but never put together a productive full season again and Veras, a fast slap-hitting second baseman, washed out after not hitting much in 2001.

The 1.0 WAR average of comparable players backs up what I expected from Travis pre-season. A tolerable season supported by below-average offense and some positional value. With already 1.1 WAR banked, Travis figures to land among the top six or seven players in the list when all is said and done. Fangraphs Depth Charts, which aggregates Steamer and ZiPS projections, expects him to finish with 2.9 WAR. The Blue Jays were very fortunate to land Devon Travis in the offseason and right now he appears to be the player who will end the team’s long-held second base deficiency.

The Toronto Blue Jays Bullpen is Taking Shape
An Early Review of the Miguel Castro and Roberto Osuna Experiment

Author: Spencer Estey

Spencer has been a baseball fan since a young age and, being from Toronto, he has always been partial to the Blue Jays. He is a statistics major at the University of Waterloo and is intensely interested in the analytic aspect of the game. Spencer follows baseball by watching countless games each season, reading various advanced analysis sites, playing in deep dynasty fantasy leagues and discussing the game with fellow fans.

Share This Post On

Submit a Comment

Your email address will not be published. Required fields are marked *

You may use these HTML tags and attributes: <a href="" title=""> <abbr title=""> <acronym title=""> <b> <blockquote cite=""> <cite> <code> <del datetime=""> <em> <i> <q cite=""> <s> <strike> <strong>