The impact of a Major League manager has been a bit of a topic this winter, as Joe Maddon bolted to Chicago for five years and 25 million of the Cubs’ dollars. As everyone knows, the Cubbies have been dreadful recently and are still looking for their first World Series title since 1908. Maddon can certainly change the culture over there, but how will it translate onto the field?
SportsCenter recently ran with this ridiculous graphic…
…saying they think he’ll contribute 16 more wins to the Chicago Cubs (“Cubs being Cubs” — funny as it is — isn’t too profound, either).
For players, we have wins above replacement. Jon Lester was worth 6.1 WAR in 2014, and projects to be worth 3-4 in 2015, so you can see where ESPN sees a 5+ win improvement over whichever replacement level pitcher he knocked out of their rotation. But for managers? It’s much tougher to quantify.
Tony LaRussa had a pretty solid quote on this kind of idea back in August:
“Our attitude as a coaching staff was, we were involved 162 games. And we felt that somewhere along the way — whether it was moving an outfielder or moving an infielder, the hitting coach tweaking hitters — we felt our job was putting people in a position to win… But I don’t know how you put a measure on that in number of games.”
So maybe we can’t quantify a manager’s win value just yet. But what about from a strategy standpoint? “…we felt our job was putting people in a position to win…”
There’s a managerial section of the 2015 Bill James Handbook, featuring a number of statistics for each manager, ranging from lineups and substitutions to pitcher usage and other tactics. How can we see which manager employed the most of these strategies in 2014?
Keep in mind, there are a ton of caveats to a manager’s “strategy.” Each team faces entirely different situations. Perhaps a team’s lineup is changing so frequently due to injuries, or the platoon advantage can’t be controlled so well because of roster construction, or whatever. This is all part of what makes quantifying a manager difficult.
But let’s see what we can do. Here are the categories we’ll use, all thanks to the Bill James Handbook: 1) lineups used, 2) percentage of players who had platoon advantage at start of game, 3) pinch-hitters used, 4) pinch-runners used, 5) defensive substitutions, 6) relievers used, 7) stolen base attempts, 8) sacrifice bunt attempts, 9) runners moving with the pitch, 10) pitchouts ordered, and 11) intentional walks ordered.
I decided to calculate z-scores for each category, which measure how many standard deviations a value is above/below the mean. This gives us an idea of how many moves of each type a manager made compared to the guys in the opposing dugouts, as defined by the criteria above.
To make it more accurate, I split up managers by league. Pitchers hitting in the National League makes a big difference, especially when it comes to pinch-hitting and bunting:
Average pinch-hitters used in AL: 111
Average pinch-hitters used in NL: 258
Average sac bunt attempts in AL: 39
Average sac bunt attempts in AL: 76
For example, the overall MLB average for pinch-hitters was 184, which was only topped by the Blue Jays and A’s in the AL. All NL teams were well over the 184 mark. By z-score methodology, that would’ve made the National Leaguers look way more PH-savvy when it’s really a fault of the rulebook. So we must compare each to their peers within their respective leagues.
Anyway, I then summed each category’s z-score for a final “strategy score,” to find which managers utilized the most and least tactics compared to their peers. I thought about weighting categories differently — say, tinkering with the lineup probably has a different impact on the game than a single pitchout — but decided to keep them all equal for now.
Without further ado, here are the results…
American League: Most Strategic Managers in 2014
1. Lloyd McClendon, Mariners (8.48)
2. Mike Scioscia, Angels (6.45)
3. Terry Francona, Indians (5.07)
4. John Gibbons, Blue Jays (3.70)
5. Robin Ventura, White Sox (1.47)
American League: Least Strategic Managers in 2014
11. Joe Maddon, Rays (-1.50)
12. Buck Showalter, Orioles (-4.79)
13. Ron Washington, Rangers (-5.27)
14. John Farrell, Red Sox (-5.92)
15. Bo Porter, Astros (-7.00)
Lloyd! The former Pirate manager finishes in first, having used an above-average amount of different lineups, platoon advantages, pinch-runners, stolen base attempts, and pitchouts, just to name a few. McClendon filled out a lineup card with 69% platoon advantages, fifth-best in the American League and a sound improvement from his days in Pittsburgh. He also called for more pitchouts than any other AL skipper (30), a category he used to dominate on the Pirates’ bench as well (led the NL with 67 in ’02 and 73 in ’03). Stealing first base wasn’t part of his arsenal this year, though.
You’re probably wondering about Maddon. I figured to see him at the top; not near the bottom. But he didn’t stick out in many categories, especially not like he has in the past. Looking at his previous numbers, Maddon typically uses many different lineups (led AL in 2012 and 2013), pinch-hitters (led six years in a row from 2008 to 2013), and stolen base attempts (led four times between ’08 and ’12). It was a down year for the Rays, and maybe a down year for Joe, too. It’ll be interesting to see what he does in a National League setting.
And perhaps we get a glimpse here of why Bo Porter was fired in Houston. He clearly didn’t do a lot behind the bench, and there were rumblings that Jeff Luhnow — one of baseball’s most forward-thinking general managers — engaged in “excessive second-guessing of (Porter’s) in-game management.” Hmm…
Anyway, how about the NL?
National League: Most Strategic Managers in 2014
1. Clint Hurdle, Pirates (9.02)
2. Bud Black, Padres (5.49)
3. Don Mattingly, Dodgers (2.33)
4. Bruce Bochy, Giants (1.45)
5. Walt Weiss, Rockies (0.56)
National League: Least Strategic Managers in 2014
11. Terry Collins, Mets (-1.51)
12. Fredi Gonzalez, Braves (-1.88)
13. Ron Roenicke, Brewers (-2.34)
14. Matt Williams, Nationals (-5.92)
15. Mike Redmond, Marlins (-6.55)
Woah, Clint! Hurdle runs away with this one. He employed the most pinch-hitters and put the most runners in motion (can recall a number of successful hit-and-run’s this year), while also ordering the most pitchouts and intentional walks. That’s the second year in a row Hurdle has used more pinch-hitters than anyone in the National League, further emphasizing the need of a strong bench. The Pirates are well on their way with Travis Snider, Corey Hart, and Sean Rodriguez, and could look even better if they lock up Jung-Ho Kang.
You’ll notice Matt Williams at #14. The first-year Nats manager was criticized for a few issues this year, especially his bullpen usage in the playoffs. He might not have known what he was doing, but won the NL Manager of the Year award anyway — there wasn’t a great storyline this year, the teams that were supposed to be good were good (STL and LAD), and an already talented team gave a rookie manager a league-leading 96 wins, so there’s your nod.
What we’ve done here is simply a look at the amount of in-game moves a manger made to try to presumably help his team win from a strategic standpoint. Keep in mind — this isn’t measuring a manager’s effectiveness. (That’s the next step, and hopefully something we can work on soon.) Not all stolen base attempts work, not all bunts are smart, etc. Hurdle thought it was a good idea to put 43 men on base via the intentional walk, but the Bill James Handbook declared 17 of those “Not Good” decisions (meaning the Pirates were unable to get a double play on the next batter, or get out of the inning without allowing any additional runs).
There’s basically no correlation between a manager’s overall “strategy score” that we calculated above and his team’s winning percentage — well, it’s a weak one but it is positive with an r-squared value of 0.1009 …
There were managers like Williams, Buck Showalter, and Brad Ausmus, who perhaps didn’t need to call for much in-game move-making and still reaped the benefit of a good winning percentage. Actually, if you remove those three from the equation, there is a much better correlation — r-squared of 0.33.
It is notable that there weren’t any high-strategy managers with a low winning percentage (see the area below Bud Black). That’s pretty much the only section of the graph that’s untouched, and what makes it stronger when you remove Williams/Showalter/Ausmus.
On the flip side, some of the lowest strategy managers produced very low winning percentages (Washington, Porter, Farrell, etc.)
This is quite preliminary, but maybe it’s a good first step. It might not be entirely possible to quantify a manager’s “on-field” contributions, but hopefully we can get closer to a better understanding sometime in the near future.
***Once again, all data courtesy of the 2015 Bill James Handbook. Highly recommend it.***