Brian Snitker and the History of Partial Season Managers (by AtlCrackersFan)

When Fredi Gonzalez got fired, the Atlanta Braves record stood at 9-28, a lowly .243 winning percentage that would have ranked the team with some of the worst in franchise history. The Braves history of midseason replacement managers provides some interesting guidance about expectations.

Before 2016, and including the 19th-century Boston teams, the franchise had made 18 midseason managerial changes.

History
Only eleven of those changes occurred when the team had a losing record. (They have fired seven managers with winning records!) In five of those eleven cases, the new manager produced a worse record, and in six cases, he produced a better record. On average, new managers achieved a winning percentage about .010 higher than their predecessors.

But that’s mostly because of a massive outlier in 1966. Bobby Bragan was fired after 111 games, his team having limped to a 52-59 record, a winning percentage of .468. Billy Hitchcock came in and finished the year with a 33-18 record, a .647 winning percentage that was .179 points higher than Bobby Bragan’s .468 mark.

If you exclude the 1966 season, the average change in winning percentage was actually -0.058 — in other words, the replacement manager usually fared worse than the man he replaced.

Over the six instances where the new manager improved on his predecessor’s mark, the new man’s winning percentage was .053 higher. Take out 1966, and the average increase in the other five cases was just .029. For the five instances when the successor manager’s team posted a worse winning mark, the average decrease totaled .041.

So, had Snitker followed precedent, his team might have been expected to have a record between 25-99 (a .201 record) and 37-87 (a .298 record). Instead, his Braves went 59-65, a .476 record, an increase in winning percentage of .233 over Gonzalez’ mark after May 18. This represents the largest change between managers in any instance of change in the dugout and only the 4th instance of a winning percentage increase exceeding .100.

What Snitker Overcame
Despite the monumental increase in winning percentage under Snitker, the Braves still finished in last place, of course, the same spot they were in on May 18. That’s a familiar story in franchise history. Only twice did a Braves team with a losing record at the time of the managerial change improve their position in the standings — 1945, when Warren Spahn and many of the league’s top stars were at war, and the aforementioned 1966.

Of all the obstacles Snitker faced, a few jump out immediately. For starters — no pun intended — the Braves ran out 14 different starting pitchers in his 124 games. For example, Jhoulys Chacin started five games before being traded on May 11, a week before Fredi was axed. Another five pitchers were traded after Snitker became manager, and Alexi Ogando got his outright release.

The Braves had a total of 35 pitchers make at least one appearance during the season, but none spent the entire season on the active roster. Everyone either spent time on the injured list, time in Triple-A, or time with another franchise. Teheran led the starters with 188 innings pitched. Williams Perez had the fifth most innings as a starter, with just 53.2 innings pitched.

On the field, 25 different men stepped between the foul lines at some point during the season. They ranged from the regulars like Freeman and Markakis to part-timers like Jace Peterson, Gordon Beckham, and the quickly forgotten Reid Brignac and Matt Tuiasosopo. After Snitker arrived, five players were traded, beginning with Kelly Johnson on June 8, and ending with Beckham on September 27. While Freeman and Markakis appeared in 158 games each, only three others had more than 100 appearances.

The Braves had 13 players make their Major League debut in 2016, ten of whom were pitchers. Four debuted under Gonzalez (Mallex Smith and three hurlers) while Snitker introduced seven pitchers and two infielders, including Dansby Swanson.

What He Accomplished
With all of the roster turmoil, it’s instructive to compare the first 37 games of the season with the final 37 games of the season.

In the first 37 games, the Braves played 21 games against teams who qualified for the playoffs, going 4-17 against them. In the final 37 games, the Braves played 15 games against teams qualifying for the playoffs, and went 7-8 against them.

Some more numbers to compare:

 Runs/gmBAOBPGIDP/gm
First 373.08.230.323.89
Last 375.35.288.409.59

Scoring improved by over two runs per game as the batting average jumped. Extra base hits almost doubled, going from 66 XBH and only 10 HR during the initial part of the season, to 118 XBH, including 36 HR, at the end. Interestingly, the number of strikeouts per game hovered around 8.25 in both parts of the season. The most significant changes were the addition of Matt Kemp on July 30, and trading away Jeff Francoeur on August 24, when just 34 games remained.

The pitching record was more mixed.

StartersRuns/gmGo < 5 IP≥ 5 runs /start
First 372.9598
Last 373.191313
RelieversRuns/gmSvBSv
First 371.9566
Last 371.54124

While the starters’ record improved significantly, from 5-17 to 14-9, by other metrics, such as runs allowed per game, the number of games where the starter didn’t make it past the 5th inning, and giving up five or more runs in a start, the starters did worse at the end of the year than they had at the beginning. The pen got better, though. Their record improved from 4-11 to 9-5, the number of saves doubled, and runs per game and walks decreased, though the number of games when the pen held the opposition hitless decreased slightly.

Other than the improvements in batting, particularly with power, it’s difficult to identify why Snitker got such dramatically improved results over Gonzalez.

Looking Ahead
With the naming of Snitker as the manager for the 2017 season, history offers both hope and caution. First: Bobby Cox totally destroys any meaningful statistical analysis, since he returned to the dugout 65 games into the 1990 season, then won 15 division championships. But Fred Haney also replaced Charlie Grimm a third of the way into the 1956 season, then managed Milwaukee into the World Series in both 1957 and 1958. And in the 19th century, John Morrill (who managed Boston to first place in 1882 and then stepped down), replaced Jack Burdock halfway through the 1883 season and took the team to first place. Morrill then kept the Boston in first in both the 1884 and 1885 seasons.

However, in most instances, the replacement manager’s success was fleeting. In 1951, Tommy Holmes replaced Billy Southworth, but Holmes lasted just 35 games in the 1952 season before Charlie Grimm took his job. In 1961, Birdie Tebbets replaced Charlie Dressen, but Tebbets was replaced at the end of the 1962 season. And in the Braves’ first season in Atlanta, 1966, Billy Hitchcock replaced Bobby Bragan in August, but then got pink-slipped fired with just three games to go in 1967.

Other Atlanta changes include Eddie Mathews replacing Luman Harris about two-thirds of the way through the 1972 season. Mathews himself got replaced by Clyde King in the middle of 1974. King didn’t complete the 1975 season, getting the axe with 27 games left on the schedule. Finally, Russ Nixon replaced Chuck Tanner 39 games into the 1988 season, only to be replaced by Cox just before the mid-point of the 1990 season. Replacements often get replaced by replacements.

One can be hopeful, but experience shouldn’t lead to great expectations. Even if Brian Snitker is the best Atlanta manager since Bobby Cox, there are plenty of other positions that badly need an upgrade. The Johns have their work cut out for them.

49 thoughts on “Brian Snitker and the History of Partial Season Managers (by AtlCrackersFan)”

  1. Great job, ACF.

    I suppose a conclusion to draw is that it’s very hard for a replacement manager to win with the same players. It’d be a very interesting study to see how much Snitker’s roster differed from Fredi’s, or if he is truly a 23% improvement over the incumbent.

  2. Since we’re on the subject of history:

    1. Mike Trout has blown past Ty Cobb with the highest total WAR before the age of 25. None of the other guys in the top-10 are even close (Mantle, Rodriguez, Griffey, Ott, Hornsby, Foxx, Vaughan, Williams).

    2. Among starting pitchers, Clayton Kershaw currently holds the best ERA+ of all time. Among all pitchers with 1000 or more innings, he’s 2nd only to Mariano Rivera.

    I wonder how high these two get in the pantheon of greats before all is said and done.

  3. I don’t get the opportunity to watch a lot of Clayton Kershaw considering the time zone and, ya know, I’m watching my team, but his dominance is fun to watch. That’s what I mean when I say that the best players should be on the main stage. You’re distracting with your own team during the season, you’ve got time zone issues, TV access issues, etc., but if you parked yourself in front of the television yesterday, you saw one of the greats. And not Aaron Blair.

  4. I’m sure Fredi would have done better if he stayed on – he couldn’t do much worse. However, my problem with Fredi’s teams was that they seemed to tank at the end of the year. If the Braves make some good acquisitions and get in good shape to start the year, I think we are on to something.

  5. So maybe letting Trevor Bauer pitch today was not a particularly great idea? Or maybe even a mind-bogglingly stupid idea?

  6. Fredi seemed to be really bad at righting the ship once things started going south. Late season collapses in ’11, ’14 and ’15, disastrous start to ’16. In each case there were mitigating circumstances, as there always are (because the manager can only do so much), but it happened enough under Fredi that it looked like a trend. The improvement from Fredi to Snitker is an aberration because how could it not be, but I do think long-term the team will win more games with Snit than it would have with Fredi, if only because Snit seems to have a basic level of competence that Fredi lacked.

  7. @4. Williams missed all of his age 24 season, as well as several others, being an ace fighter pilot for the Marines during WW2 and Korea. He put up 10.6 WAR in both his 22 and 23 age seasons.

  8. 1. Trout and Kershaw are, to date, all time greats. Unless something terrible happens they’re obvious HOF track players. They’re both pretty damned near “hit by a bus and in immediately” territory already. That said…

    2. Comparing across eras via WAR is just… I’m not sure it’s even wrong. It approaches fish-on-a-bicycle misuse of modern stats.

    3. Kershaw’s dominance is, in fact, one of those “no power in the ‘verse can stop me” type deals when he’s on. Of course, people we know have openly complained when personas non gratis said you just had to “tip you hat” to that guy sometimes.

  9. Went on a Twitter Frenzy with Braves payroll stuff. I figured I’d post here.

    Braves payroll update:

    Currently have ~56MM committed to 6 players (minus Olivera’s $): Kemp, Freeman,Julio,T-Flow, JJ,& Markakis.

    Projected arbitration numbers are 10.7MM for 7 players: Ender, Collmenter, Krol, Recker, Vizzy, Paco, Withrow

    Pre-arb guys come ~511K/each: Jace,Adonis,Winkler,Shae,Biddle,Gant, Wisler,Mallex,Blair,Mauricio,Folty, etc

    Scenario 1: No trades, there’s about 72MM w/ current guys leaving 40MM (Guessing payroll climbs to 110-115MM).
    Scenario 2: Trade Markakis, there’s 61MM, leaving 50MM to spend.
    Scenario 3: Trade Kemp, ~51MM, leaving 60MM
    Scenario 4: Hybrid (My choice): Trade Markakis, release/trade Collmenter, trade Recker, have 55MM.

    Reminder that this FA market will likely make players like Markakis more valuable: no loss of pick, cheap, no long commitment.

    Thoughts?

  10. @16
    My numbers are likely 6-7MM off as I miscued on the Olivera/Kemp debacle, so add 6-7MM for each scenario to the projected payroll.

  11. @13, Good point

    @14, Actually comparing WAR across eras is, as near as I can tell, one of the primary uses of the statistic–unless I’m mistaken, it’s the only cumulative stat that normalizes the run environment of an era, making it very useful for comparing Ted Williams to Mike Trout. And I don’t know if this applies to WAR, but Bill James developed its predecessor Win Shares explicitly to compare players from different eras.

    So we must be working from different information.

  12. @ 18 I believe you are right Edward. It can be rather effectively used to compare across eras. Despite that attempt it still can fall short in that WAR is also a counting stat. Players of Williams era and before suffer from the seasons being shorter and thus having less opportunity to accumulate WAR. If you adjust for that Williams counting stat numbers go up even more.

  13. It depends on what you mean by “compare.” I’m adapting an old Bill James argument here (I believe it was in the Politics of Glory). WAR can be a very good way to indicate that Jeff Bagwell was a really good first baseman, and so was Eddie Murray, and so was Jimmie Foxx. But it’s misleading when you try to assert that Foxx was definitely better than Bagwell, and Bagwell was definitely better than Murray, purely based on their differences in WAR. You can make those kind of ordinal ranking arguments within eras, but it’s dangerous to do it across eras. Baseball was literally different in the 1930s than it was in the 1980s.

  14. How much WAR does one accrue for badassery in the defense of freedom? Let the eggheads chew on that for a while, I say.

  15. something I’ve wondered — how does WAR look at the defensive stats of, say, 1916 Ty Cobb, vs. 2016 Ender Inciarte? I can’t imagine there’s much defensive data from 1916 to go on apart from fielding %. but doesn’t modern WAR use a whole lot more? and if that’s true, wouldn’t that make the two stats pretty different? (asking b/c I truly don’t know)

  16. I think the biggest lesson I’ve learned from this conversation is not to share any fun, eyebrow raising statistical notes because Sam will start seeing fins on Scwhinns again.

  17. @23, here’s Sean Smith’s answer:

    For most games, I have information on which fielder makes each out, and the batted ball type. Without information on the hits, I have to make an estimate. I look at each batter’s career rates of outs by position. For example, if 30% of a batter’s outs are hit to shortstop, then every time that batter gets a hit the shortstop is charged 0.3 hits. Repeat for every position. I look at batting against righthanded and lefthanded pitching separately, as switch hitters will have very different ball in play distributions depending on which side of the plate they hit from. I sum the fractional hits for every fielder, combine with plays made and errors, and get a totalzone. This is then park adjusted, and converted to runs. This method is used for all seasons before 1989, and for the dark years of 2000 to 2002.

    http://www.baseball-reference.com/about/total_zone.shtml

  18. @23

    Ted Williams use to stand in the field and practice his batting stance. He would get killed today for his defense.

    Of course, he wasn’t very popular then either.

  19. @21,

    My memory is that there were only a few official fighter pilot aces in BOTH WWII and Korea. One was Ted Williams. Another was John Glenn.

  20. @ 23,

    Most of what I have seen on historical defensive WAR for players of the “pre WAR” era is that over long periods of time, it tends to correlate well. for a shortstop, you can get put outs and assists and innings. You can also get strikeouts for his pitchers. You compare each on a per inning basis to everybody else in the Majors.

  21. Did you know that the 1998 Yankees, who won 114 games, did not have a single 30 HR hitter in their lineup? And that’s with 5 of their 9 hitters either lefties or switch hitters, and they played their home games with the short porch in right. Some interesting notes:

    -Only one regular had an OPS+ less than league average.
    -All 9 regulars hit at least 10 home runs.
    -6 regulars stole more than 10 bases, and all but two had a 65%+ success rate.
    -No starting pitcher had an ERA under 3, but no starter had an ERA higher than 4.24.
    -They had only two relievers who had an ERA less than 3, one of which was their LOOGY who only pitched 37.2 innings. The other was Mariano Rivera.
    -They used only 19 pitchers the entire season, 5 of which pitched less than 10 innings.
    -One player finished in the top 10 of MVP voting (Jeter).

    They had a deep, consistent team with few stars. I’d love to build a team similar to them.

  22. I would also like to build a team like the 1998 Yankees. In the short term, I’ll settle for not having one of the very worst players in baseball in our everyday lineup.

  23. Yeah, that’s kinda my point. It’s incredibly frustrating to seem to always have the worst player in baseball on your roster every year. Raise the floor, Coppy!

    That is the encouraging thing about Snitker. He seems to get what he’s supposed to get.

  24. Marksberry on twitter

    I don’t want to sound selfish but I really could use some prayers for my health right now. Non baseball related. Thank you guys

  25. Prayers sent csg.

    I long for the day when the Braves aren’t scared to start a 20 year old in the playoffs. I don’t like the Dodgers but I do like the way they are run.

  26. Just to be clear, csg was quoting Marksberry:

    Looks like the Cubs are going to make this a series. Good for them. Beat the Dodgers like rented mules, boys.

  27. @44 if the Cubs defeat the Dodgers, then both the AL and NL representatives will feature the franchises with the longest World Series title droughts – Indians haven’t won since 1948, Cubs (famously) not since 1908. By contrast, the Dodgers have won the world series five times since the Indians’ last championship (1959, 1963, 1965, 1981, 1988).

    I suppose, all in all, I’m rooting for the Cubs to win. Frankly, I’m just happy that the Nats and Giants lost.

  28. Marksberry went from bad to worse. Reports out that his lung collapsed due to severe dehydration during a non-baseball related procedure, and now he’s on life support. So, I guess if you’re a pray-er, you might do that again. Didn’t seem to stick last time.

  29. Awful news about Marksberry. I’ve been monitoring this beyond just his Twitter account. MLBTR has a thing on it. What kind of procedure could manifest severe dehydration? I can think of a few things that kinda sorta make sense but I’m no doctor…

Leave a Reply

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